Which kind of judicial conservative will Justice Amy Coney Barrett be, at least when it comes to cases involving the right to vote? The integrity of the current presidential election — and, with it, the Supreme Court — might depend on the answer.

The court has on its docket cases from Pennsylvania and North Carolina that will force its conservative justices, including Barrett as its brand-new member, to make a fundamental choice between restraint and activism.

Historically, there have been two types of judicial conservativism on election issues. One is characterized by Justice Felix Frankfurter, a titan of 20th-century jurisprudence, who vehemently warned the court against entering the “political thicket” for fear of becoming too entangled in partisan electoral competition.

Judicial self-restraint is the guiding principle of this form of conservativism. When the Warren court was considering whether to enter the realm of legislative redistricting in the landmark 1962 case, Baker v. Carr, Frankfurter railed against the grievous mistake he saw his colleagues making.

Some constitutional questions are better left unanswered by the court, especially those concerning electoral politics, so that the judiciary can remain removed from partisanship and thus command bipartisan respect when it answers difficult constitutional questions for which its involvement is truly necessary.

“The Court’s authority — possessed of neither the purse nor the sword — ultimately rests on sustained public confidence in its moral sanction,” Frankfurter lectured. “Such feeling must be nourished by the Court’s complete detachment, in fact and in appearance, from political entanglements and by abstention from injecting itself into the clash of political forces in political settlements.”

The other kind of judicial conservatism on election issues is exemplified by the majority opinion written by Chief Justice John G. Roberts Jr. in Shelby County v. Holder, the case that eviscerated a key section of the 1965 Voting Rights Act. The opposite of judicial restraint, Shelby County aggressively imposed upon Congress the court’s conservative conception of a constitutional principle requiring equal treatment of all 50 states — unless Congress presented convincing proof of the need to treat states differently.

The Shelby County majority wasn’t necessarily wrong in recognizing this constitutional principle. The idea of all states being equals in the Union, even after the Civil War and Reconstruction, is a core premise of the Constitution, one that conservative jurists devoted to the “original intent” philosophy of constitutional interpretation would be inclined to emphasize.

But not all correct propositions of constitutional law must be fully enforced by the court in all circumstances — especially not when doing so drags the court into the quagmire of partisan conflict over electoral procedures. It’s easy to imagine Frankfurter chastising the Shelby County court for abandoning judicial self-restraint and succumbing instead to the temptation to substitute its constitutional vision for Congress’s political judgment on how best to enforce voting rights nationwide.

Roberts has been torn between these two types of conservativism. His authorship of Shelby County indicates his occasional alignment with the aggressive form. But he also wrote the court’s opinion in last year’s gerrymandering case, Rucho v. Common Cause, which took a very Frankfurter-like stance in keeping the judiciary out of hyperpartisan electoral warfare.

Roberts channeled Frankfurter again last week when he voted with the court’s liberals to stay out of a Pennsylvania case involving the deadline for absentee ballots — a position that prevailed only because of a 4-to-4 tie. That case is back at the court, along with one from North Carolina raising the same issue. Roberts needs one more vote this time, if the philosophy of judicial self-restraint is to prevail again.

What will the new justice do?

Let’s hope she follows the Frankfurter approach, at least for now. Yes, there is a major constitutional question embedded in the Pennsylvania and North Carolina cases concerning the power of state legislatures to control election rules without interference from other institutions of state government. The Constitution provides that “each state shall appoint, in such manner as the Legislature thereof may direct, a number of electors” to select the president — and likewise that a state’s “legislature” gets to set the rules for congressional elections, unless overridden by Congress.

A conservative jurist animated by an originalist philosophy of constitutional interpretation might be tempted to weigh in on the side of that specific-sounding text. Indeed, Roberts himself might have been tempted to do so, given what he has already written on the issue in a 2015 case from Arizona.

But in the context of the pending election, and especially in the current procedural posture of these cases with deadlines already extended and no more time to un-extend them, there is extra force to the Frankfurter position of keeping the Supreme Court out of the partisan combat.

Let ballots be cast and counted, at least this year, according to the status quo established by state governments, without five or more justices attempting to impose a new national standard at the last minute.

Those would, in all likelihood, be justices nominated by a Republican president helping out a Republican president. The court has blundered into this thicket before, in Bush v. Gore, to its lasting discredit.

Read more: