There has been a good deal of talk recently about gerrymandering, set off by the attempt by Texas Republicans to redistrict their state and met by threats from Democrats to respond in kind.
Gerrymandering no longer bothers me as much as it once did. I look at the example of Texas, once a solid democrat state, now more republican.
Gerrymandering is a short-term, short-sighted attempt to freeze the present into the future. It 'works' as long as the population doesn't change in composition or beliefs. But those are always changing. And the parties change as well. The Democrats of today look nothing like the party in my youth. The gerrymanders of yesterday designed to favor the Democrats in Texas failed.
I sort of agree. If every state gerrymandered to become a one-party state, people could still move and vote with their feet. This, of course, is more costly than having equal sized electoral districts that are not gerrymandered.
A wise man once said upon the fall of the Berlin Wall: "Now people can move. They have free choice. The politics to come is pure superstructure!" He wasn't a Marxist, either.
Fair but not everyone can move. Doesn't negate the overall point that it's a zero sum game long term but still, people overly pull out the voting with your feet card.
Not everyone has to move, just enough people to give an incentive to the states with bad policy to improve their game. It's positive sum, by the way -- the winners win more than the losers lose.
What must be avoided is the Berlin Wall, where nobody moves, because the attempt to move will cost you your life.
Voting with the feet in the United States is an everyday phenomenon. California, New York, Illinois losing population. This is non-trivial.
We already have the effective Berlin Wall for a subset of the US population, namely people on probation, parole, and people required to register was more my point. The numbers are small enough it can't swing an election right now but it's also easily expanded upon given legal precedence if states start to perceive a problem with people moving. Could easily expand it out for example to anyone who every got a speeding ticket for example, etc.
Not just people, but ideas. Individuals sometimes change their thinking, generations certainly do. Parties change. Democrats once worked very hard to support the interests of their rural farmer and industrial union voters. Now they seem to despise both those groups.
If an old gerrymander no longer brings together like-minded voters, it is ready for takeover by the opposing party.
Gerrymandering leads to extremism. Rather than parties being pulled towards the center, they are pulled towards the center of *their* party, which is quite a bit more extreme than the center among the whole set of voters. This is indeed short sighted for the existing incumbents. But for the party, its still a huge win. They still get to control things and cut out all other parties.
I wouldn't say that a monopoly of a party is ok just because it will change who gets elected. It changes who gets elected for the worse.
As you mentioned, due to territorial districts system and non-uniformity of voters, the vote share doesn't really give us much to work with. E.g. if in some state 2/3 of people vote R and 1/3 vote D, and the state has 3 seats, is it reasonable to expect one of the Congress seats go to D? If voters are distributed uniformly, it's more reasonable to expect 0 seats to go to D, because in each of the districts they'd have only 1/3 of vote and thus lose. However, as you noted votes are not uniform - e.g. large cities are more D-leaning, rural areas are more R-leaning, etc. That's what gerrymandering relies on, however, if we want to find a "true" picture, going by vote shares won't give us that. The question is which boundaries are "true"? If we just mechanically divide the area of the state into N equi-populated pieces, I have no idea how to model what would be the result, but I am pretty sure it won't usually match the vote shares.
Of course, this all relies to the system where we have territorial districts. If we just said "everybody votes for the party, the party gets N% and gets to nominate a number of representatives according to the number of votes they won" that would make a different system entirely, which does not allow for gerrymandering. This btw is how Israel elections work - each party publishes a list, which are the potential Knesset members from this party, then people vote for the party they like, and then each party gets to fill as many seats as the percentage of votes they achieved (after rounding and discounting the parties that didn't pass the minimum vote barrier). It would require a much deeper change than just redrawing the maps though. And while we keep the territorial representation, I am not a fan of gerrymandering at all, but I don't think vote shares can really measure its effects. For the territorial map, every map is a gerrymandered map, some of them are just more blatant than others, but I don't see how one map can objectively be claimed "better" than the other. If we want to get rid of gerrymandering (and I think we do) we need some better argument that would allow us to establish how one map is better than the other - or abandon the territorial system altogether.
It would be cool to have an algorithm to draw maps, based on what is desirable in districts - say the minimum length of lines to divide up a state into the correct number of districts with exactly equal populations.
If we wanted, we could still let dems and Republicans change it, but only if both sides agreed. If they can't come to an agreement, then the algorithm becomes the map
You have not provided evidence to support your claim that Newsom is in a poor position to complain about Republican gerrymandering. The main issue is not fairness, it is power.
Both parties are battling for control the U.S. House of Representatives after the 2026 midterm elections and redistricting is the most potent weapon. Everyone recognizes that, in spirit, gerrymandering is bad, but in Texas, at Trump’s behest, the Republican legislature is using this weapon aggressively to maximize their party’s congressional seats. By contrast, California has engaged in a form of unilateral disarmament, outsourcing its map-drawing to an independent commission. While this does not guarantee gerrymandering will not occur, it can mitigate the severity.
California voters have taken away the ability of their legislature to shape congressional districts, while Texas has no such constraint. As such, California, as a liberal state, is effectively fighting with one hand tied behind its back.
Therefore, Newsom’s complaints about Republican gerrymandering are not hypocritical; they are the rational grievance of a leader whose rivals are exploiting a power imbalance that California itself created. His desire to “fight fire with fire” by potentially asking voters to reclaim the legislature's power to draw districts is not a cynical reversal--it is a pragmatic recognition that in this national power struggle, you cannot bring a rulebook to a knife fight.
Your comment is informative and highly relevant. That said, is it correct to say that “California Democrats have voluntarily forfeited their ability to shape congressional districts”? It seems their ability to do so was removed by the people of California through a referendum which Democrats fought against and lost.
Thank you for pointing that out. California voters, and not Democrat politicians, removed the legislature’s redistricting power. I edited my comment to reflect that. If anything, this makes the point stronger: California operates under rules imposed by its voters, while the Texas legislature faces no such constraint. So Newsom’s complaints about Republican gerrymandering aren’t hypocritical because he is operating under voter-imposed limits that Texas doesn’t share.
I agree with your point. I had the impression, perhaps false, that Newsom was threatening to circumvent the independent commission in response to what Texas is doing as opposed, for example to what Illinois or Massachusetts has done. You're trying hard to be even handed in your comments which is all too rare these days. At the risk of being a (partisan?) nit picker I'll add one more thing. I wouldn't characterize what California has done by outsourcing map drawing as "a form of unilateral disarmament." That assumes the interests of the federal Democratic Party are identical to the interests of Californians. Californians have taken steps to insure redistricting is done in the interests of the population. Newsom should showcase this as an example to be emulated by Illinois, Texas, Massachusetts, and many other states.
While the interests of the federal Democratic Party are not identical to the interests of Californians, they are very closely aligned with the interests of California’s Democratic Party which holds a substantial majority in the state.
After the 2024 election, the total House seats closely tracked the national vote totals for the Republicans and Democrats. The Republican candidates received 49.8% of the votes, the Democrat candidates received 47.2% of the votes and the results were 50.6% of the House is Republican and 49.4% is Democrat. This appears to be coincidence--likely the result of the various levels of gerrymandering offsetting each other--and future elections will vary significantly. For the 2024 election, I find it difficult to assail the overall fairness of the distribution of Republicans and Democrats in the House.
I can also understand why Democrats, who control California’s politics, as Republicans control Texas’ politics, support a temporary suspension of the Independent Commission to counter new efforts by Republican-led states at mid-cycle redistricting and gerrymandering. I hope we can eventually have a consistent set of rules and principles that every state applies to reduce the effects of gerrymandering.
“ …support a temporary suspension of the Independent Commission”
Again you are trying to have it both ways. So your champion is better than Orange Man Bad or just as unprincipled as Orange Man Bad (worse if he is going to suspend the will of the voters and the CA ballot initiative process)?
