Nuns and their supporters rally outside the Supreme Court in Washington in March 2016, as the court hears arguments in the Zubik v. Burwell case.Jacquelyn Martin / AP

It began a quarter-century ago with germ warfare in rural Oregon. That episode spurred a short-sighted state government to refuse a minor request for a religious accommodation. The resulting quarrel offered an opening for a self-confident young Supreme Court justice in a hurry to remake the law in his own image. And what it produced was a revolution in American religious-freedom law—transforming the rights of individual conscience into yet another firewall for secular wealth and ecclesiastical power.

The latest chapter began on March 23, when the Supreme Court heard arguments in a group of cases called Zubik v. Burwell. In Zubik, religiously oriented non-profit employers—such as hospitals and schools—objected to the contraceptive-care mandate of the Affordable Care Act. Their objection is not that these non-profit employers are required to provide contraceptive options to their employees. They are not. All they must do is certify their refusal. Once they fill out a government form, a third-party insurer that runs their employee plans will make the benefits available to employees—without any participation or subsidy by the employer.

But that’s not enough for the non-profits. They don’t want to fill out the form or allow the insurance company to provide the coverage. The new claim of free exercise—one which will be heard more and more—is not for accommodation, or even exemption, from the law; it is for immunity. It is for, in essence, the right to secede from civil society.

The germ-warfare beginnings of this strange career are worth recounting because they show how far off course the law of religious freedom has gone.  There is no generally recognized meaning of the term “religious freedom.” The Constitution simultaneously protects the “free exercise” of religion and bans its “establishment.” The struggle to reconcile the two has always been complicated.  By the late 1980s, when this story begins, the system was spinning upright, like a child’s top—one whose spin had developed a slight wobble. It was worrisome; it made the system harder for judges to administer. It was also somewhat arbitrary: Native Americans, for example, never received the protection their faiths deserved. But it was upright. When a top is ever-so-slightly wobbly, there’s a temptation to try to gently nudge it back to perfection. But children quickly learn that trying that will send the top spinning even further off its axis, until it spins out of control and falls. That in essence is what has happened to the law of free exercise. Without telling anyone, the Supreme Court impulsively decided to fix it. It was a mistake.

Here’s the puzzle: Everyone recognizes that “free exercise” protects religious bodies as well as individuals; but until the 1990s, by and large, the claims of organized religion—Catholic archdioceses, Jewish congregations, and other institutions with a religious cast—were negotiated in Congress or state legislatures. “Conscientious objection” to war was protected by statute, not constitutional litigation; statutes, not court cases, protected the right of Christian congregations to use wine during the sacrament of Communion even during prohibition. Anti-discrimination statutes exempted religious groups from certain discrimination claims—such as female postulants demanding entry to all-male priesthoods.

In contrast to legislation, constitutional litigation was mostly used by individuals who had no powerful denomination to protect them. For these plaintiffs, the Court created something called “the Sherbert test.” It was named after Adell Sherbert, a Seventh-Day Adventist who quit her job when the boss required her to work on Saturday, her Sabbath. The state refused to pay her unemployment compensation on the grounds that her religion was not “good cause” for leaving her job. Announced in 1963, the test asked whether the regulation was “narrowly tailored”—that is, necessary—to further a “compelling” government purpose. (In Sherbert’s case, the Court held, the regulation wasn’t “narrowly tailored”—state law recognized non-religious, but not religious, reasons as constituting “good cause.”) Sherbert mostly applied to claims by individuals like Adell Sherbert: Jehovah’s Witnesses who objected to doing war-related work, Amish parents who faced jail for home-schooling their children.

Though focused on individual rights, courts applying the test also asked whether accommodating the objector would shift the burden to a third party. A Jehovah’s Witness mother could sell pamphlets in the street, but could not make her 9-year-old to do so in violation of child-labor laws. A Connecticut law required employers to give religious employees their “Sabbath” off. The Court held that this rule burdened other employees, who also might want those days off. The Amish parents won, in part, because they weren’t seeking to exempt their teens from school, but to educate them at home. And Amish didn’t get complete immunity from the law: Amish employers, who object to Social Security, were still required to pay their employees’ Social Security payroll tax. Orthodox Jews did not have a constitutional right to keep their discount stores open on Sundays when other stores were required to close.

