Yesterday, the United States Supreme Court issued a ruling that further weakens the Voting Rights Act of 1965 — one of the most important civil rights laws in American history. Civil rights groups immediately called it a historic setback. Politicians spoke of a “slap in the face” to the heroes of the Civil Rights Movement. The case is Louisiana v. Callais, decided April 29, 2026 by the United States Supreme Court, with a 6-3 decision split perfectly along ideological lines.
If you are not American, you probably read that and felt a vague unease without being able to place exactly why it matters, or how a court ruling about a map in Louisiana connects to something as fundamental as civil rights. That’s because to understand what just happened, you need to understand gerrymandering — a US political practice that sounds technical and administrative but is, at its core, a mechanism for skewing the result of elections while keeping them real enough to call democratic.
Let’s build this up from first principles.
Before we get to the American situation, it’s worth acknowledging something that anyone who has voted in political elections understands intuitively: no electoral system is perfect. All of them introduce some distortion between what voters want and the government they actually get.
If your country uses a two-round presidential election — as France, Romania, Brazil, and many others do — you already know that the first round involves a degree of strategic voting. Your preferred candidate may have no realistic chance of winning, so you vote tactically to influence which two candidates make it to the runoff. Your sincere preference is not what you express on the ballot.
Similarly, if your parliament is elected proportionally, you know that seat allocation involves rounding. A party that wins 7.372% of the vote cannot hold fractional seats — the number of seats needs to be rounded, and that rounding, multiplied across many parties, produces a parliament that is an approximation of public opinion rather than a perfect mirror of it. Occasionally, this rounding interacts with minimum threshold rules in ways that feel slightly unfair.
These are real imperfections which produce real distortions, and political scientists study them seriously. But they share a crucial property: they are not systematically capturable by a specific actor for permanent advantage. The rounding error that slightly under-represents your party this election may slightly over-represent it next time. The strategic voting dynamics in a two-round election disadvantage different candidates in different elections. The distortions exist, but they are, broadly, noise — not signal engineered by and for the powerful.
A fair electoral system strives to minimize these unavoidable distortions. More importantly, it should be designed so that when the dust settles, the fundamental principle is respected: one person, one vote. Not just in the ritual sense (“everyone gets to walk into a polling station”) but in the substantive sense: every citizen’s vote should carry roughly equal weight in determining the composition of their government.
That is the baseline. Keep it in mind, because the American story we’re about to unfold is an excellent illustration of the Overton window in action: how far a system can gradually drift from that baseline while continuing to call itself a democracy.
The United States declared independence in 1776 and ratified its Constitution in 1788. The men who designed that Constitution faced a genuine political problem, one they resolved in a way that would poison the system for centuries.
The Southern states had large slave populations. Slaves could not vote — they were not considered people in any legal sense — but Southern politicians wanted them counted when calculating how many representatives each state got in Congress. More people meant more seats meant more power. The Northern states objected: if slaves cannot vote, why should they count toward representation?
The answer they agreed on was the Three-Fifths Compromise: each enslaved person would count as three-fifths of a free person for the purposes of apportionment. This gave Southern states significantly more congressional seats than their voting population alone would have warranted — political power derived from the bodies of people who were legally property. It was, from the very beginning, a system in which votes were not equal.
This is where the ball started rolling. The foundational architecture of American democracy was designed with an explicit carve-out from the principle of equal representation… because slaves. The Electoral College — the system by which Americans elect their President, where each state casts a block of electoral votes rather than contributing directly to a national popular vote — was partly shaped by the same compromise and the same political pressures. It is a separate mechanism from the congressional districting we are focused on here, operating at the federal level and with its own complex history, but it shares the same original sin: a design that weights some citizens’ votes more heavily than others, for reasons that have always been more about political power than democratic principle.
