How the Streets Got So Mean

When Don Mitchell was a master’s student in geography at Penn State in the late 1980s, he came across a newspaper article on homelessness that struck him. Homelessness was surging in many U.S. cities — from 1984 to 1987 the number of people living on the streets almost doubled — and the article attempted to explain the trend by looking into the characteristics of those experiencing homelessness: age, race, gender, work history, drug or alcohol abuse. That didn’t seem like a satisfactory approach to Mitchell.

“Individual characteristics don’t tell us much about causation,” says Mitchell, now a professor of cultural geography at Uppsala University in Sweden. “Some percent of homeless individuals have a substance abuse problem, for instance — but plenty of people who are housed do, too.”