The Constructivist Approach to Religion and World Politics

Review Article The Constructivist Approach to Religion and World Politics Religion and world politics is a subfield in transition. In the twenty years since political scientists rediscovered religion, scholars have struggled to advance the literature without a common paradigm. The once dominant paradigm, secularization theory, held that with economic development, religious beliefs and practices, religious organizations, and the integration of religion into other aspects of life would disappear. Of those three tenets of secularization theory, only the last one, differentiation of religion from other spheres, retains support among social scientists, and even that is contested. Samuel Huntington’s “clash of civilizations” paradigm is similarly doubted if not yet completely discarded.1 A field in transition must, therefore, address the question of how scholars of politics should approach religion after the failure of the dominant paradigms. This essay begins by explaining why secularization theory is largely discredited before reviewing the three leading approaches to rebuilding the literature, represented by recent scholarship on religion and world politics. These texts move beyond classic secularization theory, beyond the theory of a clash between culturally-rooted civilizations, and beyond appealing to scholars to “bring religion back in.”2 Instead, each work takes on the task of reconstruction. My goal in reviewing these approaches—constructivism, revised secularization theory, and religious economies—is not to set up a battle between competing paradigms. Instead, I want to highlight the innovations of the newest entry, constructivism, and suggest that, in contrast to earlier scholarship, scholars today should celebrate theoretical, methodological, and conceptual pluralism rather than aspiring to a unifying theory that will inevitably stumble upon the heterogeneity of religion. Learning from the shortcomings of the prior generation means recognizing that religion, like other aspects of culture and identity, is heterogeneous over time and space, multifaceted in practice, and its relevance to politics is dependent on context. New Approaches and Challenges What Went Wrong? The story is now a familiar one; the pioneers of the modern social sciences believed that religion was doomed.3 Weber said, “The fate of our times is characterized by rationalization and intellectualization and, above all, by the ‘disenchantment of the world.’”4 Even politicians leading religious movements claimed that religious rituals were doomed to disappear in favor of secular endeavors. In the words of David Ben Gurion, “Judaism as religious practice and tradition is Judaism of the ghetto. Judaism in a Jewish state is Judaism of labor and creativity in every field of economic and scientific endeavor, for all of man’s needs.”5 Not surprisingly, the first generation of social science studies of religion tracked these predictions. In his famous 1966 text, The Secular City, Harvey Cox noted, “The gods of traditional religions live on as private fetishes or the patrons of congenial groups, but they play no role whatever in the public life of the secular metropolis.”6 Likewise, in 1967 the renowned sociologist Peter Berger predicted that modernity would bring about a gradual decline of religious observance and religious organizations and the disappearance of the “sacred canopy,” which had prevented individuals from making decisions based on scientific knowledge, rational reasoning, and utility maximization.7 By the late 1990s, however, sociologists and anthropologists of religion had come to see the classic version of secularization theory as mistaken. Since some political scientists continue to support secularization theory, it is worth reviewing the reasons why our sister disciplines have largely repudiated it. Peter Berger’s repudiation of secularization theory was empirically driven. In an influential essay from 1999, he describes secularization theory as “essentially mistaken,” “wrong,” and “false.”8 While modernization has some secularizing effects on individual belief and social institutions, it has had many other effects as well, including the resurgence of religious identities, the adaptation of religious institutions, and a resultant pluralism both within religious traditions and within diverse societies.9 The only exception to the failure of secularization theory, he argues, is in Europe, where economic development heralded a decrease in religious belief, observance, and the influence of religious institutions. More critically, however, the sociologist Rodney Stark declares secularization theory to be wrong globally and in Europe. Drawing on an extended and detailed review of Europe’s Age of Faith, he demonstrates that the idea that Europe had a pious past is little more than nostalgia, given records of low church attendance and widespread ignorance of basic religious principles. Using 1800 as a benchmark, he shows that church membership in Britain is substantially higher today than in the past and that French Catholics are more participatory today than 200 years ago, and he cites Laurence Iannaccone’s reconstruction of church attendance rates to affirm that there are “no trends even vaguely consistent with the secularization theses.”10 Stark ends his review in dramatic fashion: “After nearly three centuries of utterly failed prophesies and misrepresentations of both present and past, it seems time to carry the secularization doctrine to the graveyard of failed theories, and there to whisper, ‘requiescat in pace.’”11 Sociologist Jose Casanova has proposed a modest preservation of one tenet of secularization theory. Of Weber’s three tenets of secularization—the decline of religious beliefs and practices, the privatization of religion and religious organizations, and the differentiation of the secular spheres (state, economy, science) from religion—Casanova argues that only the third remains valid.12 Anthropologist Talal Asad, however, levels a powerful critique of this tenet too. Once we recognize that religious actors are public, and that the public-private divide is itself historically constituted, it is impossible to confine religion’s influence to a sacred or private domain given that public religions are not ambivalent about economic policies, foreign affairs, or social relations. On theoretical grounds, then, Asad dismisses classic secularization theory.13 Asad’s argument is also borne out empirically. Using comparative quantitative indicators, political scientist Jonathan Fox demonstrates that most states, including modernized ones, are heavily involved in the regulation of religion rather than being differentiated from it.14 Instead of the world evolving toward a uniform secular modernity guided by rationality and organized around secular nation-states, most scholars now contend that the world is composed of “multiple modernities,” such that no single trajectory holds true for every society.15 And in a massively influential volume, A Secular Age, the philosopher Charles Taylor circumscribes the meaning of secularization to focus on the cultural conditions in which unbelief in religion is a viable option, and