Democracy, Civil Society, and Islam

Abstract

Particularly after the so-called Arab Spring, the relevance of the “Turkish model” for other Islamic societies has been the subject of intense debate. Put more directly, the question is whether or not democracy can flourish and be sustained in societies where Islam is the dominant faith. This chapter examines the issue making use of the most recent European Values Study data. Macro-level multivariate analyses demonstrate that, even after controlling for social and economic development, Muslim-majority societies score poorly on various comparative measures of democracy. It is proposed that one possible explanation for this is the absence of a vibrant civil society, a widely accepted prerequisite for democracy, in Islamic societies. It is shown that, by and large, the negative “Islam effect” on civil society cannot be refuted.

In the light of modern liberal democratic thought, Islam is no more, nor any less democratic than Christianity or Judaism. All three monotheistic religions, if proposed as constitutional foundations of the state, and if understood as providing an ineluctable authority for the guidance of all significant human choice, are undemocratic or non-democratic. (Binder, L. (1998). Exceptionalism and Authenticity: The Question of Islam and Democracy, Arab Studies Journal 6, 33–59)

For Islamists, democracy, expressing the will of the people, is the road to power, but it is a one-way road, on which there is no return, no rejection of the sovereignty of God, as exercised through his chosen representatives. Their electoral policy has been classically summarized as ‘One man (men only), one vote, once.’ (Lewis, B. (2003), The Crisis of Islam: Holy War and Unholy Terror (pp. 111–112). New York: The Modern Library)

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

14.1 Introduction

Ever since the founding of the secular Turkish Republic in the 1920s, the relevance of the “Turkish model” for other Middle Eastern, in fact, for other Islamic majority countries regardless of geography, has been a topic of discussion by academics, policymakers, journalists, and others. The existence of a “model” with a democratic (or at least an aspiring democratic) system of government, a secular (for some, even “too secular”)¹ constitution and laws, and finally a clear Western orientation has been a source of optimism for many. The sweeping and radical reforms of modernization (at the time, a synonym for Westernization) of the early decades of the Republic met with profound interest and even admiration on both sides of the Atlantic and beyond. The young Turkish Republic and its founder, Mustafa Kemal Ataturk, were sources of inspiration in varying degrees for a number of Muslim leaders of the period as well.² Commenting on the renewed interest in the “Turkish model” after the end of the cold war, Altunisik (2005: 45) quotes Anthony Blinkmen³: “…because of what it is – a nation of mainly Islamic faith that is secular, democratic, and modernizing – Turkey must be a leader and can be a role model for a large swath of the world.”

After 9/11 and especially after the uprisings that began in Tunisia and initiated the process that has been dubbed the “Arab Spring,” references to the so-called Turkish model have become ever so ubiquitous. In addition to the analyses in academic journals, the popular media as well as political leaders devoted considerable attention to the subject. The BBC World, for instance, debated the topic for a worldwide audience.⁴ Various polls were also conducted in a number of Arab countries to gauge public attitudes on the Turkish example. In one poll dated February 2012, no less than 72 % of the respondents in Arab countries opined that Turkey was a desirable model for the newly established regimes in Tunisia, Libya, and Egypt. However, rather remarkably, Arab citizens told the pollsters that they favored the Turkish model “because they believe Turkey is close to the Arab world in terms of culture, religion, and traditions and also because it has ‘integrated Islam into politics.”’ (Idiz 2012; italics mine).

The interesting question about the recent popularity of the “Turkish model” is its very source. One could rightly ask why Turkish and not, for example, Dutch or British or Swedish, to name but few countries whose democracies are much more advanced, consolidated, and perfect than the Turkish and whose human rights records are much more solid. After all, it is no great secret to anyone that Turkish democracy has numerous defects and shortcomings. If the idea is to establish a democratic system of government in Arab countries, would it not make more sense to look up to one of the less faulty democracies as a model? The answer to this question is, more than anything else, cultural: Turkey’s population is almost exclusively (98.3 % according to the PEW Forum on Religion) Islamic. The model, it is believed, must be one that “can integrate Islam into politics” – presumably, democratic politics.

Implicit in this approach are two assumptions from which this chapter takes off:

• (a) There is a correlation between cultural values and the system of government. In other words, culture matters for politics and political regimes as well.

