A constant question with this Supreme Court is what constraints, if any, the six Justices in the conservative supermajority see in their drive to upend constitutional law. On Thursday morning, Chief Justice John Roberts seems to have spotted his limit. In 2013, Roberts wrote the majority opinion in Shelby County v. Holder, which threw out Section 4 of the Voting Rights Act. Now, joined by the Court’s three liberals and—more tentatively and incompletely—by Justice Brett Kavanaugh, he has written the majority opinion in Allen v. Milligan, which narrowly preserved Section 2 (§2 in SCOTUS shorthand) of the Voting Rights Act. What’s more, Roberts did so in terms that suggested impatience with the maximalist demands that partisans on the right are placing on a Court they seem to feel they own. Allen v. Milligan arose from three challenges (eventually consolidated) to Alabama’s congressional-district map, which a lower court found was impermissibly racially gerrymandered. “The heart of these cases is not about the law as it exists,” Roberts wrote. “It is about Alabama’s attempt to remake our §2 jurisprudence anew.”
Alabama is not alone in attempting to remake the Court’s jurisprudence, and the Voting Rights Act is not the only target. Partisan litigants hit their mark last term in Dobbs v. Jackson Women’s Health Organization, which overturned Roe v. Wade, and in Bruen v. New York State Rifle & Pistol Association, which threw out or called into question gun-safety laws around the country. This year’s big decisions are expected to come in a steady volley over the next few weeks. Roberts’s vote in this case doesn’t necessarily mean that he will hesitate for a minute in finding that race-based affirmative action in higher education is unconstitutional, which is the issue in the tandem cases of Students for Fair Admissions v. Harvard and S.F.F.A. v. University of North Carolina. And it might not say much about his vote in certain other major outstanding cases, involving such matters as the adoption of Native American children, wedding-site designers who refuse same-sex couples as clients, and the Biden Administration’s student-loan forgiveness and border policies. However, it offers a slim, cautious hope that Roberts, and perhaps Kavanaugh, will not go to extremes in Moore v. Harper, a case involving North Carolina’s voting districts and what is known as the independent-state-legislature theory, which has the potential to destabilize the electoral system. (I wrote about Moore v. Harper in December, when the Court heard oral arguments, and my colleague Andrew Marantz has a piece on the case this week.)
If Roberts and Kavanaugh do step back from the brink in Moore v. Harper—and they may not—there’s a decent chance that the reasons they give will be similar to those offered in Allen v. Milligan. All the opinions in this case were written by conservatives: in addition to Roberts’s majority decision and Kavanaugh’s partial concurrence, there is a dissent by Justice Clarence Thomas, which Justice Neil Gorsuch joined and that Justices Amy Coney Barrett and Samuel Alito joined in part; and a separate dissent by Alito, which Gorsuch joined. The effect is to present a fight among those on the right about how far to go. Thomas, unsurprisingly, wanted to go furthest. He doesn’t believe that Section 2 of the V.R.A. should apply to congressional district maps at all.
The background of Allen v. Milligan is a redrawing of Alabama’s district lines after the 2020 census—a map referred to in the litigation as H.B. 1. (Wes Allen is Alabama’s secretary of state; when the case was argued, last October, it bore the name of his predecessor, John Merrill.) Alabama is twenty-seven-per-cent Black and currently has seven congressional seats. For a hundred and fifteen years after 1877, the state did not elect any Black members of Congress. That changed after litigation that resulted in the drawing, in 1992, of a single Black-majority district. After the 2021 redrawing, there was still only one Black-majority district; a number of groups and voters sued, alleging that the new maps perpetuated a racial gerrymander that “cracked” communities in order to keep Black voters from being fairly represented. An element of Alabama’s defense was that the new map simply gave weight to existing district lines. After extensive hearings, a federal district court in Alabama found that the map did, indeed, likely violate the Voting Rights Act. (Thomas derided those proceedings for including what he called “an impressionistic moral audit of Alabama’s racial past and present.”)
In Roberts’s view, the district court, which laid out its “careful factual findings” in a two-hundred-and-twenty-five-page opinion, did everything right. “The Court faithfully applied our precedents and correctly determined that, under existing law, HB1 violated §2,” he wrote. The key precedent was Thornburg v. Gingles, a 1986 case that offered a model for interpreting Section 2, as it had been amended by Congress in 1982. The law specifies that, while no group has the right to have its members “elected in numbers equal to their proportion in the population,” courts can look at the lack of such representation when determining whether voters have been denied equal access to political processes. In other words, having only one Black representative does not tell you everything you need to know, but it doesn’t necessarily tell you nothing, either.