so belief and unbelief coexist uneasily. This condition, and not the demise or privatization of religion, captures the difference between the place of religion in the modern and premodern Christian West.16 Yet, the possibility of unbelief and the “multiple modernities” consensus does not provide much analytical leverage for political scientists trying to explain the role of religious actors and ideas in important outcomes like democratic transition, conflict, women’s rights and representation, economic development, or tolerance. Simply “bringing religion back in” does not explain how religious actors and ideas matter. And, while the numerous calls to “bring religion back in” are significant, they are ineffective without a commensurate set of theoretical tools. This impulse may also lead to a problem of falsely assuming the salience of religion in political outcomes. Instead of simply recognizing that secularization theory was wrong or paying attention to religious actors and ideas without a coherent theoretical framework, the three approaches reviewed here attempt to chart a new path. The newest approach, constructivism, draws on social theory and cultural anthropology to theorize secularism as an analytical category and explain how (religious) ideas and actors shape political outcomes. Since this is the most recent addition to the field, it will be reviewed first and in the most depth using the texts by Barnett, Shakman Hurd, and Rofesky Wickham, and then deployed to highlight shortcomings of other approaches. The “revising secularization” approach is represented by Norris and Inglehart. The “religious economies” approach marries the rational choice approach in political science with the economic sociology of religion and is represented here by Koesel. Since all three approaches have limitations, their strengths and weaknesses will be discussed in order to suggest ways for the literature to move forward. Rather than one eliminating the others, I suggest that that the best way for the literature to avoid the mistakes of the previous generations is to focus on building middle-range generalizations and explaining real world puzzles. Constructivism What is the constructivist approach to religion and world politics? While explicitly drawing on that tradition, none of the three constructivist books reviewed here define it. Nor do any of the major review essays of the past fifteen years in comparative politics and international relations, despite constructivism’s influence in both fields.17 So let me summarize it here by synthesizing the books’ arguments along with other prominent works from cultural anthropology and social theory. Five key themes appear. First, a non-exhaustive inventory of what constitutes religion includes individual and collective rituals, ethics, canonical texts, doctrine, methods of exegesis, everyday practices, governance practices, expert behavior, organizations, identities, artifacts, sites of worship and activity such as pilgrimage, and charismatic leaders, all of which change over time. That heterogeneity means that what constitutes religion is historically specific rather than universal. This basic observation has clear implications for researchers. For researchers, the proper theoretical beginning is the instituted practices (set in a particular context and having a particular history) into which practitioners of particular religions are inducted.18 For example, to understand how contemporary religious actors like the Egyptian Muslim Brotherhood view democracy, Carrie Rosefsky Wickham does not look to the Qu’ran or the writings of the founder of the Brotherhood, Hassan al-Banna. She looks at what members of the Brotherhood say about democracy. Rosefsky Wickham traces how the Brotherhood’s references to democracy mimic that of the Mubarak regime in the 1980s. She shows how a different set of discourses about democracy emerged in the early 2000s, when reformists sought a middle position between a defense of Islamic law and the importation of political institutions from the 564 West. And, of course, the meaning of democracy to Egyptian Brothers differs from its meaning for their counterparts in Tunisia, Jordan, Kuwait, or Morocco. In other words, Rofesky Wickham follows the encouragement of the theorist Gudrun Krämer: “It is not possible to talk about Islam and democracy in general, but only about Muslims living and theorizing under specific historical circumstances.”19 For other scholars, given the heterogeneous nature of religion, understanding the religious component of an actor’s worldview necessitates local and temporally specific investigation of their ideas. Second, what constitutes religion, doctrine, or theology is an outcome of political, legal, and religious struggle grounded in the production of knowledge. As Shakman Hurd persuasively states, “To define the boundaries of the secular and the religious is itself a political decision.”20 This is readily apparent in democracies like Indonesia, Bulgaria, and Romania, where there is “soft-separation” between church and state, and religion has a privileged status delineated by the state and accompanied by economic and legal benefits. However, it is equally applicable in ostensibly secular democracies like the United States, where in order to examine whether a given religious activity enjoys protection under the first amendment, the court must determine whether the underlying belief rises to the level of legal recognition; since the court cannot broaden the reach of free exercise to all claims of religious motivation, it legitimates some claims while rejecting others.21 Given that the definition of religion and its doctrines are contingent and intertwined with power relations, a universal definition of religion becomes impossible.22 This does not imply that religion does not exist or cannot be studied. Rather, it means that religion should be studied as a discursive and embodied tradition that seeks to instruct practitioners regarding the form and purpose of a given practice; believers’ discretion to regulate and uphold certain practices while condemning and excluding other practices is a sign of political struggle and the possession of power, and not a sign of truth.23 The constructivist approach investigates the conditions under which some aspects of a tradition become defined as correct while others are incorrect, rather than assuming at the outset that the content is constant or universal. Third, given that religion is an embodied and discursive tradition and given that its content is contingent, the interests of religious actors must be defined and understood locally. Certainly, religious actors may act strategically in order to maximize their utility. Rosefsky Wickham rightly notes that Islamist movements strategically adapt to their environment, but warns against the assumption that their interests are ranked within a well-defined and stable hierarchy.24 In that respect, “strategy” and “interests” must be understood within the local context in which a group’s preferences are generated in order to explain its behavior. Rational interests cannot be determined apart from local discursive traditions, political and economic structures, and embodied practices.25 Fourth, historical