In the final weeks of the Supreme Court’s last term, the court’s conservative majority overruled two decades-old cases. The cases made headlines — not because their content was especially attention-grabbing, but because of what they may signal for the future. In both cases, liberal justices sounded the alarm on the threats they saw to other precedents. Justice Stephen Breyer even wrote in one dissent that he was left wondering “which cases the Court will overrule next.”

Breyer and the other liberal justices may soon get an answer on just how far the conservative justices are willing to go. The Supreme Court is back in session today, and it is already facing a docket full of high-profile cases, including several in which the justices have been asked to reexamine and potentially overrule past decisions. On Friday, for instance, the court announced it will review a restrictive Louisiana abortion law that’s nearly identical to a Texas law struck down by the justices in 2016. That case will be a particularly closely-watched test of whether the conservative justices are willing to start reversing liberal precedents on abortion rights. Generally, once a precedent is established, the justices say they’ll adhere to it, meaning that in cases involving hot topics like abortion or affirmative action, they’ll hand down a ruling that’s in line with what past justices have decided. But, of course, as we saw in the previous term, this isn’t always the case. Justices do sometimes overrule precedents or change them in substantial ways, and studies have shown that when that happens, their ideological preferences influence which precedents they affirm or override.

This appears to be true of the Roberts court as well, but according to a FiveThirtyEight analysis of Supreme Court decisions dating back to 1953, his court so far actually hasn’t been any more likely than previous courts to alter precedents. But there’s a key distinction: When the Roberts court does break with past rulings, it tends to do so by the slimmest of margins. Under Chief Justice John Roberts, the vast majority of precedent-altering cases have been decided by a narrow one-vote majority — and most of those cases have split along ideological lines. That’s a departure from previous courts, where there was typically greater consensus when overturning precedent. And it might also signal that Roberts — who has said that ideologically divided, 5-4 decisions undermine the court’s reputation — is actually more comfortable with divisive rulings than he lets on.

On the other hand, concern for the court’s reputation as an impartial institution might mean Roberts will move more cautiously than some of his conservative colleagues would like. “Roberts is in something of a bind,” said James Spriggs, a political scientist at Washington University in St. Louis. “There might be plenty of cases he wants to overrule. But as chief justice, he’d prefer to make big changes with a consensus, and that might lead him to vote in ways that are in tension with his ideological views.”

The Roberts court is already overturning precedents without consensus

In a system where unelected judges wield an enormous amount of power, the principle of adherence to precedent helps ensure that a single judge can’t simply decide he doesn’t like the rules and change them. And Supreme Court justices do so sparingly, perhaps in part to avoid the impression that they’re taking the law into their own hands. According to our analysis of precedent-altering decisions in the Supreme Court Database,1 they make up just about 2 percent of the total decisions handed down under Roberts, Rehnquist, Burger and Warren.

But even if these types of cases are rare, the Roberts court is handling them differently. As the table below shows, about 70 percent of the precedent-altering cases during Roberts’ tenure were decided by a thin one-vote margin, and only 14 percent were unanimous. That’s a big contrast with previous courts, where one-vote majorities only decided between 20 and 31 percent of precedent-altering cases.

One obvious culprit for this trend is the growing polarization dividing the court’s left and right flanks. For over a decade, Justice Anthony Kennedy was the court’s “swing” justice, which meant that even though he was still quite conservative, he would sometimes pivot and join the liberal justices on issues like gay rights. Nevertheless, more than half (62 percent) of the Roberts court’s precedent-altering decisions have gone in a conservative direction, as coded by the Supreme Court Database.2 And in rulings decided by one vote, the conservatives also usually came out on top. Five conservative justices (including Kennedy) were in the majority in two-thirds of these decisions; the liberal bloc, meanwhile, was only in the majority 20 percent of the time. Now with Kennedy retired, Roberts and Justice Brett Kavanaugh (Kennedy’s replacement) are at the center of the court, which means that liberal precedents may be in even more danger.