Newsom cannot unilaterally suspend the Redistricting Commission. Any change would require voter approval. That means he is not “suspending” the will of the voters but instead asking the same voters who created the Commission to decide if a temporary change is justified. There is a fundamental difference between following the democratic process by asking voters for permission and a legislature acting on its own, as is happening in Texas and other states.
As a side note, I prefer to refer to President Trump as “President Trump” or simply “Trump.”
So if you are gonna make this argument about CA, what then do you say about MA?
Moreover, what do you say about EACH of their claims that they are going to gerrymander in advance of the 2026 midterms to aid the cause of their party?
You can’t have it both ways on each of “it’s already fair” and “these fair and honest folks are just fighting back against an unfair Orange Man Bad”.
In Massachusetts in the 2024 election, Democratic House of Representative candidates received 80.19% of the statewide vote and Republicans received only 10.23%. These numbers are skewed by Republicans fielding candidates in only two of the nine districts. As a result, Democrats were effectively guaranteed at least seven seats, and their eventual sweep is not, based on Professor Friedman’s data, evidence of gerrymandering. Massachusetts’ strong Democratic bent is sufficient to explain their prevailing in the two contested races.
Regarding Democratic-led states redistricting: if Republican-led states pursue off-cycle gerrymanders, no one should be surprised that Democratic-led states would consider following suit. That may not be ideal, but it is a rational response.
My hope is that this escalation of off-cycle redistricting and gerrymandering will provide the impetus for consistent nationwide rules that discourage gerrymandering.
“My hope is that this escalation of off-cycle redistricting and gerrymandering will provide the impetus for consistent nationwide rules that discourage gerrymandering”
Given federalism and states rights in this country, your “hope” is entirely unfounded.
Given that you’ve tried to use reasonable logic for all your other claims (even where I disagree with you, you’re not saying crazy stuff), this last line is especially discordant.
Federalism limits the federal government’s role in state affairs, but when the need is great enough, states often adopt uniform rules voluntarily, the Uniform Commercial Code being a prime example. My hope may be improbable, but it is hardly impossible.
By David's (loose) calculation, California has 13 more Democratic seats than is explained by the additional Democratic population. That is, 58% of the population is Democrat, which should result in about 30 seats. But Democrats control 43 seats instead. By far the largest discrepancy in total seats among any state.
The fact that an independent commission created this map doesn't change the fact that it's heavily gerrymandered. I don't know enough about the commission or state politics to say why or how the commission resulted in such a gerrymander, but it clearly did.
Now, you can defer blame away from Newsom and the legislature to some extent, at least using their public powers. If that's all you're saying you may be right - though I don't think we can rule out Newsom and other Democrats having back channel means of influencing the commission.
But I still feel it's fair of David to say that Newsom doesn't have room to talk here. If Texas put in an independent commission that resulted in significantly better maps for Republicans, do you think that Democrats would be calling it fair? Do you think that it would be fair if Republicans had an advantage in Texas as large as California's?
Unfortunately, Professor Friedman’s “loose” calculation is not particularly useful for drawing conclusions. It may have been misleading for him to show this because we don’t elect House representatives based on the portion of voters who voted for a party. In the 2024 election in California, 60.48% of the votes went to Democratic candidates for the House, while 39.23% of the vote went to Republican candidates. When more than 60% of the votes cast go towards candidates of one party, that party should win more than 60% of the races, without the aid of any political influence. How much more depends primarily on geographic distribution and district boundaries. If Republicans and Democrats were uniformly spread across the state, Democrats would have won every race without any gerrymandering. That was not the case, but the proportionality method is still misleading.
California’s 2020 maps may embed partisan advantage, but that cannot be determined from Professor Friedman’s table. As for fairness, few maps are perfectly fair, but the fact that the majority of Republicans on the Redistricting Commission approved the 2020 maps, and Republicans did not challenge the maps in court, indicates by some reasonable measures that they were fair enough.
We're both guessing at things, because there is not now and never was a "correct" way to determine districts and boundaries. Maybe proportionality should be considered, but as you correctly identify, if every district was uniformly distributed then we would want the majority party to win in each individual district even if that means the minority party gets zero seats.
But, this also works in reverse. Democrats complaining that Texas will be more red under new maps are trying to have their cake and eat it too. Maps that don't make room for Republican seats in California aren't more or less legitimate than maps in Texas that don't make seats for Democrats. When David says that Newsom doesn't have a leg to stand on in his complaints, it's a recognition that maps in his state directly disfavor Republicans more than necessary to complete the function. It would certainly be possible to make a map with more Republican-leaning districts, but they choose not to.
We are talking about Texas now because that's where the changes are, but in every state such decisions are made regularly and often disadvantage the minority party more than necessary. I don't care about complaints from Newsom or some other blue state governor about Texas when it's blindly self serving. I'm not upset at them for having districts the way they do, just for complaining about Texas trying to do the same thing they have already done.
Your argument assumes that Professor Friedman’s table proves California is presently gerrymandering more than Texas, but his table provides no such evidence--disproportionality alone, particularly in light of the substantial majority held by Democrats, is not proof of gerrymandering. That simplistic comparison ignores critical factors like the natural geographic clustering of voters and overlooks the fact that California’s maps were approved by a bipartisan commission and never legally challenged by Republicans.
>> By contrast, California has engaged in a form of unilateral disarmament, outsourcing its map-drawing to an independent commission. While this does not guarantee gerrymandering will not occur, it can mitigate the severity.
The ProPublica article you cited was about the 2010 maps, which were a mess. By 2020 the process had evolved. The Commission unanimously approved the maps, and there were no lawsuits. That shows the system is improving.
No one claims it’s perfect, but it’s clearly more transparent. And if Republicans are right that scrapping the Commission would hand Democrats five extra seats, then the Commission is limiting partisan gerrymandering, not enabling it.
I found the media coverage of the purported gerrymandering in Texas and the response by Newsome annoying because none of the reporters made any attempt to see whether the reallocation was (a) obviously complete gerrymandering, (b) completely reasonable to rebalance for population shifts; (c) somewhere in between. Thanks for doing this analysis for the country as whole.
Another way to avoid gerrymandering is make all representatives at-large. They would end up as state representatives rather than district representatives, but I don't know how much that really matters, other than them bragging about helping their constituents in their weekly / monthly newsletters I never read.
I'll have to think about that when I'm more awake.
First, I think you must mean Massie instead of Rand Paul, since Rand Paul is a Senator, not subject to gerrymandering, and already elected at large.
Second, I've always wondered how Massie got elected in the first place. Did he start out as a more traditional boring politician and only became a principled eccentric as he aged?
Third, I think you could be right in a state with only one or two representatives, where the mainstream candidates would get most of the votes. But Kentucky has six. That might be enough for the tail end to elect one or two eccentrics.
Fourth, or maybe electing eccentrics depends more on the ratio of urban to rural voters. California has 52 representatives, but I bet most come from Southern California and the Bay Area, and I don't have the brain power to think right now about what that does for eccentrics.
ETA: Fifth, San Francisco kept switching back and forth between at-large and districts for their supervisors for a few years, and the at-large campaign always touted better minority representation, which speaks for more eccentrics rather than fewer. Whether they actually got better minority representation, I do not know.
Oh heck no! I hate party list elections. No, this is an election where all candidates for all districts appear on one single candidates ballot, and every voter gets one vote, although they could all get any number to allocate as they wish.
The big names would get the most votes. Most voters in California (52 reps!) would only know a few, and without being by district, maybe almost all would come from big cities. Kentucky might work a lot better.
Maybe California could work a variation, with, say, 5 or 10 groups of districts voting for at-large reps from the group area.
Or groups could be randomized for every election. Or district voters could also vote on moving to a neighboring group.
