The Constitutional Case for Universal School Choice in Minnesota

Executive Summary

• Few legal hurdles stand in the way of school choice programs which are limited to public schools or, for that matter, secular private schools. Minnesota pioneered the former in the mid-1980s, and the Wisconsin Supreme Court sustained the latter in 1992. However, such programs do not go far enough. In order to provide parents and children with sufficient options to meet a great range of educational needs, as well as to improve public schools through competition, school choice programs must be universal. Which is to say they must include all schools: government schools, privately sponsored secular schools, and parochial institutions.
• Opponents argue that inclusion of religious schools violates the principle of “separation of church and state.” And clearly, the religion clauses of the First Amendment of the U.S. Constitution, and those of the Minnesota Constitution, prohibit some forms of state aid to parochial schools. However, a careful analysis of the precedents of both the U.S. and Minnesota Supreme Courts suggests that a universal school choice plan can be crafted for Minnesota that will survive all legal challenges.
• The landmark 1971 U.S. Supreme Court ruling outlining the parameters of acceptable church-state interaction, Lemon v Kurtzman, requires that three standards be satisfied before a law meets the demands of the Constitution: The statute must have a secular legislative purpose; its principal or primary effect must neither advance nor inhibit religion; and the statute must not foster “an excessive government entanglement with religion.” The principle underlying the U.S. Constitution’s religion clauses and the Supreme Court’s guidelines is that the state must not favor one religion over another, or all religions over no religion. In short, the state must be neutral.
• The Supreme Court’s treatment of parochial school aid cases has sometimes been confused and inconsistent. Nonetheless, after sifting through relevant cases, a “Model School Choice Plan” is discernible. Such a plan likely would endure the scrutiny of the Court if it abided by five rules.

Rule One: Aid should not be granted directly to religious schools.

Rule Two: Religious schools must benefit only through the independent choices of parents.

Rule Three: All types of schools — public, private and parochial — must be allowed to participate.

Rule Four: Funding for students attending religious schools must not be greater than for those attending nonreligious ones.

Rule Five: Additional state regulation of religious schools, above current levels, should be minimized.

• After satisfying the requirements of the federal constitution, a universal school choice plan for Minnesota must also meet those of the state constitution. A cursory look at the Minnesota Constitution suggests that it might contain a tougher standard. However, on closer review, Minnesota Supreme Court jurisprudence clearly indicates otherwise.
• Minnesota case law is in virtual lockstep with the federal standard, and the model school choice plan outlined in the following pages does no offense to the Minnesota Constitution for fundamentally the same reasons that it does not offend the First Amendment.

---

Foreward

Real school choice — the kind that includes private and religious schools — is compelling educationally, but is it sound legally?

As Jon S. Lerner argues convincingly in “The Constitutional Case for Universal School Choice in Minnesota,” the answer is clearly yes, as long as the program is designed and implemented correctly. For example, aid should not be granted directly to religious schools. Rather, religious schools must benefit only through the independent choices of parents, who might receive help in the form of vouchers, scholarships, tax credits or the like.

Mr. Lerner, who is a second-year law student and law review member at the University of Chicago, demonstrates how such a carefully crafted plan would be consistent with key rulings of the U.S. Supreme Court, as well as Minnesota’s highest court.

His conclusion, in truth, is not terrifically surprising, as other observers and legal scholars have argued similarly. Mr. Lerner’s paper, however, pulls legal history and arguments together better than anything I’ve seen, while specifically addressing Minnesota case law and precedents.

Better yet, in the just-emerging debate about universal school choice in Minnesota, this essay will strengthen and encourage those who are devoted to providing real educational options for all Minnesota children. Obversely, it will make it measurably harder for opponents to fall back on the well-rehearsed, often deeply felt, but ultimately faulty contention that choice programs which include religious schools are necessarily incompatible with the First Amendment.

Yes, surely, this is sensitive ground, and as religious minorities, Jon and I know just how much. But I’m confident that by expanding educational liberty in ways he prescribes, we would remain perfectly true, not just to the Constitution, but to our remarkable civility as a wildly diverse people.

And not incidentally, by pursuing real choice we also would do more to improve education for more children, especially low-income boys and girls, than by any other policy step I can imagine.

Jon Lerner, also not incidentally, was born and raised in Minneapolis; spent a dozen years in its public schools, and graduated from Southwest High School. He received his undergraduate degree, cum laude, from the Elliott School of International Affairs at George Washington University, and studied for a year at the London School of Economics. As noted, he was elected this summer to the University of Chicago Law Review.

Professionally, he has served as a legislative assistant to former Sen. Rudy Boschwitz, focusing on foreign policy and defense issues. And he has been top advisor to Rep. Olympia Snowe, a Republican from Maine, at the House Foreign Affairs Committee. I thank him for this splendid, states-side paper.

“The Constitutional Case for Universal School Choice in Minnesota” is the first in a new series of American Experiment papers on education to be released in the next few months. Upcoming ones will continue discussing school-choice issues, specifically as framed in Minnesota, as well as the joint promise and threat of outcomes-based education in this state.

American Experiment members receive free copies of all Center publications, including this one. Additional copies are $4 for members and $5 for nonmembers. Bulk discounts are available for schools, civic groups and other organizations. Please note our phone and address on the previous page for membership and other information, including a listing of other Center publications and audio tapes.

As is always true, I welcome your comments.

Mitchell B. Pearlstein, President | October 1993

---

(Note: Following up on his 1993 Center of the American Experiment paper, “The Constitutional Case for Universal School Choice in Minnesota,” I asked Mr. Lerner to provide an update on recent legal changes regarding school choice. Here’s his very helpful memo. MP.)

As is typically the case, political developments on the school choice issue have overshadowed judicial developments. And, as is also frequently true, it is the judicial developments that might ultimately prove to be more important.

Governor Carlson’s endorsement of school choice that includes religious schools is a big plus for the movement. However, the Supreme Court likely provided an even bigger boost in its June 29, 1995 decision in Rosenberger v University of Virginia.

When I wrote “The Constitutional Case for Universal School Choice in Minnesota,” in October 1993, the legal principles favoring universal school choice had been generally accepted by a narrow majority on the Supreme Court. Rosenberger, however, provides the Court’s clearest ever Constitutional basis for inclusion of religious schools.

In Rosenberger, by a 5-4 decision (Justice Kennedy writing for the Court, joined by Chief Justice Rehnquist, and Justices O’Connor, Scalia, and Thomas), the Court rejected an attempt by the University of Virginia to deny funding to a Christian student newspaper. The University funded several non-religious student publications, but refused to authorize public money for use in the Christian newspaper on Establishment Clause grounds.