legacies shape the behavior of all contemporary actors, including scholars. For Michael Barnett, despite its claims to secularity, contemporary humanitarianism is predicated on categorical distinctions like civilized and uncivilized that are inherited from Christian missionaries. Historical legacies also shape IR theory rather than standing outside it. For Hurd, the limitations of IR theory when it comes to religion are a product of its origins in European traditions, including liberalism and secularization theory. This means that the challenges of studying religion in IR theory cannot be overcome by simply adding variables or building large datasets; IR theory lacks the categories of analysis necessary to explain behavior that is not premised on individualism and the division between public and private. The constructivist approach to religion, then, recognizes that political theory itself is a product of struggle and necessarily founded on categories determined by the victors of that process. Fifth and finally, social science concepts emerge from a specific (e.g., European, Catholic or Protestant) context; their utility for explaining political life in other cases must be investigated, not assumed. Daniel Philpott has made this case most forcefully in excavating the religious roots of the modern IR concept of sovereignty.26 Hurd demonstrates that the meaning of secularism and claims to the secular vary across time and space and should not be assumed to have a universal, static definition. Scholars of anthropology have made similar claims by investigating the meanings and practices of secularism in India, Egypt, and China.27 More broadly, the concepts of power, democracy, and agency have been shown to manifest differently around the globe.28 For scholars of religion and politics this means taking seriously the question of how social science concepts are understood on the ground, given that much of modern political theory is premised on the irrelevance or exclusion of faith from the public sphere. Hurd, Rosefsky Wickham, and Barnett put these tenets to good use. Hurd’s book argues that the secularist divisions between religion and politics are not fixed, as commonly assumed, but socially and historically constructed. By examining the philosophical and historical legacy of the secularist traditions that shape European and American approaches to global politics, she shows why this matters for contemporary international relations. The argument does the important work of uniting the IR literature on religion with postcolonial scholarship from cultural anthropology (especially the work of Talal Asad, Saba Mahmood, and Partha Chatterjee) and social theory (William Connolly, Charles Taylor, Edward Said, and Michel Foucault). Hurd’s book opens by defining secularism as a series of interlinked political projects underpinned by particular sensibilities, habits, and practices put into place by specific actors with interests, rather than something that emerges in a teleological manner from political development. In subsequent chapters the book gives flesh to this definition by outlining two varieties of the secular project: laicism and Judeo-Christian secularism. The laicist project sees religion as an adversary and an impediment to modern politics. It emerges from the Enlightenment critiques of religion by Immanuel Kant and others that portrayed theologies in public life as dangerous sectarianism that must be separated from the public realm of rational, deliberate argument. Kantianinspired IR theorists like Martha Nussbaum and Francis Fukuyama expand this public space to theorize the conditions under which a universal public realm will come into being. Similarly, leftists Michael Hardt and Antonio Negri delegate religion to the dustbin of history; such metaphysical traditions have no place in a world of capital and empire. Hurd demonstrates that the laicist project has had tangible manifestations in France, the Soviet Union, Turkey, and China. Its power is keenly felt in realist, liberal, and materialist approaches to international relations, all of which are predicated on the assumption that religion is either private or dying and that the Westphalian settlement marks the emergence of secular state sovereignty and the privatization of religion. Instead of taking the narrative of privatization, secularization, and sovereignty at face value, then, The Politics of Secularism in International Relations demonstrates that this set of discourses, sensibilities, and practices originated in Europe out of a particular set of historical circumstances and then laid claim to its own universality. The Judeo-Christian variant of the secular project emerges from a different set of discourses that treat modern politics as grounded in values and beliefs that originated in Christianity, or what later became known as Judeo-Christianity. It claims the secular as a unique Western achievement. While the laicist project assumes that religion is privatized or absent, the Judeo-Christian secularist project sees modern democratic states as dependent upon shared Christian, and later Judeo-Christian, values. Samuel Huntington picks up this view in arguing that Catholicism and Protestantism provide the bedrock for Western civilization. Likewise, Bernard Lewis contends that the separation of church and state is Western and Christian, but can be adopted universally. As with the laicist variant of secularism, Hurd shows how the JudeoChristian variant has tangible effects: this conception of secularism has delegitimized non-Western perspectives on religion and politics by insisting on foundational connections among particular understandings of Christianity, secularization, and democratization. Despite the empirical shortcomings of these discourses, both variants of secularism are powerful projects that shape IR theory and practice rather than standing outside of it. Both discourses of secularism, Hurd argues, are generated through opposition to Islam. Despite the shared theological, cultural, and geographic origins of JudeoChristianity and Islam, Hurd shows how prominent Enlightenment thinkers like Montesquieu used the image of “Oriental despotism” to justify western imperialism and celebrate European civil society. Montesquieu’s concept is influential in the writings of Marx, Weber, James Mill, and John Stuart Mill and serves as a negative benchmark against which Western civilization is celebrated. The effects of secularism are then demonstrated through a case study of Turkish non-accession to the European Union; Hurd argues that resistance to Turkish membership in the EU is not only a function of Europe’s attachment to a Christian identity but is also attributable to the difference between the European and Turkish variants of secularism.29 The effects of secularism are further demonstrated by showing how laicism and Judeo-Christian secularism are unable to describe the origins, interests, and goals of contemporary Islamist movements despite their being modern and middle class and originating in late-colonial interactions with European powers. Their illegibility, Hurd argues, is due to their abrogation of the secular-liberal separation of religion and policies; that separation is endogenous