Keywords

Nested cases (at the opposite ends of a negative-to-positive scale) were selected from the quantitative database (Appendices 1–4 and Supplementary Materials) for more in-depth empirical (i.e. within-case) analysis. Four countries were compared: Switzerland, Uruguay, Cuba, and Colombia. These cases are also linked to the correlated variables in the quantitative model and to the sufficient conditions in the QCA.

Cases were selected based on the “extreme case method”. This consists of a selection of cases exemplifying extreme or unusual values of one of the variables of analysis. This approach is appropriate for exploratory research (Gerring & Seawright, 2008). The QCA cases (Sect. 16.3.3) already mentioned some of the variables that characterise these countries as extreme. Section 17.5 further illustrates and compares some of the variables considered. The cases also analyse within-country variations by region (whenever possible) (i.e. more Catholic versus more Protestant areas in Switzerland, Uruguay, Colombia). Such analysis may also be a source of inferential leverage. This approach combines within-case and cross-case causal inferences to elucidate the causal mechanisms giving rise to empirical regularities (Goertz, 2017). To offer the broadest possible comparability, multiple case studies were conducted rather than a single in-depth case study.

1 Extreme Positive Case (Worldwide): Switzerland

Switzerland has the highest competitiveness score in the world (World Economic Forum, 2016), one of the lowest homicide rates (0.6 (UNODC, 2013)), and one of the lowest corruption indices in the world (85 (Transparency International, 2016)). The country also enjoys widely acclaimed political, economic, and social stability. Its population and cantonal (i.e. federated) system exhibit a mixed distribution of Roman Catholics and Protestants, making it an ideal case to analyse and compare the historical influences of these denominations. Furthermore, the author developed this research in Switzerland, which facilitated accessing and gathering primary and secondary information.

2 Extreme Positive Case (Latin America): Uruguay

Uruguay has the highest social progress score in Latin America (Porter et al., 2015). Besides Chile, it is also one of the most transparent countries in Latin America (Transparency International, 2016) and one of the continent’s most secular countries (Pew Research Center, 2014). These facts make Uruguay a worthwhile case for examining whether the factors that made Uruguay less religious have also made the country less corrupt and more equal. Consequently, this case provides vital information for understanding the interrelations between the variables under study.

3 Cuba: A sui generis Case (Communist Proxy)

Cuba is the only Communist country in the Americas. Communism and secularism are still in force in a country of contrasts: Cuba has one of the lowest homicide rates in the Americas (4.2 (UNODC, 2013)) but also one of the continent’s lowest social progress indexes (Porter et al., 2015).

The Cuban government actively suppressed religion for decades. However, since the fall of the Soviet Union, the government has increased religious liberty by opening up political space for religious belief and practice. This opening has resulted in a dramatic flourishing of religious life in recent years (Goldenziel, 2009). Further, Cuba might also serve as a proxy of former socialist countries that have tried to eradicate the influence of religion on the state through adopting Marxist ideology. Cuba currently faces corruption levels comparable to former Soviet countries (Transparency International, 2016).

4 Extreme Negative Case: Colombia

Colombia is a proxy country in Latin America. It is characterised by pervasive Roman Catholicism (i.e. with a valid Concordat, and one of the highest proportion of adults raised as Roman Catholics worldwide (Pew Research Center, 2014)). Other dominant features include high levels of corruption (Transparency International, 2016), and one of the world’s lowest equality (World Bank, 2016) and safety rankings (World Economic Forum, 2017). Moreover, the author’s first-hand experience of working in government, civil society, and international agencies in his native country provided direct insight (along with his extensive network).

5 Summary of Case Selection Criteria

Each of the selected countries serves as a proxy of a larger group of countries (Latin American Strong Catholic, Secular, Communism, Protestant or mixed Old World.). Comparing Switzerland and Latin American countries illuminates how religion and prosperity indicators evolve differently within diverse political systems and against the background of different historical roots.

Nested (extreme) cases were chosen for more in-depth empirical, quantitative and QCA analysis (see Supplementary Materials). Table 17.1 summarises the four cases.

Table 17.1 A qualitative comparison of the four case studies
Full size table

6 Methods and Data Treatment

The qualitative approach adopted here concentrated on which patterns in particular (Silverman, 2005, p. 9) might trigger the observed effects, and hence complemented the analysis of empirical correlations and QCA. The aim was to gain a greater understanding of the research problem (Zikmund, 2003, p. 111), hence complementing the statistical and QCA analyses of the previous chapters, even if this approach implies that matters are “subjective in nature” (Zikmund, 2003, p. 132).

6.1 Qualitative Method: Critical Discourse Analysis (CDA) and Religion

Discourse analysis is central to this study, mainly because discourse, knowledge, and power are closely related (Burr, 2003, p. 67; Carabine, 2001, pp. 267–268). One especially fruitful type of discourse analysis was established by the French philosopher Michel Foucault. He suggested examining the interconnections between discourse, knowledge, and power based on his key concept of “genealogy”. This, as we will see, is related to institutions and prosperity.

Foucault (1972) developed an elaborate conceptual framework for discourse analysis and proposed a rigorous methodological system to underpin such analysis. However, he did not provide a “how to guide” to genealogy. Hence, methodologies and applications vary (Carabine, 2001).

Importantly, CDA cannot be applied mechanically (Hjelm, 2014). Nor does any universal CDA methodology exist (Fairclough, 2010, p. 6). On the contrary, every analytical approach to discourse requires specific design (Hjelm, 2014, p. 860). Oswick (2012) shares similar views and sees CDA as the only discursive method for considering ideology and power. CDA bridges cultural and linguistic studies in a critical approach. Thus, it enables researchers to adequately examine the role of religion in creating, transforming, reproducing, and sustaining inequalities (Hjelm, 2014, pp. 855–58).