Under Gingles, one factor that courts are asked to look at is whether there is an alternative district map that fairly represents minority voters without, as Justice Elena Kagan put it in the Milligan oral arguments, “doing something totally crazy.” Roberts found that the lower court had properly made that determination: it found that there were possible maps that yielded two majority-Black districts that also kept together counties and communities of interest (including the historic “Black Belt,” which was named, as Roberts notes, for its rich soil, but is also home to many Black Alabamians) at least as well as the disputed map. And the districts in those alternative maps were compact—no “tentacles, appendages, bizarre shapes, or any other obvious irregularities.” (“Bizarreness is in the eye of the beholder,” Thomas wrote in his dissent—a line that might be a motto for this Court—adding that he detected what looked like “tendrils” around the city of Mobile on the alternative maps.)
Roberts also emphasized that, in other cases that followed Gingles, the Court had thrown out redrawn maps that seemed to be racial gerrymanders themselves, rather than simply being fair. He quoted from Shaw v. Reno, a 1993 case involving a disputed majority-Black district in North Carolina that wound “in snakelike fashion through tobacco country, financial centers, and manufacturing areas until it gobble[d] in enough enclaves of black neighborhoods,” and from Bush v. Vera, a 1996 case from Texas in which a “proposed majority-Hispanic district” looked like “a sacred Mayan bird.” Gingles, in other words, was effective, and the federal court in Alabama had applied it as it was meant to be applied—and yet here was Alabama asking for both the precedent and, perhaps, for the law to be thrown out. Roberts pointedly listed a raft of decisions, shaping district lines in multiple states, that had relied on Gingles.
Roberts is still the Justice who wrote Shelby. At the very end of this opinion, he writes that the concerns about race being used as a basis for districting in any sense are not new, adding, “Our opinion today does not diminish or disregard these concerns. It simply holds that a faithful application of our precedents and a fair reading of the record before us do not bear them out here.” That closing provides something of a segue to Kavanaugh’s partial concurrence, which is largely based on his unwillingness to abruptly overturn Gingles. In Kavanaugh’s view, Congress could instead change the Voting Rights Act. But he declined to join a section of Roberts’s opinion that rejected Alabama’s argument about district mapmakers needing to be “blind” to race. “The authority to conduct race-based redistricting cannot extend indefinitely into the future,” Kavanaugh wrote. And so both he and Roberts expressed some willingness to come back to the question of race and districts. But not today.
Roberts’s opinion suggests that a reason for his forbearance is that the alternative to Gingles that Alabama offers is, to put it bluntly, somewhat wacky. It involves creating multiple computer simulations of what a “race-neutral” district map might look like, and then using such maps as a benchmark against which to judge allegedly discriminatory maps. The state claimed that it had run such simulations two million times without getting any maps that included two Black-majority districts. The problem is that the state doesn’t really explain how the inputs for such algorithm-based map-drawing should be determined. For most of its simulations, for example, the state ignored municipal lines, and used population data from 2010, not 2020. Nor did it explain how a computer program would be taught to understand and weigh factors such as community or history. (Alito, for his part, quite likes the simulations.)
The discussion about the simulated maps is an unexpectedly intriguing aspect of Milligan. It gives a glimpse of the questions the Court may have to grapple with as artificial-intelligence tools evolve. Could a chatbot that generates the false texts that are known as A.I. “hallucinations” make a statement about some aspect of the judicial or electoral process and convince people that its version of reality is correct? A fairness algorithm is hard to write.
Moore v. Harper also involves redistricting, this time in North Carolina, and has potentially sweeping implications. The essence of the independent-state-legislature theory is that those legislatures are almost unconstrained in determining how elections are run, to the extent that they may be able to shape their outcomes in defiance of voters. There is a chance that the Court will decide not to decide that case, as the result of legal developments in North Carolina; the Justices may also be reflecting on the fact that, in oral arguments last December, the lawyers representing Republican state legislators did little to hide how radical their arguments were. But then neither do the Court’s most radical Justices.
Even taking a case that pushes legal boundaries can have consequences. When the Court agreed to hear Milligan, last year, it stayed the district court’s ruling. That meant that the map the lower court had found to be discriminatory was the one Alabama used in the midterms. And what the Court does in the next few weeks, in Moore and other cases, could shape the election in 2024. ♦
