> Gerrymandering is a separate issue

SCOTUS was asked to lay out a standard for courts to follow in deciding gerrymandering cases. A court can't know a priori whether a partisan skew in a particular map was deliberate or the result of geographic factors. So you can’t just treat it as a “separate issue.” Among other things, even if there’s evidence of intent, there might not be causation. You can intend to do lots of things that you don’t actually end up accomplishing.

You can kick the problem to a "neutral commission"--but that's giving tremendous power to unelected people. If "commissions of neutral experts" were a real thing that exists we could just dispense with democracy and rely on them! The Constitution is all about asking "who has the power?" but on issues like this people overlook that and get mired in "who would come up with the best answers under unrealistic assumptions about bias?"

Put differently, having legislatures draw districts might make it harder to "vote the bums out." But gerrymanders are actually pretty fragile. Democrats had massive geographic and gerrymandering advantages throughout the 1970s and 1980s: https://crystalball.centerforpolitics.org/crystalball/conten.... But Republicans still managed to break through that in 1994, and since then it's flipped back and forth quite a bit. With a "neutral commission," by contrast, you've created a body with a lot of power that can't be voted out.

Also, what target are the experts shooting at anyway? That’s difficult to tell for the same reason it’s hard for the court to tell. Are the experts supposed to be leveling natural partisan skew due to geographic factors?

> If "commissions of neutral experts" were a real thing that exists we could just dispense with democracy and rely on them!

No, we have democracy in order to choose between competing policies and ideologies.

An expert comission has a pre-defined, fixed policy - produce fair district maps, and they work to carry it out. The idea that no-one can carry out a task without bending the result to suit their personal views is absurd. If you ask some car designers to design you a saloon car, they'll design a saloon car, even if they personally prefer sports cars.

This is an important point, which I'd refine to: we have democracy in order to choose between competing goals and ideologies.

In theory, neutral experts could be used to determine the best policies towards the democratically chosen goals. And it actually works that way for the most part. The vast majority of stuff that goes through a parliament is stuff you never hear about, after all. They're like your car example. It's the few percent of topics where there's controversy where it's difficult, in practice, to find experts who will genuinely stay neutral - precisely because they're so controversial.

> This is an important point, which I'd refine to: we have democracy in order to choose between competing goals and ideologies.

I disagree with that. If you asked people what the goals should be for educating children, they’d probably be in broad agreement. But they differ widely in how they think those goals should be achieved.

Experts are only useful when there is a fairly rigorous scientific framework in place that has predictive power—the ability to predict what will work and what won’t.

Even in areas adjacent to those fields, the effectiveness of expertise quickly breaks down. Public health has elements of both medicine and social psychology, and the results were all over the place during the recent pandemic. When you get into other territory: education, welfare, etc., the real world value of expertise is even less.

> If you asked people what the goals should be for educating children, they’d probably be in broad agreement

At least in the US, this seems to be empirically false.

> You can kick the problem to a "neutral commission"--but that's giving tremendous power to unelected people

It works pretty well in Australia. Here, federal election district boundaries are drawn by the Australian Electoral Commission (AEC), an independent federal agency. I've never heard anyone remotely mainstream suggest the system is rigged or gerrymandered or politicised. (There are certain elements of "unfairness" built into the system – most notably, that the least populous state gets the same number of Senators as the most populous state, just like in the US – but the AEC didn't decide that, the Constitution says it.)

The AEC has total control over federal elections–the states have no role in decisions regarding them. For state/territory and local elections, each state and self-governing territory has its own electoral commission, similar in principle to the AEC – although some of the details, such as how exactly it is appointed, will differ. Unlike the US, in Australia, local government has zero role in running elections, even local government elections–all control of elections is at the state/territory and federal levels.

The AEC and its staff (and their state/territory equivalents) are required to be politically neutral. They are not allowed to belong to political parties or publicly support or criticise them.

The chief executive of the AEC is appointed by the government for a five year term. Although government politicians (effectively) choose the chief executive, they are required by law and custom to choose someone without a record of past involvement in politics – the current incumbent used to work in a civilian executive role for the federal law enforcement, before that as a manager for a defence contractor, and before that he was in the military – so obviously someone with some demonstrated experience in management/leadership but without any known political affiliation.

As well as its chief executive, the AEC has two other members – a chair and a third member. The chair must be an active or retired federal judge – the Chief Justice of the Federal Court of Australia gives the government a list of three judges, and the government gets to choose one from that list to become the chair. The third member is also chosen by the government, but legally must be head of another federal agency – (non-binding) tradition dictates that the government choose the head of the Australian Bureau of Statistics (ABS), the Australian Statistician. Major decisions–such as drawing electoral boundaries–must be made by the Commission as a whole, by a two-thirds majority vote.

> With a "neutral commission," by contrast, you've created a body with a lot of power that can't be voted out.

Nobody can vote out the AEC, but the government can remove its three members from office for "misbehaviour". However, the removed member can challenge the removal in the courts, so if the government tries to remove one without a strong justification, the courts are likely to strike down the attempt, whereas removing them with strong justification is likely to be upheld. To my knowledge, no removal has ever been attempted. (Also, the AEC is established by statute not constitutionally, so Parliament could amend the law to abolish the AEC and replace it with something else–highly unlikely to happen in the foreseeable future.)

I can't see why you couldn't use a similar model in the US, and why something like that couldn't work there. (You couldn't copy the exact Australian model into the US, due to various differences between their respective national constitutions, but you could translate some of the major ideas of the Australian model into something that could work in the US context.)

You’re defining gerrymandering as a deliberate act, but a looser definition includes the entire effect of geographic districts on the outcome of elections, including non-intentional effects.

The article gives an example identical to yours.