Images via

Andrew Prokop at Vox features the work of the Center for Range Voting, which developed a method to draw congressional districts with much simpler shapes.  The current North Carolina districts and the Center’s revised version are above. There’s no doubt that these new districts are, to use Jon Bernstein’s phrase, “pretty little districts.” But there’s a problem, which Prokop notes at the end of his post:

But it’s important to note that, because of their simplicity, these maps don’t take several things into account. They don’t try to keep historical neighborhoods or regions intact, don’t try to ensure representation of racial minorities, and don’t pay any attention to ensuring a balance between political parties. Still, they provide quite a contrast to the maps we have today.

To put it more bluntly, pretty little districts could actually be pretty terrible.  That is, they could be terrible at doing what districts are supposed to do: engender good representation.

What’s good representation?  Here are some answers that people regularly defend:

  • Good representation happens when representatives are beholden to specific geographical communities, who are believed to have common interests.  This is a reason to draw districts that correspond to existing cities, towns, and the like.
  • Good representation happens when the largest possible majority of people get to elect the representative of their choice.  This is a reason to draw lopsided districts with large partisan majorities.
  • Good representation happens when groups who have been historically excluded from the electoral process — like racial and ethnic minorities — get to elect the representative of their choices.  This is a reason to draw majority-minority districts (like the snakey NC-12, the subject of Shaw v. Reno and subsequent Court cases).
  • Good representation happens when a district is politically competitive, which means representatives work harder to represent “the people” because there is always a good chance they could be thrown out of office.  This is a reason to draw districts with a partisan balance close to 50-50.

Those are just a few theories.  There are others, long debated by philosophers and political theorists.   The point is, drawing pretty districts doesn’t necessarily accomplish any of these things.  In Jon Bernstein’s words, the shapes of districts “tell us exactly zilch about whether it’s a good or bad districting job.”  In fact, to accomplish one or more of those theories — and especially if you have to balance among them — weirdly shaped districts might be required.

I’m not suggesting that we shouldn’t entertain reforms to the redistricting process, such as having more independent commissions, and fewer incumbent legislators, draw the lines.  But whoever draws the lines, there’s no reason to draw straight ones.  Representation is about people, not polygons.