to IR theory, providing an epistemological stumbling block to understanding how contemporary Islamist movements could be anything other than a threat to modern politics. Hurd’s book is a crucial contribution to the new literature on religion and politics. It forces the field to rethink the relationship between secularism and religion, and it complicates the meaning of secularism by showing how it varies and is manifested. Most importantly, it demonstrates that a provincial, normative distinction between religion and politics is built into the categories of analysis in IR theory. This a priori distinction impedes understandings of important world events like the religious revival of the late twentieth century due to an unwillingness to recognize that the boundaries between religion and politics are actively contested, continually renegotiated, and demarcated as a product of power relations. At first glance, Michael Barnett’s Empire of Humanity would seem to be an odd choice for a review essay on religion and politics. Yet, as he notes in the introduction, humanitarianism is entwined with religious actors and religious ideas: Religious discourses continue to motivate, shape, and define various dimensions of humanitarianism. The importance of religion in this book is evident not only in its centrality to the narrative but also in the allegories, concepts, and metaphors that I use; it is extremely difficult to write about humanitarianism without falling under the sway of religious iconography.30 The value of Barnett’s book, then, is not that it deliberately seeks to contribute to the new literature but that its organic analysis does not foreclose the possibility that the residues of an explicitly sacred past affect the present. Barnett argues that humanitarianism is a public, hierarchical, and institutionalized project of reform dedicated to the liberation of individuals from suffering through an expanding conception of community. Liberating individuals involves progressively expanding one’s understanding of community from one’s neighbor, to one’s fellow citizens, to all those within a country’s borders, to co-religionists abroad, to potential coreligionists abroad, to anyone suffering abroad from a common enemy, to anyone simply because they are human. Barnett divides his history into three periods. The first, imperial humanitarianism, has its roots in the abolitionist movements in Britain in the late 1700s, when Christian missionaries and lay leaders organized campaigns against slavery on the grounds that it was unchristian. Likewise, missionaries worked alongside colonial rulers to save souls and civilize non-Europeans. After WWII, states began funding disaster relief to both alleviate suffering and advance their foreign policy goals. This second phase of humanitarianism contains the residues of the past; the target is suffering and the impulse is paternalistic, but within an increasingly secular discourse. In the third age, weak states come onto the international agenda alongside terrorism and ethnic conflict as major causes of suffering, demanding more than food aid to resolve. Aid agencies are also increasingly under pressure to measure their effectiveness, but this raises questions about their goal. If development is their goal, then bags of rice may be less effective than improving infrastructure. But if the goal is to alleviate suffering, as Barnett argues is the case, then a bag of rice is success. Barnett’s contribution is showing that humanitarianism is a product of faith in moral and human progress and the belief that there is a common human community. Such a belief is indebted to the influence of Christian missionaries despite the absence of a supernatural God in the new language of secular, liberal humanitarianism. This is the significance of Barnett’s analysis to scholarship on religion and politics after secularization; faith matters in places and ways that are invisible without mapping the genealogy of the present and remaining open to the possibility that that which proclaims adherence to an ethos of secular universality may be otherwise. Carrie Rosefsky Wickham’s book, The Muslim Brotherhood, explicates how the participation of Islamist actors in formal politics has impacted their world-views, goals, and organizational practices. While groups like the Muslim Brotherhood entered the political system to change it, she suggests that, paradoxically, they ended up being changed by the system. One of the most striking changes that occurred in the Brotherhood was a growing emphasis on democracy as a legitimate means of governance and on the expansion of public freedoms as the movement’s highest immediate priority. In that respect, the Muslim Brotherhood is similar to Stathis Kalyvas’ account of early Christian democrats of Europe who moderated their policies as a result of inclusion in the political process.31 Yet, in contrast to Kalyvas’s account and drawing specifically on constructivist theory, Rosefsky Wickham suggests that defining organizational preferences in terms of exogenously determined self-interest is problematic as it fails to recognize that religious organizations are often fractured by competing factions with different conceptions of the organization’s interests. Rosefsky Wickham argues that understanding religious actors means examining the forces endogenous to religious institutions that shape group behavior, including internal fragmentation, exclusion, and the competition for primacy. For example, those Islamists in Egypt that broke out of the insular networks of the movement and began to interact with other civil and political actors on a regular basis tended to embrace pluralism. Likewise, and simultaneously, those Islamists whose formative political experiences took place in prison and who made sacrifices for the Brotherhood were similarly shaped by their experiences. It was this old-guard faction, in concert with the group’s new political arm, who called the shots in the Egyptian Brotherhood after the collapse of the Mubarak regime. By contrast, in Tunisia, figures associated with the reformist wing of al-Nahda have shaped the party’s agenda since Ben Ali’s departure. al-Nahda’s top leaders stressed their desire to govern in coalition with other parties and expand employment opportunities for women, and the party was one of the first to support the alternation of men and women on party lists. Yet, the progressive policy positions articulated by al-Nahda’s senior leadership did not reflect a consensus within the party at large; what sets the Tunisian case apart is the dominant position of reformists at the apex of the party. Rosefsky Wickham shows us that the world’s most important actors have approaches to politics that are often hesitant, ambivalent, and disjointed. Without exploring what goes on within the black box of religious movements, including their temporally and intra-organizationally varied understanding of interests, we cannot understand the implications of political participation for religious movements. The constructivist approach as delineated here is an advance for a literature struggling to think beyond secularization. Yet, there is work to be done. First, all of the authors demonstrate that religion