> No, this is an election where all candidates for all districts appear on one single candidates ballot, and every voter gets one vote, although they could all get any number to allocate as they wish.
You need people to have at least as many votes as there are positions.
> The big names would get the most votes. Most voters in California (52 reps!) would only know a few, and without being by district, maybe almost all would come from big cities.
In practice this would collapse into a *de facto* party list system.
Any scheme that is not majoritarian opens Pandora's Box. Proportional representation's ability to produce stable and desirable government depends heavily on the underlying distribution of preferences in the population. If these are multi- peaked, we have the nightmare of coalition politics with the smallest party or parties wielding the greatest influence. The bigger parties acquiesce to keep the coalition alive. Think the Socialist Party in Italy, the Green Party in Germany, to say nothing of Weimar. Looking at American in the present, I am not sanguine.
The nutjobs must be disciplined inside a bigger party, non-democratically.
No, any scheme that IS majoritarian opens Pandora's Box. "Representation" is a shibboleth that elevates demoGRAPHIC non-essentials over competence and wisdom. How does ANY demographic (race, gender, age, etc.) promote the selection of officers that assure order, prosperity, and general social well-being? They do not. Only non-demographic SORTITION selects officers based on "the content of their character."
While I very much appreciate the idea of randomizing the membership of decision making bodies -- remember William F Buckley's lament that he'd rather be ruled by the first 3000 names in the Boston phone book than by the Harvard faculty -- if the underlying distribution of preferences in the population is multi-peaked, randomization will mirror that. And the smaller groups which are to make decisions will more or less reflect the whole population. Weimar in the small.
SORTITION solves the problem of GERRYMANDERING, as well as ALL OTHER ills of majoritarian elective politics. Specifically, it must be NON-STRATIFIED sortition.
I have ALREADY provided this solution in a previous post. You would do a great service to the former republic by promoting this, the ONLY solution to these problems.
I don't think there's any chance that the concept of "select a random person and give them all the representative power for about a million people" is going to be accepted by the people of the US. It's an interesting theoretical exercise but it's not going to happen.
This is NOT a "theoretical exercise." It's a common-sense solution.
Your objection would be valid if just any schmo could become an officer. The true point of your objection is the PROBLEM OF THE INITIAL POOL. The current Leftist use of sortition addresses that problem by instituting demographic strata, seeking "equal representation" in that initial pool based on race, gender, etc. – which consistently introduces social justice preconceptions. Such strata must be REJECTED.
How to populate that initial pool with only wise, competent candidates? Egalitarian fanaticism – today's atheist replacement for those enthusiasms that formerly were fenced in by dogma and creed – rises to a deafening howl on the suggestion that "wise" and "competent" can EVER be defined by ANYBODY.
The solution is to have EACH COUNTY define competence COMPLETELY ARBITRARILY in its selection of competent candidates – candidates which form the first pool of multi-round random selections. WHAAA? you say, HUH? That gonna git them nut job, Podunk fascists runnin' thangs in each county! Yes, a Mormon county might well insist on only Mormons in the initial pool; an agricultural county might insist on those holding a certain acreage. THIS element of arbitrariness is the ONLY VIABLE sense in which "representation" has any meaning. Whatever this small aspect of arbitrariness, no county is going to be so stupid as, say, exclude women or blacks; every county seeks order and prosperity. Any foolishness under this aspect will be dissolved in the later rounds of sortition.
> Whatever this small aspect of arbitrariness, no county is going to be so stupid as, say, exclude women or blacks
Why not? And why excluding non-Mormons is ok but excluding women is not? What about Jews - is excluding them ok or not? I'm pretty sure a lot of Harvard-educated people would support that if they could.
But beyond that, this sounds like proportional representation with extra steps. Which again relies on the concept of territorial "counties" - given permission to draw arbitrary counties, I could draw a county map of California which would ensure every county selects a woke leftist as their candidate. Then, completely randomly, one woke leftist out of set of 50 would be elected as California representative. But how it's better than what is there now and how it better represents 42% of California population that doesn't want woke leftists to represent them? It seems like we've got the same problem just with extra steps.
You would be right if someone could begin re-drawing customary and longstanding counties willy-nilly.
I accept that a county could abuse its arbitrary power to define competence so as to, say, exclude Jews – as was done implicitly in the past by requiring a religious oath for office. However, this county would soon find itself abandoned. We will see a demonstration of this principle if Mamdani is elected.
You are saying it like the idea of changing maps to fit one's political purpose is a completely novel invention and never before have been used. Remind me please what "gerrymandering" actually means? And most counties' maps are as arbitrary as electoral districts - somebody sometime (often not so long ago) just decided for some reason it will be so. There's no objective basis under it.
> However, this county would soon find itself abandoned
Just as Hamtramck, Michigan is abandoned now, or Oakland where people literally marched screaming "Death to America" is abandoned, or is Portland after years of Antifa riots. Or just like segregation ended because all people just abandoned the segregationist counties and refused to return until equal rights laws were instituted. This is really not how it works in real life, sorry.
> We will see a demonstration of this principle if Mamdani is elected.
We have already seen this movie though. New York before Giuliani was the fear city. It will be again (plus of course hating the Jews) if New Yorkers are stupid enough. But it won't be abandoned. Moreover, significant number of people would think the inevitable decline has nothing to do with their choices and blame the Jews for it (who probably won't leave either btw).
As I say, you would be right if county lines were changed routinely for political purposes, which is not the case.
Once the idea of SORTITION is embraced, the issues raised here become trifles. This idea has indeed become popular in the Northeast. There and elsewhere, people have become disgusted by the failures of majoritarian elective politics. However, despite this admirable motive, the recent implementations have insisted on DEMOGRAPHICALLY STRATIFIED sortition, which is a cure worse than the disease. This is so because the prevailing egalitarian fanaticism promotes the shibboleth of "representation," when it is a fictional social justice category.
No one suggests that "lines on a map" define competence.
Counties provide the initial pool because they have customary and longstanding recognition. County lines can well be perfectly arbitrary. Their only purpose is to in some way define communities whose various competencies – and totally arbitrary standard from any presumptive deontological state viewpoint – provide the initial random pool. Subsequent random choices assure that no one "arbitrary" definition of competence trumps the others.
* County lines ... only purpose is to in some way define communities whose various competencies
Counties don't do squat. They are lines on a map. If lines define competence, insist on anything, can be stupid, and have a purpose, it is because some humans act on behalf of those lines.
Once again, county lines are irrelevant. Obviously lines and counties don't make choices – the inhabitants of the county do.
These residents are not so stupid (as central state deontologists suppose) as to foul their own nest. If there is some knot-head county that insists that the initial pool be composed of transvestites or of women who are at least 1/1024 Cherokee, it will soon find its proportional contribution of candidates falls to ZERO, as its residents pack up and leave.
My idea was to have a redistricting proposal take the form of a computer program using specified inputs, probably population density, possibly town boundaries, with a maximum length of the program.
YES, they ARE missing something: Every one of these smarty-pants fixes LEAVES INTACT the root problem, which is majoritarian elective politics.
Non-stratified SORTITION resolves the gerrymandering problem, as well as ALL OTHER ills of majoritarian elective politics.
Also, "representation" is a shibboleth. Consider the following jackasses:
Rep. Sheila Jackson Lee (D-Texas) stunned attendees at a high school solar eclipse event on April 1, 2024, by claiming the rock-solid moon is a “planet” that is “made up mostly of gases.” Incredibly, she is a member of the House Science Committee’s space subcommittee.
Rep. Hank Johnson (D-Georgia) stated, “My fear is that the whole island [of Guam] will become so overly populated that it will tip over and capsize.” On July 25, 2016, in a public speech, Rep. Johnson compared Jews to “termites.”
These ignoramuses hold office strictly on the myth of "representation," that because of their black skin they best "represent" their districts.
