The principles underlying the U.S. Constitution’s establishment clause, while not yet dead, are in exceedingly poor health this morning.

To be fair, the wall of separation between church and state never really existed. But for a while, there was a fairly high fence. It protected religious institutions from entanglements with the government, and vice versa. In countries without such separation, state institutions such as schools can become instruments of the state’s preferred religion — as the writers of the U.S. Constitutions observed in England and other European countries.

England is, however, an instructive case of what can happen over time. The entanglement between the two institutions initially bolsters the church, but genuine religious beliefs are harder to impose than is nominal adherence to those beliefs. So Americans became more religious over time as they voluntarily brought church attendance and religious teachings into their lives. But the English became less so, even while maintaining state financial assistance for church schools to this day.

In the United States, that high fence of separation between church and state existed from approximately the mid-1940s through the mid-1980s. It transformed our essentially Protestant public schools into secular institutions attended by a cross-section of the population, including strongly religious families. That seeming contradiction of religious upbringing plus nonreligious schooling was, in fact, entirely consistent with Thomas Jefferson’s reasons for embracing a “wall of separation” to avoid government involvement that could corrupt free religious practice, while also protecting the government against church influence.

Churches and related religious institutions benefit from this arrangement in three key ways. First, the government stays away from the internal affairs of churches. While this can lead to fraud and abuse, it also protects religious liberty. Second, the government grants churches freedoms denied to other institutions, including the freedom to discriminate. Third, because “the power to tax involves the power to destroy,” churches are given an extraordinary number of tax benefits. This was outlined in an analysis published in The Washington Post in 2013:

When people donate to religious groups, it’s tax-deductible. Churches don’t pay property taxes on their land or buildings. When they buy stuff, they don’t pay sales taxes. When they sell stuff at a profit, they don’t pay capital gains tax. If they spend less than they take in, they don’t pay corporate income taxes. Priests, ministers, rabbis and the like get “parsonage exemptions” that let them deduct mortgage payments, rent and other living expenses when they’re doing their income taxes. They also are the only group allowed to opt out of Social Security taxes (and benefits).

As part of this exceedingly hands-off approach, church-affiliated institutions were not, in the past, eligible to participate in some government programs. For instance, while private religious colleges could receive aid to help students fund their education, the court prohibited state aid directly to religious K-12 schools.
Today’s Supreme Court does not share Jefferson’s vision. In fact, the court has been sawing away at the high fence for decades. In cases posing challenges to state funding of religious institutions, the court has steadily permitted greater and greater financial entanglements.

In fact, in a case three years ago called Trinity Lutheran v. Comer, the court pronounced that states can, under some circumstances, be constitutionally required to fund religious institutions, pursuant to the free-exercise clause of the First Amendment.

Will the Supreme Court’s Trinity decision lead to the spread of school voucher programs?
When providing a public benefit (in that case, state grants for playground resurfacing), the state cannot make religious status an impediment to receipt of that benefit — at least where the benefit is not directly supporting religious practice.

Even from this conservative court, the Trinity Lutheran decision was a bit of a surprise. Earlier, in 2004, in a case called Locke v. Davey in which the majority opinion was written by the very conservative Chief Justice William H. Rehnquist, the court found no constitutional impediment to a state prohibiting a college scholarship from being used directly to support religious practice, by excluding students pursuing a “degree in devotional theology.”

With those key precedents, the court today decided a case involving a neo-voucher law that had been adopted in Montana. The law used tax credits to create a funding mechanism for small vouchers (about $500 each) to help pay for private school tuition. Because the Montana state constitution includes a “no aid” clause that prohibits direct or indirect state support for church-controlled schools, the Montana Department of Revenue only allowed the law to go forward on the condition that religious schools be excluded.

A lawsuit called Espinoza v. Montana Department of Revenue challenged that ruling and made its way to that state’s Supreme Court. That court struck down the entire neo-voucher law, thus avoiding the possibility of anti-religious discrimination raised in Trinity Lutheran v. Comer, while also avoiding a violation of the Montana constitution. That should have ended the matter, but the U.S. Supreme Court decided to weigh in.

Before discussing the court’s decision in the Espinoza case, it’s important to step back and consider the unusual ideological extremity of the current Supreme Court. The court was designed as a moderating institution. One fundamental reason for this is that the more-extreme views of any single justice typically will have only a minimal impact on the court’s final decision. The court will only take the law as far as the “fifth vote” or “swing vote.” If four justices wanted, for instance, to declare all affirmative action programs unconstitutional, but the fifth vote in that case wanted to allow limited affirmative action programs under narrow circumstances, then the latter becomes the new legal standard.

The court also is designed to be somewhat insulated from political pressures, with justices appointed for life. One result is that presidents can ultimately be surprised by their appointments. We saw this, for instance, with Eisenhower appointee Justice William J. Brennan Jr. ending up to the left of Kennedy appointee Justice Byron R. White.

Yet the Supreme Court has now become almost as ideologically predictable as the Congress. And the ideological pendulum has become a one-way bulldozer — a process that started during the Nixon administration. The so-called swing justice went from Lewis F. Powell Jr. in the 1980s to Sandra Day O’Connor, to Anthony M. Kennedy to, now, Chief Justice John G. Roberts Jr. All of these were Republican appointees, but they’ve become more conservative with each new retirement. So the court has moved further and further to the right. There’s now a reliable five-justice majority on issues ranging from school vouchers and affirmative action to border control and deregulation.

