On Friday, December 4, The University of Pittsburgh’s Mellon-Sawyer Seminar Information Ecosystems: Creating Data (and Absence) from the Quantitative to the Digital Age was joined by Bill Rankin, an Associate Professor of the History of Science at Yale University. Professor Rankin’s research focuses on the relationship between science and mapping, the environmental sciences and technology, architecture and urbanism, in addition to methodological problems of digital scholarship, spatial history, and geographic analysis. His prize-winning first book, After the Map: Cartography, Navigation, and the Transformation of Territory in the Twentieth Century, was published by the University of Chicago Press in 2016. Professor Rankin is also an award-winning cartographer, and his maps have been published and exhibited widely in the U.S., Europe, and Asia.
Rankin talked with the Sawyer Seminar Participants, who are faculty and students at the University of Pittsburgh and Carnegie-Mellon University, about cartography, election mapping, and the contemporary U.S. political landscape. Amid the many reactions to and characterizations of the historic 2020 Presidential election, this meeting helped the Seminar participants understand how and why election mapping continues to play an increasingly crucial role in the electoral process; in particular, Rankin’s talk touched generatively about the concept of “purple states” or “purple places.”
Purple has been, in recent years, offered as a more representative complication to the simple binarism of “blue,” or liberal, and “red,” or conservative states. The “red” versus “blue” state discourse began as a simple, visual way for newscasters to characterize a state’s partisan tendencies over long durations of time. And while we do know that not everyone in a blue state is liberal and not everyone in a red state is conservative, Rankin argues that these characterizations still wield massive power over how we think about states as coherent, homogeneous spaces. Rankin’s own work has proven time and time again that portraying state, county, city, or neighborhood boundaries as fixed and rigid, and the populations they contain as homogenous, is quite common in public-facing maps—and such maps can have serious impacts on budgeting and subsequent funding prioritization and disbursement. In turning, for example, Texas from a purely red state to a our post-election maps, we begin to more accurately characterize the reality of living there: some people are “red,” but not everyone. But in shading the entire state varying shades of purple, the maps still rely on ultimately simplified, binaristic understandings of place and politics. In other words, it’s not true that people in a purple state are each equally half liberal and half conservative, and it’s likewise untrue that towns or cities in purple states are exactly half liberal and half conservative.
This impulse in mapping to go purple, Rankin discussed, is not incompatible with desires to further complicate or contextualize maps as better representative of any places’ inherent complexities; Rankin suggests, however, that maps can (and are!) doing more work to establish new metrics for “accuracy” in mapping, as it relates to elections or broad-strokes characterizations of U.S. people and places as a whole. Rankin shared images of election maps that emphasize different types of information than we’re accustomed to in such vizualizations. Take, for example, “wind maps,” which show the changes, rather than static outcomes, in voting activity over the last four years. Another useful example is Rankin’s 2010 reimagining of U.S. Census demographics maps; rather than displaying population according to county, Rankin uses zip codes for a more granular view of population that tracks more accurately with population density. Furthermore, each population map is one of a pair, showing not only the relative minority or majority status of a given racial group, but also an absolute measure of where those populations actually live. In doing so, Rankin’s map elucidates what Census maps very often completely obscure: there are many places, mainly but not always urban, where minoritized racial groups are actually the majority. Recontextualizing this information has important implications for how we understand ourselves, our towns, our cities, and our nation: the U.S. Census questions are far from neutral and the answers represented by this data is inherently reductionist, but reworked maps like electoral turnouts or Rankin’s Census prove something important about mapping; complexities like time, change, and self-reported racial identities are not too complex for visualizations.
After his talk, a Seminar participant asked Rankin about this tension between complexity and accuracy in mapping. The conventional wisdom is that maps cannot “handle” too much complexity, and, of course, in some ways this is very true. It has, for example, been argued by more than a few that big city transit maps are sometimes so complex as to be illegible. To this, Rankin helpfully responded that it is less an issue of how to show complex information on a map, and more of what kind of complexity a map could most effectively offer. There are so many things about being a human and existing in space-time that seem all but impossible to standardize and visualize. But there are also active, glaring omissions from public-facing maps that we already know how to visualize—perhaps not with crystal-clear, precise accuracy, but, as Professor Rankin’s work shows us, we need to try anyway. Take the Sentencing Project’s 2020 map of African American Felony Disenfranchisement Rates, i.e. the percentages of Black populations in each state who cannot vote due to a felony-level conviction. In 2020 in Tennessee, Wyoming, and Florida, over twenty percent of voting-age Black adults were disenfranchised; disenfranchisement came up throughout our talk as something traditional electorate mapping simply does not consider. A standard, red-and-blue—or even purple—electorate map is unlikely to contend with this complexity, which is so utterly essential to understanding contemporary elections.
Professor Rankin is currently working on two new book projects, one on the spatial history of the environmental sciences since 1960, and another a methodological-theoretical take on mapping and visual communication. As we ended our Seminar talk, Rankin mentioned that there are no easy answers to the questions we’ve raised and will continue raising about space, maps, elections, and, relatedly, who we are. In the massive imagined and literal geographics of the United States, it can feel absolutely overwhelming to live through a historically unique moment in time—collectively and privately. I took great comfort, even in our current lack of concrete solutions, that folks as brilliant and careful as Rankin are still hard at work in their pursuit.