There are three angles from which the Rosenberger decision should be read to support universal school choice. The first is clear, and lends reinforcement to the existing principles favoring universal choice. The two others offer the possibility of even bolder legal justification for inclusion of religious schools.

The first angle is the solid reaffirmation of the neutrality principle in Establishment Clause cases. As you will recall from the 1993 paper, the neutrality principle is an interpretation of the First Amendment’s clauses on state involvement with religion that dictates that the state may engage in programs and activities that have the effect of benefiting religion, so long as the state does not directly provide funding to the religious institution, and the program is crafted so that benefits are extended to recipients without regard to their religious preference.

Critics of this approach, including the four dissenters in Rosenberger (Justices Souter, Stevens, Ginsburg, and Breyer), contend that the principle of separation of church and state requires that no aid flow from the state to a religious institution. However, Rosenberger clearly demonstrates again that the “neutrality” approach is ascendant over the “no aid” approach.

In Rosenberger, the neutrality principle dictated that University funds which went to third party printers could be used to publish religious student newspapers on an equal basis with non-religious papers. The school choice parallel is clear. If state education funds go to parents rather than schools, the neutrality principle permits parents to use the funds at religious schools on an equal basis as non-religious schools. The Rosenberger opinion could be read, more controversially, to go well beyond past uses of the neutrality principle, however. This case presents two other angles of great interest to school choice advocates.

In the 1993 paper, I speculated that once the Establishment Clause was deemed not to prohibit inclusion of religious schools in a school choice plan, it might indeed be unconstitutional to exclude them. The neutrality principle demands that the state neither advance religion nor inhibit religion.

In a concurring opinion, Justice O’Connor wrote that “withholding access [to religious entities] would leave an impermissible perception that religious activities are disfavored . . . .” She also wrote that, “if a state refused to let religious groups use facilities open to others, then it would demonstrate not neutrality but hostility toward religion.”

Under this reasoning, once a school choice plan was opened to participation by secular private schools, if private schools were denied access to the program simply because of their religious affiliation, this too would indicate an impermissible hostility toward religion.

The second controversial angle that could spring from Rosenberger pertains to the oft-expressed fear of some conservatives that a voucher program that included religious schools could lead to greater state regulation of those schools.

The Rosenberger case was a tangle between the First Amendment’s Establishment Clause and Free Speech Clause. The students claimed that the state support of secular publications and denial of support to religious publications was the kind of content-based viewpoint discrimination that has been routinely struck down under the Free Speech Clause. The University defended on the ground that the Establishment Clause demanded such discrimination because of the separation of church and state principle.

Well, in this clash of Constitutional principles, the Free Speech Clause won. Justice Kennedy’s majority opinion suggested that if the state interceded to ensure that all student publications met some baseline standard of secular orthodoxy, this would “eventually raise the specter of governmental censorship.” He concluded that “this course of action . . . would risk fostering a pervasive bias or hostility toward religion, which could undermine the very neutrality the Establishment Clause requires.”

This raises serious doubts about the constitutionality of excessive government “strings” tied to inclusion of religious schools in a voucher plan. If, for example, the state mandated that any religious school that participated in a voucher program could not teach sectarian values, this would likely implicate Free Speech Clause concerns, and raise Justice Kennedy’s “specter of governmental censorship.”

Thus, at its clearest point, the Supreme Court’s Rosenberger decision must be seen to validate the neutrality principle that underlies the constitutional support for universal school choice. However, it also could plausibly be seen to lay the foundation for much bolder claims that: (1) a school choice plan that includes private secular schools must include religious schools; and (2) inclusion of religious schools in a voucher program does not open the door to state domination of religious schools.

---

(I) Introduction

“Let’s give choice a chance!” That’s how Wisconsin Supreme Court Justice Louis J. Ceci began his concurring opinion in a case upholding the legality of the Milwaukee Parental Choice Program.¹ The battles being waged in state legislatures and through ballot initiatives between those who want to give real school choice a chance, and those who would prevent such plans, are by now familiar. However, there is another theater in the school choice debate that has seen only a few skirmishes to date, but promises to bring major clashes in the future. That battleground, as Justice Ceci suggests, will be the courts.

The opportunity for parents, especially those with modest incomes, to decide where their children are educated will be won principally in voting booths and state houses — and in Minnesota, with the help of pathbreaking econometric research, soon to be released.² But it’s also essential that such inevitable wins be confirmed in the courts, as opponents will move immediately to those venues. As this paper makes clear, a properly designed Minnesota school choice program surely can survive constitutional challenges both federally and at the state level.

Roughly speaking, there are three varieties of school choice plans, varying as to the type of schools that are permitted to participate. One allows choice among government schools only. Minnesota is to be commended for pioneering this expansion of educational opportunity, under the leadership of Gov. Rudy Perpich in the mid-1980s. Here in Minnesota, most parents can send their children to most public elementary or secondary schools in the state, regardless of the family’s proximity to the school. The legality of this type of school choice is unquestioned, as there are no constitutional provisions requiring students to attend schools within arbitrary state-drawn district lines.

However, the usefulness of this option is strongly questioned. Eminent scholars who support expanding choice in education, such as Milton Friedman, John Chubb, Terry Moe, John Coons and Stephen Sugarman, have all forcefully argued for inclusion of private schools.³ And it’s no wonder. Since the source of so many of the failures in education today is the structure and management of government schools themselves, spurred by their insulation from market forces, choice plans which include only government schools necessarily miss the mark. Such plans fail to provide parents with the range of choices required to cause government schools to reform themselves.

Which leads to the second variety of school choice plans: Those that allow parents to use state funds to send their children to private, nonreligious schools. In 1990, Wisconsin became the first state to approve such a plan. It did so on a very small scale, limiting the program to approximately 1,000 students from low-income families in the city of Milwaukee.⁴ The Wisconsin plan was immediately challenged in court, alleged to be in violation of three generally procedural provisions of the Wisconsin Constitution.⁵ In March 1992, the Wisconsin Supreme Court’s decision in Davis v Grover upheld the plan as valid under the Wisconsin Constitution.

The Davis decision is significant in two ways. One is its dismissal of state constitutional hurdles to school choice plans that include private, nonreligious schools. While the 50 state constitutions differ considerably, the provisions under which the Milwaukee plan were challenged are common to many of them, including Minnesota’s.⁶ It’s certainly true that the high courts of two states may and sometimes do differ in their interpretations of identical state constitutional provisions. Nonetheless, state supreme courts frequently borrow legal reasoning from one another, particularly on procedural matters, and there is no reason to suspect that the Minnesota Supreme Court would arrive at a conclusion different from that of its neighbor to the east.

