Jonathan Ernst / Reuters

Midway through Monday’s oral argument in Janus v. American Federation of State, County, and Municipal Employees, Justice Sonia Sotomayor asked U.S. Solicitor General Noel Francisco, “[H]ow much is there unionization in the general corporate sector … or private sector?”

“I don’t have that number,” Francisco replied.

Francisco cited very few facts, in fact, even though he was asking the Court to reverse a 40-year-old precedent that allows public-employee unions to collect “agency fees” for the cost of representing non-member employees in collective bargaining. Along with William Messenger, staff attorney for the National Right to Work Legal Defense Foundation, he assured the justices that reversing that case—called Abood v. Detroit Board of Education—would cause no real problems for the states, their employees, or the unions those employees chose to represent them.

The record didn’t support that assurance, simply because … well, there is no record in this case. There is simply the claim, a longtime staple of conservative legal thinking, that Abood was wrong; there is the unspoken corollary that conservatives now at last have five votes and can get rid of it.

In legal terms, that’s a curious assertion. Courts claim to follow a principle called stare decisis, meaning that cases, once decided, are not to be overturned simply because new judges come on the Court, or new parties win elections, or newly tenured law professors think they were wrong; the radical step of voiding precedent is saved for cases that have been proven unworkable or unjust in the years since they were decided.

Brown v. Board of Education in 1954 reversed not one but three venerable Supreme Court decisions that spanned more than a decade—Plessy v. Ferguson (1896), Cumming v. Richmond County Board of Education (1898), and Berea College v. Kentucky (1908). Those cases had held that state laws requiring racial segregation were constitutional—as long as facilities for the races were “separate, but equal.”

The NAACP Legal Defense and Education Fund, founded in 1940, did not rush into court arguing they had always been wrong; instead, the LDF brought case after case demonstrating that facilities made “separate” by race could never, in any meaningful sense, be “equal.” By 1954, the facts the LDF cited convinced the justices that the segregation trilogy rule simply did not work; and thus, the Court unanimously overturned it.

Ordinarily, a case testing important constitutional questions would arise out of a trial of some sort, in which the two parties would present factual evidence—at least documents and affidavits, if not expert witness testimony—supporting their side. In Brown, for example, the NAACP plaintiffs called psychologists Kenneth and Mamie Clark to testify that their work with black children showed negative effects of segregation on their self-esteem (as measured, for example, by their choice of white dolls over dolls with African American features). The Clarks were cross-examined before a judge; the states defending segregation presented their own expert psychologist, Columbia University Professor Henry Garrett, to defend segregation. Brown’s record was rich in evidence about the nature of segregated schools and their effects on students. Confronting that evidence, the Court had a basis to conclude that the previous cases should be overturned.

There’s none of that in Janus.

Until 2010, all sides had regarded Abood as settled law. Then, in a public-employee-union case that did not present the issue, Justice Alito, writing the majority opinion, announced that Abood was entirely wrong  and should be overturned.

In 2013, a group of home-health workers brought a case called Harris v. Quinn challenging the agency fees. Defeat for public-employee unions was avoided, however, when the majority decided that home-health workers couldn’t be covered by union contracts at all. Abood survived.

Then the powerful anti-union advocacy network brought a case called Friedrichs v. California Teachers Association, challenging “agency fees” in teachers’ unions. They filed that case in district court, then immediately asked the court to dismiss their own case, so they could appeal it without a trial and thus get it before the seemingly receptive Supreme Court without delay. The union told the court, “The Unions have not moved for judgment on the pleadings, as they would prefer to ground the judgment in a factual record.” But the court decided not to assemble that record; it granted the challengers’ motion to lose without a trial.

That case reached the Supreme Court in the fall of 2015. After argument on November 1, the anti-union forces seemed to be on the verge of victory. Then, in February 2016, Justice Antonin Scalia died unexpectedly. Without his deciding vote, the Court was tied and affirmed the Ninth Circuit, 4 to 4. Abood, improbably, still lived.