For elections to the House of Representatives — the lower chamber of Congress that makes federal law — the entire USA is divided into geographic districts, each of which elects one representative. The idea, in its honest original form, was reasonable: a large, diverse country needs local representation. A farmer in rural Kansas and a factory worker in Detroit have different concerns, and it makes sense for each to have a representative who is specifically accountable to their community, who knows their roads and their schools and their specific problems.
To achieve this, the country is divided into districts of roughly equal population. After every ten-year census, districts are redrawn to reflect population changes — people move, cities grow, rural areas shrink. Each district elects one member of the House of Representatives by a simple plurality: whoever gets the most votes wins, even if that is 40% in a three-way race.
Today, there are 435 districts, each containing roughly 750,000 people. Of course, the geographical size of districts is vastly unequal: the entirety of Alaska is one district, the same as a Manhattan district which covers a few city blocks — so they each get one representative in the House.
Everything so far feels perfectly reasonable, until you consider the practicality of it: somebody has to draw the district borders.
The word itself is over two hundred years old. In 1812, the governor of Massachusetts, Elbridge Gerry, signed a law that created a state senate district so grotesquely shaped — contorted into a long, winding form to capture specific towns and exclude others — that a newspaper cartoonist compared it to a salamander and called it a “Gerry-mander.” The governor’s name became the word for the practice.
The basic insight is simple and devastating: if you control where the district boundaries are drawn, you can largely predetermine who wins each district, regardless of how people vote.
There are two core techniques, and they work in combination:
A skilled gerrymander combines both: pack as many of your opponents as densely as you can in the lowest possible number of districts (hence seats), and distribute the remaining opponents into systematic minorities by cracking them everywhere else; this leads to a map where you win many districts by modest margins, while your opponent wins a few districts by huge margins. The result: you win 70% of the seats with 50% of the votes, while your opponent wins 30% of the seats with 50% of the votes.
Now, you might ask: how could this ever seem legitimate? The uncomfortable answer is that for a long time, in a political culture where parties were understood as legitimate actors with interests worth protecting, there was a broadly accepted — if never quite openly stated — norm that the party that won an election got to use the levers of power available to it. Drawing districts was a lever of power. Using it to your advantage felt, to those doing it, like ordinary politics rather than systematic disenfranchisement. Both parties did it when they had the chance. It’s rough, it’s self-serving, but it’s the game everyone is playing.
This was always wrong, measured against first principles. But it was normalized over so many election cycles that the wrongness became invisible to those inside the system.
For most of American history, gerrymandering was limited by a practical constraint: it was hard to do precisely. Politicians and their allies would draw maps based on rough knowledge of where their supporters lived, general demographic trends, and intuition. The manipulation was real but imprecise. You could tilt a map, but you couldn’t engineer it.
Two things changed this. First, detailed census data became increasingly granular, providing block-by-block population and demographic information that allowed map-drawers to understand exactly who lived where. Second, and more transformatively, computing power arrived.
By the 1990s, it was possible to sit at a computer, load a state’s demographic data, and run thousands of simulated district maps, evaluating each one for its likely partisan outcome. By the 2010s, this had become extraordinarily sophisticated — optimization algorithms that could generate maps guaranteed to produce a specific partisan result with high statistical confidence, maps that looked, to the untrained eye, like reasonable geographic divisions, but which were in fact precision instruments for predetermining election outcomes.
The 2010 census redistricting cycle — controlled in many key states by Republicans after a wave election — produced maps of a sophistication and deliberateness that represented a qualitative leap from previous gerrymandering. In North Carolina, Pennsylvania, Michigan, and other states, maps were drawn that virtually guaranteed Republican congressional majorities regardless of the statewide vote share. In some cases, a party could win 55% of the statewide vote and lose the majority of seats. The gap between votes cast and power obtained had never been wider, or more deliberately engineered.
While partisan gerrymandering — manipulating maps to benefit a party — has always been a feature of American politics, a specific and particularly ugly variant attracted the most sustained legal pushback: racial gerrymandering, the deliberate manipulation of district lines to dilute the voting power of racial minority groups.