Equally importantly, CDA is ethically committed to exposing the processes through which discourses disseminate social structures that perpetuate hegemony (i.e. maintain and assert the interests of dominant groups or classes) (Fairclough and Wodak 1997; Wodak as cited in Heracleous, 2006, p. 4) and suppress alternative constructions (Hjelm, 2014, p. 860).

CDA is one of the most used methodologies for analysing language and texts in a wide range of fields, including management and organisation studies (The Editors, 2010, p. 1192). However, studies employing CDA are scarce in the sociology of religion, and the valuable research potential of CDA has been underused as a result (Hjelm, 2014; Moberg, 2013). Therefore, Hjelm (2014) has strongly encouraged sociologists of religion to apply CDA, due to the obvious lack of critical approaches to religion that take inequality and hegemony seriously (p. 857).

6.2 Critical Discourse Analysis Protocol

Fairclough (1992), Fairclough (2010), and Gee (2011) have developed stringent guidelines for applying Critical Discourse Analysis (CDA). According to Fairclough (2010) selecting a CDA methodology involves developing a theoretical framework relevant to a specific research project than choosing from an existing repertoire of methods (Fairclough, 2010, p. 225).

What makes CDA unique is its division of analysis into three, cross-fertilising aspects: (1) Textual analysis, (2) Analysis of discourse practice, and (3) Analysis of social practice (including the relationship to existing hegemonies) (Hjelm, 2014, p. 861; Fairclough, 1992, p. 73). These categories are of central interest for this study as they help produce a comprehensive understanding of the impact of religion on the three levels of analysis:

The first level of analysis (textual) involves closely studying a range of texts such as political constitutions or Christian documents including The Holy Bible. Especially the Scriptures are central to analysing Christianity, as the text has remained largely unaffected by time for centuries (i.e. King James or Martin Luther versions). In contrast, Christian traditions have changed over time and vary among denominations.

The second level (analysis of discourse practice) considers the paradigms and public discourse of Protestants, Roman Catholic Church-State leaders, and government and civil society stakeholders.

The third level (analysis of social practice) explores the coherence between formal and de facto discourses, including the “commonsensical” language of social practices. Typically, producing documents for their own sake neither relates to nor explains social practice. Nevertheless, it is possible to analyse their coherence and influence in institutions. Comparing lived reality beyond formal discourses allows identifying patterns of social acceptance related to corruption and prosperity.

6.3 Data Treatment

Sixty semi-structured interviews were conducted and analysed using memos and open coding. Protestant and Roman Catholic Church-State leaders, government, academics, and civil society stakeholders were interviewed in a snowball system between 2015 and 2017. Interviews were conducted in English, German, and French, in the German and French-speaking regions of Switzerland. Interviews in Latin America were conducted in Spanish and English (i.e. English speakers in the Colombian Caribbean). All interviews were translated into English before coding. For reasons of scope, not all the data or stakeholders interviewed appear in the main body of this book. Table 17.2 shows the total number of interviews conducted in each country.

Table 17.2 Interviews conducted in this research
Full size table

Interviewees were mostly religious leaders or educated and informed stakeholders. Many stakeholders interviewed were also university professors or academic researchers. Protestants, Catholics, or non-religious interviewees were also asked about their different ideologies.

6.3.1 Protecting Interviewee Identity

Before interviewing, stakeholders agreed to release their data on condition of anonymity. Consequently, the discussion offered in this book and the corresponding public datasets only contain stakeholders’ general institutional affiliations for comparison/triangulation purposes (e.g. government officer, university expert-lawyer, Catholic priest, Protestant preacher). Other sensitive data possibly enabling identification were omitted.

6.3.2 Coding

Two qualitative researchers independently coded the data to align code application and code definitions (a total of three independent researchers). Moreover, the author took position in a separate memorandum (Appendix 5).

6.3.3 Triangulation

Triangulation is essential for crosschecking information through multiple procedures and sources. It produces either corroboration or convergence in the case of agreement or divergence in the case of disagreement.

The notion of identifiable Roman Catholic or Protestant traditions requires gathering theoretical and historical data as well as qualitative data from interviewees. Although interviewees shared their perceptions of religious traditions, these perceptions do not provide robust evidence of historical linkages or ramifications. Present-day interviewees can only share their impressions of what their country has inherited from its religious background.

Therefore, findings were triangulated using a triple strategy: (1) Diverse stakeholder perspectives were included in each code (e.g. interviews with a government officer, a Catholic priest, a Protestant pastor); (2) public documents or discourses; and (3) other materials (e.g. historical, theoretical, textbooks).

The next four chapters provide background information on the religion–prosperity–institution nexuses for each of the four cases. Each case study includes charts that compare the different stakeholder principles identified in terms of (1) Prosperity, (2) Corruption, or (3) Church-State relations. Each chart is followed by a three-level CDA analysis.

Case details vary depending on historical importance, the information relevant for comparison, and scope. More information is deliberately provided for Switzerland, since this is the only European country compared and contrasted with the three selected Latin American countries. Such detailed background information and three empirical analyses concern prosperity, corruption, and church-state relations. Consequently, the case of Switzerland also draws more detailed conclusions from its three empirical analyses. By comparison, each Latin American case contains briefer background information and merely one empirical analysis (Uruguay: prosperity; Cuba: corruption; Colombia: prosperity). The respective conclusions are synthesised in Chap. 22.

Each case begins by characterising the background information and the main features of the prosperity–corruption–religion nexus. This is followed by empirical analysis.