Environment as Politics

New drawings of the relation between residential density and voting behavior.

Graphic of United States flag representing 2016 presidential election
[The Open Workshop for Places Journal]

One lesson of the U.S. presidential election is that we should forget about red and blue states, North and South, coastal coffeeshops and heartland diners. The geographic divide in American politics is closer to home. If you want to predict how someone will vote, ask, How near are your neighbors?

Residential density has long played a role in shaping American politics. In the recent election, 49 of the 50 highest density counties voted for Hillary Clinton, and 48 of the 50 lowest density counties chose Donald Trump (nearly the same split as for Barack Obama and Mitt Romney four years earlier). 1 In “blue” California, the agricultural towns of the Central Valley swayed Republican, while in “red” Texas the big cities voted Democratic. Across the spectrum of purple hues, a high-resolution map of population density closely matches voting results.

Chart of population density related to popular vote, 2016 U.S. presidential election
Hover over the image to explore it with the magnifying glass. (Or click here to see the large version.) Population density related to vote margin in the 2016 U.S. Presidential Election. The furthest left column shows the average density in counties that voted for Clinton by a margin of 70 percent or greater, while the furthest right column shows the average density in counties that voted for Trump by a margin of 70 percent of greater. Each swatch represents one acre that is emblematic of how space is organized in these counties. [The Open Workshop for Places Journal]

There are reasons for this that go beyond identity politics. Urban density has social and economic advantages that make cities attractive to liberals and that also condition liberal values over time. Living among diverse neighbors can reduce fear and resentment, as everyday interactions break down stereotypes and misconceptions of ‘the other.’ (Which is not to ignore that cities have their own problems with racial and economic segregation.)

One recent study compared voters who switched from Obama to Trump with those who switched from Romney to Not-Trump. Controlling for “age, race, education, income, gender, party identification, concern about rising immigration, racial resentment, and worries about personal finances,” the authors found that a vote for Trump was correlated with fear of rising diversity. They concluded that the election was a “clash over the openness of society,” 2 which may explain why cities produce liberal voters. Density politics is arguably at the core of racial acceptance.

Or, as The Stranger put it, two weeks after the election:

Look at the faces of the people sitting with you in the cafe, or bar, or free clinic lobby, or library. Look across the aisle of the bus, or subway, or light rail train you’re riding. Look at the drivers stuck in freeway traffic with you. Look at the young and old of all colors and creeds who share this city with you, some sleeping, not far from you, under tarps and highway overpasses. This is the American city. You are fortunate to be here, inside one of the most powerful machines we have for defeating fascists. 3

Chart of population density related to popular vote, 2016 U.S. presidential election
Hover over the image to explore it with the magnifying glass. (Or click here to see the large version.) Typical spatial organizations in 51 American counties, with counties that voted heavily for Trump in the center and counties that voted heavily for Clinton on the left and right of the chart. Population density is represented on the vertical axis. [The Open Workshop for Places Journal]

Understanding the relation between space and politics is now a basic requirement for civic literacy. Can the typical American voter tell you how dense her or his community is? How land, housing, and infrastructure are organized at different scales? To illuminate this, my office examined the relation between density and the popular vote across all 3,144 counties. We then illustrated 51 counties, representing a range of geographic locations and political affiliations, to understand their varied spatial organizations. So rarely are the spaces in which we live depicted when Americans discuss politics. We wanted to show that landscapes are not neutral, that they reflect social and economic relationships that shape (and are shaped by) our politics.

In counties that voted overwhelmingly for Clinton, households are 215 feet apart on average; in strong Trump territory, they are nearly half a mile distant.

If you live in a Clinton stronghold, one of the 11 counties which the Democratic candidate won by a margin of 70 points or greater, the average density is about 2.1 people per acre. That includes high-density Manhattan, the Bronx, San Francisco, and Washington, D.C., but also rural counties in Mississippi and part of the Pine Ridge Indian Reservation in South Dakota. If you were to distribute those households evenly, sprinkling New York apartments through the Badlands, you would have neighbors 215 feet in every direction.

Conversely, there are 201 sparsely populated counties that Trump won by a margin of at least 70 points. If households in Trump country were evenly spaced, they would be 2,412 feet apart — nearly half a mile. The tipping point, where the two candidates split the vote, is a hypothetical county where neighbors live 608 feet apart. 4

Chart of population density related to popular vote, 2016 U.S. presidential election
Plotting the 2016 election results in all U.S. counties, with high-density counties on the left and low-density counties on the right. In counties near the tipping point, where the average distance between households is about 608 feet, a vote for Clinton was equally likely as a vote for Trump. [The Open Workshop for Places Journal]

These drawings were inspired by the biologist-geographer Patrick Geddes who, in 1909, argued that physical environments shape human activity, and that people, in turn, act on environments through their labor. Geddes drew a geographic section that moves from mountain to foothills, through plains and riparian edges, down to the water. In each of these settings, Geddes identified the most adapted forms of labor and tools, from the miner’s pickaxe to the fisherman’s net, believing that “the kind of place and the kind of work done in it deeply determine the ways and institutions of its people.” 5 For Geddes, our lifestyles, and therefore our politics, emerge from negotiation with the environment.

