Stephanie Muravchik and Jon Shields, by their own admission, were “aghast” at the election of President Donald Trump. So, like good political scientists, they set out to understand the causes of the catastrophe: they interviewed residents of three out of 206 “pivot” counties that voted twice for Barack Obama and then for Trump. The result is Trump’s Democrats.

One of the three counties in question had fairly predictable reasons for voting Trump: the traditionally Democratic Ottumwa in Wapello County, one of Iowa’s 31 pivot counties, is the site of a now-closed John Deere factory. Another, Elliott County in the hills of eastern Kentucky, was more unusual: unlike dozens of Appalachian counties, Elliott did not abandon its ancestral Democratic allegiance in 2008 or 2012. (Obama carried eight and two Kentucky counties respectively in those years, not two and one as the authors say.) For their third choice, Muravchik and Shields cheated just a little: the Providence suburb of Johnston is actually just outside Kent, the one pivot county in Rhode Island.

Thankfully, Trump’s Democrats does not contain the identically structured interviews and dubious quantitative analysis which are fashionable in social science. Instead, Muravchik and Shields seem simply to have driven into each locale and schmoozed with dozens of

Subscribe for access This article is reserved for subscribers.