A legislature chosen by sortition would instantly become a rubber stamp, as they’d have no ability to control the civil service (which is most of what congress does).
Politics is about getting people to do what you want, mostly by using other people you’ve already managed to get to do what you want. It’s a fairly difficult set of skills, and VERY SMART people with GOOD IDEAS, tend to be very very bad at it. As importantly, when it comes to getting bureaucrats (whose job is also mostly politics) to do what you want, the ability to make their lives better or worse is vital. Individual politicians can’t do that, you need coalitions (themselves built on threats, favours and patronage); long-term incumbency is thus very important, as is a seniority/party system. Choosing people at random every two years would turn them into a joke. They’d also get shaken down by helpful experts (really lobbyists and bureaucrats, but these are really experts and 90% of the time they all agree with each other), so would vote for whatever was put forward.
The point of elected politicians is that they extend the network of patronage into the ordinary population by giving everyone a favour they can return/offer/withhold in the form of a vote, so government has to take ordinary people into account to an extent (not so you can pick the person whose GOOD IDEAS you like because they’re COMMON SENSE). Sortition destroys that, but terrifyingly might legitimise the state to the point that it can care even less about public welfare than dictatorships have to.
Mr. jumpingjacksplash, you haven't read closely. Your objections are a celebration of the Leviathan state.
>>A sortitioned legislature would be a "rubber stamp" to civil service experts. "They’d [...] get shaken down by helpful experts".
No, non-stratified sortition guarantees COMPETENT candidates in the initial pool. They are self-sufficient, NOT beholden to experts.
>>"[Y]ou need coalitions [and] long-term incumbency".
Your recommendation is exactly the problem: Favor-swapping and immunity from removal is the essence of "politicking" in its worst sense.
>>"[E]lected politicians [...] extend the network of patronage into the ordinary population by giving everyone a favour they can return/offer/withhold in the form of a vote, so government has to take ordinary people into account to an extent".
Your recommendation is exactly the problem: Patronage and political spoils consolidates power and leads to beggar-thy-neighbor increasing debt – further aspects of "politicking" in its worst sense.
>>"Sortition [...] terrifyingly might legitimise the state".
False. Sortition annihilates politics and the state machinery that it supports.
I learned some things in this post and in the comments that are important to me. As someone who leans right and gravitates towards like minded commentary I thought that the redistricting villains were clearly on the Democratic side and Texas, out of necessity, was starting to fight fire with fire. Although I still feel the "outrage" expressed by Democrats is intellectually dishonest, David's table indicates that partisan redistricting is more evenly distributed than I had thought although Democrats are still slightly more guilty. California truly deserves bragging rights here provided Newsom doesn't undermine what its citizens have achieved. I was unaware of that until I read Omar's comment.
“Republican votes clear 30%, but are distributed so uniformly that they are locked out of the possibility of representation. Though there are more ways of building a valid districting plan than there are particles in the galaxy, every single one of them would produce a 9–0 Democratic delegation.”
Did you draw any Red congressional districts? Ones that reach the population threshold? By the way, the paper I referenced is several years old—perhaps the situation has changed. But I can personally confirm it was impossible then—this is easy to show by solving a small-ish linear program. The authors also proved it in a different way.
I did not — I don't have the data. But the article does not say that the impossibility applies in general, only that there have been past elections where it did. Note that:
"Meanwhile the Clinton/Trump election had a nearly identical 35.3% R vote share but produces the possibility for as many as two districts (built from towns) with a Trump majority."
So my conclusion was overstated. But the map by towns that I showed has much more clustering than the two they show, so I would be very surprised if one could not find continuous clusters of towns that would give a Republican majority district. I can't be sure without population density data — the red area could be all tiny towns.
There are lots of free districting software with web interfaces. Like playing with Microsoft paint. You can try your hand at making a red MA district here, for example: https://davesredistricting.org/maps#home
By the way, even if it is now mathematically possible to draw a red district, that does not mean it was likely to have been drawn by a neutral mapmaker. There’s a whole literature out there on randomly drawn plans (ideally sampled from a reasonable target distribution), and that is a better way to arrive at intent—if the enacted map was an outlier in the distribution. Here’s one such paper: https://www.pnas.org/doi/abs/10.1073/pnas.2217322120
I would be very interested in you, David Friedman, writing an essay that explores alternatives to representation that is not based on geography. I think the electoral process of selecting representatives for Congress in the USA is reprehensible. What do you think?
My favorite plan is true representation. Each voter chooses a representative. Each representative casts a number of votes equal to the number of people he represents. You can change your representative at any time.
I don't know if it would work better but it is more elegant.
I've had the same idea, but I wonder how fast that would turn into a dictatorship (one representative having a majority of the votes), which might lead to unpleasant consequences. One nice thing about democracies as we know them is that they are quite slow at making decisions, as they need to corral a majority of legislators to make things happen, which gives their subjects a chance to vote with their feet before it's too late.
I too wondered about one legislator having too many votes, and solved that by requiring passage by both votes and headcount. A sort of combined House and Senate all in one.
I solved this in my Chartertopia, ha ha. The reason districts have to be drawn and redrawn is because each legislator gets one vote in the legislature, so districts must be equally sized, population-wise.
Each district elects three representatives, who cast as many votes in the legislature as they won in the election. I call it proxying. One voter is selected at random to proxy all remaining votes; he's the amateur.
It does treat the count of voters as being a good proxy (!) for population. I ran some comparison a few years ago on turnout and population, and they did vary, but not by more than 10%. I no longer remember how I made the calculations or what data I used.
It eliminates the need for redrawing districts and for a census. It also encourages people to vote, since voting for a minority political viewpoint still allocates some votes outside the mainstream candidates, even if only to the amateur.
It reduces the incentive for fraud, since a few fake votes in a close election no longer spell the difference between winner gets all and loser gets none. At best, a candidate can be boosted from 4th no-show to 3rd, but that also gives more votes to the amateur, who now gets the true 3rd place votes instead of the true 4th place votes. That's a lot less gain for the same risk.
The kernel problem for sortition is the INITIAL POOL. How to keep "amateurs" and total ignoramuses OUT so that only competents can be selected? My book resolves that problem.
I don't know if you're right or wrong, but it's way more complicated than the analysis you did. In that map of MA, are there even enough voters in any of the two red county blocks to make a district? It's not at all surprising to me that a 2:1 voter ratio would lead to a sweep, even without gerrymandering.
What would be more interesting is a scatter plot of voter ratio vs seat ratio to look for outliers.
Gerrymandering no longer bothers me as much as it once did. I look at the example of Texas, once a solid democrat state, now more republican.
Gerrymandering is a short-term, short-sighted attempt to freeze the present into the future. It 'works' as long as the population doesn't change in composition or beliefs. But those are always changing. And the parties change as well. The Democrats of today look nothing like the party in my youth. The gerrymanders of yesterday designed to favor the Democrats in Texas failed.
I sort of agree. If every state gerrymandered to become a one-party state, people could still move and vote with their feet. This, of course, is more costly than having equal sized electoral districts that are not gerrymandered.
A wise man once said upon the fall of the Berlin Wall: "Now people can move. They have free choice. The politics to come is pure superstructure!" He wasn't a Marxist, either.
Fair but not everyone can move. Doesn't negate the overall point that it's a zero sum game long term but still, people overly pull out the voting with your feet card.
Not everyone has to move, just enough people to give an incentive to the states with bad policy to improve their game. It's positive sum, by the way -- the winners win more than the losers lose.
What must be avoided is the Berlin Wall, where nobody moves, because the attempt to move will cost you your life.
Voting with the feet in the United States is an everyday phenomenon. California, New York, Illinois losing population. This is non-trivial.
