The Voting Rights Act is dead. The law, very likely the most consequential civil rights statute Congress has ever passed, died on April 29, 2026. It was not a natural death. Congress did not repeal it, having concluded, for example, that it was no longer needed. On the contrary, Congress has reauthorized the statute four times since its initial passage in 1965, and has only expanded, not narrowed its scope. The perpetrator was instead the Supreme Court—or, more accurately, its six Republican-appointed members. In a deeply disingenuous decision, Justice Samuel Alito claimed only to be “updating” the statute. In fact the Court radically rewrote the law to eliminate the protections it has long secured for minority voters throughout the nation.

The decision in Louisiana v. Callais may be the Roberts Court’s most radical and far-reaching yet, rivaled only by the 2022 elimination of the right to abortion. It will almost certainly usher in a bleaching of the nation’s legislative bodies—federal, state, and local—unlike anything this country has seen since the “Redemption” of 1873–1877 ended Reconstruction and returned white supremacists to power throughout the South. In theory Congress can correct the Court’s egregious error, because it purports to merely interpret Congress’s statute. Indeed, Congress has done so before. If Democrats take Congress and the presidency in 2028, restoring the Voting Rights Act should be at the top of their agenda. But the very gerrymandering the decision invites will make getting there more difficult than ever, as Republicans in red states will in the meantime be free to cement their partisan advantage by disenfranchising black voters with impunity.

*

The Fifteenth Amendment to the Constitution, adopted shortly after the Civil War, forbids racial discrimination in voting. If the newly freed slaves were to be citizens, their votes had to be protected. But in the decades that followed, states—especially in the South—came up with a seemingly unending variety of “race-neutral” mechanisms to disenfranchise black voters, such as literacy tests, poll taxes, and property requirements. Even when black residents managed to overcome these obstacles and cast a vote, states strove to deny them any meaningful say in electoral outcomes by drawing districts that ensured they lacked a majority.

When Congress enacted the Voting Rights Act of 1965, it sought to remedy these forms of discrimination in several ways. Section 2 permitted individuals to sue states or local governments for practices that deny minority voters an equal opportunity to elect representatives of their choice, including districting that dilutes minority influence and rules that erect barriers to registration and voting. Sections 4 and 5 identified a handful of states and jurisdictions that had persistently and systematically denied black citizens access to the ballot, and required those jurisdictions to obtain affirmative federal approval before making any change to their electoral systems. The statute worked, and over time minority citizens managed to obtain some measure of representation—until the Roberts Court got involved.

In 2013, in Shelby County v. Holder, the Court struck down the remedies offered by Sections 4 and 5, ruling that singling out jurisdictions based on their voting records from the 1960s no longer made sense in the twenty-first century. Current remedies should demand proof of current discrimination, the justices reasoned. The Court reassured readers that Section 2 afforded remedies where current discrimination justified them.

Then, eight years later, in Brnovich v. Democratic National Committee (2021), the Court began unraveling Section 2. At issue was the provision’s application to practices—such as registration requirements and rules governing voting—that disproportionately impeded minority voters from accessing the ballot but could not be shown to be motivated by racist intent. In a 6–3 decision written by Justice Alito, the majority claimed only to be clarifying the statute, but in fact rendered it virtually impossible to challenge such discriminatory obstacles to voting. Since that decision, not a single lawsuit has successfully challenged a restriction on voting under the Voting Rights Act.

All that remained was Section 2’s prohibition on vote dilution, the practice of arranging voters so that their votes, even if cast, do not matter. Vote dilution can be very effective. Imagine a state in which 20 percent of voters are black and there are ten congressional seats. The legislature might draw a map that gave black voters a majority—or close to it—in two districts and white voters a majority in the other eight. That map would ensure that black voters could elect representatives of their choice in 20 percent of the districts. But the legislature might also divide the pie so that black voters make up 20 percent of the voters in every district—in which case, if people voted along racial lines, none of the candidates preferred by black voters would make it to Congress, and white voters would effectively choose 100 percent of the delegation. 

In 1980, in Mobile v. Bolden, the Supreme Court interpreted Section 2 to provide a remedy for such dilution only when the parties could prove that a legislature intentionally sought to dilute minority votes. That standard was almost impossible to meet, for two reasons. First, legislatures rarely announce their discriminatory intent. Second, states can easily come up with pretextual justifications for drawing almost any district line, because such lines can be based on many vague and largely subjective factors, from whether a district is compact to whether it protects incumbents or serves a “community of interest.” So two years later, after widespread criticism of the Bolden decision, Congress amended the Voting Rights Act to make clear that it forbids any districting practice that “results in” denying minority voters an equal opportunity to elect their favored candidates.

