Definition

Gerrymandering is the manipulation of legislative district lines to predetermine electoral outcomes. Representatives choose their voters rather than the other way around.

The term dates to 1812. Massachusetts Governor Elbridge Gerry signed a redistricting bill that produced a salamander-shaped state senate district. A Boston newspaper merged “Gerry” and “salamander.” The word stuck, and so did the practice.

Two mechanics dominate. Packing concentrates the opposing party’s voters into a small number of districts, wasting their votes on lopsided wins. Cracking splits them across many districts, so they fall short in each. Both produce delegations that bear no resemblance to the state’s underlying political composition.

The feedback loop that makes representative government work runs like this: voters choose representatives, representatives govern, voters reward or punish them at the next election. Gerrymandering severs one link. The incumbent’s party draws the map, selects the electorate, and converts a genuine accountability mechanism into a procedural ritual.

(For the prior question of who gets to vote at all, see the companion entries on Qualified Democracy and Off-Cycle Elections. Gerrymandering sits downstream of both. The first two filters determine who votes. This one determines how those votes translate to seats.)

The Current Moment: Louisiana v. Callais

On April 29, 2026, the Supreme Court ruled 6-3 in Louisiana v. Callais that Louisiana’s congressional map, which created a second majority-Black district, was an unconstitutional racial gerrymander under the Fourteenth Amendment.

Justice Kagan, in dissent, called the ruling’s practical effect “all but a dead letter” for Section 2 of the Voting Rights Act.

She is not wrong that the section is now a dead letter. She is wrong about whether that is a problem.

Section 2, as the regime had applied it for sixty years, required courts to draw majority-minority districts as an affirmative remedy. The statute does not say this. The courts read it in. The Callais majority is walking that judicial invention back and returning the text to what it actually says: a prohibition on drawing lines “to afford minority voters less opportunity because of their race.” That is a ban on intentional discrimination, not a mandate to engineer outcomes by race. Chief Justice Roberts put it plainly in 2007: “The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.” That is not a new position. The court is simply extending it to the map.

What Callais did not do: strike down Section 2 altogether. The access cases (voter ID, registration, polling-place hours) proceed under the same standards as before.

The downstream effect on the 2026 map was immediate. Louisiana suspended its May primary. Alabama and Tennessee called special sessions. Florida signed a new map adding four GOP-leaning seats. Mississippi and South Carolina are queuing up. Projections suggest a swing of one to nine House seats toward Republicans before the cycle is out.

That is the surface story. Here is the deeper one: the same problem recurs under different labels every decade. Strike down racial gerrymandering and you get partisan gerrymandering. Strike down partisan gerrymandering and you get incumbent-protection gerrymandering. The line-drawing power itself is the temptation. Whoever holds the pen at redistricting time gets to design the political battlefield they compete on. Remove one excuse and the incentive remains. The structural defect is not in how lines are drawn. It is that lines must be drawn at all.

Two Ideas Worth Taking Seriously

1. Jefferson’s Hundreds

In the last fifteen years of his life, Jefferson grew convinced that the Founding Fathers had missed something essential. The federal structure was sound, the state structure was sound, but power had not been pushed far enough down.

In letters to John Tyler (1810) and Samuel Kercheval (1816), Jefferson laid out a plan to divide every county into “wards” or “hundreds,” each small enough that every child could walk to a central school within it, each with its own justice of the peace, constable, and militia captain. He called these “elementary republics” and said the article was “nearest my heart.”

Paul Rosenberg has written about this, calling it Jefferson’s Hundreds, and arguing that Jefferson saw the wards as the only durable check on consolidation. The logic: citizens who hold real authority over the things nearest them remain habituated to self-government. Citizens who do not become subjects of distant administrators. Republican habits atrophy within a generation or two.

Jefferson believed by the end of his life that the experiment was being lost. He wrote in 1821 that “it is not by the consolidation, or concentration, of powers, but by their distribution, that good government is effected.” The wards never came.

This is where gerrymandering gets its teeth. When the federal House controls so much of daily life, every congressional district becomes existentially valuable. The line-drawing fight becomes total war because the prize is real. Strip the federal layer back to its original scope and the stakes of any congressional map collapse. The fight is not worth having if the office isn’t worth buying.