as an object of study is extraordinarily multifaceted. Given the constructivist commitment to religion’s inherent fluidity and heterogeneity, scholars working in this tradition must grapple with the question of whether or not there can be a consistent approach to religion, let alone a generalizable argument about its effects. Hurd’s most recent work contends that there cannot: “Religion is too unstable a category to be treated as an isolable entity, whether the objective is to attempt to separate religion from law and politics or design a political response to “it.”32 Rosefsky Wickham and Barnett, however, try to bypass this problem by focusing on selfproclaimed religious actors and their behavior. Such a strategy is advantageous for explaining behavior but invites the question of whether a religious actor is concurrent with religion. Second, there is no single constructivist approach to religion. Clifford Geertz, Pierre Bourdieu, Max Weber, Michel Foucault, and Ludwig Wittgenstein are all typically categorized as constructivist yet differ in their definitions and analysis of culture and ethnicity. A Foucaultian approach to religion, for example, would investigate the origins and effects of fields of knowledge about religion.33 A student of Wittgenstein, meanwhile, might focus on how inherited discursive and embodied practices constitute religious traditions apart from any external definition.34 A student of Bourdieu might focus on the production of religious and political fields and their effects on social relations.35 Nor is there always a single approach from any one theorist. A student of Geertz might draw on his early work on culture as a system of symbols or his later work on culture as performance.36 Such diversity is not unique to the constructivist tradition, but it is challenging for those building an alternative to the secularization and religious economies approaches. Third, while constructivists have disaggregated religion, they have yet to do so with the foremost modern political actor: the state. Political parties, courts, legislatures, executives, and bureaucracies all interface with religion differently, yet constructivists have not explored the content and implications of different modes of religious governance. Scholarship on religion and politics has moved beyond the normative question of whether religion should be in the public sphere to the causal question of how religion and religious actors shape state policy, nationalism, human rights, and democratization. How do sub-state political institutions—international organizations, the executive, the courts, the legislative and political parties, and the state administration—govern differently? What are the implications of religious governance for human rights, public health, economic development, corruption, and democratic consolidation? These questions merit consideration in future work if this approach is to gain traction. Nonetheless, Hurd, Barnett, and Rosefsky Wickham are to be commended for developing a new approach to the old question of how religion (and secularism) matter in world politics. Revising Secularization Rather than discard secularization theory, some scholars have sought to revise it. Pippa Norris and Ronald Inglehart argue in Sacred and Secular that the traditional secularization thesis needs to be broadened to examine attitudes outside of Europe and data beyond Stark’s measures of church attendance in order to see global trends. They argue that secularization persists among secure post-industrial nations. Meanwhile, religiosity persists and is even growing among vulnerable populations: “People who experience ego-tropic risks during their formative years (posing direct threats to themselves and their families) or socio-tropic risks (threatening their communities) tend to be far more religious than those who grow up under safer, comfortable, and more predictable conditions.”37 They contend that the advanced societies have declining birth rates while the religious ones have high fertility rates, leading to a growing gap between the two groups, but not to the total failure of secularization theory. Norris and Inglehart also seek to narrow secularization theory. Whereas the classic version held that rational calculation and scientific knowledge would erode religiosity, they find the reverse: “Societies with greater faith in science also often have stronger religious beliefs.”38 (The authors offer no theoretical explanation for this finding; see below for my interrogation of this ostensible paradox.) Whereas the classic version held that the trends were global, they argue that the “distinctive worldviews that were linked with religious traditions have shaped the cultures of each nation in an enduring fashion.”39 Finally, whereas secularization theory predicted a steady decline in religious belief alongside economic development, the authors restrict the trend to postindustrialized societies; agrarian and industrialized societies are said to retain their traditional values. Norris and Inglehart marshal a prodigious amount of data, including survey data from almost eighty countries, other indicators from 191 countries, and time series data from 1980 to 2001. Their empirical chapters demonstrate that using the indicators of religiosity specified above, group means are different across agrarian, industrial, and postindustrial societies. They show that disaggregated measures of development, including income inequality, urban population, illiteracy, education, access to mass communication, AIDS cases, infant mortality, access to water, immunization, number of doctors, and population growth, are correlated with religiosity in the direction predicted. Then, focusing first only on European cases and then only on industrialized cases, the authors show a consistent trend toward decreasing church attendance. Their analysis of different types of societies shows a clear decrease in religious participation by birth cohort in postindustrial societies, but not in agrarian or industrial societies. To explore the puzzling persistence of high levels of religiosity among postindustrial Ireland, Italy, and the U.S. and the increased religiosity in Japan, the authors test the argument, articulated by proponents of the religious economies school, that high levels of religiosity are a product of the competition between churches. They find “no support to the claim of a significant link between religious pluralism and participation.” 