“When followers of a particular sect enter into commercial activity as a matter of choice,” the Court explained, “the limits they accept on their own conduct as a matter of conscience and faith are not to be superimposed on the statutory schemes which are binding on others in that activity.” (Though these claimants lost in court, both these disputes were eventually solved by the legislatures. Congress created a scheme to allow Amish employers to opt out of Social Security for their Amish employees only; state legislatures slowly did away with Sunday-closing laws.)

A quarter-century later, however, “religious freedom” has morphed from a shield defending the meek into yet one more sword demanding obeisance to the haughty. Officials in Greece, New York, begin public town council meetings with explicitly Christian prayers. The implicit demand of citizens who must do business before the council is for passive adherence—especially when the clergy ask them to bow their heads, to recite prayers, or to accept Jesus. In 2014, the Supreme Court held that citizens must accept this indignity—to do otherwise would trespass on the religious rights of the officials. They have chosen to display their piety. The citizens’ rights, the Court said, are trumped by the religious right of the officials themselves to display their faith—“to show who and what they are.” And in 2015, the Court held that the owners of large for-profit corporations may demand some sort of religious exemption to prevent their employees from making contraceptive choices of which their employers don’t approve.

Some of the states, meanwhile, are legally guaranteeing the religious “rights” of employers, contractors, schools, and businesses to discriminate against legally married same-sex couples and LGBT people in general. Under a North Carolina statute, county clerks—who, in the marriage context, embody the power of the state—may refuse to issue licenses to same-sex couples. In Kentucky, a county clerk went to jail rather than allow the issue of same-sex licenses; the legislature has now ratified her lawless behavior by statute. “Religious freedom” measures—aimed at licensing discrimination—are pending in other states.

All these measures transfer power from the less powerful party to the more powerful one. All of them explicitly privilege one set of religious beliefs, centered around those of conservative Christianity. All of them simply brush aside the consciences of the less powerful parties—the employees who wish to make their own decisions about contraception, the couples who seek only to enjoy their constitutional right to form families.

In the new world of religious freedom, the less powerful just don’t count. As a lawyer for Hobby Lobby told the Court in 2014, “not all burdens are created equal.”

How did this happen? The answer is a cautionary tale of how the best intentions of good people—lawyers, legislators, and judges—can produce results no one foresaw.

In 1981, Bhagwan Shree Rajneesh, an Indian holy man, purchased the 64,000-acre Big Muddy Ranch near Antelope, Oregon. The plan was to make “Rancho Rajneesh” into a huge commune with a hotel, theater, concert hall, and retreat center. Unfortunately, Rajneesh had failed to research Oregon’s demanding land-use laws, which forbade any such development unless the land was inside an “urban growth boundary.” Once the cultists learned of this, they went to work to become a city. First, they persuaded the county commission to allow them to incorporate the Ranch as the city of “Rajneeshpuram.” Then they sent cult members into nearby Antelope (pop. 39) to take over the city council there. In both cities they imposed the heavily armed “Rajneesh Peace Force” as the official law-enforcement agency. They closed previously public roads and facilities to non-members. In Antelope, they began harassing the original residents to make them leave.

All the while, the leaders of the cult proclaimed themselves simple pilgrims seeking the American dream of salvation and self-sufficiency. If a government action stood in their way, they proclaimed it the product of bigotry and religious hatred. When circumstances called for it, they changed the organization of the cult, even its central beliefs, to fit the individual challenge they faced.

As the crisis deepened, a local official asked Dave Frohnmayer, the Attorney General of Oregon, for a formal opinion on whether the cities’ conduct was legal. Frohnmayer’s response: a “city” in which one religious corporation owned all land and controlled all roads, and permitted only members of the religion itself to become residents, was, under the law, not a city at all. “Our legal system requires that the pathway to religion be private and internal to each pilgrim's mind and soul,” Frohnmayer wrote. “The state and federal constitutions do not permit the road to Damascus to be paved with public funds.”

That opinion set off a state of warfare between the Rajneesh followers and the state government. At one point, the cult decided its best course would be to seize control of the entire Wasco County government, where the “cities” were located. It first tried to import enough homeless voters to sway the election; when that failed, it hatched the germ warfare plan. County officials who visited the ranch began to fall ill with food poisoning. Cult members slipped salmonella bacilli slipped into the salad bars of restaurants around the county, sickening dozens. It was a dry run for a plan to immobilize much of the population on election day—keeping them away from the polls.

