The correlates of religion and state: an introduction

ABSTRACT

This volume examines the Religion and State Round 3 (RAS3) dataset which includes information of State Religion Policy (SRP) for 183 countries and self-governing territories between 1990 and 2014. This collection includes a general discussion of the dataset but most of the contributions take these data and use them to analyse the intersections between SRP and a wide range of phenomena including civil wars, terrorism, international conflict, attitudes towards religion and state, constitutional promises of religious freedom, and religious mobilisation. While this volume constitutes the formal primary presentation and analysis of the RAS3 dataset, it is intended to be the beginning of an ongoing discussion.

KEYWORDS:

• State Religion Policy
• Religious Freedom
• Religion and State Project

This volume presents the first comprehensive analysis of the Religion and State project (RAS3) Round 3 dataset, which covers 1990 to 2014 with data on multiple aspects of State Religion Policy (SRP). I discuss the details of the dataset in my contribution to this volume. In this introduction, I address the broader scope of the Religion and State (RAS) project and the essays in this volume.

In brief, this volume is intended to achieve two goals. First, to provide a basic presentation of RAS3 and a basic analysis of the trends in SRP over the 25-year period covered by the dataset. Second, to demonstrate the large variety of ways in which the RAS data can be used to address a broad spectrum of issues and question. Each contribution represents the work of a different author or set of authors approaching a wide variety of relevant sociological and political science research agendas.

A brief history of the RAS project

In order to assess the impact of RAS, it is worthwhile to briefly address the project’s history. I first conceived of the RAS project in 2000 and first began collecting data in late 2002. At that time, there was little cross-national data on religion other than religious demography variables. The few political science studies based on what data existed tended to focus on conflict (Fox 2002; Rummel 1997) and often focused on Huntington’s (1996) clash of civilisations argument rather than directly on religion (Henderson and Tucker 2001; Russet, et. al., 2000). Others focused on using survey-data in a cross-national context but were hampered by the fact that even the broadest surveys such as the World Values Survey covered only a portion of the world’s countries (e.g. Barro and McCleary 2003; Norris and Inglehart 2004). Also, few of these studies appeared before 2002. Grim and Finke (2006: 3) echo this sentiment:

Religion receives little attention in international quantitative studies. Including religion in cross-national studies requires data, and high-quality data are in short supply […] For the study of religion … [existing] data are not merely old, they are often lacking. This lack of data has hampered both research and theory.

In the past decade, this situation has shifted dramatically and the RAS project has contributed to this renaissance in cross-national empirical study of religion. The essential goals of the project have been consistent since its inception. First, to provide accurate data on SRP which can be used in quantitative analysis. Second, to use those data to provide an overview of the general worldwide trends in SRP over time. Third to use the data to examine the correlates of SRP. That is, what influences how governments deal with religion and how does SRP influence other important political, economic and societal phenomena? Fourth, to provide a resource for researchers interested in the empirical study of religion and state.

RAS1 covered 1990 to 2002 and included 62 core variables measuring various aspects of SRP. My work based on RAS1 focused on falsifying secularisation theory – the prediction that religion will either become weaker or disappear in modern times (Fox 2008). I also looked at related issues such as whether democracies really have separation of religion and state, whether countries in practice observe their constitutional clauses on religion (Fox 2011, 2012).

RAS2 expanded the data to cover 1990–2008 and the number of core variables to 111 types of SRP. My core work based on this round focuses on my secular-religious competition perspective, which argues that SRP is to a great extent determined by competition between secular and religious political actors. (Fox 2015) I discuss this theory in more detail in my contribution to this volume. My analysis of RAS2 continued to address the issues I addressed using RAS1 but I added a focus on the causes of religious discrimination (Basedau et al. 2017; Fox 2016; Fox and Akbaba 2014, 2015; Fox, Bader, and McClure 2017).

RAS3 expands the data to cover 1990–2014 and the number of core variables to 117 for 183 countries and self-governing territories. As I discuss in my contribution, these data allow for a more sophisticated and nuanced analysis of the competition between religious and secular political actors over SRP and reveal that this competition is one of the primary determinants of SRP.

The RAS project is not the only project to increase the data available to researchers on religion in the social sciences. Several of the authors in this volume have also collected data that contribute to significant progress in the field. Roger Finke produced the Association of Religion Data Archives (ARDA) dataset, which focuses on religious freedom, both societal and governmental (Grim and Finke 2011). Davis Brown produced the Religious Characteristics of States dataset which includes improved religious demographic information as well as basic information in SRP covering the 1800 to 2015 period (Brown 2017; Brown and James 2017). Luis F. Mantilla (2016) collected a dataset on Catholic political parties which he also uses in this volume.

The correlates of SRP

As I note above, my work on the correlates of SRP has focused on the issues of democracy, constitutionalism, separation of religion and state, and religious discrimination. However, much of the work on the correlates of SRP has been carried out by the larger scholarly community using the various iterations of RAS.