40 Likewise, they use pooled surveys from 1995 to 2001 to examine whether there has been a resurgence of religiosity in post-Communist Europe. Again, the authors find that any changes are in the direction predicted by their theory and against the religious markets theory. Chapters eleven and twelve are new to the 2011 edition and use new data to reaffirm the importance of existential security, including in the former Communist countries. Sacred and Secular has received a great deal of attention, including some criticism. First, scholars worry that aggregating survey data to produce a single data point for a national mean can seriously distort the underlying distribution of the data.41 If the data are based on individual-level indicators, then the analysis should focus on the individual level rather than national means. In other words, if no test is made for the relationship between individual religiosity and individual existential security, then the converse proposition, that those with low levels of existential security have low levels of religiosity, can not be rejected. Second, some of the categorization is surprising. The cases of postindustrial societies are limited to the top twenty-three most affluent states and societies in ranked order, which means that all but one, Japan, are European. This is not a convincing way to assess whether secularization theory is valid beyond Europe. Moreover, in the logic of the theory, a ranked indicator makes less sense than one based on the dominant type of economic production. This raises the question of whether the categorization was tweaked to fit the theory; ostensibly post-industrial states with high levels of religiosity, like Greece, South Korea, Portugal, and the Czech Republic, are categorized as industrial. Third, despite defending secularization theory, the book does not discuss why religious institutions endure in our secular age, nor does it assess the privatization of religion or religious organizations, nor does it investigate the differentiation of the secular spheres from religion. Despite claims to being comprehensive, then, the text does not assess two of the three pillars of Casanova’s influential theory of secularization. The constructivist literature suggests two additional concerns. Where are history and the institutions that shape individual preferences? For example, the theory of existential security fails to explain low levels of religiosity in the world’s largest, by population, country, China; more plausible than the secularization thesis is that the Cultural Revolution and policies of the Communist Party have shaped public attitudes. Likewise, why has religious engagement in the Muslim world been so much greater since the 1970s than in the 1950s and 1960s? The Islamic Revival, arguably the most powerful religious and political movement of the twentieth century, cannot be explained by the existential security hypothesis but can be readily explained by the existence of transnational ideas and social movements.42 While the authors point to religious factors to explain the lack of political development in the Muslim world, scholars of the Middle East like Michael Ross contend that the region’s underdevelopment has everything to do with the democratic obstruction that accompanies oil-based economies.43 Finally, the constructivist literature suggests that despite claims to universality, the book’s origins in the European Values Survey mean that some of the indicators render religious practices elsewhere invisible and their implications inexplicable. For example, the dependent variable is measured using individual and collective religious participation defined in Christian terms: religious values (i.e. the significance of God in one’s life) and religious beliefs regarding heaven, hell, and the existence of a soul. The authors recognize the limits of these indicators for Buddhism, Confucianism, and Shintoism, for which frequency of attendance in public ritual is a poor measure of religiosity. Likewise for animist and folk religions that are not organized around church attendance or doctrines directing individuals to think about God, heaven, or hell. Yet, the book refrains from developing other measures or restricting the theory to Christian societies. This is despite empirical outcomes that go unexplained, like the correlation between high levels of religiosity and support for scientific progress in Muslim societies. The authors’ unacknowledged Christian conception of the relationship between science and religion leads them to falsely assume this tension exists in other faiths despite the fact that Islamic scholars generally do not see any tension between scientific progress and faith, since both come from God.44 So, how should scholars working in the tradition of the revised secularization theory rebuild the field after the failure of the dominant paradigm? Among the books reviewed here, Norris and Inglehart’s task is both the most straightforward and the most difficult. Revising secularization theory demands better specification of the conditions under which individuals, congregations, societies, and nations decrease their religious beliefs, practices, and memberships and of how religious organizations wither away. While survey data may be useful for beliefs, they tell us little about the practices of congregations, states, or other social organizations. Given that the underlying hypotheses are psychological as well as social, it would also help to better develop the causal mechanisms by which greater existential security would lead to a secular worldview. Theoretically, the authors are heavily rooted in the European case even as they draw on an increasingly large dataset. Given that the problems of secularization theory, especially the inability to account for religious revivals in the nineteenth and twentieth centuries, are due to an enduring academic Eurocentrism, it would behoove proponents of secularization to explore the meaning and trajectory of secularism in Asia, Africa, and Latin and South America using other data. It would also be helpful to follow the literature in sociology and anthropology and gain greater distance from classic secularization theory; although the authors recognize that economic downturns and state policies affect religiosity, their theory is still deterministic: “our theory predicts that the importance of religion in people’s lives will gradually diminish with the process of human development.”45 Sacred and Secular is most convincing in its explanations for declining religious observance in Europe and the rise of the welfare state; beyond the borders of Europe, however, its explanatory power is limited. Religious Economies The most conspicuous alternative to the new secularization theory is the religious economies school. More than the Islamic revival or even the attacks of September 11, 2001, this approach has helped bring religion and politics into conversation with the rest of political science. Anthony Gill’s 2001 article in the Annual Review of Political Science was the first major article to call for bringing religion back into comparative politics (Daniel Philpott did the same for international relations in 2000).46 Given that International Political Economy and Comparative Political Economy are often seen as a single field, the religious economies approach has the potential to revolutionize the way that scholars approach religion. Yet, a review of recent work raises hard questions about the wisdom of replacing secularization theory with another, equally grand approach to studying religion. Originating in the work of the economist Laurence Iannaccone and the sociologists Rodney Stark and William Sims Bainbridge, this school seeks to develop a general theoretical understanding of religion on the basis of microfoundations of individual choice. The baseline assumption is that religious people find value in believing in a religious creed because it provides comfort through ideas about salvation and peace of mind.47 