Gerrymandering... uhhh... redistricting:

April 26, 2002 9:19 AM

Gerrymandering... uhhh... redistricting: A fascinating piece from the Economist describing how both of the major US parties are becoming ever more efficient in redefining electoral boundaries to suit themselves at the expense of the electorate. Check out the maps.
posted by pascal (8 comments total)

In Virginia, it is all the more interesting since the Governor and Attorney General can and happen to be from separate parties. A judge recently overthrew a redistricting effort penned by the Republican controlled state senate, the Republican AG then said the state would appeal, and the Democratic Governor said, "ummm, you might want to consult me on that first."
posted by machaus at 9:28 AM on April 26, 2002

Nice article.

I met these two really young kids in DC in a party once - they apparently worked for a company that advises the democratic party on redistricting. This was early 2001 and they were almost jumping up and down with excitement. They said whatever gets recommended, it is sure to be challanged by the other party in the court of law and they are very likely to be hauled before the court to testify. This guy said that finally he would have something that he can wear like a badge of of honour everywhere. ...One day he would like to be a senetor ...

It seemed to be that redistricting is so accepted as a part of the political fabric that even the newbies into politics (I am sure on both sides) dont question it any longer.
posted by justlooking at 9:54 AM on April 26, 2002

Weirdly shaped districts like these are signs that a crime has been committed, writes the Economist, but that's only true in the general moral sense of the word. They surely should know better -- gerrymandering was strictly litigated away in the beginning of the 19th century, but returned in the late 20th as an ad hoc solution to racial disparities in the electorate, particularly in Southern states covered by the Voting Rights Act. Most recently, in the 1990s, there were several cases where Congressional districts were brought to the Supreme Court, either because they were created to favor winning a minority representative, or because they failed to do so.

(Several redistricting rulings showing the trend; a lawyerly overview.)

The redistricting commission was indeed popular at one point, but the trouble was that commissions -- whether made up of legislative subcommittees, or appointed emeritus legislators and judges -- produced maps that despite painstaking efforts at balance (or at the very least protestations of same), at least one of the legislative parties would object and take the commission plan to court. Wisconsin and Illinois both went through painful court cases regarding their redistricting after the 1990 census. The Illinois case eventually ended up in Federal district court, and a panel of judges drew the map "themselves" (with professional consultants), which is the origin of Illinois's Fourth Congressional District, the weird caliper-shaped one.

The Economist article does not once contain the word "court", "judge", "ruling", etc., which shows that by that magazine's standards, there was appallingly little research.
posted by dhartung at 9:59 AM on April 26, 2002

I have an easy solution to this problem: legislate a maximum fractal dimension to the boundaries of each electoral district.

Where's the mathematician lobby when you need it?
posted by ook at 10:05 AM on April 26, 2002

About two months before this spring's primary election, a federal panel of judges threw out our GOP lawmakers' first swing at redistricting here in Pennsylvania -- over a disputed area where 19 people pushed the map into illegality. So everyone ran around like chickens sans heads, until that same panel said this May's election could happen as scheduled using the illegal map. Wha? Democracy at its finest.
posted by krewson at 10:15 AM on April 26, 2002

Re-districting has many interesting facets, and the article reveals a number of assumptions people make about democracy.

For one, there is an assumption made about geographic compactness somehow correlating to "validity" of a district, and that an odd-shaped district is presumptively invalid - in fact, the Economist article outright says so: "Weirdly shaped districts like these are signs that a crime has been committed."

(Actually, they're not. There's nothing illegal about weird shapes.)

Second, there is an assumption that the split of a delegation ought to closely mirror the political composition of the state. I think it's fair to say that such an outcome is desireable, but not the ultimate goal. We have a representative, majoritarian-based government. Our protection from the tyranny of the majority comes in the form of our limits on government enshrined in our Constitution - not in our particular adherance to any particular districting plan.

Third, there's an assumption that districting to favor one party over the other presumptively dis-empowers the minority party every election. Simply not true. The more heavly skewed the delegation is away from the true composition of the electorate, the more subject it is to a sudden shift in the electorate mood. That's because to skew the delegation, you have to have a number of closely-drawn districts. All it takes to lose a closely-drawn district is a handful of controversial votes that activate the opposition party's base, and suddenly it's 1994 all over again.

The irony is that the disctricts held up as examples by the article aren't new - they've been around for years - and they're not even the worst ones. Illinois' 4th has been around for at least one, if not two previous census cycles. Florida's Thrid District, recently the subject of a court challenge, stretches from Jacksonville to Orlando, and is meant to "benefit" the African Americans by having them send Corrine Brown to Washington.

Why does it matter that these are old districts? Because it shows that even when the balance of power shifts from one party to the other, the prime object of redisricting - at least this cycle - has not been to make great advances against the other party. On the contrary, both parties have fought hard this year to preserve the status quo above all else.

Instead of creating more skewed delegations this cycle, the line-drawers have been padding existing margins and reducing the number of competitive seats - in other words, protecting incumbents. To me, this is a far greater threat to democracy than creating a number of closely-matched swing seats. Unfortunately, the author of the article chose to bury that point under a bunch of maps and charts, sound the alarm about "crabs" and "embryos."
posted by mikewas at 12:52 PM on April 26, 2002

i'm don't see the irony, but i haven't had my chocolate milk today either.
posted by tolkhan at 1:11 PM on April 26, 2002

Hrm, its an interesting problem. One I don't think you can really ever "solve", that is to say I don't think you can really make it 'fair' in any true sense.

The only question you can as is how you want to optimize the situation. Should minorities preference be highlighted? should we optimize for 'close' races, meaning that we put about 50/50 rep/dem in each one? should we just optimize for geographic symmetry?

Nither of those is truly 'fair' in the most pure sense of the word.

Here's my solution, let people pick what district they want to be a part of when they vote. In fact, don't even call them 'districts' call them 'races' (not race=ethnicity, race=contest). So you can chose to vote for who ever you want.

I think it would work because it would make the election entirely unpredictable, and thus no one could 'game' it.


Another solution would be to use ascension voting, and pick the top X reps. It would be difficult for a state like California, you might want to split something like that up into 5 or 6 blocks based on geography.
posted by delmoi at 4:42 PM on April 26, 2002