But while Friedrichs was pending, Illinois Governor Bruce Rauner had asked a federal court to decide that he didn’t have to follow Illinois’s state-employee-union statutes any more, since they were probably going to be unconstitutional any day now. The federal court responded that  having to follow state laws isn’t an “injury” to a governor, so Rauner had no “standing”; in fact, there was no case.

At that point, Mark Janus, an Illinois state social worker who opposes agency fees, asked to join the case. The court allowed him to do so, even though, legally, there was no case for him to join—and then dismissed Janus’s claims because of Abood. He appealed; this week his case reached the Supreme Court, encumbered by no more facts than was Friedrichs.

In Monday’s argument, Francisco’s casual ignorance of labor statistics was not the only gap in the advocates’s knowledge. Messenger, the Right-to-Work fund lawyer, assured the justices that unions who represent non-members don’t incur any additional expense by doing so; he gave no source for this, simply his assertion that “there’s no reason why” it shouldn’t be so. Upending the labor law of 23 states, the District of Columbia, and Puerto Rico would be no big deal either, he said. “I submit the contracts will survive.” When Justice Elena Kagan asked him whether the contracts in those states had “severability” clauses (which would make the disruption slightly less), he admitted, “I couldn’t find a number for the public sector,” but added that he had “anecdotal” experience that many contracts do have such clauses.

Justice Stephen Breyer pointed out that California, in a brief supporting the union, contended precisely the opposite. Messenger dismissed that claim without really answering it, and pointed out that contracts eventually expire anyway.

David Franklin, who argued for the state of Illinois in defense of its “agency fee” law, told the Court that Messenger had been wrong to focus on the cost of contract negotiations. Unions also must represent workers in grievance proceedings, he noted, and that costs money: “We don’t know what percentage of the union’s activities are wrapped up with grievances.” In fact, he said, grievance-representation costs “can be three times, six times, seven times as much … [as] the line for collective bargaining. So to decide this case in an evidentiary vacuum on the basis of assumptions about how that speech breaks down or how those expenses break down would in our view be irresponsible.”

When David Frederick, representing the union, rose to argue, Kagan asked him about Messenger’s assurance that an anti-union victory wouldn’t disrupt labor relations. Frederick responded that, “Intangibly, there are plenty of studies that show that when unions are deprived of agency fees, they tend to become more militant, more confrontational, they go out in search of short-term gains that they can bring back to their members and say stick with us.”

Chief Justice John Roberts jumped in: “Well, the argument on the other side, of course, is that the need to attract voluntary payments will make the unions more efficient, more effective, more attractive to a broader group of their employees. What’s wrong with that?”

What’s wrong with that as a factual matter may be nothing; it may be everything. Without a record—without a good look at the studies Frederick referenced, and maybe the cross-examination of their authors—we simply have no idea what economic theory and history tell us about union behavior. But what’s wrong with this argument as a matter of law is that it’s being carried on in the dark, with no more grounding in facts than the average afternoon radio call-in show. A responsible course—as even Solicitor General Francisco seemed to admit—would be to remand the case for the creation of a factual record to supplant some of the airy theorizing the advocates (and the justices) engaged in.

But that would allow public-employee unions to carry on for another year or two. And that’s what’s really wrong with the whole Janus proceeding. The conservative justices don’t even try to hide it: The case is really about politics—about their feeling that public-employee unions are too powerful and that the policies they favor are hurting the country and they are all Democrats and they need to be stopped right away.

Justice Anthony Kennedy, who rarely hides his true thoughts, summed up what is really going on. When Franklin, for Illinois, suggested that states had an interest in negotiating with a “stable, responsible, independent counter-party,” Kennedy all but exploded:

It can be a partner with you in advocating for a greater size workforce, against privatization, against merit promotion, against—for teacher tenure, for higher wages, for massive government, for increasing bonded indebtedness, for increasing taxes? That’s—that’s the interest the state has?

A few minutes later, Kennedy asked Frederick, representing the union, the question that seems to be foremost on his mind. If the Court rules against the union, “the unions will have less political influence, yes or no?”

“Yes,” Frederick said, “they will have less influence.”

Kennedy replied: “Isn’t that the end of this case?”