Individual consumers of religion seek to maximize their spiritual satisfaction through interactions with producers of religion. Consumers and producers then interact within a marketplace that is defined by competition between firms (churches) and by the government, which regulates the entry and exit of firms in the marketplace.48 Religious economies accounts posit that religious actors have fixed and clearly defined preferences and that they will select strategies for behavior that maximize their utility within given constraints.49 Iannaccone and Gill credit Adam Smith for laying the foundation for the economics of religion. The Wealth of Nations contains a lengthy discussion of the role of religious institutions in education and the factors that determine the behavior of clergy. The purpose of the clergy “is not so much to render the people good citizens in this world, as to prepare them for another and better world in the life to come.”50 Selfinterest motivates clergy just as it does secular producers, and market forces constrain or motivate churches just as they do secular firms. As a result, the benefits of competition, the inefficiencies of monopoly, and the problems of government interference in the market shape religious actors’ behavior. Smith makes other distinctions that continue to shape the religious economies literature. Religions fall into two categories: the strict or austere ones popular among the common people and the liberal or loose ones adopted by elites. Clergy’s interests are to maximize their authority with the population, which is obtained by convincing the population of the certainty of their doctrine.51 In an impressive account that merges deep area knowledge with the religious economies framework, Karrie Koesel uses this interests-based theory to explain the behavior of religious actors in Russia and China. Koesel begins with the potent observation that 80 percent of the Russian population identifies as religious and there are approximately 300 million Chinese that identify as religious, a group four times larger than the membership of the Chinese Communist Party. Yet, the political significance of religious ideas and organizations is undertheorized in both cases. This is an especially pressing concern for students of authoritarianism, since Koesel argues that “Religion and the authoritarian state represent competing centers of authority.”52 She observes that religious actors in Latin America, Eastern Europe, and Asia played prominent roles in the democratization struggles during the Third Wave. Religious communities have strong associational ties outside of the authoritarian states, contain dense networks that cut across ethnic, geographic, and linguistic cleavages, and have transnational ties, resources, places to meet, regular donors, and charismatic leaders. In other words, religious groups have a vast network of resources that can be used to undermine authoritarian regimes. So what are the consequences of growing religiosity in authoritarian regimes? Koesel’s answer is backed by lengthy, impressive fieldwork in two difficult languages. Against the assumption that religion and secular authority are in competition, she finds that cooperation is far more prevalent than one would expect. Religious and local government actors frequently form partnerships in order to improve governance, secure protection in an uncertain political environment, and gain access to much-needed resources. Both sides seek stability and behave pragmatically in advancing their interests, which are defined as being rooted in financial needs, their need for power, and their desire for prestige. Koesel’s empirics demonstrate grounding in local politics in both cases. For example, a Muslim community on the outskirts of Shanghai welcomed money from the local government to repair its mosque, in return for developing the site as a museum and tourist attraction that brought in revenue to the surrounding businesses. The same local government also built a temple for an unrecognized religious cult in order to attract tourists from Taiwan. A Buddhist temple in a commercial district of Shanghai borrowed money from the local government to pay for much-needed repairs after the Cultural Revolution and in return transformed part of the temple into a tourist site and commercial space whose rent goes directly to the local authorities.53 These examples highlight how political elites and religious leaders collaborate to advance their mutual interests. Despite the assumptions of conflict, then, the reality on the ground is often cooperation. One of the strengths of Koesel’s account is that it does not draw on the full religious economies approach, which, as she makes clear, stumbles in explaining behavior in religious markets that are neither perfectly open nor perfectly closed.54 A 2012 review essay by Anna Grzymala-Busse further critiques the religious economies approach on the grounds that religious doctrine shapes individual political views and doctrinal differences are more sticky or salient than other identities. Grzymala-Busse contends that doctrinal differences are also more potent than other identities because “religion is concerned with the supernatural; everything else is secondary.”55 It is on these grounds that she suggests the religious economies school may overstate the degree to which consumers of religion will shop for another faith. While church shopping is an accepted practice in American Protestantism, it is unclear that this is common in any other religion. This is why Grzymala-Busse levels the compelling accusation that, “without a clearer appreciation of doctrine (other than as strict practice) or how it matters, the political economy literature operates within unacknowledged boundary conditions of American Protestantism.”56 Grzymala-Busse then mounts an empirical charge against the religious economies literature: changes in the structure of religious markets in the post-communist world have not led to the predicted increases in religious observance. Grzymala-Busse’s goal in these critiques is to offer an alternative approach to the new secularization and religious economies explanations, grounded in doctrine and the relationship between church and state. Drawing on the history of Ireland and Poland, she argues that a close alignment of national and religious identities may increase both religious observance and state regulation, an explanation that runs counter to the religious economies school. Conversely, where the nation (church) historically opposed the state, secularism is more likely than high levels of observance, as in the cases of France and the Czech Republic. This historical and institutionalist account helps to explain why churches’ influence can be so powerful even while popular majorities disapprove of their involvement in politics. These critiques of the religious economies school raise hard questions about its status as a viable alternative to secularization theory. That said, the constructivist literature suggests that Grzymala-Busse’s account misses its own carefully specified target by relying on a definition of religion, via doctrine, that is as rigid as the accounts she laments. Grzymala-Busse’s definition of religion, “a public