This was not a subtle problem. After the Civil War theoretically extended rights to Black Americans, Southern states systematically used every tool available — including district manipulation — to prevent black citizens from translating their numbers into political power. Black communities were cracked across districts, packed into single seats, and otherwise arranged so that their votes would never threaten white political dominance. This continued, in various forms, for nearly a century after emancipation.
The Voting Rights Act of 1965 was the legislative response. Its Section 2 established a legal standard: if a racial minority group was large enough, geographically concentrated, and politically cohesive, the state could not draw districts that diluted their collective voting power. In practice, this meant states could be required to draw majority-minority districts — districts where a racial minority group formed the majority of voters, ensuring they could elect a representative of their choice.
This was an important but fundamentally limited way to patch the system. It addressed the most egregious racial manipulations, but it could not address partisan gerrymandering more broadly. What’s worse, it also carried within it a tension that ended up being weaponized against it: it required map-drawers to think explicitly about race when drawing districts, which would later be reframed as itself a form of racial discrimination.
To understand why the racial gerrymandering story became so politically charged, you need to understand one demographic fact about American politics: in the South, race and party affiliation are very highly correlated.
This was not always the case. For nearly a century after the Civil War, Black Americans voted predominantly Republican — the party of Lincoln, the party that had nominally freed the slaves. Southern whites voted predominantly Democrat — the party that had defended the Confederacy and later constructed the Jim Crow system of racial segregation.
This alignment flipped during the 1960s. The Democratic Party, under President Lyndon Johnson, pushed through the Civil Rights Act of 1964 and the Voting Rights Act of 1965. Johnson reportedly said, upon signing the Civil Rights Act, that he had just delivered the South to the Republican Party for a generation. He was right, and it was longer than a generation. The Republican Party pursued what became known as the “Southern Strategy” — a deliberate appeal to white racial resentment that gradually converted the South from solidly Democratic to solidly Republican.
The result, by the 1980s and accelerating through the 1990s and 2000s, was a near-perfect racial-partisan alignment in Southern states. Black voters vote Democratic at rates above 85-90%, and white voters in many Southern states vote Republican at comparable rates. Across much of the South, a map that packs black voters into as few districts as possible and cracks them everywhere else is simultaneously a map that packs Democratic voters and cracks them everywhere else.
In these states, race and party are so closely correlated that you can barely separate them statistically. And that overlap became the terrain on which a long, sophisticated legal battle was fought.
Here is where the story becomes particularly sharp, and where it helps to hold two things in your head simultaneously.
Over the decades following the VRA (the Voting Rights Act), Democrats used racial gerrymandering law as one of their few available tools to constrain Republican map manipulation in the South. The logic was genuine and legally supported: if you draw a map that systematically dilutes Black voting power, you are violating the VRA. Because black voters vote overwhelmingly Democratic, requiring majority-minority districts also tended to produce Democratic seats. The civil rights tool and the partisan tool happened to point in the same direction, and Democrats took full advantage of that.
Republicans developed a response to this that is, depending on your perspective, either a principled constitutional argument or an extraordinarily cynical maneuver — or, most accurately, both simultaneously. The argument went as follows: the Fourteenth Amendment’s Equal Protection Clause prohibits the government from making decisions based on race. When the VRA requires states to draw majority-minority districts, it forces them to sort citizens by race. That is itself racial discrimination — reverse discrimination, discrimination against white voters who are now excluded from a district on racial grounds. Race-consciousness, even remedial race-consciousness, violates constitutional equality.
The argument is not invented, there is a real constitutional tension here. But notice what it achieves in practice: it takes the one legal mechanism that reliably constrained racial gerrymandering in the South and reframes the constraint itself as the violation. The remedy becomes the crime; the protection becomes the discrimination.