We already have the effective Berlin Wall for a subset of the US population, namely people on probation, parole, and people required to register was more my point. The numbers are small enough it can't swing an election right now but it's also easily expanded upon given legal precedence if states start to perceive a problem with people moving. Could easily expand it out for example to anyone who every got a speeding ticket for example, etc.
Not just people, but ideas. Individuals sometimes change their thinking, generations certainly do. Parties change. Democrats once worked very hard to support the interests of their rural farmer and industrial union voters. Now they seem to despise both those groups.
If an old gerrymander no longer brings together like-minded voters, it is ready for takeover by the opposing party.
Gerrymandering leads to extremism. Rather than parties being pulled towards the center, they are pulled towards the center of *their* party, which is quite a bit more extreme than the center among the whole set of voters. This is indeed short sighted for the existing incumbents. But for the party, its still a huge win. They still get to control things and cut out all other parties.
I wouldn't say that a monopoly of a party is ok just because it will change who gets elected. It changes who gets elected for the worse.
An important insight, I think.
As you mentioned, due to territorial districts system and non-uniformity of voters, the vote share doesn't really give us much to work with. E.g. if in some state 2/3 of people vote R and 1/3 vote D, and the state has 3 seats, is it reasonable to expect one of the Congress seats go to D? If voters are distributed uniformly, it's more reasonable to expect 0 seats to go to D, because in each of the districts they'd have only 1/3 of vote and thus lose. However, as you noted votes are not uniform - e.g. large cities are more D-leaning, rural areas are more R-leaning, etc. That's what gerrymandering relies on, however, if we want to find a "true" picture, going by vote shares won't give us that. The question is which boundaries are "true"? If we just mechanically divide the area of the state into N equi-populated pieces, I have no idea how to model what would be the result, but I am pretty sure it won't usually match the vote shares.
Of course, this all relies to the system where we have territorial districts. If we just said "everybody votes for the party, the party gets N% and gets to nominate a number of representatives according to the number of votes they won" that would make a different system entirely, which does not allow for gerrymandering. This btw is how Israel elections work - each party publishes a list, which are the potential Knesset members from this party, then people vote for the party they like, and then each party gets to fill as many seats as the percentage of votes they achieved (after rounding and discounting the parties that didn't pass the minimum vote barrier). It would require a much deeper change than just redrawing the maps though. And while we keep the territorial representation, I am not a fan of gerrymandering at all, but I don't think vote shares can really measure its effects. For the territorial map, every map is a gerrymandered map, some of them are just more blatant than others, but I don't see how one map can objectively be claimed "better" than the other. If we want to get rid of gerrymandering (and I think we do) we need some better argument that would allow us to establish how one map is better than the other - or abandon the territorial system altogether.
It would be cool to have an algorithm to draw maps, based on what is desirable in districts - say the minimum length of lines to divide up a state into the correct number of districts with exactly equal populations.
If we wanted, we could still let dems and Republicans change it, but only if both sides agreed. If they can't come to an agreement, then the algorithm becomes the map
You have not provided evidence to support your claim that Newsom is in a poor position to complain about Republican gerrymandering. The main issue is not fairness, it is power.
Both parties are battling for control the U.S. House of Representatives after the 2026 midterm elections and redistricting is the most potent weapon. Everyone recognizes that, in spirit, gerrymandering is bad, but in Texas, at Trump’s behest, the Republican legislature is using this weapon aggressively to maximize their party’s congressional seats. By contrast, California has engaged in a form of unilateral disarmament, outsourcing its map-drawing to an independent commission. While this does not guarantee gerrymandering will not occur, it can mitigate the severity.
California voters have taken away the ability of their legislature to shape congressional districts, while Texas has no such constraint. As such, California, as a liberal state, is effectively fighting with one hand tied behind its back.
Therefore, Newsom’s complaints about Republican gerrymandering are not hypocritical; they are the rational grievance of a leader whose rivals are exploiting a power imbalance that California itself created. His desire to “fight fire with fire” by potentially asking voters to reclaim the legislature's power to draw districts is not a cynical reversal--it is a pragmatic recognition that in this national power struggle, you cannot bring a rulebook to a knife fight.
Your comment is informative and highly relevant. That said, is it correct to say that “California Democrats have voluntarily forfeited their ability to shape congressional districts”? It seems their ability to do so was removed by the people of California through a referendum which Democrats fought against and lost.
Thank you for pointing that out. California voters, and not Democrat politicians, removed the legislature’s redistricting power. I edited my comment to reflect that. If anything, this makes the point stronger: California operates under rules imposed by its voters, while the Texas legislature faces no such constraint. So Newsom’s complaints about Republican gerrymandering aren’t hypocritical because he is operating under voter-imposed limits that Texas doesn’t share.
I agree with your point. I had the impression, perhaps false, that Newsom was threatening to circumvent the independent commission in response to what Texas is doing as opposed, for example to what Illinois or Massachusetts has done. You're trying hard to be even handed in your comments which is all too rare these days. At the risk of being a (partisan?) nit picker I'll add one more thing. I wouldn't characterize what California has done by outsourcing map drawing as "a form of unilateral disarmament." That assumes the interests of the federal Democratic Party are identical to the interests of Californians. Californians have taken steps to insure redistricting is done in the interests of the population. Newsom should showcase this as an example to be emulated by Illinois, Texas, Massachusetts, and many other states.
While the interests of the federal Democratic Party are not identical to the interests of Californians, they are very closely aligned with the interests of California’s Democratic Party which holds a substantial majority in the state.
After the 2024 election, the total House seats closely tracked the national vote totals for the Republicans and Democrats. The Republican candidates received 49.8% of the votes, the Democrat candidates received 47.2% of the votes and the results were 50.6% of the House is Republican and 49.4% is Democrat. This appears to be coincidence--likely the result of the various levels of gerrymandering offsetting each other--and future elections will vary significantly. For the 2024 election, I find it difficult to assail the overall fairness of the distribution of Republicans and Democrats in the House.
I can also understand why Democrats, who control California’s politics, as Republicans control Texas’ politics, support a temporary suspension of the Independent Commission to counter new efforts by Republican-led states at mid-cycle redistricting and gerrymandering. I hope we can eventually have a consistent set of rules and principles that every state applies to reduce the effects of gerrymandering.
“ …support a temporary suspension of the Independent Commission”
Again you are trying to have it both ways. So your champion is better than Orange Man Bad or just as unprincipled as Orange Man Bad (worse if he is going to suspend the will of the voters and the CA ballot initiative process)?
Newsom cannot unilaterally suspend the Redistricting Commission. Any change would require voter approval. That means he is not “suspending” the will of the voters but instead asking the same voters who created the Commission to decide if a temporary change is justified. There is a fundamental difference between following the democratic process by asking voters for permission and a legislature acting on its own, as is happening in Texas and other states.
As a side note, I prefer to refer to President Trump as “President Trump” or simply “Trump.”
He IS trying to circumvent the commission, surely? If the commission controls the districting, how can it be otherwise? Am I missing something?
So if you are gonna make this argument about CA, what then do you say about MA?
Moreover, what do you say about EACH of their claims that they are going to gerrymander in advance of the 2026 midterms to aid the cause of their party?
You can’t have it both ways on each of “it’s already fair” and “these fair and honest folks are just fighting back against an unfair Orange Man Bad”.
In Massachusetts in the 2024 election, Democratic House of Representative candidates received 80.19% of the statewide vote and Republicans received only 10.23%. These numbers are skewed by Republicans fielding candidates in only two of the nine districts. As a result, Democrats were effectively guaranteed at least seven seats, and their eventual sweep is not, based on Professor Friedman’s data, evidence of gerrymandering. Massachusetts’ strong Democratic bent is sufficient to explain their prevailing in the two contested races.
Regarding Democratic-led states redistricting: if Republican-led states pursue off-cycle gerrymanders, no one should be surprised that Democratic-led states would consider following suit. That may not be ideal, but it is a rational response.