It was a crazy plan, and it blew up in their faces. Meanwhile the cult was riven with factional disputes and internal poisoning plots as well. Before too long the Oregon State Police were searching the ranch and the cult’s leader was in federal custody. But before the end, the cult sent out a team of assassins with orders to kill the U.S. attorney, a prominent reporter, and Dave Frohnmayer, the attorney general. They failed; but being the target of an assassination plot can, as Samuel Johnson said in another context, serve to concentrate the mind wonderfully.

That concentration explains what happened next. During the final days of the cult, the Oregon agency handling unemployment insurance received a low-key claim from two drug-abuse and alcohol counselors in Roseburg, a small town 250 miles away from the dispute. The two—Galen Black, an Anglo and Al Smith, a member of the Klamath Tribe—had been fired from their jobs for participating in a ritual of the Native American Church. Though peyote religion is not drug use (any more than Catholicism is alcohol abuse), their employer refused to pay unemployment. The two claimants appealed to the state’s administrative tribunal, where they won on narrow grounds—grounds that had almost no implications for Oregon law.

At this point the attorney general’s office entered the fray with barely concealed fury. Though the dispute was a very ordinary one—the leading religious freedom cases of the 60s and 70s were unemployment cases—Oregon now drew a line in the sand. “Drug use” was misconduct. “Drug use” was illegal. “Drug use” was a threat to social order. If peyotists got their payments, then pot users and acidheads would be next. When the State Board of Pharmacy issued a rule providing that peyote was not a “drug,” the attorney general convened an emergency meeting and forced the Board to back down. When Native American advocacy groups from outside the state tried to arrange a settlement, the state insisted on terms so punitive that the two plaintiffs refused to sign.

There was no need for any of this. Peyote religion has roots nearly 10,000 years olds, and is mostly centered in the American Southwest and the north of Mexico. In its present form it has coexisted with American society for more than a century. It is the most tightly regulated religion in America: no one can grow, harvest, sell, or possess peyote without the approval of state and federal governments. Peyotists cooperate with the system and seek to avoid conflicts with the law. They agree that the sacrament is not to be eaten by children; they object to illicit use of “Grandfather Peyote” far more strongly than does law enforcement. A major aim of peyote congregations nationwide is to help worshipers recover from alcoholism and drug addiction.

But these explanations fell on deaf ears, and it’s easy to see why. Frohnmayer and the Oregon state officials had heard all that before. Freedom of religion is the American way. We pledge to obey the law. We seek only a small accommodation. Those who criticize us are bigots.  That was the refrain of the Rajneesh cult, while it transformed itself over and over into new and deadlier forms, cold-bloodedly broke the law, used state power aggressively against outsiders, and violated all its pledges of good behavior. To state officials, it seemed, peyotism—a faith thousands of years old—was no more to be trusted than a New Age cult. Assurances could be broken; church doctrine might change. Today peyote, they reasoned, and tomorrow it might be marijuana or LSD. Oregon dug in on the Maginot Line to fight the last war. The state took the case to the Supreme Court not once but—despite anguished pleas from clergy and legal groups—twice.

At 1 First St NE, this small dispute ran into a brash new justice, 54-year-old Antonin Scalia. He shared the general concern that the law of religious freedom was unstable. But unlike many other judges, he was sure he personally, without help from anyone, could right the top with one firm shove.

Without notice to parties or anyone else, Scalia persuaded his fellow justices to change the issue from a real question of “unemployment compensation” to an imaginary one of crime. All Smith and Black had asked for was their unemployment. Scalia blandly asserted that the two men were criminals. Smith and Black, he wrote, “urge us to hold, quite simply, that when otherwise prohibitable conduct is accompanied by religious convictions, not only the convictions but the conduct itself must be free from governmental regulation.” This was a bald-faced invention: In fact, Smith and Black asked, “quite simply,” for their unemployment compensation. The state of Oregon had never charged them with any crime. There was, in fact, a great deal of question whether peyote use was unlawful at all; anyway, under Oregon law, violations of criminal law were explicitly irrelevant to unemployment compensation.