Recognizing that Congress had repudiated its prior interpretation, the Court in Thornburg v.Gingles (1986) acknowledged that the law does not require proof that a legislature intended to discriminate, but only that its map had the effect of doing so. In order to establish a violation, the Court explained, plaintiffs had to prove the existence of three objective “preconditions.” First, they needed to show that a minority group was sufficiently large and compact that the state could create a majority-minority district consistent with traditional districting principles. Second, plaintiffs had to show that members of the minority group tended to vote as a bloc, so that they would be able to elect a candidate of their choice if they comprised the majority in a given district. And finally, the plaintiffs were required to demonstrate that the majority also voted along racial lines, thereby frustrating minority voters’ ability to elect their preferred candidate. Once those preconditions were met, the lower courts were to consider the “totality of circumstances,” including other forms of past and ongoing discrimination in the state, to determine whether minority voters’ influence had been diluted. The overall focus, as Congress had made clear, was results—not intent.

There matters stood for forty years. Indeed, just three years ago, in Allen v. Milligan, the Court ruled that Alabama had violated Section 2 by diluting the state’s black vote. Black Alabamians, the plaintiffs showed, were sufficiently numerous to make up majorities in two districts, and Alabamians voted along racial lines—yet the state voting map had only one majority-black district. In its appeal to the Supreme Court, Alabama tried to reinterpret Section 2 to demand a showing of intentional discrimination. But Chief Justice John Roberts, Justice Brett Kavanaugh, and the three liberal justices rejected the state’s reasoning: as Roberts wrote for the majority, it was “patently clear” that because Section 2 liability “turns on the presence of discriminatory effects, not discriminatory intent,” a plaintiff need not demonstrate a “purpose of racial discrimination.” Justice Alito dissented. The standard for establishing a Section 2 violation, he argued, should be reinterpreted in light of subsequent developments to make plaintiffs’ task substantially more difficult. But only Justice Neil Gorsuch joined him. 

*

Louisiana v. Callais arose out of a challenge to Louisiana’s 2022 congressional map, which the state had redrawn after the 2020 census. That map created five districts in which white voters (largely Republican) were comfortably in the majority, and only one district where black voters (largely Democratic) had a majority. African Americans make up about one third of the state’s population, but under the 2022 map they were only able to elect one sixth of the state’s congressional delegation. (The one district in which black voters were a majority had itself been drawn in the 1990s precisely to avoid violating the Voting Rights Act.) In 2022 a federal court found that the plaintiffs were likely to succeed on their claim that the state was obliged to create a second black-majority district, and two years later, after a series of unsuccessful appeals, Louisiana adopted a new map that did just that.

No sooner had the state legislature passed that map, however, than Republicans challenged it as a “racial gerrymander,” arguing that it violated the Fourteenth Amendment, which the Supreme Court has interpreted to prohibit states from drawing districts predominantly on the basis of race. The state responded that race did not in fact “predominate” in the drawing of the new map, because the state sought principally to protect as many incumbents as it could while also complying with the district court’s ruling. In any event, it maintained, states are permitted to draw maps on the basis of race when seeking to remedy a likely violation of the Voting Rights Act.

The Court heard argument on the case last term but set the case for reargument this term, directing the parties to address whether remedying a Voting Rights Act violation permits racially motivated map-drawing. In prior cases the Court had repeatedly assumed that remedying a Voting Rights Act violation was a “compelling interest” that justified using race in constructing a new map, but it had never squarely decided the issue.

On Wednesday the Court handed down its decision. As a pure formality, it stated that a Voting Rights Act violation can indeed justify using race in drawing a remedial map. But that ruling was beside the point because, the Court concluded, the original 2022 decision finding that Louisiana had likely violated the Voting Rights Act by creating only a single black-majority district was itself wrong. The Court identified no error in the district court’s application of the longstanding Gingles test for Voting Rights Act violations. Instead, much as Alito had urged in his dissent three years earlier, the majority “updated” the test, and then found that plaintiffs in the original suit had failed to predict and satisfy this entirely new test.

It is no longer sufficient, the Court now insists, for plaintiffs to show that voters cast their ballots along racial lines and that minority voters are numerous and compact enough to constitute a majority-minority district. It is not enough, in other words, to prove that minority votes have indeed had their influence diluted. Now plaintiffs must show that it is possible to draw a map with an additional majority-minority district that fully satisfies all of a state’s districting goals, including protecting incumbents and maximizing partisan advantage. Justice Alito justified these new requirements by claiming that, without them, “plaintiffs would fail to demonstrate that the State’s chosen map was driven by racial considerations.” In other words, plaintiffs must prove not just dilutive effects, but racist intent—exactly what Congress rejected when it amended the statute to correct the same error in 1982.

No one will be able to satisfy this test. If Louisiana says, for example, that it wants to protect all its incumbents, any map that allows for the election of anyone new will be thrown out. If it says it wants to maximize Republican advantage, any alternative map that increases the ability of black residents to elect someone of their choice will fail to fulfill the state’s purpose, because black Louisianans tend to vote for Democrats. What’s more, Alito continued, to demonstrate racial bloc voting, plaintiffs must now show that it is race, not politics, that divides voters—an impossible task in practice, since black Americans overwhelmingly vote Democratic.