My hope is that this escalation of off-cycle redistricting and gerrymandering will provide the impetus for consistent nationwide rules that discourage gerrymandering.
“My hope is that this escalation of off-cycle redistricting and gerrymandering will provide the impetus for consistent nationwide rules that discourage gerrymandering”
Given federalism and states rights in this country, your “hope” is entirely unfounded.
Given that you’ve tried to use reasonable logic for all your other claims (even where I disagree with you, you’re not saying crazy stuff), this last line is especially discordant.
Federalism limits the federal government’s role in state affairs, but when the need is great enough, states often adopt uniform rules voluntarily, the Uniform Commercial Code being a prime example. My hope may be improbable, but it is hardly impossible.
“These numbers are skewed by Republicans fielding candidates in only two of the nine districts.”
So your numbers are entirely meaningless, and your 7 seat logic circular.
By David's (loose) calculation, California has 13 more Democratic seats than is explained by the additional Democratic population. That is, 58% of the population is Democrat, which should result in about 30 seats. But Democrats control 43 seats instead. By far the largest discrepancy in total seats among any state.
The fact that an independent commission created this map doesn't change the fact that it's heavily gerrymandered. I don't know enough about the commission or state politics to say why or how the commission resulted in such a gerrymander, but it clearly did.
Now, you can defer blame away from Newsom and the legislature to some extent, at least using their public powers. If that's all you're saying you may be right - though I don't think we can rule out Newsom and other Democrats having back channel means of influencing the commission.
But I still feel it's fair of David to say that Newsom doesn't have room to talk here. If Texas put in an independent commission that resulted in significantly better maps for Republicans, do you think that Democrats would be calling it fair? Do you think that it would be fair if Republicans had an advantage in Texas as large as California's?
Unfortunately, Professor Friedman’s “loose” calculation is not particularly useful for drawing conclusions. It may have been misleading for him to show this because we don’t elect House representatives based on the portion of voters who voted for a party. In the 2024 election in California, 60.48% of the votes went to Democratic candidates for the House, while 39.23% of the vote went to Republican candidates. When more than 60% of the votes cast go towards candidates of one party, that party should win more than 60% of the races, without the aid of any political influence. How much more depends primarily on geographic distribution and district boundaries. If Republicans and Democrats were uniformly spread across the state, Democrats would have won every race without any gerrymandering. That was not the case, but the proportionality method is still misleading.
California’s 2020 maps may embed partisan advantage, but that cannot be determined from Professor Friedman’s table. As for fairness, few maps are perfectly fair, but the fact that the majority of Republicans on the Redistricting Commission approved the 2020 maps, and Republicans did not challenge the maps in court, indicates by some reasonable measures that they were fair enough.
We're both guessing at things, because there is not now and never was a "correct" way to determine districts and boundaries. Maybe proportionality should be considered, but as you correctly identify, if every district was uniformly distributed then we would want the majority party to win in each individual district even if that means the minority party gets zero seats.
But, this also works in reverse. Democrats complaining that Texas will be more red under new maps are trying to have their cake and eat it too. Maps that don't make room for Republican seats in California aren't more or less legitimate than maps in Texas that don't make seats for Democrats. When David says that Newsom doesn't have a leg to stand on in his complaints, it's a recognition that maps in his state directly disfavor Republicans more than necessary to complete the function. It would certainly be possible to make a map with more Republican-leaning districts, but they choose not to.
We are talking about Texas now because that's where the changes are, but in every state such decisions are made regularly and often disadvantage the minority party more than necessary. I don't care about complaints from Newsom or some other blue state governor about Texas when it's blindly self serving. I'm not upset at them for having districts the way they do, just for complaining about Texas trying to do the same thing they have already done.
Your argument assumes that Professor Friedman’s table proves California is presently gerrymandering more than Texas, but his table provides no such evidence--disproportionality alone, particularly in light of the substantial majority held by Democrats, is not proof of gerrymandering. That simplistic comparison ignores critical factors like the natural geographic clustering of voters and overlooks the fact that California’s maps were approved by a bipartisan commission and never legally challenged by Republicans.
>> By contrast, California has engaged in a form of unilateral disarmament, outsourcing its map-drawing to an independent commission. While this does not guarantee gerrymandering will not occur, it can mitigate the severity.
—————————
But it didn’t. See ProPubluca’s article about California’s “independent “ redistricting: https://www.propublica.org/article/how-democrats-fooled-californias-redistricting-commission
Actually, it did.
The ProPublica article you cited was about the 2010 maps, which were a mess. By 2020 the process had evolved. The Commission unanimously approved the maps, and there were no lawsuits. That shows the system is improving.
No one claims it’s perfect, but it’s clearly more transparent. And if Republicans are right that scrapping the Commission would hand Democrats five extra seats, then the Commission is limiting partisan gerrymandering, not enabling it.
Fair enough. But we’ll see.
I found the media coverage of the purported gerrymandering in Texas and the response by Newsome annoying because none of the reporters made any attempt to see whether the reallocation was (a) obviously complete gerrymandering, (b) completely reasonable to rebalance for population shifts; (c) somewhere in between. Thanks for doing this analysis for the country as whole.
Another way to avoid gerrymandering is make all representatives at-large. They would end up as state representatives rather than district representatives, but I don't know how much that really matters, other than them bragging about helping their constituents in their weekly / monthly newsletters I never read.
It would make representatives more dependent on their parties, thus eliminating the possibility of principled eccentrics like Rand Paul.
I'll have to think about that when I'm more awake.
First, I think you must mean Massie instead of Rand Paul, since Rand Paul is a Senator, not subject to gerrymandering, and already elected at large.
Second, I've always wondered how Massie got elected in the first place. Did he start out as a more traditional boring politician and only became a principled eccentric as he aged?
Third, I think you could be right in a state with only one or two representatives, where the mainstream candidates would get most of the votes. But Kentucky has six. That might be enough for the tail end to elect one or two eccentrics.
Fourth, or maybe electing eccentrics depends more on the ratio of urban to rural voters. California has 52 representatives, but I bet most come from Southern California and the Bay Area, and I don't have the brain power to think right now about what that does for eccentrics.
ETA: Fifth, San Francisco kept switching back and forth between at-large and districts for their supervisors for a few years, and the at-large campaign always touted better minority representation, which speaks for more eccentrics rather than fewer. Whether they actually got better minority representation, I do not know.
I thought you were proposing something like a party list system.
Oh heck no! I hate party list elections. No, this is an election where all candidates for all districts appear on one single candidates ballot, and every voter gets one vote, although they could all get any number to allocate as they wish.
The big names would get the most votes. Most voters in California (52 reps!) would only know a few, and without being by district, maybe almost all would come from big cities. Kentucky might work a lot better.
Maybe California could work a variation, with, say, 5 or 10 groups of districts voting for at-large reps from the group area.
Or groups could be randomized for every election. Or district voters could also vote on moving to a neighboring group.
Hadn't thought of all those possibilities.
> No, this is an election where all candidates for all districts appear on one single candidates ballot, and every voter gets one vote, although they could all get any number to allocate as they wish.
You need people to have at least as many votes as there are positions.
> The big names would get the most votes. Most voters in California (52 reps!) would only know a few, and without being by district, maybe almost all would come from big cities.
In practice this would collapse into a *de facto* party list system.
> You need people to have at least as many votes as there are positions.
I don't follow. Voters get one vote now. Why do they need more, just because there are more names on the list?
ANY scheme that preserves majoritarian elective politics must FAIL. Gerrymandering is just one of its many problems.
Any scheme that is not majoritarian opens Pandora's Box. Proportional representation's ability to produce stable and desirable government depends heavily on the underlying distribution of preferences in the population. If these are multi- peaked, we have the nightmare of coalition politics with the smallest party or parties wielding the greatest influence. The bigger parties acquiesce to keep the coalition alive. Think the Socialist Party in Italy, the Green Party in Germany, to say nothing of Weimar. Looking at American in the present, I am not sanguine.