I would be remiss if I did not mention here the most momentous and egregious event in this process of building the current ultraconservative court: the unprecedented obstruction of President Barack Obama’s nominee Merrick Garland.

Scalia died unexpected in February 2016, and Obama put forward Garland’s nomination in March. But Senate Majority Leader Mitch McConnell (R-Ky.) refused to allow any confirmation hearings, citing the presidential election to take place eight months later. President Trump eventually appointed Justice Neil M. Gorsuch to the seat, maintaining the conservative five-justice majority (which was soon strengthened with Justice Brett M. Kavanaugh replacing Justice Anthony M. Kennedy). Had Garland replaced Scalia, the swing justices (depending on the issue) would have been moderate-liberal justices Stephen G. Breyer, Elena Kagan and Garland — with Justices Ruth Bader Ginsburg and Sonia Sotomayor reliably to their left. Even though Garland was generally seen as a moderate judge, this would have been the court’s most momentous shift to the left since the 1960s. Decisions like that handed down today would have looked far different.

But the Espinoza decision was itself decades in the making. The legal landscape for vouchers supporting private religious schools has changed 180 degrees, corresponding to the shift in the makeup of justices on the Supreme Court. Vouchers for religious schools have moved from being broadly understood to be constitutional forbidden in 1970s to constitutionally allowed in 2003, via the Zelman decision, to now arguably constitutionally required, at least under the Montana circumstances.

Let’s return to that high fence mentioned earlier. Once the Supreme Court decided to hear the Espinoza case, we were left to hope that it would at least leave in place a speed bump of separation between church and state.
What we got instead is a shotgun marriage between church and state.

The court’s decision this morning held that application of the “no aid” provision in the Montana state constitution was barred by the Constitution’s free-exercise clause. The Montana provision prohibited any direct or indirect aid to a school controlled by a “church, sect, or denomination.” Like the dispute at issue in the Trinity Lutheran case, this prohibition was based on status as a religious institution, rather than a religious use. But the court’s Espinoza majority opinion — written by Roberts and joined by the other four conservative justices — also minimizes that distinction, which was important in Locke and potentially crucial to the decision in Trinity Lutheran:

“None of this is meant to suggest that we agree with the Department [of Revenue] that some lesser degree of scrutiny applies to discrimination against religious uses of government aid. Some Members of the Court, moreover, have questioned whether there is a meaningful distinction between discrimination based on use or conduct and that based on status. [Citations to Justices Gorsuch and Thomas]. We acknowledge the point but need not examine it here. It is enough in this case to conclude that strict scrutiny applies under Trinity Lutheran because Montana’s no-aid provision discriminates based on religious status. (Page 12, with internal citations omitted.)”

Here’s more critical language, from pages 13-14 of the court’s slip opinion (with internal citations omitted and emphasis placed on one sentence):

“Locke differs from this case in two critical ways. First, Locke explained that Washington had “merely chosen not to fund a distinct category of instruction”: the “essentially religious endeavor” of training a minister “to lead a congregation.” Thus, Davey “was denied a scholarship because of what he proposed to do — use the funds to prepare for the ministry.” Apart from that narrow restriction, Washington’s program allowed scholarships to be used at “pervasively religious schools” that incorporated religious instruction throughout their classes.”

By contrast, Montana’s constitution does not zero in on any particular “essentially religious” course of instruction at a religious school. Rather, as we have explained, the no-aid provision bars all aid to a religious school “simply because of what it is,” putting the school to a choice between being religious or receiving government benefits.

At the same time, the provision puts families to a choice between sending their children to a religious school or receiving such benefits. Second, Locke invoked a “historic and substantial” state interest in not funding the training of clergy, explaining that ‘opposition to … funding “to support church leaders’ lay at the historic core of the Religion Clauses.” … But no comparable “historic and substantial” tradition supports Montana’s decision to disqualify religious schools from government aid.

The court concludes, “[I]t is clear that there is no ‘historic and substantial’ tradition against aiding such schools comparable to the tradition against state-supported clergy invoked by Locke.” (p. 16). But this historical focus was a side note in the court’s earlier Locke decision. What Roberts did this morning was to limit Locke to its unique facts, marginalizing its usefulness as a precedent.

But note that sentence in bold from the passage above. A state’s constitution is given meaning by the state’s courts. Imagine if, upon remand, the Montana Supreme Court issues a new decision, saying something like: “In view of the U.S. Supreme Court’s decision, we interpret the ‘no-aid’ provision in our state constitution to prohibit any direct or indirect financial support to religious instruction in church-controlled schools.”

This would allow a voucher law that provides support to religious schools but not to religious education. Implementing or enforcing that funding mechanism would entangle the state with the operations of the religious schools, but it would seem doable. Would the Roberts court find that approach to be constitutional, or would it further restrict the reach of Locke?

What’s clear for now is that the long-standing “tuitioning” voucher systems in Maine and Vermont, which are limited to nonreligious private schools, cannot stand. There will have to be a shotgun wedding between church and state in those two states.