The second significant item springing from the Wisconsin litigation is the absence of a challenge on federal constitutional grounds. It’s not clear exactly how one might successfully challenge the federal constitutionality of the Milwaukee program, and that’s just the point. The entire Wisconsin education establishment brought suit against this plan, and decided that its best shot at invalidating it was to forego any mention of the U.S. Constitution.

Thus, it’s clear that inclusion of private, nonreligious schools would pass muster in the Minnesota courts. But this should not come as a big surprise, nor is it earthshaking in its significance. By far the biggest battle will be over the third variety of choice plans: “universal choice,” which encompasses government schools, privately sponsored secular schools, and parochial institutions.

No choice program can hope to achieve widespread use, significantly serve children, or succeed in bringing competition to bear on government schools unless it includes religiously based private schools. In Minnesota, for example, 85 percent of all private schools are religiously affiliated.⁷

Nearly everyone would agree that parents should be free to educate their children in accordance with their religious beliefs, whatever they might be. Today that freedom is exercised only by the few who either can afford private education, or are lucky enough to receive scholarships or private financial aid. Such an opportunity should be no less available to families of modest income.

Hence, just as in legislatures, the real school choice battle in the courts will be over inclusion of religious schools. But clearly, as quickly as the words “government” and “religion” are mentioned within paragraphs of each other, profound questions about the separation of church and state are provoked. Which is to say, the future of school choice pivots on overcoming thoroughly legitimate concerns about offending this exalted principle of constitutional law.

The religion clauses of both the U.S. and the Minnesota Constitutions are where opponents of school choice will immediately turn to challenge any plan that would allow state funding to go to parents who choose to educate there children in sectarian private schools. The critical provisions of the U.S. Constitution are the “Establishment” and “Free Exercise” Clauses of the First Amendment, which read: “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof. . .”⁸ The Minnesota Constitution has a similar provision, and an additional provision specifically concerning state aid to sectarian schools.⁹

Despite these obstacles, a comprehensive school choice plan, one that includes religious schools, can be drafted in a way that comports with both federal and Minnesota constitutional standards. Barring a significant departure from judicial precedents, such a plan would be upheld by the United States and Minnesota Supreme Courts.

(II) The Lemon test

The landmark U.S. Supreme Court ruling outlining the parameters of acceptable church-state interaction came in its 1971 decision in Lemon v Kurtzman.¹⁰ Though the so-called “Lemon test” has been much criticized by members of the court and by commentators,¹¹ it remains the focal point of Supreme Court Establishment Clause jurisprudence. It’s no exaggeration to say that every parochial school aid case that the Court has heard in the past two decades has turned substantially on Lemon principles. Thus, it’s there that the legal case for school choice must begin.

There are three prongs to the Lemon test, each of which must be satisfied before a statute meets the demands of the Establishment Clause, as interpreted by the Supreme Court: First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion; and third, the statute must not foster “an excessive government entanglement with religion.”¹²

Before dissecting the three prongs in detail, it’s useful to briefly identify the principle that underlies the Constitution’s religion clauses and the Lemon guidelines. That principle, in short, is that government must be neutral, both between varying religions and between religion and non-religion.¹³

Just as it is important to decipher exactly what “neutrality” entails, it’s equally critical to note what it does not. Neutrality certainly does not mean that the state must be hostile to religion. Nor does it mean that state funds can never be used on behalf of religion or religious institutions. Indeed, state funds are frequently put to such use. Even the strictest of separationists would not withhold police and fire department protection from places of worship, or refuse to allow religious organizations to use public roads. Similarly, federally subsidized student loans are routinely used by students attending religiously oriented universities, and state assistance goes to hospitals and nursing homes with religious affiliations. Thus, the metaphoric “wall” separating church and state has always been rather porous.

At root, the Establishment Clause and the Lemon guidelines are intended to prevent government from favoring one religion over another, or all religions over no religion, either through the use of its taxing and spending powers, or via a symbolic stamp of approval. Therefore, properly understood, “neutrality” in the church-state context means that government must have no interest, pro or con, in the religious decisions made by individuals in accordance with their own conscience.

The first means adopted by the Supreme Court for safeguarding the neutrality principle is the requirement that government actions have a secular purpose. The Court has never invalidated a state program to assist parochial schools on this basis, and there is no reason to suspect that a school choice program that included religious schools would even be disputed on grounds of the “purpose” aspect of the Lemon criteria. All that is required to satisfy the purpose doctrine is to show a plausible and controlling secular purpose behind the legislation. Improving education, providing parents and children with greater educational opportunities, and saving taxpayer dollars would all fit the bill.

The remaining two prongs of the Lemon test provide more formidable hurdles, however. It’s not difficult to envision school choice plans that could be designed in such a way that their primary effect advanced religion, or that excessively entangled government and religion. To be constitutional, the plan must be structured to avoid both of these hazards.

In its treatment of several parochial school aid cases following Lemon, the Supreme Court has attempted to offer some guidance as to legislation that meets these constitutional standards. Viewed generously, the Court has created more confusion than clarity.¹⁴ One prominent scholar was less generous, concluding that the Court’s efforts amounted to a “hodge-podge” of decisions derived from “Alice’s Adventures in Wonderland.”¹⁵ It is into this morass that those seeking to craft a constitutionally viable school choice plan must delve.

(III) Primary effects

On several occasions, the Supreme Court has invalidated all or part of state programs that provided various forms of assistance to parochial schools or to students attending them because they violated the second element of the Lemon test, the “effects” prong.¹⁶ Remember that the requirement here is that the law’s “principal or primary effect must be one that neither advances nor inhibits religion.”¹⁷ In each of those cases, the Court has found that government funds were being used to impermissibly advance religion.

However, other decisions, and particularly more recent ones, have rejected challenges to legislation based on the “effects” prong. Through these confusing and sometimes contradictory holdings, it’s possible to discern ways in which a universal school choice plan could survive the “effects” test. In doing so, the parameters of a model school choice plan begin to take shape. “Model” in this context does not necessarily describe the best kind of school choice plan; politicians and voters will decide that. Rather, it describes a universal school choice plan that is most immune from legal challenges.

1. The direct/indirect aid distinction

The first critical distinction to be made is between plans under which state funds go directly to religious schools, and those that result only in indirect financial benefits to those schools. Bearing in mind that government funds must not be allocated in ways that advance religion, it’s possible to develop constitutionally safe school choice programs under either system. However, many more restrictions would be required under a direct aid scheme, so much so that most choice advocates would find their desired outcome elusive.

