Given the authors' scholarly reputations, the rich datasets on which they were able to draw, the qualitative vignettes that add texture to the findings, and the suppleness of the prose, American Grace is an important book that will contribute mightily to shaping the future research agenda in our sub-discipline. As evidenced by reviews in the prestige press, the study also will influence public discourse about religion as a social and political factor in American life. Although the volume offers many provocative ideas that warrant careful examination, space limits force me to confine my comments to what I consider the most important argument of the book and also suggest the major weaknesses of the volume. Brevity should not disguise my admiration for the project nor for the volume's intellectual power and elegance.

In my view, the central contribution of American Grace is raising the right question about American Exceptionalism in religion. The important issue is not that the United States has such a high level of religiosity in spite of its economically advanced status, a question that has provoked decades of scholarly interest, but rather how intense and diverse religiosity coexists with a stable polity. When we consider the United States in a comparative framework, that is one of the core puzzles about its longevity. Political systems in societies with strong confessional attachments often fail due to the nonnegotiable nature of conflicts over first things. In a book that does many useful things, Robert Putnam and David Campbell provide a theory that gets us closer to solving this puzzle.

The “Aunt Susan” effect, a concept perhaps destined to match the fame of “Sheilaism” in Robert Bellah et al.'s (1985) Habits of the Heart (Berkeley, CA: University of California Press), is a contextual explanation that draws on Gordon Allport's seminal theory about the social conditions that promote intergroup harmony. Putnam and Campbell hypothesize that religious differences among Americans are muted — or at least channeled into non-destructive paths — by the frequency of intimate contact and social interaction with family members, neighbors, and acquaintances who embrace different religious traditions. In an age when religion has moved from an ascriptive identity to a matter of free choice and rates of religious intermarriage have soared, most contemporary families are religiously heterogeneous to an appreciable degree. By intermingling with people across religious lines, the authors suggest, Americans come to appreciate that their religious differences need not produce hostility. Learning that your friendly neighbor belongs to an “exotic” religious tradition is another mechanism that reinforces religious tolerance: the “My pal Al” effect. This is yet another case of the personal becoming political.

Note that I suggest the “Aunt Susan” effect is not a solution to the puzzle of American Exceptionalism in religion and politics but only a theory that “gets us closer” to comprehending the coexistence of strong pluralistic religiosity and stable democracy. As they readily acknowledge, the authors have not conducted a definitive test of the hypothesis but rather have generated a few models that behave as if the theory were true. But even more fundamentally, the hypothesis appears time-bound. Long-term data on rates of interreligious marriage deployed by the authors suggest that Americans were not all that likely to encounter family members and friends of another religion until relatively recently in American history. When Americans had more religiously homogeneous social networks, what prevented the emergence of a pillarized society in which confessional differences became gaping political fissures?

Perhaps, as pluralists have suggested, the diversity of the United States along many lines — class, section, race, ethnicity, etc. — encouraged cross-cutting rather than reinforcing cleavages, preventing religious differences from structuring political conflict. Maybe the incredibly fine differentiation of religious attachments in the United States, a society in which even most counties lack a majority religion according to the decennial survey of church membership, undermines the capacity of any religious tradition to mobilize successfully on a broad scale. From a cultural perspective, the secret of national cohesion could be a syncretic “civil religion” that places national unity above sectarian loyalty by endowing the state with transcendent qualities. Structuralists in the fashion of Morton Grodzins contend that the division of authority in the United States across three levels and three branches of government also diffuses conflict by providing multiple arenas of contestation. Another structural perspective with a Tocquevillian heritage insists that the American regime of religion and state, the liberal model that denies religious authority to the state and political authority to the churches, has largely insulated the United States from the tensions that arise in polities that fuse state and church. As the authors recognize (by citing some of these factors), it is unlikely that any single explanation can account fully for the long-term persistence of the American political system in spite of vigorous religious attachment and heterogeneity. Reinforced by these plausible hypotheses, the “Aunt Susan” effect is an original and provocative theory with strong political science roots. It merits further analysis.

American Grace is ambitious in scholarly reach but even more so in design. Most notably, the authors have opted to write a book that will appeal to scholars and the educated public alike. In an age when so much scholarship is directed at narrow audiences of academic specialists, Putnam and Campbell admirably have aimed for a much broader readership. Much like Putnam's previous Bowling Alone or Bellah et al.'s Habits of the Heart, the plain language, qualitative interludes, and straightforward argument of the book gives social science a good name. I doubt there's a sentence, table, or graph in the book that would be inaccessible to any intelligent citizen who wishes to gain greater insight into religion in American life. (I'd be tempted to assign the book to my classes except students might then wonder why I can't write so clearly myself!) The book is a fine model for any aspiring public intellectual.

And yet there is some unavoidable tension between the appeal to a public readership and academic audiences. I doubt many of the former will delve into some of the fascinating tables and graphs that illuminate the findings, but that is not my major concern. I suspect that scholars reading the book as I did may frequently want to see more than the authors give us, even in the footnotes. At many points, I expected the authors to link their findings to the political science and sociology literatures, yet they did not. The discussion of party images in Chapter 12 cried out for nesting in the concept of issue ownership, just as the discussion of God images in Chapter 13 should have integrated prior research by Andrew Greeley and others. Even the concept of “bridging social capital” presented in a book by a fellow named Putnam was curiously understated in the chapter on religion and neighborliness. Scholars also may wonder why the authors so strongly embrace the “God gap” hypothesis partly on the basis of a 0.10 “correlation” between religious attendance and Republican identification (372) until they discover that the correlation coefficient is actually a standardized regression coefficient.

I also think that the authors were, at times, too trusting of their data. Americans routinely over-report socially desirable behavior, such as church attendance and charitable donations, yet the authors sometimes present such data as if they were uncomplicated by measurement error. Considering the pattern of mass theological ignorance documented by the Pew Forum on Religion and Public Life and Steven Prothero, among others, I am highly skeptical that most people select a congregation for reasons of theology and religious belief (170). In view of the dominance of gender complementarity in traditionalist churches, I doubt that the apparent growth of feminist sentiments among frequent churchgoers is as deep as the authors report (240). Based on Robert Wuthnow's research, there is reason to add a large dose of skepticism when assaying claims about the widespread embrace of religious diversity by churchgoers (521). I think these patterns reflect the tendency of survey respondents to compartmentalize or to embrace general statements but then draw back on specific applications when confronted by religious “considerations” (pace Zaller, John, and Stanley Feldman. 1992. “A Simple Theory of the Survey Response.” American Journal of Political Science 36:579–616). In a book characterized by modesty and caution, a volume that resolutely avoids overstatement and unqualified claims of causality, these exceptions stand out. Social scientists cannot be as trusting as Putnam and Campbell portray American churchgoers.

In view of the magnitude and quality of the project, these are small problems, but not fatal flaws, indications that even the best social science has shortcomings. Having committed to make their survey data available to other scholars, the authors have equipped critics with the means to further test and refine some of the arguments in the book. All in all, American Grace is a stimulating book that considerably advances our understanding of the role of religion in American life and politics. In the ultimate tribute that authors should hope for, scholars will be arguing about the claims and insights of this tour de force for years to come.