Scalia, however, used the rewritten “facts” as a means to do away with the Sherbert test for good and all. As long as a law did not intentionally single out religion, he now wrote, the free exercise clause did not protect those whose practice was “burdened”—or even outlawed entirely. Dissenters could beg the legislature for relief; but the courts weren’t open to them.  “It may fairly be said,” Scalia admitted, “that leaving accommodation to the political process will place at a relative disadvantage those religious practices that are not widely engaged in.” But that was an “unavoidable consequence of democratic government.”

One basic tenet of judicial craftsmanship is that a court should decide only the case in front of it. A common criticism of Roe v. Wade, the landmark abortion case decided in 1973, is that the Court reached far beyond the facts to create, in essence, an entire legal framework for adjudicating any issue relating to abortion regulation.

Employment Division v. Smith was, in that sense, Scalia’s Roe. As it arrived at the Court, the case had almost no implications for organized religious groups; it attracted little attention outside Indian country and fewer than half a dozen amicus briefs. But the sudden announcement of an entirely new system of religious freedom—without limiting or qualifying language—stripped from even mainstream groups a set of protections they had taken for granted. Even though the major religious groups often succeeded by negotiating with lawmakers, they had always had, or believed they had, the Constitution in their back pockets. If government went too far into their affairs, they calculated, they could go to court. Now, courtesy of the devout new justice, the courts were to be closed to them as well as to obscure peyote adherents.

The reaction from organized religions was immediate and furious. Douglas Laycock, a religious scholar respected by all sides of the debate, told a congressional hearing that “Many judges, bureaucrats, and activists have taken Smith as a signal that the Free Exercise Clause is largely repealed, and that the needs of religious minorities are no longer entitled to any consideration.” Every faith and every church body faced danger from hostile majorities and government regulators, he said: “No church is big enough or tough enough to fight them all off, over and over, at every level of government. The situation is even more hopeless for individual believers with special needs not shared by their whole denomination.” At the insistence of religious groups, Congress repudiated the decision by passing the Religious Freedom Restoration Act, which ordered state and federal courts to apply the Sherbert test once again.

The early history of RFRA, in retrospect, ought to have been a warning that the two halves of free exercise—the protection of institutional religion on the one hand and of lonely dissent on the other—were going to pull apart in the wake of Smith. The coalition of mainstream religious groups that backed RFRA began their campaign by very carefully and explicitly excluding the real losers in Smith—Anglos and Native people who worship Grandfather Peyote. (In fact, the House report accompanying RFRA explained that even after it passed, government would still be able to criminalize peyote religion.)

Thanks to the persistence of Indian people, a different coalition passed the American Indian Religious Freedom Amendments of 1994, to make sure that peyote religion was protected. But the betrayal by the mainstream groups was a warning of what was to come—the transformation of “free exercise” into the property of the powerful.

Once they had walked away from Native religion, however, religious groups insisted on a statute as broad as possible. Congress often overrules or corrects the Supreme Court in statutory questions; RFRA, instead, curtly told the Court it didn’t understand the Constitution. In Smith, the statute says, “the Supreme Court virtually eliminated the requirement that the government justify burdens on religious exercise imposed by laws neutral toward religion”; however, in the judgment of Congress, “the compelling interest test as set forth in prior Federal court rulings is a workable test for striking sensible balances between religious liberty and competing prior governmental interests.” From now on, the statute commanded, the courts were to apply that test in all free exercise cases, state or federal.

It was a twist of the lion’s tail, and the lion roared back. In 1997, The Supreme Court decided that the RFRA was valid for federal law but was too broad to apply to the states. Congress might pass targeted statutes, but could not apply Sherbert to the states across the board. After the Court held states immune from RFRA, 21 state legislatures enacted state RFRAs of different scope and force. Congress passed a more focused bill, the Religious Land Use and Institutionalized Persons Act of 2000, extending free exercise protection to church groups in state zoning disputes and also to inmates in state prisons. The bifurcation of claims was continuing. Prisoners are often the loneliest practitioners of lonely minority religions. Religious landowners, on the other hand, usually have access to power. (A guaranteed exemption from land-use laws, remember, was what the Rajneesh cult had sought.)

The stated aim of these federal and state RFRAs was to restore the law of religious freedom as it had been before Scalia’s brainstorm. But anyone who has ever read W.W. Jacobs’s classic story, “The Monkey’s Paw,” knows that what is resurrected is often not the same as—and is often a terrible parody of—what has died.

