“If you wish to make an apple pie from scratch, you must first invent the universe.”
That’s a famous quote from the late Carl Sagan, an astronomer who popularized scientific thought on topics like the place of the human species in the universe.
It is somewhat surprising that of Sagan’s 600 published scientific papers, none include a mention of Bloomington, Indiana, as the exact center of the known universe.
Compensating for Sagan’s error of omission are two key notions from that famous quote—“pie” and “from scratch.” Those two ideas have some current relevance in Bloomington’s civic life.
Both ideas are relevant to the work of the city’s redistricting advisory commission, which met for the first time this past week.
The commission was supposed to be seated 18 months ago, at the start of 2021, but was not appointed until mid-June of this year.
Baking districts from scratch?
The city council’s ordinance that established the commission describes it as an “independent” commission. The idea is that the commission’s work is supposed to be independent of the interests of city council incumbents.
If an independent redistricting commission wants to take the approach of baking the new council districts “from scratch,” they could start with a blank map, then apply elements of a general pie-making recipe for city council districts, which are supposed to include: compactness; contiguity; population balance; and various “communities of interest.”
Under the ordinance establishing the commission, the “communities of interest” are not supposed to be split across more than one city council district. The communities of interest listed out in the ordinance include: “political subdivisions, neighborhoods, school districts, historic districts and other areas where residents share common traits and concerns.”
But based on last week’s first meeting, the commission’s work won’t be all that independent from city council incumbents, and won’t be “from scratch.”
In fact, city council incumbents will have their own employee in the kitchen, in the form of council attorney Stephen Lucas, who will be in a position to add flavor to the commission’s work. At the first meeting, Lucas dumped some rotten fruit into the mixing bowl, by giving the commission bad advice about who their chair had to be.
Lucas told commissioners that under the ordinance, the chair had to be the member who was not affiliated with the Democratic Party or the Republican Party. Long story short: That is not true. And it was only at the end of that first meeting, that the incorrectness of Lucas’s advice emerged. But the commission took the path of least resistance, and ratified Lucas’s direction about the chairship. The choice of chair should have been their own to make, from scratch.
Lucas’s wrong advice also deprived the commission of a chance to learn a bit about each other’s deliberative styles, and perspectives—by allowing them to work through their first decision: Who will serve as chair of the five-member group?
Another choice already made before the first meeting involves the GIS map drawing tool that city staff and Lucas worked to create, to help support the commission’s work.
This key question had already been answered for the commission before it met for the first time: What kind of map should the city-staff-developed GIS tool display, when the software is launched by a user?
The apparent answer was to show, as a default initial view, the existing city council district boundaries. That seems like the right choice, only if you want to put your thumb on the scale in favor of preserving the current district boundaries as much as possible. Preserving as much of the current configuration as possible is not a legal requirement, and appears to be at odds with the explicit direction of the local ordinance, which states:
(4) Where it does not negatively impact the above criteria, districts shall be drawn to encourage political competition.
An alternative map-drawing tool, which MGGG Lab at Tufts University released at The B Square’s request, starts with a blank map of precincts.
Are council districts like a pie?
Comparing the re-drawing of city council districts to baking a pie would be just a colorful analogy for most cities in America. But it’s worth reflecting on one city where the idea of a pie is baked right into the local law.
In Ann Arbor, Michigan, the city charter mandates (emphasis added):
The five wards [city council districts] should each have the general character of a pieshaped segment of the city with the point of such segment lying near the center of the city so as to make each ward [district] a very rough cross section of the community population from the center outward.
It’s a question worth chewing on: Should each of Bloomington’s six city council districts reflect the general diversity of the whole community’s population?
It’s certainly possible to draw a map where each of Bloomington’s districts touch the outside crust and the meet in the middle, without a central, crust-free piece of pie. That’s Map B at the top of this column.
A consideration that should probably figure into the mix is one big difference in local government between Ann Arbor and Bloomington. In Ann Arbor, 10 of the 11 city council seats are tied to geographic districts of the city, while only one (the mayor’s seat) is elected at large. In Bloomington, three of the nine council seats are elected at large, while the mayor is not even a member of the council.
Bloomington’s current city council map includes one central district (District 6), that does not touch any of the non-city parts of the county. That suggests Bloomington’s districts have been considered historically as a way to capture distinct communities of interest.
Otherwise put, it has historically not been considered important for each of Bloomington’s six districts to reflect the full diversity of Bloomington’s population. And that’s likely because the three at-large council seats already represent the whole geographic area of the city. Why would full diversity need to be echoed in each district?
But looking at the current map of city council districts, it’s not apparent what the distinct communities of interest are, that each district is supposed to capture. In fact, the map looks pretty gerrymandered—but it’s not clear to what end. In 2012, the redistricting process produced only minor changes from the previous map.
If Bloomington is not looking to define its districts in terms of pie-shaped wedges, what are the distinct communities of interest that should shape decisions about districts?
The communities of interest defined in the ordinance include “political subdivisions.” What does that mean for Bloomington?
Except for two, all of the city precincts can be divided north-to-south between Bloomington and Perry townships. Richland and Van Buren townships have one precinct apiece.
Of course, it would be impossible to avoid splitting Perry and Bloomington townships across more than one city council district—that’s a function of population. Specifically, it would not be possible to dump all the Bloomington or Perry township precincts into a single district, because that single district would have about three times the ideal population.
But the current city council district map splits both of those townships across five different districts, which is just one short of the maximum split. Working east to west here’s how Bloomington and Perry townships break down under the current map:
- Perry Township is split between District 3, District 4, District 6, District 5 and District 1.
- Bloomington Township is split between District 3, District 4, District 2, District 6 and District 1.
Under the city’s redistricting ordinance, it would be desirable—all other things being equal—to reduce the number of township splits, from five to some smaller number. Given the constraints of population balance, any reduction below three would be impossible.
Can a population balance be achieved with a split of each township across no more than three city council districts? Yes. That’s the Map C shown at the top of this column.
That map puts a lot of weight on townships as communities of interest. But it also weights “core” and “periphery” as significant geographic factors. Map C shows not just one, but two central city council districts that do not touch any non-city area.
Conclusion: Slicing the pie before it’s baked?
I hope that the redistricting advisory commission takes some time to deliberate on the communities of interest they think are most important, before considering any maps.
For example, considered as communities of interest, do commissioners think townships are a higher or lower priority than historic districts or neighborhood associations? How heavy a weight, if any, should the Indiana University campus have as a community of interest?
If townships are given the greatest possible consideration, that would basically mean slicing Bloomington’s political pie along the boundary between Bloomington and Perry townships (3rd Street), so that the remaining pieces are north-south chunks. Those north-south pieces would need to be further divided into three districts each.
In some sense, that choice would amount to slicing the pie, even before it’s baked.
But that is true of any choice that the redistricting advisory commission might make along these lines—about how much weight to give different communities of interest.
What I hope to see at future meetings is some deliberation by the commissioners about how much they value different communities of interest and why. Those deliberations should come before they start trying to draw new maps themselves, or evaluating the merits of maps that others have submitted.
The dates of the next meetings of the redistricting commission are set: July 25 (7:30 p.m.) and Aug. 9 (9:30 a.m.).