This pattern is troubling to some, in part because it signals that the current court is willing to make big changes without a consensus, which was rarer under previous chief justices. Leah Litman, a law professor at the University of Michigan, told me that she’s concerned that these types of decisions “could undermine the court’s legitimacy.” Chief Justice Earl Warren, for instance, famously negotiated with his more conservative colleagues to obtain a unanimous decision in Brown v. Board of Education because he thought it would give the decision greater authority. Ideologically divided decisions in high-profile cases run a higher risk of being seen as politically motivated — such as when the Roberts court ruled that the government can’t restrict political spending by corporations.

Some liberal precedents could be in trouble

Of course, none of this bodes well for more liberal precedents on issues like abortion or affirmative action, since there’s plenty of evidence to suggest that if justices do overturn a precedent, they’re more likely to dismantle one they disagree with ideologically. “Conservatives will generally uphold conservative precedents and overturn liberal precedents, and liberals will do the opposite,” said Jeffrey Segal, a political science professor at Stony Brook University and the co-author of a book on the Supreme Court and precedent.

But precedents aren’t overturned only because of the justices’ ideological leanings — other factors play a role, too. Sometimes there are philosophical differences, as in the case of Justice Clarence Thomas, who is notoriously skeptical of loyalty to precedent for precedent’s sake alone. In a concurring opinion a few months ago, he wrote, “When faced with a demonstrably erroneous precedent, my rule is simple: We should not follow it.” Thomas is not necessarily alone in this regard, although he is unusually open about his views. Other justices have also varied on how much they seem to lean on precedent. According to one study using data from the Burger and Rehnquist courts, Justice Ruth Bader Ginsburg was more likely to show deference to precedent than Kennedy. The stature of the precedent in question can matter, too. For instance, one study found that justices were more likely to overrule a precedent that they disagreed with if it was also still active and influential, as was the case when the conservative justices overturned a longstanding ruling involving public sector unions in 2018.

There may also be moments when a wing of the court feels more emboldened to overturn precedents — like when the court’s ideological center of gravity shifts. In 2006, for instance, Justice Sandra Day O’Connor, then the court’s “swing” justice, was replaced by Samuel Alito, a reliable conservative. The following term, the court issued five precedent-altering decisions — more than in any other single year since — including a case where the conservatives voted to allow a ban on a specific method of abortion for the first time. Ryan Black, a political science professor at Michigan State University, said it’s possible that this year could be similar. “It’s the same pattern — the median moved right, and the conservative wing of the court is stronger and more homogenous, so it’s not unreasonable to expect they might want to strike while the iron is hot,” he said.

But Roberts may emerge as a defender of precedent

There’s a possibility, though, that Roberts will emerge as an unlikely savior of liberal precedents — or at least work to postpone their demise. Given his pivotal role at the court’s center and as the chief justice, he’s likely to be more influenced by concerns about the court’s institutional reputation and his own judicial legacy at a moment when public opinion of the court is more polarized than at any other point in the past 20 years. And that could lead to some unexpected alliances with the liberal justices, in this term and beyond.

Take a case from last term, where the court had the opportunity to overturn a 1997 ruling that gives federal agencies the power to interpret their own regulations — a longstanding conservative bogeyman. Instead of voting with the conservatives to overturn it, Roberts cast his lot in with the liberals and instead voted to narrow the precedent’s scope. Litman told me that the liberals may similarly work with Roberts this term if they see an opening to avoid sweeping changes to precedents they support.

But Roberts could also join his fellow conservatives to start overturning more high-profile liberal precedents, and if he does, some experts say he risks jeopardizing the court’s status as a neutral arbiter. According to Spriggs’s research, people are less likely to think favorably of a narrowly decided, precedent-altering ruling than a unanimous one. “Consensus makes the ruling seem more neutral and less partisan,” he said.

Then again, more recent studies by Michael Salamone, a political science professor at Washington State University, have suggested that when the issue is controversial, the size of the majority might not matter very much. In his experiments, he found that if the issue was something high-profile, like abortion or gun control, the case’s margin didn’t matter — people’s views of the ruling were guided by their politics on the issue.

Roberts is an avowed skeptic of political science, so chances are he’s not looking to these studies for guidance. And regardless of how he decides to approach upholding precedent this term, it seems safe to say that at least some precedents are likely to fall by the wayside, given the court’s recent track record. But it’s also entirely possible that the Supreme Court’s long-awaited conservative revolution also may not be coming as quickly as some on the court’s right wing would like.