Here’s a new take on the age-old problem of political gerrymandering, where incumbent politicians choose their voters before their voters choose them: What if we could re-draw political boundaries based on our own natural affiliations to nearby cultural centers and regions instead?
That, in part, is the ingenious germ of inspiration behind CommonCensus.org, a project of Michael Baldwin, a 25-year-old English teacher and former political science student living in Brazil. His site asks visitors to input their home address, and then walks them through twelve screens asking what nearby towns and cities they have the strongest affiliation with. As a bonus, he’s also inviting people to identify what sports teams they are most loyal to.
The result so far is a multicolored map of the United States that shows the “sphere of influence” of dozens of major cities, along with other maps showing the fan base of the local teams in five different sports. For a New Yorker like me, it’s fascinating to discover that NYC’s influence spreads deep into north-central Pennsylvania, but when it comes to baseball, the Boston Red Sox own half of upstate New York!
This is a classic example of “emergence,” where the seemingly disconnected efforts of many small actors combine to produce a larger intelligence.
Baldwin is updating his maps as more people come to the site and add their data. A few days ago, he had a little more than 16,000 responses, which, he warned visitors, meant that his maps were still prone to statistical inaccuracies. As of yesterday, he had over 24,000 responses, and he promises to keep producing finer iterations of his maps–plus local maps–as more individual data comes in. He admits that his system could be gamed by someone who wanted to input false data, but he has cleverly built in some safeguards (including a deliberately slow series of survey pages) to prevent deliberate spoiling of the information he is collecting.
While Baldwin’s sports maps give the site a playful tone, they are a brilliant way to draw in participants. And he clearly has a political purpose. As he says on his FAQ page, it is
to educate people about what geographic groups Americans ‘feel’ they belong to, as opposed to where politicians and post offices say they belong to. This matters because every day people categorize Americans into geographical shapes that are not representative. Nobody has ever before created a detailed map of the ‘sphere of influence’ of every city on local, regional and national levels. CommonCensus does not espouse any particular political viewpoint, but imagines that the kind of data it creates should be invaluable to the debate on Congressional redistricting, to anyone involved in urban planning or studying markets, and to anyone with an interest in geography or demographics. It is also the first systematic way to exactly determine the limits of a metropolitan area without relying on guesswork or one person’s opinion.
I emailed him a few questions.
Q: First, just for the techies: What platform is your site built on? Have you considered mashing your data with the Google Maps API?
A: For the website, PHP and MySQL. Actually, if you click on one of the five most recent contributors, it shows their answers using Google Maps. But for the main maps, Google Maps doesn’t work because it only allows you to add icons, not change colors of areas, etc. I do the map processing in a program of my own I wrote in Visual Basic. Ideally I would write it in something the web server could compute, but I’m still experimenting with the graphics, and VB lets you prototype things quickly.
Q: Second, when you have enough data to make the local maps visible, how do you imagine the overlay between someone’s local hometown and their congressional district will be made visible? I’m intrigued by what you suggest will be a disjunction, but I don’t quite see how you will use your data to create local communities of identity that will somehow congeal into a population pool matching the size of a congressional district. The fact that we don’t think of ourselves as living in CD34 but instead in a town or a collection of towns isn’t new…so I’m wondering if you can be clearer on how you envision your maps energizing the gerrymandering debate in a fresh way.
A: You’re right that the sizes of congressional districts don’t match the sizes of local communities at all…because they can be neighborhoods in some cities or entire states in other places. I think the local-area data will be very interesting because there are a lot of congressional districts of about that size, that are particularly susceptible to gerrymandering, and CommonCensus shows how community boundaries can be drawn by ‘the will of the people’, and not by computer algorithms (based on population statistics alone) or politicians.
I think that comparing a CD with the shapes of counties or even cities doesn’t always inspire indignation, because in a lot of the US, county shapes are rather arbitrary too, as well as some city limits. Nobody from my part of Upstate New York would ever say they identified with Oneida County–they’d say the ‘Utica-Rome Area’ or ‘Mohawk Valley’, and that doesn’t show up in a line on any map, so the main point of reference we have culturally is useless for coming up with CD’s.
Certainly there are more issues in congressional districting, such as the fact that there have to be a certain number per state and have relatively equal populations. I also think the very idea of boundaries drawn ‘by the will of the people’ is exciting in itself, and will stimulate people to think if something similar couldn’t be done by governments one day? After all, if a website can calculate cultural areas from people’s votes, couldn’t a government do something similar to generate CD’s.
Obviously it wouldn’t be exactly the same algorithm I use, and I am aware that various algorithms have already been proposed. This could just add another piece to the puzzle. Furthermore, before I started this site, I had actually never seen a CD map in my life. If my website gets enough publicity (which seems to be starting, little by little), I’m glad that I can use it to focus people’s attention on the shapes of their so-called “representative” areas, and how easy it is for a guy with a computer to show how wrong a lot of them are.
Baldwin plans to keep updating his sites maps as more data flows in, and to eventually expand its gambit beyond the United States. True to the logic of his “emergent” model, he promises no “final map,” just a “current map.” “After all,” he notes, “cities grow and change—a new neighborhood might rise to prominence in New York City, or the sphere of influence of New Orleans might have changed after Hurricane Katrina.” Eventually, he will also use different shades of color around cities to show their degrees of influence on particular areas.
It is intriguing to imagine other uses of this process. Could Westchester County, where I happen to reside, invite its residents to register where they live and the degree of danger they feel from living just miles from a nuclear power plant, Indian Point? If enough people participated, wouldn’t that information affect the debate over its continued operation? Could a municipality use a CommonCensus approach to gather information from residents about zoning issues, or whether to allow a Walmart to come to town? Could activist groups with large memberships use it to involve their members in seeing where they live and where local chapters are needed?
Is this e-government, e-democracy, or do we need another word to describe a project like CommonCensus? It’s not about using technology to make it easier for government and recipients of services to communicate, nor is it devoted to enabling direct conversation between citizens. And yet, by cleverly enticing people to quickly share some personal data about their connections to their physical environment, CommonCensus is producing a fresh form of public self-awareness, and may help open a new debate about the meaning of representation in an age when old political boundaries mean less.