The affirmative case here is straightforward. Return authority to the ward level and the schoolteacher down the street has standing in decisions about her school. The firefighters’ union that mobilized 400 votes to put the county commissioner in office has a county commissioner who actually governs something they care about. The parish benevolence fund operates without a federal agency determining who qualifies. Time and attention now consumed by federal administrative capture go back to family, neighborhood, trade. Jefferson’s Hundreds is not nostalgia. It is the scale at which accountability is possible.

(The forthcoming companion entry on Decentralization / Jefferson’s Hundreds develops the ward-level argument at length.)

2. Use county lines

A more modest reform, scoped to the existing system: stop drawing congressional districts and apportion House seats to counties instead.

Counties already exist. Their boundaries are reasonably stable across decades. They correspond to functioning local governments and recognizable communities of interest. They were not designed by partisans for partisan ends in the way congressional districts are.

The mechanic: each county elects its allotted House seats at-large from within the county. A county entitled to three seats runs a single county-wide election where the top three vote-getters go to Washington. Voters evaluate all three representatives directly. No internal sub-districts. No lines to draw.

Apportionment of seats to counties uses the same method already applied to apportion House seats to states. Math is math. Run it once after each census. Counties below the threshold for a full seat cluster with neighbors by a published geographic rule, applied mechanically. Nothing for a legislature to negotiate.

What this kills:

• The decennial redistricting fight. Gone. Counties already exist, populations are counted, the formula is published.
• Packing and cracking. You cannot pack voters into a district that does not exist. You cannot crack a county across districts when the county is the district.
• Incumbent-protection mapping. An incumbent in a multi-member county faces every voter in the county, every cycle, alongside competitors from their own party. No safe seats carved by friendly cartographers.

What it changes that is worth naming:

• Single-member representation goes away in populous counties. Los Angeles County would send roughly 14 representatives, all elected county-wide. Some voters dislike not having “their” specific congressman. That is a fair objection. The counter is that the single-member district was always a fiction in safe seats anyway. A 14-member county delegation under ranked-choice voting gives proportional representation across the county’s actual political diversity, which is more honest than a 14-way packed-and-cracked map where the outcome is settled before anyone votes.
• Minority representation follows from proportionality, not cartography. Under multi-member ranked-choice, any cohesive group above the seat threshold elects a representative. A 20 percent Black population in a five-seat county elects one of five seats without anyone drawing a majority-minority line. Race never enters the map. Outcomes follow voter coalitions. This is the clean answer to Kagan’s “dead letter” concern: you do not need racial gerrymandering to get minority representation if your electoral mechanic is proportional rather than winner-take-all.
• Counties were also drawn by state legislatures, sometimes with political intent. Some in the South are themselves gerrymandered. So this is not a magic solution. Fixed boundaries that are merely imperfect beat fluid boundaries that are weaponized every ten years.

Why This Matters

Gerrymandering is a symptom of consolidation. The line-drawing fight exists because the federal seat is worth fighting for. Expand what the federal layer does and the cartographic arms race intensifies in proportion. This is not a partisan argument. Both parties gerrymander wherever they hold state legislatures. It is a structural argument: the temptation is inherent to the power.

Callais removes one input (race) from the line-drawing process. It does not remove the line-drawing process. The structural problem survives the ruling intact. Whoever holds the pen at redistricting time still gets to choose the next decade’s political reality.

The two reforms above point in the same direction: take the pen away. Jefferson’s Hundreds does it by shrinking what the federal layer does, so the pen matters less. County-based apportionment does it by replacing congressional districts with units that already exist and cannot be redrawn for advantage.

Either is more honest than the current arrangement, where every ten years the country watches state legislatures design their own electorates and then listens to the losing side explain why this particular gerrymander was uniquely illegitimate.

You already know what voluntary structures still work at the scale where accountability is real: the county commission, the school board, the parish council, the fire district. The question is whether the federal layer will leave enough oxygen for them to matter.

Further Reading

Paul Rosenberg, “Decentralizing the Failed American Republic,” LewRockwell.com, 2014. The Jefferson’s Hundreds framing carried in Sam’s adjacent register, grounded in the primary letters.

Hans-Hermann Hoppe, Democracy: The God That Failed (2001). The structural critique of why centralized democracies produce these distortions, and why the incentive to gerrymander is downstream of the scale problem, not the character of the participants.

For the broader framework of suffrage as the first filter and turnout as the second, see the companion essay on Suffrage vs. Turnout.

For the scale layer of the same argument, see the companion entry on Jefferson’s Hundreds.

For the upstream question of whether a polity has the substrate to sustain self-government in the first place, see the companion entry on The National Question.

« Back to Glossary Index