In parallel to my analyses of RAS1 and RAS2, other researchers used these data in combination with many different datasets to analyse a wide variety of political, societal, and economic phenomena. These research topics include but are not limited to democracy and democratisation (Cesari 2016; Driessen 2014), church–state relations (Grzymala-Busse 2015), repression (Sarkissian 2015) conflict, (Basedau et al. 2011; Brown 2017; Henne 2012), Terrorism (Saiya 2016), gay rights (Carlo-Gonzalez, McKallagat, and Whitten-Woodring 2017), women’s rights (Ben-Nun Bloom 2015; Htun and Weldon 2015; Sweeney 2014), religion and economics (Schneider, Linsbauer, and Heinemann 2015), migration (Kolbe and Henne 2014), globalisation (Ben-Nun Bloom, Arkian, and Sommer 2013), corruption (Sommer, Ben-Nun Bloom, and Arkian 2013), and social capital (Traunmuller and Frietag 2011).

This volume is intended to continue, strengthen, and expand upon this tradition. In each contribution, the author or authors use RAS3 in combination with additional data to expand our knowledge of the interaction between SRP and a number of important political and social phenomena. In doing so, many of them also provide a basic analysis of interesting segments of RAS3.

As most of the earlier political science-based cross-national studies of religion focused on conflict, and this type of study has remained popular, it is not surprising that several of the contributions focus on various aspects of religion and conflict. Basedau and Schaefer-Kehnert examine the influence of religious discrimination and other aspects of SRP on armed conflict in sub-Saharan Africa. They find that while sub-Saharan Africa has the lowest levels of religious discrimination of any world region, these levels increased between 1990 and 2014, especially in East Africa. This rise in religious discrimination has coincided with a rise in armed conflict especially theological and religious identity conflicts.

To analyse this relationship, they use RAS3 in combination with the UCDP/PRIO armed conflict data and two datasets collected by Basedau: the Religion and Rebels dataset (Basedau, 2017) and the RCDC dataset (Basedau, et. al., 2016). They find that religious discrimination is linked to theological conflict, which they define as ‘an armed conflict in which the warring factions differ regarding the role of the religion in the state.’ However, a more in-depth analysis including several case studies reveals that this relationship is more complex. In many cases, religious discrimination is part of ‘a wider context of problematic state-religion relations, which may fuel the rise of religious extremists.’ It also reveals that this discrimination-conflict relationship involves contextual conditions and indirect causal mechanisms.

Overall, this study has wide implications. It demonstrates that conflict mechanisms proposed in non-religious contexts also apply to religious conflicts. It also has implications for the wider literatures on the link between discrimination and conflict.

Brown examines the role of religion in interstate armed conflict and argues that neoclassical realism is the paradigm best able to explain religion’s role in international conflict. Within this context, religious identity can generate ingroup-outgroup differences. This can influence foreign policy, including intervention on behalf of religious kin in other countries. The involvement of religious issues in a conflict can make it more intractable. It can be used as a rallying point to strengthen commitment to the state. Also, religious beliefs can influence policy preferences.

Brown examines this relationship using the Militarized Interstate Disputes (MID) dataset in conjunction with RAS3. Brown finds a deep connection between SRP and MIDs. ‘Merely having an official establishment of religion is sufficient to raise the state’s propensity toward this outcome.’ In addition, state support for religion in isolation and a combined index of support for religion, regulation of the majority religion, and discrimination against religious minorities also predict a state’s first use of force. In fact, ‘when all three dimensions are combined into a single variable, that variable’s relationship is stronger than those of any of the individual variables’. This suggests that the three dimensions of state-religion work in tandem.’ This study is important because quantitative studies of religion and international conflict that do not rely on demographic or identity data are rare. Thus, these findings increase our knowledge of the influence of religion in the international arena as well as strengthen the awareness of religion variables in the international conflicts literature.

Henne examines the influence of government interference in religious institutions on terrorism asking whether government control and interference with religious institutions represses terrorism and extremism or increases grievances and, accordingly, has the opposite effect. Henne uses RAS3 combined with the Global Terrorism Database to address this question. Henne takes advantage of the ability of disaggregate RAS3’s composite indicators to focus on the extent to which the government funds religion and restrictions that are specifically targeted at religious institutions or clergy. He finds that restrictive SRPs result in more deaths from terror in a state. While there have been previous studies of the influence of religion on terror, many of them focus on whether religious identity or ideology influences levels of terror by terror groups (e.g. Henne 2012). Henne’s study adds to our knowledge of how SRP influences terror and finds that repressive policies often have the opposite of the intended effect. Thus, the best policy to limit religious terror and extremism is religious freedom. That being said, some specific SRPs do seem to inhibit terror. These include restricting minority clergy (such as chaplains) from military bases and providing funding for religious schools.

The rest of the contributions to this volume look at a broader range of the correlates of SRP. Mantilla examines the impact of SRP on the presence and success of political parties that use Catholicism to mobilise voters. Like Henne, he examines the impact of restrictive policies but uses a different type of dependent variable. Mantilla finds that laws restricting religious policies parties tend to increase Catholic-based mobilisation. Support for Catholicism has mixed results. Financial support for religion decreases this mobilisation but friendlier religion-state religions increases it.