What came back, in fact, is much more closely attuned to the wishes of the powerful than the pre-Smith law. The movement of free-exercise rights from the meek to the rich parallels other such migrations in constitutional law. In the 19th century, the Fourteenth Amendment was framed as a guarantee of racial equality and political openness; within a generation, shrewd lawyers had transformed it into a charter of corporate privilege. The Civil Rights, antiwar, and women’s movements of the 1960s launched what legal scholars call a “rights revolution,” in which courts intervened against police actions, government-agency decisions, and even legislation that transgressed constitutional rights or acted to oppress or exclude minorities. Those victories attracted covetous eyes among the rich and powerful. If minorities and protesters could use the Constitution to block government, then why should the rich, and powerful corporations, not do the same?

In the area of free speech, the result is familiar. Under the Roberts Court, free speech is now less important as a tool of self-government and civic equality than as a protection for entrenched wealth. The wealthy may spend unlimited amounts to influence elections, and now may even contribute directly to an unlimited number of federal candidates.

Increasingly, these forces can also use First Amendment free speech as a tool to block government regulation. In a two-year period, the D.C. Circuit held that tobacco companies had a First Amendment right to refuse to put graphic warning labels on cigarette packs; that employers’ First Amendment rights blocked an NLRB rule requiring them to post notices about employees’ rights; and that manufacturers could refuse to disclose whether their products are made with “conflict minerals” mined by armed rebel groups in the Congo. (These cases have now been limited by the full court, but the anti-regulatory First Amendment argument is still attractive to other courts, and will almost certainly come before the U.S. Supreme Court soon.)

This is a natural consequence of a system in which access to courts requires deep pockets. Individuals fighting for their rights must often represent themselves, or settle for whatever free representation they can find. Powerful organizations and rich people, by contrast, hire legions of lawyers who search the cases day and night for their clients’ advantage. When a new area of law opens up, these faithful legal retainers head out for it, and stake unexpected and creative claims.

Thus, the business-friendly free speech argument soon, by gravitational force, began to create a business-friendly free exercise doctrine. The legislative history of RFRA makes virtually no reference to for-profit corporations. Nonetheless, two decades later the Supreme Court decided in Burwell v. Hobby Lobby Stores that RFRA protected even large commercial corporations against the Affordable Care Act’s requirement that employee health-insurance policies cover all medically safe methods of contraception.

Under the ACA and regulations implementing it, religious bodies (“houses of worship”) are entirely exempt from the contraceptive requirement.  Religious non-profits are not exempt: They are “accommodated.” They need not pay for contraceptive coverage themselves; they simply certify their objection, and the third-party insurer provides the coverage without charge to employee or employer.  For-profit companies, however, were expected to provide and pay for the employee coverage just like every other part of a medical policy.

The owners of Hobby Lobby Stores, a conservative religious family, wanted a veto over coverage of some methods of contraception. They were far from the only objectors, however. One who brought a similar religious claim, Michael Potter of Eden Foods, tipped his anti-regulatory hand when he told Salon, “I don’t care if the federal government is telling me to buy my employees Jack Daniel’s or birth control. What gives them the right to tell me that I have to do that?”

Unless libertarian philosophers F.A. Hayek and Robert Nozick have donned pontifical robes, that is a political belief, not a religious belief. It says, with a straight face, that government regulation offends my religion.

The Supreme Court, 5-4, upheld Hobby Lobby’s religious-freedom claims. Neither Hobby Lobby nor the Court majority paid even lip service to the conscience rights of the employees. The government argued that, whatever the Green family might wish for their employees’ sex lives, those employees had a countervailing right to decide for themselves which methods of contraception are moral.

Only the objectors’ rights counted, Scalia said during oral argument. “If they wanted you to balance the interest of the religious objector against the interest of other individuals,  they made no reference to that in RFRA at all.” Hobby Lobby’s counsel, Paul Clement, told the justices that “not all burdens are created equal”—burdens on employees are allowed, he explained; those on employers are not.

