American Grace: How Religion Divides and Unites Us

James Croft reviews Robert D. Putnam and David E. Campbell's new book, in The Humanist:

American religion is a conundrum. Americans manage to combine deep religious devotion with wide religious diversity, all the while remaining remarkably tolerant of each other. What factors have shaped the current religious landscape in the United States? What characteristics do people of faith have, in comparison to those of other faiths and those with none? And what explains America’s unique combination of diversity, devotion, and tolerance?

These are the questions that Robert Putnam, Harvard professor of public policy and the sociologist who shone a spotlight on American community in Bowling Alone, and David Campbell, professor of political science at Notre Dame, set out to answer in their book American Grace. In short, the authors seek to provide a definitive snapshot and analysis of the state of contemporary religiosity in America. They ask about the relationship between religion and politics, between religion and civic values, whether religion plays a divisive role or brings people together and, in the opening chapters, how America got to where it is today, religiously speaking. The breadth of the book’s ambition, along with its hefty dimensions (the main text runs to 550 pages) and steepled hands on the cover, convey the intention clearly: this is to be the new bible for sociologists of religion.

The majority of the book is based on two large surveys (3,108 participants in the first, and 1,909 in the second) conducted in 2006 and 2007. The sample drawn for the first was representative of the population of the United States and was randomly selected. The second followed up with as many of the individuals surveyed in the first as possible, and asked most of the same questions. Therefore, the authors argue, it is possible to see how some measures change (like church attendance) between one year and the next. This second survey is important because it enables the authors to “test” whether one variable alters with another: if making a friend of another religion coincides with a warmer view of that religion, for example, then one might plausibly hypothesize that making friends with people of another faith leads to warmer feelings for others of that same faith. This is not enough to establish causality, but it does give useful hints that would not emerge without the second survey.