and collective belief system that structures the relationship to the divine and supernatural,” originates in Geertz’s work on culture as a semiotic system, which is taken up by the sociologist Christian Smith and borrowed by Gill.57 Like Smith and Gill, Grzymala-Busse posits a coherent, unitary, and temporally consistent object called religion. As explained above, this definition is problematic on at least two grounds. First, defining religion as a belief system ignores the fact that religion as a social-scientific category is grounded in the political production of knowledge. Grzymala-Busse says that religious claims are absolute and non-negotiable: “Because it is a belief system that cannot be disconfirmed, the claims of religion on politics can be absolute and irrefutable. . ..”58 Furthermore, because doctrine is assumed to be durable, potent, concerned primarily with the supernatural, and because doctrine demands a form of loyalty that is oppositional and in competition to states, Grzymala-Busse assumes it is distinct from politics: “Because it is public and collective, religion and its claims continually come up against politics, and vice versa.”59 Yet, Grzymala-Busse’s account of what religion is (the definition), what religion does vis-à-vis politics (opposition), what religion claims to be (absolute and supernatural), what religion wants (agency), and who religion represents (transnational communities) runs counter to decades of research in anthropology and sociology; scholars of religion have long demonstrated that doctrine is highly negotiable, mutually constitutive with secular power rather than separate from it, highly fluid over time rather than rigid, often concerned with everyday practices rather than the supernatural, and as much complementary to the demanded loyalty of the state as subversive of it. One of the central findings of Koesel’s book is the irrelevance of doctrine to the behavior of religious actors: “spiritual matters only peripherally enter the negotiations and only then among select religious actors. Instead, the bargaining games typically evolve around pragmatic political issues as each side tries to maximize money, power, and prestige.”60 All this leads to a second problem with this definition of religion: a universal definition of what religion is or what religion does is not feasible. Asad’s influential book, Genealogies of Religion, makes this point best: “There cannot be a universal definition of religion, not only because its constituent elements and relationships are historically specific, but because that definition is itself the historical product of discursive processes.”61 What constitutes religion in a given time and place is an outcome of ongoing struggle and cannot be assumed to be consistent with what religion is at other junctures. This means that the common definition of religion developed by Stark and Bainbridge, advanced by Gill, and adopted by Grzymala-Busse merits reconsideration in light of the research by scholars working in the constructivist tradition. Conclusion It is an exciting time to be a scholar of religion and politics. All of the work reviewed here is pushing beyond the flawed paradigms of the past toward new and meaningful understandings of the power of religion in world politics. This work is also looking beyond Western Europe, based on the implicit or explicit recognition that the European experience maps poorly onto the development, significance, and complexities of religion and world politics. All three approaches offer a useful set of tools to scholars trying to rebuild the literature. The revising secularization approach builds on decades of survey research and scholarship to explain the uncoupling of church and state and declining church attendance in Europe, and the increasing number of people that profess no world religion. The strength of this approach is its ambition and intuitive appeal; the change from agrarian and industrial economies, directed by authoritarian regimes, to postindustrial economies, backed by representative democracies, undoubtedly has effects on culture. Those effects, however, need to be better specified, with more attention to variation and non-European experiences, and put into conversation with institutions if this approach is to endure. It would also behoove proponents of secularization to take seriously the contemporary literature on the sociology of religion, rather than just the field’s founders. The religious economies approach draws on powerful assumptions from rational choice theory and methodological tools from economics to explain the persistence of religiosity in some post-industrial states and the behavior of religious organizations (firms) elsewhere. The strength of this approach is its deductive method, theoretical parsimony, and integration with political economy work in other fields. The challenge for this approach is to go beyond the observation that religious actors are competitive and have interests to more novel findings that draw on empirical studies beyond Europe and Christianity, and to build conceptions of religion that travel beyond American Protestantism. The chief advantage of the constructivist approach is that it avoids the problems of the other two by recognizing the internally and geographically heterogeneous nature of religion, placing any analysis of religion’s influence in time and context, questioning secularism, and recognizing that religion exists within a matrix of power relations, which is shaped by the modern state. Drawing on cultural anthropology and social theory, the three constructivist books reviewed use that understanding to explain the moderation and immoderation of the Muslim Brotherhood, the origins and transformation of global humanitarianism, and the power of secularism in shaping IR theory. Other scholars working in the constructivist tradition have explained the roots of sectarianism and its utility for authoritarian regimes,62 the substance of religious nationalism and the reasons for its evolution,63 how historical legacies shape minority integration,64 and how state formation is generated in spaces of disorder.65 Clearly, constructivism is an approach that travels. Yet, constructivism’s strength is also a constraint; its attention to heterogeneity, context, and local conditions means authors may eschew universal generalizations in favor of middle range theory. In an age of failed theory, however, that might be just what the field needs. Under what conditions are sectarian identifications likely to emerge, then diminish? How are transnational affections manufactured and maintained in the digital age? How do subnational political institutions govern religion? What are the implications of religious governance for human rights, public health, economic development, corruption, and democratic consolidation? How might the religious virtues of brotherhood, generosity, and opposition to greed be harnessed in the political struggle against economic inequality? How can public religious organizations help strengthen social solidarities and democratic political institutions? Constructivists are well placed to address these questions and further contribute to the revival of scholarship on religion and politics.