Erin Schaff / Reuters

Among the many strange and worrying truths about American elections, one has a tendency to get lost: The path to the presidency can run not just through battleground states but also through the Supreme Court.

Back in 2000, it was the Supreme Court in Bush v. Gore that put an end to a month-long post-election battle between the Democrat Al Gore and the Republican George W. Bush over who would be awarded the state of Florida’s Electoral College votes and, in turn, the presidency. This time, were the outcome of the 2020 election to fall to the Court, the situation could be far messier, and at stake would be the legitimacy of both the Court and the entire American electoral process.

The Supreme Court already plays an outsize role in policing the American election process, compared with the role played by high courts in other advanced democracies in setting rules for political competition. Despite the Court’s decision last year in Rucho v. Common Cause to keep federal courts out of the business of limiting partisan gerrymanders, the Court has been enmeshed in the political thicket for decades.

The U.S. Constitution, unlike newer constitutions from similar countries, does not contain many election ground rules. Instead, it has general language, such as in the Fourteenth Amendment, which guarantees “equal protection of the laws.” The Supreme Court and lower courts have used the constitutional text as a hook to decide some basic issues, such as whether election districts need to have roughly equal numbers of people to assure fairness in voting power, and whether the government can require people to disclose their identities when they spend money to influence how others vote in elections.

Aside from these general rulings, the Supreme Court and lower courts have become more and more involved in disputes over election procedures, balloting, and vote counting. According to statistics I compiled for my new book, Election Meltdown: Dirty Tricks, Distrust, and the Threat to American Democracy, the amount of election-related litigation keeps rising, and is now at nearly triple the rate of litigation in the period before the 2000 election. The 2018 election saw the largest number of election-related cases since at least 1996 (the first year for which I have been keeping records), which is all the more shocking given that litigation rates in midterm-election years tend to be lower than in presidential-election years. With Democrats bringing ever more lawsuits challenging restrictive voting practices put in place by Republican legislatures and elected officials, there’s every reason to believe that another record will be set in 2020, and that the most important of these cases will end up before the Supreme Court.

No one knows which cases will make it to the Court during the 2020 election season, but based on its generally conservative track record of late, it is a good bet that the Court will allow all but the most egregious efforts at voter suppression to go through. A recent example involved a law in North Dakota requiring voters to produce identification with a residence address on it, which uniquely burdened Native Americans living on reservations. In 2018, the Court refused to block this law, despite a total lack of evidence that the state had a good reason to impose it. That same year, the Court gave a green light to Ohio’s tough voter-purge practices. In addition, the Court’s conservative majority has been cutting back on protections of the Voting Rights Act for the last decade, most significantly when it killed off a key provision of the act in the 2013 Shelby County v. Holder case.

Most of these disputes over voting rules and elections, and Bush v. Gore itself, featured a Supreme Court divided 5–4 between the Court’s conservatives and liberals, with the conservatives coming out on top. When angry Democrats confronted the late Justice Antonin Scalia about the Court handing the 43rd presidency to Bush, Scalia told them to “get over it.” For the most part, people did get over it, with the Court’s legitimacy not taking a serious hit after the case.

But increased polarization and other changes since 2000 have altered the landscape, and it is not clear that things will go as smoothly for the Supreme Court or the nation if the Court ends up in the position of determining the outcome of a presidential election again.

To begin with, although the Court’s conservatives prevailed in Bush v. Gore, two of the “liberal” dissenters were Republican-appointed justices, John Paul Stevens and David Souter. Today, all of the Court’s conservatives (Samuel Alito, Neil Gorsuch, Brett Kavanaugh, John Roberts, and Clarence Thomas) were appointed by Republican presidents, and all of the Court’s liberals (Stephen Breyer, Ruth Bader Ginsburg, Elena Kagan, and Sonia Sotomayor) were appointed by Democratic presidents. People have begun thinking and talking about “Republican justices” and “Democratic justices,” and public opinion about the Court now seems to diverge along party lines.

All of this tension over the Court and its role in American democracy was heightened by Senate Majority Leader Mitch McConnell’s 2016 decision not to give a confirmation hearing to Judge Merrick Garland, whom Obama had nominated to succeed the late Justice Scalia. Instead, McConnell ran out the clock, denying Garland his chance for a vote to get on the high court and giving President Trump the opportunity to fill the seat with the Scalia-like Gorsuch. Although McConnell justified his delaying tactics by citing the “Biden rule” not to confirm a Supreme Court justice in the last year of a presidency, he has now reversed himself and says he would hold a confirmation hearing should a justice leave the Court in the run-up to the 2020 election.

In this context, there is uncertainty about whether the Court can play a unifying and finalizing electoral role as it did, at least partially, in 2000. In my new book I describe a number of scenarios that could lead to election meltdown and put the fate of the 2020 presidential election before the Court. For example, suppose that Russian hackers target the power grid on Election Day in a major Democratic city in a swing state (think Detroit or Milwaukee). The outage disrupts voting in that city, depressing Democratic turnout and causing the state to tilt to Trump in the Electoral College and pushing him to victory. Democrats might go to court to demand a revote or some other remedy, and the issue could come before the Supreme Court. How likely is it that Democrats accept a 5–4 Republican/Democrat split decision siding with Trump over his Democratic opponent and denying the chance for a revote?

One of the key features of a functioning democracy is that it is capable of conducting fair elections, with fair means for resolving election disputes. It is essential for such a democracy that losers accept election results as legitimate, vowing to fight harder the next time for political power.

With gaping polarization, not just in society but on the Supreme Court, it becomes difficult to imagine that liberals could simply “get over it” if they were once again on the losing side of a 5–4 decision choosing another Republican president following an election meltdown. Things could get very ugly very quickly.

This story is part of the project “The Battle for the Constitution,” in partnership with the National Constitution Center.

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