The Supreme Court’s Establishment Clause precedents indicate that under a direct aid plan, the state would be obliged to ensure that its funds were not used for specifically religious activities. Prior to Lemon, the Court approved state laws authorizing reimbursement of transportation costs to parents of children attending religious schools,¹⁸ and the lending of secular textbooks to children attending religious schools.¹⁹ Such programs are still valid today under the second prong of the Lemon analysis. Though they entail a form of direct assistance, the aid is restricted to inherently nonreligious activities, and therefore does not have the effect of advancing religion.

It follows that in piecing together a school choice plan, two problems arise if the program features direct aid to religious schools. First, schools that the Court has characterized as “pervasively sectarian” would have to be excluded. The Court has held that where aid flows to a school in which “a substantial portion of its function[s] are subsumed in the religious mission . . . aid normally may be thought to have a primary effect of advancing religion.”²⁰ Several reasons have been given for this position, among them that such direct aid programs provide a symbolic link between government and religion, and at least imply state support of the particular denomination operating the school; and that such programs provide a direct state subsidy not simply to the secular educational mission, but to the primary religious mission of those institutions.²¹

The second difficulty is that the legislation would have to specify that no state funds could be used for sectarian purposes, and safeguards would have to be put in place to monitor the use of the funds. Though one would hope that most religious schools facing such onerous government demands would elect to opt out of the program, such monitoring of course could be done. Even where accomplished, however, use of these mechanisms engender the risk of violating the “excessive entanglement” prong.

Thus, we have arrived at Rule One of the Model School Choice Plan: Aid should not be granted directly to religious schools.

The question immediately arises as to what alternative funding mechanism is preferable. Minnesotans can take satisfaction in knowing that the answer in large part is provided by a homegrown case.

The Supreme Court set a benchmark for instances of “indirect” assistance to religious schools, and broke new ground, in the 1983 case of Mueller v Allen.²² The law in question in Mueller was a Minnesota statute that allowed all parents to deduct tuition, textbook and transportation expenses from their gross income on state tax returns for their children attending elementary and secondary schools.²³ The Minnesota law allowed a tax deduction of up to $500 for each child in grades K through 6, and up to $700 for each child in grades 7 through 12. Given that most parents of public school children incurred no tuition expenses, and that at the time 96 percent of Minnesota children in private schools attended religiously affiliated ones, it was clear that the bulk of the tax deductions would go to parents of children in sectarian schools.²⁴ A group of Minnesota taxpayers brought suit against the Minnesota commissioner of revenue and Minnesota parents who had taken the tax deduction for expenses incurred in sending their children to parochial schools. The suit alleged that the law amounted to a provision of financial assistance to sectarian institutions, and thereby violated the Establishment Clause.

Justice Rehnquist’s (he had not yet been elevated to Chief Justice) majority opinion in Mueller set forth a crucial distinction between direct aid and indirect aid to parochial schools. He wrote: “It is true, of course, that financial assistance provided to parents ultimately has an economic effect comparable to that of aid given directly to the schools attended by their children. It is also true, however, that under Minnesota’s arrangement public funds become available only as a result of numerous private choices of individual parents of school-age children.”²⁵ The Court concluded: “Where, as here, aid to parochial schools is available only as a result of decisions of individual parents no ‘imprimatur of state approval’ can be deemed to have been conferred on any particular religion, or on religion generally.”²⁶ Thus, under the Mueller Court’s view, this type of indirect aid is consistent with the neutrality principle embodied in the Establishment Clause and the effects prong of Lemon, i.e. it neither advances nor inhibits religion.

The Court has applied this reasoning in subsequent decisions as well. In 1986, in Witters v Washington Department of Services for the Blind,²⁷ a student applied for aid under a Washington state program to help provide vocational assistance to the blind.²⁸ At the time, the student was attending a Bible college, being trained for the ministry. The Washington Supreme Court denied aid to the student, citing the Establishment Clause and the primary effects prong of Lemon.²⁹ However, the U.S. Supreme Court unanimously overturned that ruling, and found no impermissible advancement of religion. Writing for the Court, Justice Thurgood Marshall elaborated on the individual choice aspect stressed by Justice Rehnquist in Mueller. Justice Marshall made a very telling comparison: “[A] State may issue a paycheck to one of its employees, who may then donate all or part of that paycheck to a religious institution, all without constitutional barrier; and the State may do so even knowing that the employee so intends to dispose of his salary. . . . In this case, the fact that aid goes to individuals means that the decision to support religious education is made by the individual, not by the State.”³⁰

In one of the major decisions of its most recently completed term in June, the Court cited Mueller and Witters with approval in Zobrest v Catalina Foothills School District.³¹ In Zobrest, the Court found no violation of the Establishment Clause where a public school district was asked to place one of its employees in a Roman Catholic high school to serve as a sign language interpreter for a deaf student, pursuant to a federal program to assist disabled school children.³² The language used by the Court had a familiar ring: “By according parents freedom to select a school of their choice, the statute ensures that a government-paid interpreter will be present in a sectarian school only as a result of the private decision of individual parents.”³³

Hence, the Court’s jurisprudence suggests a very strong judicial preference for “indirect” aid to religious schools. Within a universal school choice plan, this could be accomplished in several ways. It could come through a much expanded version of the Minnesota tuition tax allowances upheld in Mueller, or through the “scholarship certificates” given to each school-age child under Coons’ and Sugarman’s plan,³⁴ or by way of the “educational vouchers” long championed by Friedman.³⁵

In any case, as a corollary to Rule One, we can now establish Rule Two of the Model School Choice Plan: Religious schools must benefit only through the independent choices of parents.

2. Scope of eligibility

While the individual choice aspect is vital, it’s not sufficient by itself to secure a school choice plan against First Amendment challenges. The Court has held that more is needed to establish neutrality and prevent the primary effect of advancing religion. One additional factor in determining constitutional viability is the scope of eligibility. Ironically, it’s those who advocate the most inclusive, broad-based versions of school choice who are on the surest legal footing in this area.

What must be avoided is structuring program eligibility in such a way that skews the plan toward religious institutions or religious people. In rejecting a New York state law that specifically provided financial aid to nonpublic schools only, the Court clearly distinguished between programs that benefit all schoolchildren, and those benefitting only children in private (mostly religious) schools.³⁶ The Court found the New York program impermissibly skewed toward religion, because of its exclusion of public schools and the students attending them.

The Court remained consistent in another Establishment Clause case which presented the opposite predicament. When a state university which made its meeting facilities generally available to all student groups tried to prohibit the use of its buildings by religious student groups on Establishment Clause grounds, the Court said no.³⁷ Noting that there were over 100 student groups at the school (the University of Missouri at Kansas City), the Court significantly stated: “The provision of benefits to so broad a spectrum of groups is an important index of secular effect.”³⁸ Thus, far from dooming the program, it was the exclusion of religious groups from a program with an otherwise secular purpose and effect that was prohibited by the Court.

