How to Police Gerrymanders? Some Judges Say the Courts Can’t.

WASHINGTON — Courts decide vexing legal matters and interpret opaque constitutional language all the time, from defining pornography and judging whether a search or seizure is unreasonable to determining how speedy a speedy trial must be.

And then there is the issue that some judges increasingly say is beyond their abilities to adjudicate. It was on display again last week, in North Carolina.

The North Carolina Supreme Court said that it could find no way to determine when even egregious gerrymanders — in this case, lopsided partisan maps of the state’s General Assembly and its 14 congressional districts — cross the line between skewed but legal and unconstitutionally rigged. In addition, the justices said, any court-ordered standard “would embroil the judiciary in every local election in every county, city and district across the state.”

The effect was to give the Republican-led legislature carte blanche to draw new maps for 2024 elections that lock in G.O.P. political dominance, even though the state’s electorate is split almost evenly between the two major parties.

Under its current court-ordered map, North Carolina now elects seven Democrats and seven Republicans to the U.S. House. Maps drawn by the Republican legislature could mean 10 Republicans to four Democrats, or possibly 11 to three. Without judicial review, the only remedy is to vote the dominant party out using maps drawn to keep them in power.

The 5-to-2 decision, which fell along party lines in a court led by Republicans, pointedly threw out a ruling by a Democrat-led court only a few months earlier that said such lines could — and should — be drawn. In that respect, the North Carolina ruling reinforced what seems to be a hardening partisan divide between jurists who believe unfair political maps should be policed and ones who do not.

The U.S. Supreme Court also split along partisan lines in 2019 when it ruled 5 to 4, after decades of dithering, that it could not devise a legal standard to regulate partisan gerrymandering, though it suggested that state courts could.

It is hard to separate party allegiance from jurists’ positions, said Paul M. Smith, the senior vice president of the Campaign Legal Center, a nonpartisan advocacy group that litigates voting rights issues.

“One explanation would be that the courts decide cases about elections based on who will be helped,” he said. “On some days, I’m cynical enough to believe that.” Whether that consciously figures in court decisions, though, is less easy to say, he added.

Nate Persily, a Stanford Law School professor and expert on election law and democracy, said that any standard for judging partisan gerrymanders has to be above reproach.

“The response is always going to be that you’re picking winners and losers,” he said. “Unless we come up with some sort of clear mathematical test, I respect the argument that judges’ political preferences might creep into the process.”

Passing judgment on a legislature’s constitutional authority to set political boundaries can be a fraught exercise. In 1962, one U.S. Supreme Court justice, Charles Evans Whittaker, who had heard the historic redistricting case Baker v. Carr, suffered a nervous breakdown during the court’s deliberations and skipped the final vote.

But some say that just because it is hard to create fair district lines does not mean it cannot be done.

“I think that’s intellectually dishonest and intellectually lazy,” Rebecca Szetela, a political independent and a member of the Michigan Independent Citizens Redistricting Commission, said in an interview. “We had a commission made up of 13 randomly selected voters of varying educational backgrounds, and somehow we were able to come up with fair standards.”

The Michigan commissioners drew their first set of maps after the 2020 election, following orders not to give any party a “disproportionate advantage.” They relied on several statistical metrics to meet that standard. But overall, they decided that an acceptable range for the statewide ratio of votes to seats won would fall within five percentage points of their calculation of the state’s political preferences: 52 percent Democratic, 48 percent Republican.

In practice, Ms Szetela said, the maps hewed closely to the calculated partisan divide.

Still, some experts say that it is impossible to construct a standard that will be reliably fair.

Daniel H. Lowenstein, an election-law expert at UCLA School of Law, said that would-be regulators of partisan gerrymanders by and large know little of how politics really works. He said that he picked up such an education during the 1970s while working in the California Secretary of State office, and later while running the state’s Fair Political Practices Commission.

“There’s nothing in the Constitution that says elections have to be fair,” he said, “and that’s a good thing, because different people all have different concepts of what it means to be fair.”

Peter H. Schuck, professor emeritus of law at Yale wrote a detailed analysis on the topic, “The Thickest Thicket,” in 1987. “I just don’t see any objective criteria that would be authoritative in assessing whether a gerrymander ought to be upheld or not,” he said.

A few other state courts have set standards for partisan gerrymandering and applied them. Pennsylvania was the first state to strike down partisan gerrymanders in 2018, and the Alaska Supreme Court upheld a lower-court decision last month stating that gerrymandered State Senate seats violated the State Constitution’s equal protection clause.

Many voting rights advocates say the same computer-driven advances that enable today’s extreme gerrymanders also make it possible to easily spot them.

In particular, software programs can now generate thousands and even millions of maps of hypothetical political districts, each with small variations in their borders. Using statistical measures, those maps can be compared to a map being contested to gauge their partisan slant.

In actual court cases, the technique has shown that some gerrymandered maps produce more lopsided partisan outcomes than 99 percent and more of the hypothetical ones.

Measures of partisanship have improved, as social scientists employed data analytics to tease out the partisan impact of map changes. One yardstick, called the efficiency gap, gauges how much the votes of one party are wasted when its voters are disproportionately packed into one district or carved up among several. Another, partisan bias, measures the effectiveness of a gerrymandered map by calculating how many seats the same map would give each party in a hypothetical election where voters were split 50-50.

There are many others, and each has its shortcomings. For example, voters sort themselves geographically, with a lopsided share of Democrats packed in cities and Republicans in rural areas, for reasons that have nothing to do with partisan skulduggery. And some metrics are useful only in particular situations, such as in states where party support is closely divided.

In a 2017 hearing in a Wisconsin partisan gerrymander case, Chief Justice John G. Roberts Jr. called such metrics “sociological gobbledygook.” But if so, much of American jurisprudence carries the same label, said Nicholas Stephanopoulos, a Harvard University law professor who has been a leading advocate of standards to judge partisan gerrymanders.

“In any voting rights case, people have to calculate racial polarization, which is a far more complex calculation than the efficiency gap,” he said. “You have to calculate the compactness of districts. You have to estimate voting patterns for minority voters and white voters.”

“Tests involving some matter of degree are just ubiquitous in constitutional law,” he added, and nothing makes a partisan gerrymander case any different.

Mr. Stephanopoulos and others also say that drawing a line between permissible and illegal political maps is not all that difficult. Courts make similar judgments in lawsuits claiming racial bias in redistricting, he noted. After the one-person, one-vote ruling in 1964, judges quickly set a limit — 10 percent — on how much political districts could deviate from the new requirement to have substantially equal populations.

Some gerrymandering yardsticks have already been suggested. For example, a political map might be assumed constitutional unless measures of partisanship uniformly argued against it. At that point, the body that drew the map would have to demonstrate another compelling reason for the way boundaries were drawn.

Critics like Professor Lowenstein argue that any dividing line between unfair and fair maps will have an unwanted consequence: Every subsequent map may be drawn to extract as much partisan gain as possible, yet fall just short of the legal standard for rejection.

“The ultimate question,” Professor Schuck said, “is how crude a fit should a court be willing to accept?”

Then again, he pointed out, the U.S. Supreme Court and the North Carolina Supreme Court have answered that question: Future political maps, they have ruled, can be as crude as their makers want them to be.

“Declining to apply a rule is still going to validate or invalidate what politicians have done,” he said. “There’s no total innocence, no virginity, as it were.”