The Supreme Court and redistricting in Alabama

 In an interesting turn, the Supreme Court rejected Alabama's latest district lines.  There are few legal or constitutional guidelines for congressional districts, and the Court rejected the lines for dilution of the racial vote, which was interesting.  Let's unpack.

Prior to what we call the reapportionment revolution, which is highfalutin jargon for a series of court cases beginning with Baker v. Carr, states could do essentially whatever they wanted with their districts.  They could have such egregious malapportionment that a district could have 1000 times more people than an adjacent district and according to the Court, that was kosher.  We stopped doing that about 60 years ago, when we started requiring equal population.  We also require "contiguity," which means that you need to be able to get from anywhere in the district to anywhere in the district without leaving the district.  Beyond that... nothin'.  Except for the weird and semicontradictory requirements on race.  According to the 1965 Voting Rights Act, we have some requirements about racial representation, which the courts have interpreted to mean that where convenient, states must draw "majority-minority" districts, which are districts in which a majority of the population is of a racial or ethnic minority based on the premise that racial and ethnic minorities do not get elected except from constituencies that are a majority of that subpopulation.  But, we also have a requirement that states aren't supposed to use race as the predominant consideration because the Court doesn't care about logical consistency.

Anyway, every plan gets challenged on the tension.

Here's the thing about a majority-minority district.  It will always be an overwhelmingly Democratic district.  Now, do you remember how partisan gerrymanders work?  Pack-and-crack.  You pack the opposing party's voters into inefficient supermajorities in a small number of districts, thereby "wasting" a bunch of their voters.  You know what does that really well?  Majority-minority districts.  So, there is an argument to be made that majority-minority districts, nationwide, help Republicans by packing Democrats.

Regardless, Alabama has seven districts, about 1/4 of the state is African-American, and there was one African-American majority district.

Is that kosher?

I do not have an opinion on majority-minority districts, specifically.  My works on redistricting more generally have argued for the creation of redistricting plans with districts that are as homogeneous as possible, because homogeneous districts allow each partisan/ideological slice of the population to be represented as precisely as possible.  Pure Democratic districts, where possible, pure Republican districts, where possible.  One could extend the logic to any other characteristic deserving of representation, and while I did not do so in print, it would not do damage to the core ideas in my models.  I just didn't take it in the direction of majority-minority districts.

The concept of a majority-minority district is based on the notion that descriptive representation leads to substantive representation.  Descriptive representation means having officials who have demographic markers similar to your own.  Substantive representation means having representatives whose beliefs match your own, or who pursue your interests.  (That gets into delegates vs. trustees, but this is already going somewhere else.)  Does descriptive representation lead to substantive representation?  The research on that is less than clear, but let's move on anyway because preferences can be treated as having their own value.

There is another point.

Seven.  Think about that number.  Maybe I should use the Arabic numeral.  7.

1 + 1 + 1 + 1 + 1 + 1 + 1

That is not a lot.  You can visualize seven.  You cannot visualize the number of districts from California, nor Texas nor Florida.  Alabama?  You can visualize that.  The human mind evolved to visualize a certain range of numbers, and above that, we require symbolic representations.  Why?  Because our ancestors weren't managing 52 bananas.  Seven bananas?  Sure.  52?  No.  We'd fight over 7 bananas, but we could not distinguish, visually, between 52 and 40 and 60.  That was just a lot of fucking bananas.  We can do this and more with symbolic math, but our monkey brains don't work that way, visually.

More still, how many ways are there to dole out 7 bananas?

Fuck off, that was funny.

[Ducks rotten fruit...]

Not many.  I do not simply mean that 7 is a prime number, which is interesting, but not germane.  Consider the time signature, 7/8.  Cool time signature.  You can chop up the 7 in a bunch of different ways, but the interesting things come a) from the variation across measures, b) polyrhythms, and c) a mix of quarter notes, eighth notes, and so forth.

Sure, you can literally chop up a banana like that, and divide seven bananas in even more ways, but with seven districts?

Suppose you can allocate one district per subpopulation.  A maximum of seven subpopulations can get a district.  Give one subpopulation two districts, and a maximum of five remaining subpopulations can get a district.  See where I'm going here?  How many different subpopulations are there?  Shall we ask Professor Crenshaw?

The more districts a state has, the more options you have to dole out bananas districts, but with a small state, you are really limited.  One district, instead of two, out of seven.  The Court said nopesies.

The Court ruled that the legal status of a majority-minority district requirement for African-Americans creates a specific requirement, and you can argue amongst yourselves about whether or not that requirement should still hold sixty years after the VRA, but if one takes, as I do, the principle that ideal districts are homogeneous districts, a limiting factor is the number of districts within a state.

I am skipping over quite a lot.  Given geography, can one parse anything like intent in Alabama to avoid a second majority-minority district?  Is it really race, or is it actually party?  Given Cooper v. Harris, is the Court bound to reject a partisan defense?  In the end, though, with seven districts, this is small bananas potatoes.  My argument has consistently been that our primary normative goal should be to avoid partisan and ideological biases, and this is best done through bipartisan gerrymanders.  That will mean, in some cases, majority-minority districts, but only through coincidence.  Yet the pursuit of majority-minority districts can easily aid the very pack-and-crack strategies that produce the biases at the national level that I seek to avoid, all for the sake of an empirically questionable relationship between descriptive and substantive representation.

I keep missing Friday jazz posts, so it's a twofer today.  First up, some 7/8 jazz from the odd meter man himself, Dave Brubeck.  "Unsquare Dance," from Time Further Out.  Then, the title cut from Pierre Dørge's Jazz Is Like A Banana.  Fuck off, it's still funny.