A half century later, a group of young architects affiliated with the breakaway CIAM group Team X published the Doorn Manifesto, which updated Geddes’s valley section to show corresponding dwelling types. 6 Team X contended that housing should be conceived as habitat, a negotiation of local circumstances, environment, and social relationships. Their “scales of association,” from city to town, village, and farm, matched density to particular social relations and patterns.

Typical land use in counties that voted for Hillary Clinton
Democratic America. How one acre of space is organized in typical counties that voted strongly for Clinton. (9 A-B-C): Washington, DC, Manhattan, NY, and San Francisco, CA; (8 A-B-C): Arlington, VA, St. Louis, MO, and Providence, RI; (7 A-B-C), King, WA, West Hartford, CT, and Los Angeles, CA. [The Open Workshop for Places Journal]

Typical land use in counties where the vote was split in the 2016 U.S. presidential election
Divided America. How one acre of space is organized in typical counties where the presidential vote was split fairly evenly. (6 A-B-C): Kanawha, WV, Dakota, NE, Forest, MS; (5 A-B-C): Atlantic, NJ, Boone, MO, and Pierce, WA; (4 A-B-C), Guilford, NC, Douglas, KS, and Concord, NH. [The Open Workshop for Places Journal]

Typical land use in counties that voted for Donald Trump
Republican America. How one acre of space is organized in typical counties that voted strongly for Trump. (3 A-B-C): Jefferson, KS, Cattaragus, NY, and Nobles, MN; (2 A-B-C): Garrett, MD, McHenry, ND, and Union, MS; (1 A-B-C), Leslie, KY, Roberts, TX, and LaSalle, LA. [The Open Workshop for Places Journal]

Today, many geographers, planners, and designers understand density on a continuum. There is no clean line between city, suburb, exurb, town, region, and hinterland; there are only multifaceted, networked territories. Our drawings aim to update the valley section to probe the relationship between environment and density politics. We still do not have adequate language to fully describe all these landscapes, but we can examine their spatial characteristics.

It’s remarkable that even as the internet disperses information and enables us to form online communities across great distances, our politics are still highly correlated with physical environments. Who we are is largely defined by where we are. For architects and urban designers, this is an important reminder that space is and always has been political, from the days of the valley section to the postmodern stage of Trump. 7

Author’s Note

Research and drawing by Elizabeth Lessig and Cesar Lopez.

I would also like to thank Andrew Kudless and Adam Marcus, who curated the exhibition “Drawing Codes: Experimental Protocols of Architectural Representation,” January 17 to February 4, 2017, at the California College of the Arts, which provided the foundation for this research.

Notes
  1. In the 2016 election, the exceptions were high-density Richmond, Virginia, which voted for Trump, and low-density Kenedy and Culberson counties, on the Texas-Mexico border, which voted for Clinton. For analysis of the 2012 election, see Emily Badger, “How Suburbs Change: More Density Means More Democrats,” The Washington Post, May 2, 2014, and Dave Troy, “The Real Republican Adversary? Population Density,” November 19, 2012. A similar relation between density and voting behavior was noted in the 2015 general election in the United Kingdom. See Joe Sarling, “Can Population Density Predict Voting Preference?,” CityMetric, September 14, 2015.
  2. Sean McElwee and Jason McDaniel, “Fear of Diversity Made People More Likely to Vote Trump,” The Nation, March 14, 2017.
  3. Dan Savage and Eli Sanders, “The American City Is a Machine That Kills Fascism,” The Stranger, November 21, 2016.
  4. Our model uses 2016 election results at the county level (sourced from Townhall.com) and the most recent U.S. census data. Population density is calculated using census figures for people per square mile, averaged across all land area in the county, while “nearest neighbor” is calculated using housing units per square mile.
  5. Patrick Geddes, “The Valley Section from Hills to Sea,” lecture at The New School of Social Research, 1923, published in Cities in Evolution, New and Revised Edition (London: Williams and Norgate, 1949).
  6. Alison Smithson, Team 10 Primer (London: Studio Vista, 1968), 75. See also Annie Pedret, “Preparing CIAM X 1954-1955,” Team 10 online.
  7. See, for example, the recent issue of Log (39), which includes a special section on the role of architecture as a political medium in the Trump era. For recent articles on Places, see Reinhold Martin, “The Demagogue Takes the Stage,” Places Journal, March 2017, https://doi.org/10.22269/170328, and Jeremy Till, “Reality in the Balance,” Places Journal, January 2017, https://doi.org/10.22269/170124.
Cite
Neeraj Bhatia, “Environment as Politics,” Places Journal, April 2017. Accessed 17 Nov 2017. https://doi.org/10.22269/170418

If you would like to comment on this article, or anything else on Places Journal, visit our Facebook page or send us a message on Twitter.