The Court agreed that Hobby Lobby was entitled to refuse the statutory coverage. The majority said that the government might try providing for-profit companies with the non-profit accommodation. The companies, however, did not commit to accepting that accommodation—and the non-profits themselves had already rejected it. A few days after the Hobby Lobby decision, in fact, the same Court majority entered an injunction blocking the religious non-profit accommodation. In a dissent for herself and the other two female Justices, Justice Sonia Sotomayor pointed out that only days before, “the very Members of the Court that now vote to grant injunctive relief concluded that the accommodation ‘constitutes an alternative that achieves all of the Government’s aims while providing greater respect for religious liberty.’” Her pen dipped in acid, she added, “Those who are bound by our decisions usually believe they can take us at our word. Not so today.”

One truly radical part of Hobby Lobby has largely escaped notice. Under the ACA, Hobby Lobby had a ready option that would not require it to be complicit with contraception in any way. It could simply drop its employer-provided coverage. The company would pay a tax penalty to the government that would amount, roughly, to the cost of maintaining its own plan. Employees could seek individual insurance on ACA-approved health exchanges.

But the plaintiffs rejected that option. The “companies have religious reasons for providing health-insurance coverage for their employees,” Alito wrote. The companies’ “religious beliefs govern their relations with their employees.”  In other words, the company’s religion took precedence over both the governmental scheme and the needs and beliefs of the employees. Government was unable to protect employees’ statutory rights; employees were to accept only such coverage as their employers chose for them; nothing more, nothing less.

The brings us to Zubik. Unlike Hobby Lobby, the challengers in Zubik are religious non-profits. Any fair assessment of free exercise issues would require a balancing of their religious rights against the government’s interest in ensuring access to a full range of health benefits for employees. And the government had already struck such a balance. Under regulations issued by the Department of Health and Human Services, religious corporations are required only to notify the government of their objection. The government then sends an order to the third-party insurer, instructing it to provide the benefits without cost to the employer. The Court in Hobby Lobby had written approvingly of this accommodation as “an alternative that achieves all of the Government’s aims while providing greater respect for religious liberty.”

But that’s not good enough, the challengers say. Even checking the form is oppression, because it will lead to contraceptive availability. Since churches are exempt, nonprofits must be entirely exempt: no certification, and no contraception for the employees. (Though this question is still in the pipeline, by the logic of Hobby Lobby, for-profit businesses must surely be entirely exempt as well.)

“[F]acilitation of sin,” the objectors’ brief argues, “is itself a sin.”

In the new world of “free exercise,” “sinners” have no religious rights. If female employees want contraceptive coverage, they must buy their own policies, at significant expense, on the ACA health exchanges. That cost is a burden the employer has a religious right to impose. It is, in effect, a kind of new feudalism, in which institutions can wall off their internal affairs from the government and, with the sanction of the state, establish their own religions inside the walls.

It’s a radical argument. Douglas Laycock, the religious-freedom scholar who helped draft RFRA, filed an amicus brief in Zubik on behalf of the Baptist Joint Committee for Religious Liberty. “I had never before filed a brief in support of the government in a case about the free exercise of religion,” he wrote in a recent Washington Post op-ed.

In the brief, he noted that the claim in Zubik is not for accommodation or even for exemption—it is for control over the actions of others. “Petitioners have been wholly exempted from providing contraception themselves,” he wrote. “Petitioners’ claim is not just that they should be relieved from providing contraception—a claim we would support without reservation. Petitioners claim that their secular insurers should be exempt as well. And that claim reaches too far.”

But the Court has received that radical claim with the utmost sincerity. During oral argument, the four conservative justices wondered aloud why the female employees really need coverage as part of their employee health plan. Why, they asked, can’t they get some separate policy, or pay for an entirely different plan on the ACA exchanges?

On March 29, the Court plunged deeper into a legal thicket of its own making. Apparently deadlocked over how to respond to the objectors’ claims, it issued an extraordinary order containing a new regulatory scheme that some justices apparently want to implement. In the new setup, the objecting non-profits would notify their insurers that they want insurance that does not cover contraception. The insurers would then offer contraceptive coverage to the employers that would somehow not be part of the employer’s plan but would still by a part of their coverage.

The idea is incoherent, and the Court’s authority to implement it is, to say the least, unclear. It is a deeper plunge into the quicksand the Court entered a quarter-century ago with Smith.

And there is no reason to believe that the new demands for “free exercise” will stop there. State governments themselves seem poised to demand religious rights against sexual minorities, and later, perhaps others. The argument seems frivolous today; but not long ago, the claims in Zubik would also have seemed outlandish.

The top has spun out of control, and no one can say where its strange career will carry it next.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.