This raises the interesting question that some commentators have explored but no court has yet addressed. Namely, if in fact the Establishment Clause does not prohibit inclusion of religious schools in a school choice plan, is it permissible to exclude them?³⁹ After all, the state must neither advance nor inhibit religion. Through its Free Exercise Clause jurisprudence, the Supreme Court has long recognized the “fundamental interest of parents, as contrasted with that of the State, to guide the religious future and education of their children.”⁴⁰ In many instances, both the current school structure, and choice plans that exclude religious schools, effectively deny low- income parents the opportunity to freely exercise their religious beliefs in the education of their children. Only under a fully inclusive school choice plan is the free exercise right preserved.

In exploring the depths of the First Amendment, that proposition is open to much debate. For now, though, it is clear that in the area of eligibility, the effects prong of the Lemon test may be satisfied by making tuition assistance available to schools and students without regard to the religious-secular or private-public nature of the school in the program.

Hence, Rule Three of the Model School Choice Plan: Allow all schools — public, secular private, and parochial — to participate.

3. Financial incentives

The Court has expressed similar concerns about parochial school aid programs creating a financial incentive for parents to opt for religious education for their children.⁴¹ A school choice plan, for example, that made it less expensive for students to attend religious schools than nonreligious ones would create such a financial incentive, and would clearly run afoul of the mandate of avoiding the primary effect of advancing religion. In upholding the use of public funds in parochial schools in Witters, Justice Marshall emphasized this point, stating: “[The Washington program] does not tend to provide greater or broader benefits for recipients who apply their aid to religious education, nor are the full benefits of the program limited, in large part or in whole, to students at sectarian institutions.”⁴²

The Court’s insistence on financial neutrality in the parochial school aid context raises interesting questions of economic liberties. When matters of taxation and government subsidization are considered, it’s surely the case that the current system of elementary and secondary education creates a profound incentive for parents to abstain from religious education. One eminent legal scholar, Prof. Richard A. Epstein of the University of Chicago Law School, has noted that courts have mistakenly tended to look only at the benefits to individuals who use state aid in religious schools, while ignoring the costs they are required to bear.

Professor Epstein suggests that while the free exercise of religion does not exempt religious people from taxation, it does allow them to share equally in the benefits of public expenditures. The current regime, therefore, penalizes those taxpayers who choose to provide a religious education to their children by demanding two payments from them. Epstein argues that this distorts the balance between religious and nonreligious persons, and makes a mockery of the neutrality principle by forcing the former to subsidize the latter.⁴³

Thus, while the Court’s holdings require that there be no financial inducement toward religion, another virtue of universal school choice plans is that they could eliminate the perverse incentives in the current system. A choice plan that provided all schools with an equal amount of funding on a per-pupil basis would accomplish this. It’s noteworthy, however, that that would not be necessary in most cases. Most private schools could receive less funding than public schools and the resulting incentives would be the same as if the funding were equal. This is so because the difference in operating expenditures between public and private schools is so routinely wide. In Minnesota, in 1991-92, combined government expenditures for each student in public schools statewide averaged $5,194; in Minneapolis, it was $7,102.⁴⁴ By contrast, combined parish, family, and private expenditures per student in Minnesota Catholic elementary schools in 1991-92 was $1,919.⁴⁵

In either case, what is important is adherence to Rule Four of the Model School Choice Plan: Funding for students attending religious schools is no greater than for those attending nonreligious ones.

Accordingly, an analysis of the Supreme Court’s reasoning in its treatment of cases arising under the second prong of the Lemon test permits a simple conclusion: Where aid goes to parochial schools indirectly, through the private choices of individuals, and those choices are not skewed toward the sectarian over the secular, the religion clauses of the Constitution are not violated. Furthermore, where aid is transmitted in this manner, unlike with direct aid to religious schools, there is no need to exclude pervasively religious schools from the program or to forbid religious activities. Even in those instances, there is no state-initiated advancement of religion.

(IV) Excessive entanglement

A universal school choice plan could pass all of the secular purpose and advancement of religion legal hurdles and still be ruled unconstitutional if it fails the third prong of the Lemon test. In striking down a New York program that used federal funds to pay the salaries of public school teachers who volunteered to teach remedial classes in parochial schools, the Supreme Court held that: “Even where state aid to parochial institutions does not have the primary effect of advancing religion, the provision of such aid may nonetheless violate the Establishment Clause owing to the nature of the interaction of church and state in the administration of that aid.”⁴⁶ The third prong of Lemon prohibits a law from fostering excessive government entanglement with religion.

The key to this doctrine is the manner in which the school choice program is administered. Again, there are important differences between direct and indirect aid programs, and again direct aid schemes are more legally problematic.

Recall that under the Lemon analysis of direct aid programs, no state funds can be used for sectarian purposes, as this would fail the effects test. In the above case, the Court recognized that it was possible to prevent state-paid teachers from engaging in specifically religious activities, but doing so would entail thorough state supervision. Such supervision would result in too much state administrative entanglement with religion.⁴⁷ Thus, one reaches what Justice White, in his dissenting opinion in Lemon itself called an “insoluble paradox.”⁴⁸ If a method is devised to successfully avoid government funds supporting religious activities, it will still probably be struck down because its procedures are likely to foster impermissible entanglement.

This paradox is not insoluble, indeed, it does not exist, under the indirect aid plans. Where aid flows to sectarian schools through the independent decisions of individuals, there is no need for them to abstain from religious activities, and thus, no need for ongoing government monitoring. Consequently, in Mueller, the entanglement prong was quickly and easily dismissed, and in Witters it was not even considered.⁴⁹

However, even where aid is indirect, religious schools are not completely out of the regulatory woods. In the modern regulatory state, one can underestimate neither the number nor the intrusiveness of bureaucratic requirements that governments at all levels place on those who participate in state-sponsored programs. In fact, the greatest fear of many of the most ardent school choice supporters is that such plans could provide governments with the means of socializing private schools. A less recognized, but no less ominous fear must be that at some level of state regulation, the universal school choice plan would excessively entangle government and religion, and thereby pose an affront to the Establishment Clause.

Since the courts have not thoroughly addressed the question of entanglement in the indirect parochial school aid cases, the precise level of regulation that would face constitutional difficulty is unclear. However, considerable guidance can be gleaned from the host of areas in which the state currently regulates sectarian schools. Under various federal and state laws, religious schools are already required to meet minimum curricular, health, safety, hiring and employment requirements.⁵⁰ None of the laws requiring religious schools to meet these standards have been found to excessively entangle government and religion.

