Jordan S. Ellenberg states a paradox: “The voters are supposed to choose their representatives, but in many cases, the representatives are choosing their voters.” (Ellenberg 350). He is talking about gerrymandering.
When partisan representatives control district maps, they are choosing their voters and controlling the election outcomes.
Gerrymandering skews the weight of one’s vote, more or less, so that it is no longer proportional– one person one vote– or, in other words, some votes are worth more than others.
The Virginia Supreme Court considered gerrymandering and released an opinion on 8 May 2026. The case, Scott v. McDougle, made national headlines.
The court held:
In this case, the Commonwealth submitted a proposed constitutional amendment to Virginia voters in an unprecedented manner that violated the intervening-election requirement in Article XII, Section 1 of the Constitution of Virginia. This violation irreparably undermines the integrity of the resulting referendum vote and renders it null and void. For this reason, the congressional district maps issued by this Court in 2021 pursuant to Article II, Section 6-A of the Constitution of Virginia remain the governing maps for the upcoming 2026 congressional elections. (Scott v. McDougle 30).
The case arose from an upcoming referendum in which Virginia voters approved an amendment to the Virginia Constitution, authorizing “the General Assembly to redraw congressional districts outside of the regular decennial-census redistricting as a response to other states that also redistrict outside of decennial-census redistricting or court-ordered redistricting.” (Scott v. McDougle 4).
People opposed to a referendum vote on the proposed amendment sued in Circuit Court (a trial court). The court ruled in their favor on 5 February 2026, and defendants appealed to the Virginia Supreme Court the same day. The referendum vote was on 21 April 2026. The Virginia Supreme Court invalidated the vote in its opinion of 8 May 2026. (Scott v. McDougle).
The Virginia Supreme Court did not decide the issue of whether gerrymandering were permissible; it decided whether the procedure allowing gerrymandering in this case were permissible.
The court was not asked to decide the substantive question: whether gerrymandering is legal, and, being a court, did not consider the extra-legal, philosophical question of whether gerrymandering should be allowed.
Gerrymandering is “[t]he practice of drawing district lines to secure advantage for yourself or your co-partisans.” (Ellenberg 347). Agreement on how to do it or how to regulate it is nonexistent. Although a court only rules on judiciable issues presented to it, Ellenberg is free to express his opinion that he disagrees with it:
We have tried to find a numerical standard for “fairness” and we have failed. The reason is that we’ve made a basic philosophical mistake. The opposite of gerrymandering isn’t proportional representation, or efficiency gap zero, or adherence to any particular numerical formula. The opposite of gerrymandering is no gerrymandering. (Ellenberg 386).
This language might sound familiar. In an unrelated legal opinion, United States Supreme Court Chief Justice John Roberts opined on another difficult practice to employ and regulate: “The way to stop discrimination on the basis of race is to stop discriminating based on race.” (Parents Involved 748).
When I read Ellenberg, who is the John D. MacArthur Professor of Mathematics at the University of Wisconsin, I reflect on my math education. Math was not taught properly. Ellenberg wrote Shape: The Hidden Geometry of Information, Biology, Strategy, Democracy, and Everything Else about practical, and interesting, applications of math. He says:
Talking to people about the dark art of gerrymandering is a form of teaching math, and math has intrinsic purchase on the human mind, especially when it’s twined up with other things we care deeply about: power, politics, and representation. (Ellenberg 420).
Had I known math would have enhanced my relationship with things I “care deeply about”, I would have paid more attention.
The partisan district map-makers create districts to their own party’s advantage, and “[t]he implication here is that the first duty of a political party is to protect its own interests against the whims of a potentially hostile electorate.” (Ellenberg 347). This misconception of their duty reminds me of the prosecutor who believes that his first duty is to convict rather than to seek the truth.
Ellenberg argues that district maps should be drawn by neutral parties, and “if you don’t know any neutral parties, you can just program a computer to act like one.” (Ellenberg 386). The goal is “to generate maps automatically, and in large number, by a mechanical process that has no preference between parties because we didn’t code it to have one.” (Ellenberg 386).
Consciously or not, Ellenberg is channeling legal and political philosopher John B. Rawls. Rawls believed that political decisions about structure and order should be made by neutral actors:
The original position, with the feature I have called the ‘veil of ignorance’, specifies this point of view. In the original position, the parties are not allowed to know the social positions or the particular comprehensive doctrines of the persons they represent. They also do not know persons’ race and ethnic group, sex, or various native endowments such as strength and intelligence, all within the normal range. We express these limits on information figuratively by saying the parties are behind a veil of ignorance. (Rawls 15).
Ellenberg argues that the goal of proportional representation is a “philosophical mistake”, and, accepting that map-drawing is unavoidable and unavoidably not proportional, rephrases the question we should ask about it: “Does this districting tend to produce maps similar to the ones a neutral party would have drawn?” (Ellenberg 386). Ellenberg further refines the question to ask about proposed district maps: “Does this districting tend to produce outcomes similar to a map randomly selected from the set of all legally permissible maps?” (Ellenberg 387).
Computers can create all the possible and permissible maps for a given state if provided the number of districts in which it must be divided and all the rules governing districting, and they can do it behind a veil of ignorance. From this ensemble of maps, Ellenberg argues we should eliminate the outliers (the statistical extremes), and randomly select a map from the rest. (Ellenberg 389).
The randomly selected map could be the new de jure district map, or, at least, it could be used to evaluate the fairness of the district map proposed by humans.
Another paradox is that people who cannot read maps, relying exclusively on GPS, are endeavoring to be cartographers, but a real cartographer does not draw the map how he wants to draw it.
I. Librarian
Works Cited:
Ellenberg, Jordan. Shape: The Hidden Geometry of Information, Biology, Strategy, Democracy, and Everything Else. Penguin Press, 2021.
Parents Involved in Community Schools v. Seattle School District No. 1. 551 U.S. 701. Supreme Court of the United States. 28 June 2007.
Rawls, John. Justice as Fairness: A Restatement. Edited by Erin Kelly, Harvard University Press, 2001.
Scott v. McDougle. Supreme Court of Virginia, 8 May 2026.
You Might Like:
Post Two – SWEET WAIST OF AMERICA
Post Five – FROM CARRIE FISHER
Post Six – REPHRASE THE QUESTION
Leave a Reply