The nutjobs must be disciplined inside a bigger party, non-democratically.
No, any scheme that IS majoritarian opens Pandora's Box. "Representation" is a shibboleth that elevates demoGRAPHIC non-essentials over competence and wisdom. How does ANY demographic (race, gender, age, etc.) promote the selection of officers that assure order, prosperity, and general social well-being? They do not. Only non-demographic SORTITION selects officers based on "the content of their character."
While I very much appreciate the idea of randomizing the membership of decision making bodies -- remember William F Buckley's lament that he'd rather be ruled by the first 3000 names in the Boston phone book than by the Harvard faculty -- if the underlying distribution of preferences in the population is multi-peaked, randomization will mirror that. And the smaller groups which are to make decisions will more or less reflect the whole population. Weimar in the small.
We have met the enemy and he is us!
I remember Buckley's comment.
I answer your objection in my reply to MostlyCredibleHulk below.
SORTITION solves the problem of GERRYMANDERING, as well as ALL OTHER ills of majoritarian elective politics. Specifically, it must be NON-STRATIFIED sortition.
I have ALREADY provided this solution in a previous post. You would do a great service to the former republic by promoting this, the ONLY solution to these problems.
https://store.mises.org/Constitution-of-Non-state-Government-Field-Guide-to-Texas-Secession-P11264.aspx
I don't think there's any chance that the concept of "select a random person and give them all the representative power for about a million people" is going to be accepted by the people of the US. It's an interesting theoretical exercise but it's not going to happen.
This is NOT a "theoretical exercise." It's a common-sense solution.
Your objection would be valid if just any schmo could become an officer. The true point of your objection is the PROBLEM OF THE INITIAL POOL. The current Leftist use of sortition addresses that problem by instituting demographic strata, seeking "equal representation" in that initial pool based on race, gender, etc. – which consistently introduces social justice preconceptions. Such strata must be REJECTED.
How to populate that initial pool with only wise, competent candidates? Egalitarian fanaticism – today's atheist replacement for those enthusiasms that formerly were fenced in by dogma and creed – rises to a deafening howl on the suggestion that "wise" and "competent" can EVER be defined by ANYBODY.
The solution is to have EACH COUNTY define competence COMPLETELY ARBITRARILY in its selection of competent candidates – candidates which form the first pool of multi-round random selections. WHAAA? you say, HUH? That gonna git them nut job, Podunk fascists runnin' thangs in each county! Yes, a Mormon county might well insist on only Mormons in the initial pool; an agricultural county might insist on those holding a certain acreage. THIS element of arbitrariness is the ONLY VIABLE sense in which "representation" has any meaning. Whatever this small aspect of arbitrariness, no county is going to be so stupid as, say, exclude women or blacks; every county seeks order and prosperity. Any foolishness under this aspect will be dissolved in the later rounds of sortition.
> Whatever this small aspect of arbitrariness, no county is going to be so stupid as, say, exclude women or blacks
Why not? And why excluding non-Mormons is ok but excluding women is not? What about Jews - is excluding them ok or not? I'm pretty sure a lot of Harvard-educated people would support that if they could.
But beyond that, this sounds like proportional representation with extra steps. Which again relies on the concept of territorial "counties" - given permission to draw arbitrary counties, I could draw a county map of California which would ensure every county selects a woke leftist as their candidate. Then, completely randomly, one woke leftist out of set of 50 would be elected as California representative. But how it's better than what is there now and how it better represents 42% of California population that doesn't want woke leftists to represent them? It seems like we've got the same problem just with extra steps.
You would be right if someone could begin re-drawing customary and longstanding counties willy-nilly.
I accept that a county could abuse its arbitrary power to define competence so as to, say, exclude Jews – as was done implicitly in the past by requiring a religious oath for office. However, this county would soon find itself abandoned. We will see a demonstration of this principle if Mamdani is elected.
You are saying it like the idea of changing maps to fit one's political purpose is a completely novel invention and never before have been used. Remind me please what "gerrymandering" actually means? And most counties' maps are as arbitrary as electoral districts - somebody sometime (often not so long ago) just decided for some reason it will be so. There's no objective basis under it.
> However, this county would soon find itself abandoned
Just as Hamtramck, Michigan is abandoned now, or Oakland where people literally marched screaming "Death to America" is abandoned, or is Portland after years of Antifa riots. Or just like segregation ended because all people just abandoned the segregationist counties and refused to return until equal rights laws were instituted. This is really not how it works in real life, sorry.
> We will see a demonstration of this principle if Mamdani is elected.
We have already seen this movie though. New York before Giuliani was the fear city. It will be again (plus of course hating the Jews) if New Yorkers are stupid enough. But it won't be abandoned. Moreover, significant number of people would think the inevitable decline has nothing to do with their choices and blame the Jews for it (who probably won't leave either btw).
As I say, you would be right if county lines were changed routinely for political purposes, which is not the case.
Once the idea of SORTITION is embraced, the issues raised here become trifles. This idea has indeed become popular in the Northeast. There and elsewhere, people have become disgusted by the failures of majoritarian elective politics. However, despite this admirable motive, the recent implementations have insisted on DEMOGRAPHICALLY STRATIFIED sortition, which is a cure worse than the disease. This is so because the prevailing egalitarian fanaticism promotes the shibboleth of "representation," when it is a fictional social justice category.
Counties are lines on a map. How exactly do lines on a map "define competence"?
Oh, you mean people, who somehow "represent" those lines on a map, do the actual defining. You're back to square one.
No one suggests that "lines on a map" define competence.
Counties provide the initial pool because they have customary and longstanding recognition. County lines can well be perfectly arbitrary. Their only purpose is to in some way define communities whose various competencies – and totally arbitrary standard from any presumptive deontological state viewpoint – provide the initial random pool. Subsequent random choices assure that no one "arbitrary" definition of competence trumps the others.
There you go again. I haven't been very clear.
* have EACH COUNTY define competence
* a Mormon county might well insist
* an agricultural county might insist
* no county is going to be so stupid
* County lines ... only purpose is to in some way define communities whose various competencies
Counties don't do squat. They are lines on a map. If lines define competence, insist on anything, can be stupid, and have a purpose, it is because some humans act on behalf of those lines.
Which humans? Who selected them?
Once again, county lines are irrelevant. Obviously lines and counties don't make choices – the inhabitants of the county do.
These residents are not so stupid (as central state deontologists suppose) as to foul their own nest. If there is some knot-head county that insists that the initial pool be composed of transvestites or of women who are at least 1/1024 Cherokee, it will soon find its proportional contribution of candidates falls to ZERO, as its residents pack up and leave.
What do you think of proposals for algorithmic districting, like this one:
https://rangevoting.org/SplitLR.html
Are they missing something? I can see how maybe ideally you’d have “representative” districts. But this is always going to be a vale judgement.
My idea was to have a redistricting proposal take the form of a computer program using specified inputs, probably population density, possibly town boundaries, with a maximum length of the program.
YES, they ARE missing something: Every one of these smarty-pants fixes LEAVES INTACT the root problem, which is majoritarian elective politics.
Non-stratified SORTITION resolves the gerrymandering problem, as well as ALL OTHER ills of majoritarian elective politics.
Also, "representation" is a shibboleth. Consider the following jackasses:
Rep. Sheila Jackson Lee (D-Texas) stunned attendees at a high school solar eclipse event on April 1, 2024, by claiming the rock-solid moon is a “planet” that is “made up mostly of gases.” Incredibly, she is a member of the House Science Committee’s space subcommittee.