Thus, the only regulations that would potentially invalidate a school choice plan on First Amendment grounds would be those over and above the substantial regulations already in place. In the Milwaukee program, for instance, in addition to the standard regulations, the state is authorized to conduct financial audits and performance evaluations of participating private schools.⁵¹ It’s very unlikely, however, that these additional regulations would be sufficient to cause legal problems if placed on religious schools.⁵²

Additionally, it must be remembered that if a court was to invalidate a school choice program because of excessive entanglement, there is more than one judicial remedy or legislative solution available. Instead of killing the program, the excessive regulations could always be repealed.

What emerges from the analysis of the third part of the Lemon test, therefore, is Rule Five of the Model School Choice Plan: Additional state regulation of religious schools above current levels should be minimized.

(V) The Minnesota Constitution

For a universal school choice plan to survive legal challenges in Minnesota, yet another hurdle must be overcome. Though it would be somewhat anomalous for a program to meet the exacting church-state separation standards of the U.S. Constitution, yet fail under a state’s constitutional requirements, such a situation would be far from unique. Many states have provisions in their constitutions that strengthen, or otherwise supplement the freedoms guaranteed in the federal Bill of Rights. The Minnesota Constitution has a provision similar to the religion clauses in the First Amendment, and, additionally, it has one specifically referring to aid to sectarian schools. A Minnesota universal school choice plan must be able to meet objections based on these portions of the state constitution, and a careful examination reveals that it can.

Article I, Section 16 of the Minnesota Constitution is the rough equivalent of the Establishment and Free Exercise Clauses of the U.S. Constitution.⁵³ The Minnesota Supreme Court has always construed them the same way, and has always followed the U.S. Supreme Court’s jurisprudence as binding on its treatment of state-based legal actions on church-state issues. Thus, any school choice program that meets federal standards will not have any difficulty under Minnesota’s version of the religion clauses.

However, there is more to the story — sort of. Another provision in the Minnesota Constitution appears to present an obstacle to inclusion of religious schools. Article XIII, Section 2, one of the more concisely written provisions of the Minnesota Constitution, reads: “Prohibition as to aiding sectarian schools. 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 of any particular Christian or other religious sect are promulgated or taught.” This section is only sort of a factor for two reasons: One, it’s so seldom turned to by the Minnesota courts that it can almost be considered dormant; and two, even if resurrected for the legal battle against school choice, it would be surmounted by a program tailored along the lines of the model school choice plan.

The sectarian schools provision was added to the Minnesota Constitution in 1877. No Minnesota court has ever used it as a basis to strike down a program under which state assistance flowed to parochial schools. In fact, there have been only two cases in which the section played a significant part in decisions of the Minnesota Supreme Court, and none since 1975.⁵⁴ Last year the section arose in a case before the Minnesota Court of Appeals, and that court relied heavily on the state Supreme Court’s reasoning in its two earlier applications.⁵⁵ If the sectarian schools provision were used to challenge a universal school choice plan, the Minnesota Supreme Court would very likely review its two earlier decisions for guidance, and it is to them that we must turn.

The first use of the sectarian schools provision came in 1970. In Americans United v Independent School District No 622,⁵⁶ the Minnesota Supreme Court examined a constitutional challenge to a state law that provided funding for transporting children to parochial schools. The U.S. Supreme Court had recognized the validity of this type of program 23 years earlier,⁵⁷ and the issue therefore gave the Minnesota court the opportunity to distinguish the federal standard from that of the state. The court stated that the “limitations contained in the Minnesota Constitution are substantially more restrictive than those imposed by U.S. Const. Amend. I.”⁵⁸ However, it chose to uphold the state law, and ultimately relied on the U.S. Supreme Court’s judgment that such busing programs do not “support” parochial schools.⁵⁹

Last year, 22 years after the Americans United case, the Minnesota Court of Appeals characterized that decision as “reject[ing] an interpretation of the Minnesota Constitution that would prohibit any indirect or incidental benefit to religiously oriented institutions, even if an institution is so pervasively sectarian that some aid to religion results.”⁶⁰

The second case in which the sectarian schools provision was applied was the 1975 case of Minnesota Higher Education Facilities Authority v Hawk.⁶¹ In Hawk, the Minnesota Supreme Court found that direct aid to private, religiously affiliated Minnesota colleges in the form of tax-exempt revenue bonds did not violate the Minnesota Constitution. The court based its judgment on a similar U.S. Supreme Court holding, placing significance on the fact that the colleges that benefited were essentially secular in nature.⁶²

Thus, just as the U.S. Supreme Court has treated the effects prong of the Lemon test, Minnesota courts have drawn a clear distinction between direct and indirect aid. Last year, the Minnesota Court of Appeals upheld a state program that paid the expenses of high school students taking classes at religiously affiliated colleges and universities in Minnesota. Under the program, the state paid tuition and other fees directly to the colleges. Citing Hawk, the Court of Appeals approved this direct funding program to all colleges except those that were “pervasively sectarian.”⁶³ This is perfectly consistent with the U.S. Supreme Court’s treatment of direct aid cases.

It appears, therefore, that despite the perception of an additional burden placed on aid to sectarian schools by the Minnesota Constitution, and despite having made a fleeting notation of the tougher standard, Minnesota case law is in virtual lockstep with the federal standard. Indirect aid has been distinguished from direct aid; busing and tax exemptions have been explicitly approved; and due deference has been given to the relevant U.S. Supreme Court decisions. As is frequently true, perhaps even more significant is the dog that did not bark. In challenges to Minnesota laws that provided indirect aid to religious schools through tuition tax credits, the sectarian schools provision was never even mentioned by the courts.⁶⁴

There is no good reason for this to be otherwise. Article XIII, Section 2 prohibits public money from being used for the “support” of religious schools. A universal school choice plan, designed along the lines outlined here, in no way uses public money to “support” religious schools. The purpose of the plan is not to “support” religious schools. It’s not crafted in a way that encourages students to opt for religious schools. It includes every public and nonreligious private school in the program. And most importantly, any “support” that ultimately goes to religious schools does so as a result of the independent choices of parents, not from any decision by the state. Therefore, the model universal school choice plan does no offense to the Minnesota Constitution for really the very same reasons that it does not offend the First Amendment.

(VI) Conclusion

It’s clear that as long as a few essential factors are accounted for, a comprehensive school choice plan will conform to constitutional standards, and should therefore be sustained by the Supreme Court of the United States as well as Minnesota’s. As the U.S. Supreme Court’s jurisprudence of the last 25 years has demonstrated, however, prevailing standards are rather fluid, particularly in the area of government involvement with religion. Thus, it is useful to take a brief look at where the situation stands on the current Court and where it might be heading.

