Inventing American Religion

In Inventing American Religion, Robert Wuthnow challenges the common belief that public opinion polls about religion provide truly descriptive and generalizable information on Americans and American religion.

Wuthnow provides a detailed history of the development of academic surveys and public opinion polling. Academic surveys began as in-depth ethnographic studies which could provide nuance regarding questions about how people are religious and the role of religion, religious resources, and religious relationships in their lives. Surveys provide context, so that the meaning of religion is not divorced from its role in lived experiences. However, academic surveys take considerable time, money, and expertise.

In contradistinction, public opinion pollsters inquiring about religion produce results which are quicker, cheaper, and allegedly generalizable to all Americans. Polls could show us at a glance that 96% of Americans believe in God and around 40% attend religious services during any given week. In producing those statistics, however, Wuthnow contends that pollsters also produced a concept of religion that is a distorted version of a complex and nuanced reality (13).

Wuthnow identifies four topics of concern with polling and how it shapes our understanding of religion: “generalizations, categories, concepts, and contexts” (193). First, as opposed to academic ethnographies, the point of polling is to make generalizations about all Americans. Such generalizations, however, ignore the diverse mosaic of Americans and our religions—and, in so doing, privilege white Christian religiosity over other religious lifestyles. This occurs not only due to the relative sizes of these populations, but also due to the nature of the questions themselves. For instance, questions about beliefs are inherently biased towards Protestants, as many religions place little or no emphasis on the content of belief. Even the question about belief in God lacks any specificity about the nature of God. Likewise, questions about attendance atreligious services are irrelevant for some religions. Thus, Wuthnow contends that public opinion polls engage in Christian-norming in their attempt to generalize.

Second, polls create or shape categories for classifying people. In some cases, this creates distinctions that are not meaningful to the respondents themselves, meanwhile perhaps ignoring distinctions that are meaningful to them. Polls thus attempt to “simplify a complex world so that it can be studied through comparisons among a relatively small number of categories” (200), as opposed to current practices in the social sciences which examine complexity and fluidity.

Third, polls define or redefine concepts through the way questions are asked and answers interpreted. Secularization, for example, has become a matter of declining church attendance, as opposed to its original meaning. As Wuthnow explains, “Secularization was once understood to have taken place over several centuries and to have consisted of changes in the institutional authority of religious organizations. Polls redefined it. It now meant a measurable decline from year to year in religious attendance and belief in God. Secularization was happening if polls said it was” (200).

Finally, polls ignore the contexts in which people live their lives. Unlike ethnographies, polls do not allow us to peer into people’s relationships with their neighbors, co-workers, and so on. The interactions between individuals and the ways these influence their religious behavior are ignored. Polls present a snapshot of the static, unchanging identities of isolated individuals rather than a moving picture of interconnections and variations. And because response rates to polls are at historic lows (sometimes as few as 9%), there is concern that what little may be accurate about that snapshot is difficult to generalize to the rest of America.

As one of the leading researchers and scholars in the sociology of religion, Robert Wuthnow is perfectly positioned to provide considerable insight into the process and production of sociological research into religion. This work provides interesting historical context to some of the most famous studies in the field. This book will be of interest to those who are interested in the history of those famous studies, those who are interested in polling and sociological research methods, and those who are interested in the social construction of knowledge itself. Readers will be challenged to reconsider what they think they know about what it means to be religious.

V. Jacquette Rhoades is associate adjunct professor of sociology at the University of Indianapolis.

V. Jacquette Rhoades

Date Of Review:

October 23, 2017