Buckley examines the relationship between SRP and attitudes related to the place of religion in public life. He considers how all of the following influence SRP: public support for religion in government, public norms, variation of views within a population on religion in government, and regime type. Buckley combines RAS3 with survey data from the WVS programme to address this issue. This adds to a growing literature combining RAS3 with the WVS data on a wide variety of topics. These include the causes of religious conversion (Barro, Hwang, and Rachel 2010), individual religiosity (Fox and Tabory 2008; Muller 2009; Muller and Neundorf 2012), Western attitudes towards Muslims (Helberg 2014), and preferences towards social insurance (Scheve and Stasavage 2006). Other survey data have been used on conjunction with RAS3 to examine the influence of SRP on European integration (Grotsch and Schnabel 2012; Minkenberg 2009), social capital (Traunmuller and Frietag 2011), and national belonging (Trittler, 2017).

Yet, to my knowledge, this is the first study to test this relationship using RAS3. In fact, all of these previous studies look at how SRP influences some social or political attitude. Buckley looks at how attitudes towards religion in society influence SRP. Buckley also takes advantage of the ability of disaggregate RAS3’s composite indicators and focuses on two controversial aspects of SRP: state financial support for religion and laws on sex and gender. He finds that SRP and public attitudes towards religion are linked but that this plays out differently in democracies and non-democracies. Specifically in democratic countries when there is consensus between religious and non-religious individuals on SRP issues the government tends to follow that consensus. In non-democratic countries, SRP increases when there is a lack of consensus over SRP among individuals.

Mataic and Finke examine the discrepancy between constitutional promises of religious freedom and actual religious freedom. The general compliance gap literature, which examines other similar compliance gaps, focuses on economic development, governance structures, and global connections as explanations for compliance gaps. Mataic and Finke propose a ‘religious economy’ explanation as an alternative case of the compliance gap in religious freedom. The religious economy approach focuses on the competition between religious brands for religious consumers. ‘When the state and the dominant religion(s) form an alliance, the state will be under increased pressure to restrict the activities of the minority religions perceived as unwanted religious competitors.’ This pressure can come from religious institutions and clergy as well as general societal pressure. Accordingly, where societal discrimination and prejudice is present, there is more likely to be a compliance gap in religious freedom.

Mataic and Finke combine RAS3 with the RAS Constitutions dataset to examine the extent to which a compliance gap in religious freedom is present and assess its causes using a variety of data. This involves looking at 21 types of promises of religious freedom in constitutions and RAS3’s indexes for religious discrimination against minorities and the regulation of the majority religion. They found no support for economic development and global connections as explanations for the religious freedom compliance gap. Governance, as measured by free and open elections as well as an independent judiciary, were significant predictors for the gap between promises and practices but did not predict absolute levels of religious freedom. Societal discrimination was a strong predictor of absolute levels of government-based discrimination against religious minorities and gaps in compliance for the religious freedom of both minorities and the majority group.

Molle takes a different approach than the other studies by integrating RAS3 into a time series analysis of a single country, Italy. He asks the question of what influences levels of support for Italian Eurosceptic populist political party, the League Prima gli Italiani. While SRP does not directly influence voting for the League, the stable religion-state relations in Italy provide a political context in which nativist, anti-immigrant, anti-Muslim, and anti-EU bases are more likely to emerge. Molle adds to a young and vibrant research area on religion and the rise of populism that was explored in a special issue (46/3) of Religion, State & Society (DeHanas and Shterin 2018).

Conclusions

My contribution to this volume focuses on the competition between secular and religious forces in society. I argue that this is one of the most important aspects of religious politics. While several of the other contributions touch on this issue, all of them focus on other aspects of religion and politics. This range highlights the complexity and diversity of religion’s interaction with the political. Its impact is found in multiple and often crosscutting influences, many of which are significant in and of themselves. Yet when combined, these influences are multiplied.

The diversity of topics examined in this collection demonstrates not only the ability to apply the RAS3 data in multiple contexts, but also the increasing diversity of social science issues which have entered the collective research agenda. Each contribution provides a thorough discussion of the literature related to its chosen topic, which amounts to a broad and up to date review of the state of the cross-national quantitative literature on religion and politics as well as a review of the relevant theoretical and comparative literatures. This collection helps to both highlight and make understandable sense of the multifaceted and diverse relationship between religion and politics. While this volume constitutes the formal primary presentation and analysis of the RAS3 dataset, I consider it a contribution to an ongoing discussion rather than the last word. As was the case for past rounds of RAS, it is my hope that researchers will continue to use RAS3 as a resource to build knowledge on a wide variety questions and issues.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Jonathan Fox

Jonathan Fox is the Yehuda Avener Professor of Religion and Politics at Bar Ilan university in Ramat Gan, Israel. He has published extensively on various topics in religion and politics. His most recent books include An Introduction to Religion and Politics: Theory and Practice, Second Edition (Routledge, 2018) and The Unfree Exercise of Religion: A Worldwide Survey of Discrimination against Religious Minorities (Cambridge University Press, 2016).