Rep. Hank Johnson (D-Georgia) stated, “My fear is that the whole island [of Guam] will become so overly populated that it will tip over and capsize.” On July 25, 2016, in a public speech, Rep. Johnson compared Jews to “termites.”
These ignoramuses hold office strictly on the myth of "representation," that because of their black skin they best "represent" their districts.
A legislature chosen by sortition would instantly become a rubber stamp, as they’d have no ability to control the civil service (which is most of what congress does).
Politics is about getting people to do what you want, mostly by using other people you’ve already managed to get to do what you want. It’s a fairly difficult set of skills, and VERY SMART people with GOOD IDEAS, tend to be very very bad at it. As importantly, when it comes to getting bureaucrats (whose job is also mostly politics) to do what you want, the ability to make their lives better or worse is vital. Individual politicians can’t do that, you need coalitions (themselves built on threats, favours and patronage); long-term incumbency is thus very important, as is a seniority/party system. Choosing people at random every two years would turn them into a joke. They’d also get shaken down by helpful experts (really lobbyists and bureaucrats, but these are really experts and 90% of the time they all agree with each other), so would vote for whatever was put forward.
The point of elected politicians is that they extend the network of patronage into the ordinary population by giving everyone a favour they can return/offer/withhold in the form of a vote, so government has to take ordinary people into account to an extent (not so you can pick the person whose GOOD IDEAS you like because they’re COMMON SENSE). Sortition destroys that, but terrifyingly might legitimise the state to the point that it can care even less about public welfare than dictatorships have to.
Mr. jumpingjacksplash, you haven't read closely. Your objections are a celebration of the Leviathan state.
>>A sortitioned legislature would be a "rubber stamp" to civil service experts. "They’d [...] get shaken down by helpful experts".
No, non-stratified sortition guarantees COMPETENT candidates in the initial pool. They are self-sufficient, NOT beholden to experts.
>>"[Y]ou need coalitions [and] long-term incumbency".
Your recommendation is exactly the problem: Favor-swapping and immunity from removal is the essence of "politicking" in its worst sense.
>>"[E]lected politicians [...] extend the network of patronage into the ordinary population by giving everyone a favour they can return/offer/withhold in the form of a vote, so government has to take ordinary people into account to an extent".
Your recommendation is exactly the problem: Patronage and political spoils consolidates power and leads to beggar-thy-neighbor increasing debt – further aspects of "politicking" in its worst sense.
>>"Sortition [...] terrifyingly might legitimise the state".
False. Sortition annihilates politics and the state machinery that it supports.
We should just have proportional representation in Congress. It's simple and fair.
“It is clear, however, that the governors of California and Massachusetts are in a poor position to complain about Republican gerrymandering.”
But of course that is only true if you go by facts and vote percentages relative to representation in Congress.
But they are in fact in excellent position to do that given how our media and elite culture work.
As the lack of pushback from the MSM those governors have gotten demonstrates.
I learned some things in this post and in the comments that are important to me. As someone who leans right and gravitates towards like minded commentary I thought that the redistricting villains were clearly on the Democratic side and Texas, out of necessity, was starting to fight fire with fire. Although I still feel the "outrage" expressed by Democrats is intellectually dishonest, David's table indicates that partisan redistricting is more evenly distributed than I had thought although Democrats are still slightly more guilty. California truly deserves bragging rights here provided Newsom doesn't undermine what its citizens have achieved. I was unaware of that until I read Omar's comment.
Not taking any position on the topic, but you might find this interesting.
https://www.youtube.com/watch?v=Lq-Y7crQo44
Regarding Massachusetts:
“Republican votes clear 30%, but are distributed so uniformly that they are locked out of the possibility of representation. Though there are more ways of building a valid districting plan than there are particles in the galaxy, every single one of them would produce a 9–0 Democratic delegation.”
https://www.liebertpub.com/doi/full/10.1089/elj.2018.0537
That cannot be true, given the map I showed.
Did you draw any Red congressional districts? Ones that reach the population threshold? By the way, the paper I referenced is several years old—perhaps the situation has changed. But I can personally confirm it was impossible then—this is easy to show by solving a small-ish linear program. The authors also proved it in a different way.
I did not — I don't have the data. But the article does not say that the impossibility applies in general, only that there have been past elections where it did. Note that:
"Meanwhile the Clinton/Trump election had a nearly identical 35.3% R vote share but produces the possibility for as many as two districts (built from towns) with a Trump majority."
So my conclusion was overstated. But the map by towns that I showed has much more clustering than the two they show, so I would be very surprised if one could not find continuous clusters of towns that would give a Republican majority district. I can't be sure without population density data — the red area could be all tiny towns.
There are lots of free districting software with web interfaces. Like playing with Microsoft paint. You can try your hand at making a red MA district here, for example: https://davesredistricting.org/maps#home
By the way, even if it is now mathematically possible to draw a red district, that does not mean it was likely to have been drawn by a neutral mapmaker. There’s a whole literature out there on randomly drawn plans (ideally sampled from a reasonable target distribution), and that is a better way to arrive at intent—if the enacted map was an outlier in the distribution. Here’s one such paper: https://www.pnas.org/doi/abs/10.1073/pnas.2217322120
I would be very interested in you, David Friedman, writing an essay that explores alternatives to representation that is not based on geography. I think the electoral process of selecting representatives for Congress in the USA is reprehensible. What do you think?
My favorite plan is true representation. Each voter chooses a representative. Each representative casts a number of votes equal to the number of people he represents. You can change your representative at any time.
I don't know if it would work better but it is more elegant.
I've had the same idea, but I wonder how fast that would turn into a dictatorship (one representative having a majority of the votes), which might lead to unpleasant consequences. One nice thing about democracies as we know them is that they are quite slow at making decisions, as they need to corral a majority of legislators to make things happen, which gives their subjects a chance to vote with their feet before it's too late.
I too wondered about one legislator having too many votes, and solved that by requiring passage by both votes and headcount. A sort of combined House and Senate all in one.
Fascinating. Gerrymandering is abhorrent.
I think this is a finer version of proportional representation, yes? Instead of the parties making the electoral lists, the voters do.
I solved this in my Chartertopia, ha ha. The reason districts have to be drawn and redrawn is because each legislator gets one vote in the legislature, so districts must be equally sized, population-wise.
Each district elects three representatives, who cast as many votes in the legislature as they won in the election. I call it proxying. One voter is selected at random to proxy all remaining votes; he's the amateur.
It does treat the count of voters as being a good proxy (!) for population. I ran some comparison a few years ago on turnout and population, and they did vary, but not by more than 10%. I no longer remember how I made the calculations or what data I used.
It eliminates the need for redrawing districts and for a census. It also encourages people to vote, since voting for a minority political viewpoint still allocates some votes outside the mainstream candidates, even if only to the amateur.
It reduces the incentive for fraud, since a few fake votes in a close election no longer spell the difference between winner gets all and loser gets none. At best, a candidate can be boosted from 4th no-show to 3rd, but that also gives more votes to the amateur, who now gets the true 3rd place votes instead of the true 4th place votes. That's a lot less gain for the same risk.
Of course it will never happen.
Chartertopia, you say: "One voter is selected AT RANDOM... ."
– Close, but no cigar. The answer is full-on randomization, specifically NON-STRATIFIED SORTITION, as I explain here:
https://store.mises.org/Constitution-of-Non-state-Government-Field-Guide-to-Texas-Secession-P11264.aspx
The kernel problem for sortition is the INITIAL POOL. How to keep "amateurs" and total ignoramuses OUT so that only competents can be selected? My book resolves that problem.
I don't know if you're right or wrong, but it's way more complicated than the analysis you did. In that map of MA, are there even enough voters in any of the two red county blocks to make a district? It's not at all surprising to me that a 2:1 voter ratio would lead to a sweep, even without gerrymandering.
What would be more interesting is a scatter plot of voter ratio vs seat ratio to look for outliers.