Five members of the current Supreme Court have expressed displeasure with the Lemon test, albeit from varying angles.⁶⁵ Nonetheless, it still survives. The Lemon analysis played a significant part in both Establishment Clause cases the Supreme Court heard in its 1992 Term, Zobrest v Catalina Foothills School District, and Lamb’s Chapel v Center Moriches School District. A scathing critic of Lemon, Justice Scalia noted that “like some ghoul in a late-night horror movie . . . Lemon stalks our Establishment Clause jurisprudence once again, frightening the little children and school attorneys . . .”⁶⁶ Despite this criticism, there is no unity of opinion on the Court on what standard would replace Lemon, and consequently no majority to scrap it.

Within the prevailing Lemon framework it’s possible to discern with varying degrees of confidence how the members of today’s Court are likely to approach a universal school choice case (which is sure to be brought before the Supreme Court shortly after its approval in any state).

Based on their prior decisions and writings, three Justices on the current Court appear as though they would be rock solid in favor of upholding a universal school choice plan similar to the model outlined here. Chief Justice Rehnquist authored the Mueller decision, and has long championed the indirect aid approach. Justice Scalia has favored a much narrower view of the Establishment Clause, which would limit its prohibitions to laws that signify state embrace of particular religions.⁶⁷ In his short time on the high bench, Justice Thomas has joined Justice Scalia in every Establishment Clause case, and there is every reason to think that he too would support the model school choice plan.

A fourth Justice, Justice Kennedy, appears very likely to sustain a universal school choice plan as well. Justice Kennedy is also uncomfortable with Lemon, and he favors a standard that is both easier to apply and in a sense less hostile to religion. In what has become known as the “coercion test,” Justice Kennedy has suggested that the Establishment Clause is violated only when state action coerces anyone to support or participate in a religion, or when the state directs benefits to religion in such a degree that it establishes a de facto state religion.⁶⁸ Neither of those limitations is remotely approached by the model school choice plan.

Newly retired Justice Byron White would most assuredly have provided the fifth and deciding vote in favor of universal school choice. Justice White was the Court’s most vigorous opponent of the “strict separation” approach, dissenting in every case in which Lemon was used to strike down assistance to parochial schools. Justice White’s absence will be felt in this area, and supporters of school choice will have to turn to the remaining members of the Court to find one Justice to fill out a majority opinion.

Justices Blackmun and Stevens are virtually certain to oppose a universal school choice plan. The dissenting opinion that they joined in Mueller rejected tuition tax deductions. It’s hard to conceive of one who adopts that view of the Establishment Clause supporting school choice that includes religious schools.

Justice Souter should also be viewed with suspicion. He alone joined Justice Blackmun in dissenting in Zobrest on Establishment Clause grounds. On the positive side, in Zobrest, Justice Souter concurred with the statement: “When government dispenses public funds to individuals who employ them to finance private choices, it is difficult to argue that government is actually endorsing religion.”⁶⁹ However, in the same case, his broad view of the Establishment Clause did not permit the state to provide a sign language interpreter to a deaf student in a religious school, under a program to assist in educating disabled children.

Perhaps prophetically, the famous statue of “Justice” — blindfolded, with scales in one hand, and sword in the other — is a woman. It’s the two women on the Supreme Court who could well hold the fate of universal school choice in their hands. The newest member of the Court, Justice Ruth Bader Ginsburg, has written little on church-state issues, and there are few clues as to her approach or temperament on them. Justice Sandra O’Connor is harder to peg than the other members of the Court. She refused to join the majority in Zobrest, dissenting on technical grounds.⁷⁰ Yet she was part of the five Justice majority in Mueller, and has expressed displeasure with the lack of clarity in the Lemon test.

Justice O’Connor has suggested modifying the Lemon doctrine so that it turns, broadly speaking, on whether a law amounts to a government “endorsement” of religion or a particular religion. She has defined her “endorsement test” as “preclud[ing] government from conveying or attempting to convey a message that religion or a particular religious belief is favored or preferred.”⁷¹ This analysis led Justice O’Connor to support the Court’s invalidation of a law authorizing a moment of silence in public schools on the ground that the law lacked a secular purpose and had the effect of endorsing religious prayer in the public schools.⁷² The same cannot be said of a universal school choice plan, however, and it seems quite likely that Justice O’Connor would find no impermissible endorsement of religion in such a plan.

Though contentious fights surely lie ahead, the battle for school choice indeed seems very winnable in the courts. Americans rightly take great pride in knowing that their highest law guarantees religious freedom. Still today, only a minority of countries can say the same. Americans should also take pride in the fact that the Constitution does not stand in the way of parents — parents of all income levels — deciding the best course for their children’s education.

There are a great number of problems in education in America today. Many are convinced that bringing competition and greater consumer choice to bear on public schools will not only increase the liberty of all Americans, but will fundamentally force all schools to improve. Thankfully, we live under laws that do not foreclose that long-awaited innovation.

---

¹ Davis v Grover, 480 NW2d 460, 477 (Wis 1992) (Ceci concurring).

² See, for example, forthcoming econometric study by University of Minnesota Prof. Stephen A. Hoenack, commissioned by the Choice in Education Foundation, to be released November 16, 1993 by Center of the American Experiment.

³ Milton Friedman, The Role of Government in Education, in Robert Solo, ed, Economics and the Public Interest 123-44 (Rutgers, 1955); John Chubb and Terry Moe, Politics, Markets, and America’s Schools (Brookings, 1990); John E. Coons and Stephen D. Sugarman, Education by Choice: The Case for Family Control (California, 1978).

⁴ The Milwaukee Parental Choice Program, Wis Stat Ann ¤ 119.23 (West 1990), provides that up to one percent of the students enrolled in the Milwaukee Public Schools may attend, free of charge to their families, private, non-sectarian schools within the city of Milwaukee. Family income must not exceed 175 percent of the poverty level, and participating private schools must meet defined performance criteria, submit to financial audits, and meet state health, safety, and discrimination laws that apply to government schools.

⁵ One issue was whether the plan was a private or local bill in violation of article IV, section 18 of the Wisconsin Constitution; the second was whether the plan violated the establishment of uniform school districts requirement of article X, section 3 of the Wisconsin Constitution; and the third concerned the “public purpose doctrine,” which requires that public funds be spent only for public purposes. See Davis v Grover, 480 NW2d at 462-63.

⁶ MN Const, Art XII, ¤ 1 (private bills); MN Const, Art XIII, ¤ 1 (uniform public school system); MN Const, Art X, ¤ 1 (public purpose doctrine).

⁷ In Minnesota in the 1990-91 school year, 85 percent of private schools had religious affiliations, while 91.7 percent of the students enrolled in private schools were in religious schools. “Information on Minnesota Nonpublic Schools,” Minnesota Department of Education (1990-91).