The Louisiana v. Callais ruling handed down yesterday is the culmination of this argument. The state of Louisiana had been ordered by lower courts, under the VRA, to draw a second congressional district to remedy the dilution of Black voting power. It did so. But then a separate group of plaintiffs — identifying themselves as “non-African American” voters — sued, arguing the new map was an unconstitutional racial gerrymander. Yesterday, the Supreme Court agreed with them.
The majority opinion essentially holds that states can almost never use race explicitly when drawing districts, even to remedy the effects of racial discrimination, unless they can prove intentional racial discrimination — not just that the map produces racially discriminatory outcomes. A state attorney can label any map a “partisan” exercise rather than a racial one, and given that the Court has now ruled partisan gerrymandering is beyond judicial review, the practical effect is that even maps which obviously dilute minority voting power are unchallengeable, as long as the word “partisan” appears in the justification.
Now let’s step back for a minute and take stock of how we got here.
Was the VRA’s requirement for race-conscious remedies a blunt instrument? Clearly so. Did Democrats benefit politically from it? Certainly, measurably so. Is the Republican argument that race-consciousness in redistricting violates constitutional equality formally coherent? It certainly is, and independently so. And yet, the outcome — the practical, measurable outcome — is that the legal system has now removed almost all constraints on drawing maps that predictably eliminate minority political representation, provided those maps are labeled partisan rather than racial.
The net outcome is deeply Orwellian: we judge maps by label, not by their effect, in order to obtain racial injustice in the name of racial neutrality.
Now step back even further, and forget yesterday’s ruling altogether. Forget all the precedents and the legal doctrine and the 39 years of VRA redistricting jurisprudence. Forget the incremental steps that brought us here.
Start from first principles: one person, one vote. Every citizen’s vote should carry roughly equal weight in determining the composition of their government.
Now look at what the American system produces, in sequence.
District lines are drawn by the party in power in each state legislature. That party draws lines to maximize its own seats. Computing power makes this optimization near-perfect. The resulting maps predetermine congressional outcomes with high statistical confidence before a single vote is cast. The party that draws the maps wins more seats than its vote share would suggest. Those excess seats translate into more power in Congress. That congressional power influences presidential elections and Senate confirmations. Senate confirmations determine who sits on the Supreme Court. The Supreme Court rules on whether the maps are legal.
Yesterday, a Supreme Court composed of six justices appointed by Republican presidents and confirmed by Republican Senate majorities ruled, 6-3, that maps designed to eliminate minority (and coincidentally, Democratic) voting power are legally permissible. The three dissenting justices were appointed by Democratic presidents.
The American political game has gradually eroded the democratic fundamentals until it ended up with a structurally corrupt electoral system. No individual is corrupt in the ordinary sense. Every judge is ruling according to legal reasoning they genuinely hold. Every politician drawing maps is operating within the law as it has been defined. Every individual decision is formally defensible. No laws are broken, no bribes change hands. The system is functioning exactly as designed — or rather, exactly as it has been incrementally redesigned, over decades, by the people whose power depends on it functioning this way. The decision system itself is a victim of packing and cracking - each decision is locally defensible, and yet the systemic outcome is corrupt.
Maps produce the congressional and electoral majorities that determine who sits on the Supreme Court, which rules to protect the maps. This is not a malfunction, it’s a positive feedback loop built into the architecture.
If you compare today’s ruling to yesterday’s legal landscape, it looks like a step in an ongoing legal debate. If you compare it to the principle of one person, one vote, it looks like something else entirely: a self-reinforcing system in which the formal procedures of democracy are meticulously preserved while its substantive purpose — a government deriving its just powers from the consent of the governed — is converted, incrementally and lawfully, into a permanent advantage for those who already hold power.
That is what the news from yesterday actually means. Not a legal dispute about Louisiana’s congressional map. A data point in the long arc of a democratic system learning to perpetuate itself against the people it was built to serve. It’s climate change, not weather.