Religiosity is often seen as a prime mover of American public opinion. In Protestant-Catholic-Jew, Will Herberg spoke of “tripartite” religious traditions as the cause of political dynamics. More recent scholarship has broadened the concept of religiosity to distinguish tradition from the specific features of one's religious beliefs, the dynamics of one's religious behaviors, and a believer's sense of belonging to a particular faith or religious movement — an approach commonly called the “3 B's” (see Geoffrey Layman, The Great Divide). Paul A. Djupe and Brian R. Calfano's God Talk: Experimenting with the Religious Causes of Public Opinion is, quite ambitiously, a critique of this entire approach to religion and politics scholarship. Religiously motivated public opinion dynamics often arise, change, and recede faster than do individuals' beliefs, behaviors, and belonging. This premise in place, the authors submit that it is necessary to rethink the nature of religion itself — from viewing it as a rather static dispositional feature (individual and attitudinal) to a more dynamic understanding of religion as a communication source, accounting for the extent to which individuals are exposed to messages with religious cues, and the extent to which individuals adopt said messages. As such, God Talk involves using fairly common theoretic approaches to understanding political communication (see e.g., John Zaller. The Nature and Origins of Mass Opinion) in a subfield where political communication has been somewhat neglected.

According to Djupe and Calfano, this theoretical omission has largely been the product of methodological limitations. Religion and politics scholarship has largely relied on observational methods, while experimental approaches are better suited to identify religious communication dynamics. The experiments in the book are creative and grounded in public opinion research, and the authors take pains to address a number of the critiques that are often leveled against experimental behavioral research. For example, instead of relying strictly on samples of American college students, Djupe and Calfano go to great lengths to diversify the religious composition of their participants. God Talk finds creative ways to implement experiments by surveying congregations directly, by cooperating with religious political action committees, and even by soliciting samples in Great Britain. Likewise, God Talk guards against the critique that the design lacks generalizability. For example, the treatments in Chapter Two are drawn directly from the Republican Party operative David Kuo's own examples of religious cues used in crafting political speech, and Chapter Six is designed to mimic an ultimatum issued by Focus on the Family founder James Dobson. While experimental designs might have benefited by incorporating large-scale content analytic work like David Domke and Kevin Coe's The God Strategy or Roderick P. Hart, Jay P. Childers, and Colene J. Lind's Political Tone, the treatments are, on the balance, thoughtful and grounded in political realities.

The authors unpack the argument in three main sections, each corresponding to a different communication source. First, God Talk investigates the impact of politicians' use of religious language, finding (generally) that certain religious cues can be leveraged to influence vote choice among certain message recipients. Next, Djupe and Calfano turn their attention to religious elites, again finding that religious cues can shape public opinion, depending on how the message is framed and characteristics of the message recipient. Finally, the authors examine the types of statements made within the congregation through the use of survey experiments. Instead of testing explicitly politicized communication from the pulpit, the authors test exposure to inclusive and exclusive religious values.

This brief synopsis glosses over the tremendous complexity of the argument, and a number of nuanced findings that are important for any scholar of religion or public opinion. For example, in Chapter Two, Djupe and Calfano uncover an interaction between “God talk” language and gender — when male candidates use religious cues, support for the candidate goes up (as expected), while the opposite is true for female candidates. Another fascinating example comes in Chapter Nine, where the evidence shows that priming religious values can shift public opinion, though the direction of the change depends on whether “inclusive” or “exclusive” values are primed. Findings like these are important, though the sheer range of experimental findings detracts from the theoretic parsimony of the argument. While it is clear that religious communication matters, the book misses an opportunity to advance a systematic theory about for whom religious language exerts its most profound effects, and why some messages are more effective than others. For example, while the book tests religious statements based on Kuo's cues, it doesn't test why some of these cues turn out to be more persuasive than others. Instead, the theoretic mechanisms responsible for the effectiveness of God talk tends to shift from chapter to chapter: The idea of “process cues” is persuasively advanced in Chapter Four as an explanation for preference change, but process cues are abandoned in Chapter Nine in favor of values priming. Elsewhere, the book explores moderators for “God talk” including trust, in-group commonality, threat, demonstrating shared values to establish credibility, the role of political knowledge, and the role of numerous religious moderators (85, 71, 159, 94, 160). It is unclear why political knowledge moderates some types of communication effects, while respondents' guidance from religion moderates others. In short, God Talk provides convincing evidence that religious communication matters, but stops short of telling the reader how religious cues work across the varied contexts studied.

Djupe and Calfano make a convincing case that “God talk” is politically important, and the book is successful in promoting experimental methods in religion and politics research. God Talk also offers a serious theoretical challenge to dispositional accounts of religion and political behavior (163). The book is less persuasive on this count, in part because effects are almost universally contingent on some religious moderator, suggesting that religious communication cannot be understood unless one first carefully scrutinizes the “3 B's” as they exist in the electorate. In this way, I might characterize God Talk as building on the shoulders of the dispositional approach to religion and public opinion, rather than a complete rejection of it. Message exposure and message adoption need to be scrutinized to a greater extent than has been the case, but enduring religious beliefs and behaviors are — and will remain — an important part of the story. This said, the authors themselves acknowledge that God Talk is not the final word on religious communication and public opinion. The book makes an argument about how theoretical accounts of religion and public life are grounded in methodological and ultimately epistemological choices, and for this reason is an important addition to the subfield.