⁸ US Const, Amend I.

⁹ See MN Const, Art I, ¤ 16; Art XIII, ¤ 2.

¹⁰ 403 US 602 (1971).

¹¹ See, for example, Lemon v Kurtzman, 403 US at 668 (White dissenting); Wolman v Walter, 433 US 229, 262 (1977) (Powell concurring in part and dissenting in part); Eric J. Segall, Parochial School Aid Revisited: The Lemon Test, The Endorsement Test and Religious Liberty, 28 San Diego L Rev 263 (1991).

¹² Lemon, 403 US at 612-13 (citations omitted).

¹³ The Court has repeatedly enunciated this view. For a recent example, see School District of the City of Grand Rapids v Ball, 473 US 373, 382 (1985).

¹⁴ For a compendium of the muddiness of these guidelines, see Wallace v Jaffree, 472 US 38, 115 (1985) (Rehnquist dissenting).

¹⁵ Philip B. Kurland, The Religion Clauses and the Burger Court, 34 Cath U L Rev 1, 10 (1984).

¹⁶ See, for example, Committee for Public Education & Religious Liberty v Nyquist, 413 US 756, 776 (1973); Meek v Pittenger, 421 US 350, 365 (1975); Wolman v Walter, 433 US 229, 251 (1977); School District of the City of Grand Rapids v Ball, 473 US 373, 397 (1985).

¹⁷ Lemon, 403 US at 612-13.

¹⁸ Everson v Board of Education, 330 US 1 (1947).

¹⁹ Board of Education v Allen, 392 US 236 (1968).

²⁰ Hunt v McNair, 413 US 734, 743 (1973).

²¹ Ball, 473 US at 385.

²² 463 US 388 (1983).

²³ Minn Stat ¤ 290.09 subd 22 (1982).

²⁴ Mueller, 463 US at 401.

²⁵ Id at 399.

²⁶ Id at 399 (citations omitted).

²⁷ 474 US 481 (1986).

²⁸ Wash Rev Code ¤ 74.16.181 (1981).

²⁹ Witters, 474 US at 482.

³⁰ Id at 487-88.

³¹ 113 S Ct 2462, 2466 (1993).

³² Individuals with Disabilities Education Act (IDEA), 20 USC ¤ 1400 (1991).

³³ Zobrest, 113 S Ct at 2467.

³⁴ Coons & Sugarman, Choice, at 31 (cited in note 3).

³⁵ Friedman, The Role of Government in Education (cited in note 3).

³⁶ Committee For Public Education & Religious Liberty v Nyquist, 413 US 756, 782 n 38 (1973).

³⁷ Widmar v Vincent, 454 US 262 (1981).

³⁸ Id at 274.

³⁹ Peter K. Rofes, Public Law, Private School: Choice, the Constitution, and Some Emerging Issues, J L & Educ 503, 520-24 (1992).

⁴⁰ Wisconsin v Yoder, 406 US 205, 232 (1971), exempting Amish parents and children from state law requiring school attendance until age 16. See also, Pierce v Society of Sisters, 268 US 510, 535 (1924), rejecting mandatory public school attendance, stating famously, “The child is not the mere creature of the State.”

⁴¹ See Nyquist, 413 US at 786; Witters, 474 US at 488.

⁴² Witters, 474 US at 488.

⁴³ Richard A. Epstein, Testing the Boundary Between Church and State, The Wall Street Journal A9 (December 23, 1992).

⁴⁴ “School District Profiles,” 1991-92, Minnesota Department of Education. These figures include capital expenditures and debt service outlays ($539 for statewide average, $637 for Minneapolis).

⁴⁵ This figure excludes approximately $300 per student in state aid for transportation and textbooks. See, A Comparison of Education Expenditures, School Choice News 1 (May 1993).

⁴⁶ Aguilar v Felton, 473 US 402, 409 (1985).

⁴⁷ Id at 410.

⁴⁸ Lemon, 403 US at 668 (White dissenting).

⁴⁹ Mueller, 463 US at 403; Witters, 474 US at 489 n 5.

⁵⁰ Note, The Increasing Judicial Rationale for Educational Choice: Mueller, Witters, and Vouchers, 66 Wash U L Q 363, 388 (1988).

⁵¹ Wis Stat Ann ¤ 119.23(9)(a).

⁵² See Mueller, 463 US at 403-04 n 11 (distinguishing among the entanglement analyses applied to various types of assistance).

⁵³ MN Const, Art I, ¤ 16: Freedom of conscience; no preference to be given to any religious establishment or mode of worship. . . . The right of every man to worship God according to the dictates of his own conscience shall never be infringed; nor shall any man be compelled to attend, erect, or support any place of worship, or to maintain any religious or ecclesiastical ministry, against his consent; nor shall any control of or interference with the rights of conscience be permitted, or any preference be given by law to any religious establishment or mode of worship . . . nor shall any money be drawn from the treasury for the benefit of any religious societies or religious or theological seminaries.

⁵⁴ See Americans United, Inc v Independent School District No 622, 288 Minn 196 (1970); Minnesota Higher Education Facilities Authority v Hawk, 305 Minn 97 (1975).

⁵⁵ Minnesota Federation of Teachers v Mammenga, 485 NW2d 305, 307 (Minn App 1992).

⁵⁶ 288 Minn 196 (1970).

⁵⁷ See Everson v Board of Education, 330 US 1 (1947).

⁵⁸ Americans United, 288 Minn at 213.

⁵⁹ Id at 208.

⁶⁰ Mammenga, 485 NW2d at 308.

⁶¹ 305 Minn 97 (1975).

⁶² See id at 113, citing Hunt v McNair 413 US 734 (1973).

⁶³ Mammenga, 485 NW2d at 309-10.

⁶⁴ See Minnesota Civil Liberties Union v State, 302 Minn 216 (1974); Minnesota Civil Liberties Union v Roemer, 452 FSupp 1316 (1978).

⁶⁵ For a chronology see Lamb’s Chapel v Center Moriches School District, 113 S Ct 2141, 2150 (1993) (Scalia concurring).

⁶⁶ Id at 2149.

⁶⁷ Id at 2151

⁶⁸ Allegheny County v American Civil Liberties Union, 492 US 573, 655-57 (1989) (Kennedy concurring in judgment in part and dissenting in part).

⁶⁹ Zobrest, 113 S Ct at 2474 (Blackmun dissenting).

⁷⁰ Zobrest, 113 S Ct at 2475 (O’Connor dissenting).

⁷¹ See Wallace v Jaffree, 472 US 38, 70 (1985) (O’Connor concurring).

⁷² Id at 77-79.