Putting aside the fact that the Court’s job is not to “update” statutes but to apply them, these are not mere “updates.” They effectively resurrect the intent requirement that Congress explicitly repudiated in 1982. In response to a withering and frankly unanswerable dissent by Justice Elena Kagan making exactly this point, Alito objected that his decision does not literally require proof of intent—just facts that “give rise to a strong inference that intentional discrimination occurred.” But in practice there is little if any daylight between those two concepts. And because the Court’s revised test allows legislatures to excuse even maps that deny black voters any opportunity to elect their preferred candidates if they say they’re advancing their partisan goals or protecting incumbents, successful claims will be a null set unless a legislature is so stupid as to announce, “No, we weren’t seeking to advantage our party, just to disenfranchise black voters for the sake of it.”

*

The decision effectively overturns the interpretation the Court has given the Voting Rights Act in nearly a dozen precedents over forty years, ever since the first time Congress rebuffed it for reading the law to require proof of intent. Of course, the Roberts Court has not been overly bothered by stare decisis, the obligation to follow precedent. But until now it has mostly broken from precedent on constitutional matters—as with its rulings on abortion and affirmative action. The Court has always been more open to reconsidering and reversing prior constitutional decisions, where, given the near impossibility of amending the Constitution, the Court is the only body that can correct errors. But stare decisis has always had much stronger force when it comes to interpreting statutes, because if the Court gets something wrong, Congress can fix it with simple legislation—and if Congress has not seen fit to change a statutory interpretation, there is no basis for the Court to do so. 

Alito himself has called statutory stare decisis a “nearly impregnable…shield.” (I guess his emphasis was on “nearly.”) Three years ago, when Kavanaugh voted with Roberts and the three liberal justices to reject Alabama’s invitation to rewrite Section 2, he invoked “stringent stare decisis” and noted that Congress had “not disturbed” the Court’s prior reading for “37 years…even as they have made other changes to the Voting Rights Act.”

So what drove the majority to ignore the rules they generally follow? Frustratingly, Roberts and Kavanaugh, whose votes made the difference, don’t offer any explanation for their about-face from the Alabama case just three years earlier. Alito’s claim that “important developments” since Gingles justify the rewrite, meanwhile, begs the question of why, if Congress doesn’t think new “developments” call for a rewrite, the Court gets to rewrite the statute itself. The developments Alito cites—namely the “great strides” black voters have made, the emergence of a “full-blown two-party system” in the South, the Court’s decision in Rucho v. Common Cause (2019) that federal courts will not review claims of partisan gerrymandering, and the increased use of computers in the redistricting process—do not change the meaning of the statute Congress enacted in 1965 and amended in 1982. These are excuses, not justifications.

The only plausible explanation for the conservative justices’ ruling is their own ideological opposition to tests that turn on discriminatory effect rather than intent. This is the same majority, after all, that invoked a “colorblind” vision of the Equal Protection Clause to invalidate affirmative action. From this perspective, an antidiscrimination statute that turns on disparate effects will inspire discomfort, because it will sometimes require government officials to take race into account in order to avoid creating a racially discriminatory effect.

In the view of the Republican appointees, it is evidently worse for legislators to consider race than it is to deny black voters any meaningful say in the electoral process. That is, needless to say, a debatable view. Would Congress be a better place with no black representatives from the South, even from states with substantial black populations? Apparently the six Republican justices think so. But that’s not their call; it’s Congress’s. Congress chose otherwise, and the Court’s duty is to respect that choice, not supplant it. 

Alito suggests at times that his reinterpretation is designed to conform the statute to the Constitution, but this is unpersuasive. The Court has long interpreted the Fifteenth Amendment to prohibit only intentional discrimination in voting, but it has also long held that Congress, in passing laws to enforce the Fifteenth Amendment, can go further and adopt prophylactic rules that prohibit race-neutral practices that have the effect of discriminating. For example, the Court has held that literacy tests do not violate the Fifteenth Amendment because they are race-neutral on their face; they apply to all would-be voters. But the Court has also upheld Congress’s ability to forbid all literacy tests by statute. Section 2 of the Voting Rights Act does the same thing, and the Court has never before questioned its constitutionality. Thus the Fifteenth Amendment provides no basis for departing from the long-standing statutory test.

When Callais is coupled with Rucho, the partisan gerrymandering decision, it unleashes legislatures to brazenly draw districts to maximize the influence of the majority party, even if doing so eliminates any meaningful vote for minority voters. Rucho green-lit the blatant partisan gerrymandering underway today in states like Texas, California, Florida, and Virginia. But until Wednesday there was at least some limit; states could not dilute the voting influence of sizable minorities.

Now that check is gone. Already Southern states are considering redrawing their maps to eliminate districts with black majorities before the 2026 midterms, and more will certainly do so before the 2028 presidential election. The United States, going forward, will become even more polarized than it is today. Legislators will pick their voters, rather than vice versa. And in the name of “interpreting” a law designed to safeguard the voting rights of African Americans, the Court will have put minority voters exactly where they were when the first “Redemption” brought an end to Reconstruction: relegated to the sidelines of democracy.