Pippa Norris and Ronald Inglehart tackle two pressing empirical and theoretical questions about religion and politics: Why have most wealthy countries become less religious, and why have most poor countries become more religious? Their answer will be no surprise to those who have followed Inglehart's research over the years: The socioeconomic and external security conditions in which people grow up strongly affect their propensity to be religious. Those who were socialized during conditions of “existential security” have less need for religion; those whose formative years were marked by conditions of insecurity (to themselves, their families, or their communities) have much more need for religion. As with Inglehart's earlier work on postindustrial values (The Silent Revolution: Changing Values and Political Styles Among Western Publics, 1977), the main variable is socioeconomic well-being and security when young.

The book is an effort to show that despite appearances to the contrary, secularization theory is viable. Much recent scholarship has argued the contrary, that secularization theory, rather than God, is dead. The theory has been challenged by the facts of extensive religious participation and its political influence in the United States, as well as the rise of religious parties and movements across the globe, including in Europe. Even if people do not attend religious services, the majority in many wealthy countries express a belief in God. Modernization and industrialization, with their emphasis on rationality and science, do not eradicate religiosity. Yet Norris and Inglehart want to make a strong case for secularization theory. They do so by presenting “a revised version of secularization theory” (p. 4), which holds that people turn to religion when their world is insecure. Faced with poverty and violence, religion trumps science.

The book also disputes the theory of religious markets, derived from rational choice theory and supply-side economics. This theory holds that the demand for religion is constant, while the supply of it, and hence societal religiosity, varies according to the extent of competition among faiths.

The authors' argument is based on two axioms. The first is that societies differ in their level of security and that those conditions may vary over time within any given society, and across any given society at particular times. The second is a “cultural traditions” axiom (pp. 17–18), which implies that even as countries become secular, they become secular in Protestant or Catholic or Islamic ways (though there are negligible differences on attitudes toward major moral issues across religions): “Even in highly secular societies, the historical legacy of given religions continues to shape worldviews and to define cultural zones” (p. 17).

The book's first chapter develops six hypotheses, which the authors claim are suggested by the two axioms. The first one, and the book's main hypothesis, which the reader might also confuse with their theory, is that “the experiences of growing up in less secure societies will heighten the importance of religious values, while conversely, experience of more secure conditions will lessen it ” (p. 18, italics in original). This hypothesis relies on a psychological view of religion: that religions reassure individuals that despite all the difficulties and threats they face, “things work out” (p. 19).

Norris and Inglehart develop several more hypotheses (pp. 20–24); the second is identical to their second axiom: that even as societies secularize, the dominant religion will be discernible in the values of the society. The third hypothesis is that as religious values lose their importance, participation in religious services and organizations drops. The fourth is that participation in religious organizations is related to participation in civic organizations; the fifth is that existential security reduces religious values, which in turn reduces fertility rates; and the sixth is that “greater religious pluralism and also greater religious freedom will both increase religious participation” (p. 24). The authors are taking advantage of their 76-country data set to test a variety of hypotheses that have been dominant in the literature in the past 20 years (the World Values Survey and the European Values Survey panels from 1981 to 2000).

The main conclusion is a striking contrast to the Huntington thesis: There is no clash of civilizations; there is, instead, a clash of demographics: Rich people are more secular and have fewer children, while poor people are more religious and have more kids. There are more poor people in the world having more kids, and fewer rich people in the world having fewer kids; therefore, “the world as a whole now has more people with traditional religious views than ever before ” (p. 25; cf. p. 64; italics in original).

The authors use generational data; as a result, they do not expect religiosity to fluctuate with immediate economic conditions and disasters but that different generations in the same country will have been exposed to different levels of security during their formative years, and that the values acquired during those years will hold (pp. 28–29). Unfortunately, they undermine this thesis with frequent comments to excuse poor correlations and regression results, including that peoples' religiosity can be deeply affected by immediate events, such as September 11. More puzzling is that for a theory that depends on time as a key factor, they often pool their data, thereby washing out the variance in existential security of different generations within the same country (e.g., pp. 68, 235).

Perhaps the book's most interesting finding is presented in Chapter 6: What differentiates the West from the rest of the world, and most especially from Islamic societies, is the level of support for “gender equality and sexual liberalization” (p. 149), not support for democratic governance. Their survey data show that while there is a broadly similar level of public opinion support for democratic governance, there is a striking gap in attitudes directly affecting women and sexual mores.

The statistical analyses employed by the authors cannot do away with the glaring anomaly of high religiosity in the United States. According to their theory, the United States should, because of its high level of human development, rank low. It does not. Some inconsistency is evident in the way they try to do away with the anomaly by discrediting the statistics. In order to discredit the religious market theory, they point out that religiosity is still high in two countries where a religion had a virtual monopoly, Ireland and Italy (p. 94), but it is only for the U.S. case that they argue that the survey data are systematically inflated due to the “social desirability bias concerning churchgoing in American culture” (p. 91). However, one could say the same of Italy and Ireland. Further, given that the presumed bias only increases religiosity by a few percentage points, they turn to ad hoc explanations, such as immigration to the United States by Hispanics and Asians, suggesting that those immigrants are more religious than the rest of the population, and therefore distort the underlying trend toward decreasing religiosity. However, if that is the case, European countries should see similar effects. They do find a correlation between poverty and frequency of prayer in the United States, with the poor, on that measure, being “almost twice as religious as the rich” (p. 108, italics in original). Yet rich Americans still present an anomaly: 47% of them claim to pray daily.

Operationalization of some variables is problematic. For instance, Norris and Inglehart claim to debunk the part of secularization theory that holds that a “rational worldview” undermines religious beliefs (pp. 8–9, 67). However, they do so by measuring attitudes toward the benefits of science, not by measuring attitudes toward rationality in government, society, and daily life. European societies have experienced some of the worst excesses of scientific discoveries, and so it may not be surprising that they do not tend to think that scientific advances will help humanity (pp. 67–68).

The major failing of Sacred and Secular is that it wavers on two of its key assertions. First, it is ambiguous about the mechanism by which “existential security” affects individuals: Sometimes it is by way of diffuse communication within a society, sometimes by way of personal experience. Second, the authors caution that the socialization effect can be reversed by current events. In the end, readers may find the book frustrating, because it raises far more questions than it answers. There are so many country-, region-, and era-specific exceptions to their rule that one wonders at times if the theory is anything more than a statistical artifact. The book is, nevertheless, essential reading for those engaged in the debate about secularization theory and about religiosity throughout the world, and it invites further theorizing and research on these subjects.