• (b) The relationship between Islamic values and democracy as a system of government is problematic. Put differently, it is assumed that Islam is less conducive to democracy compared to other major religions.

The first assumption has been argued, explicated, discussed, debated, and evidenced so widely and competently that there is no need for us to spend any time on it here. To get to the roots of the culturalist approach, one can go as far back as Plato and Aristotle. Suffice it to note that the present author is also convinced that cultural values set limits to the sustainability of any legal-institutional structure.

Currently, the debate on the second assumption does not suffer from any lack of popularity either.⁵ The burning question is whether or not a society whose value system rests on the principles of Islam can build up and sustain a secular and democratic system of government. The discussions and widespread doubts about Islam’s compatibility with democracy are based upon both empirical observations and theoretical considerations. We begin with the former and briefly review the evidence about the relationship between religious, more specifically Islamic, values, and political regimes. In the second part of the chapter, we discuss some of the theoretical arguments which inevitably take us to the core values of Islam.

14.2 Empirical Evidence

A review of the existing literature based on empirical data reveals two main approaches to the subject: macro (country)-level evidence and micro (individual)-level evidence. While the former utilizes aggregate data about the characteristics of countries, the latter is almost exclusively based on survey data.

14.2.1 The Macro Level

A number of international organizations routinely calculate indices of democracy that score and rank countries on how “democratic” and “free” they are. Many of these indices are highly correlated among themselves. While the best known and most widely used democracy and freedom index is the one put out by the Freedom House, Germany’s Bertelsmann Foundation and The Economist, among others, also calculate similar indices. Furthermore, individual academics have contributed to the rather popular field of “index of democracy” construction (e.g., Bollen 1993) providing the analyst with a wide choice. Scholars working on the correlates of democracy routinely use these indices – and most commonly the Freedom House Index – as a measure of the degree of democracy in a country.

In our attempt to assess the relationship between Islam and democracy at the macro level, we use the democracy index calculated by Alexander and Welzel⁶ and which is a combination of the Freedom House Index and Rule of Law Index from the World Bank’s Good Governance Project (Kaufmann et al. 2008). The index is known as “the Effective Democracy Index” and, as noted by its authors, takes into account not just the theoretical existence or nonexistence of freedoms and human rights in a country as stipulated by its constitution and laws but also how effectual these written rules are in actual reality.⁷

No sophisticated analyses of the Welzel-Inglehart Index are needed to get a view of the overall picture. The index ranges between 0 (Libya, Sudan, Turkmenistan) and 97.10 (Finland). Of the 88 countries in the upper half of the distribution of Effective Democracy Index, only four are Muslim-majority countries: Turkey (26.91), Kuwait (29.05), Senegal (30.21), and Mali (32.92). More strikingly, only one Muslim-majority country (Mali) barely made it to the top 40th percentile, while none even approached the top quartile.⁸ It may also be worth noting that there are 38 Muslim-majority countries in the lower half of the index.

Although a casual glance at the Effective Democracy Index (or any other measure of democracy for that matter) easily reveals the rather unfavorable positions of Muslim societies on a democracy and/or freedom scale, examining the relevant correlation and regression coefficients will be more instructive.

Scholars who have investigated the relationship between Islam and democracy at the macro level have used either the proportion of Muslims in a given country (Muller 1995; Fish 2011) or a binary variable (usually with 50 % as the cutting point) to define Muslim-majority countries (Norris and Inglehart 2004; Anckar 2008). Neither one of these measures is perfect. However, we believe that the proportion of Muslims is a comparatively better measure despite its shortcomings. Furthermore, it is comforting to know that the correlations we are about to report are robust regardless of the measure used.

The simple bivariate linear correlation coefficient between the percentage of Islamic population⁹ and the Effective Democracy Index is no less than −.50 (n = 166). Clearly, this is not a finding that one can dismiss readily.¹⁰ A look at the scattergram with percent Muslim on the horizontal and effective democracy on the vertical axes reveals that a second degree polynomial will provide a better fit to the data. A regression equation with effective democracy as the dependent and percent Muslim plus percent Muslim squared as the independent variables yields an adjusted R squared of 0.27 with all coefficients statistically significant at 0.01 level or better.

Borooah and Paldam (2006: 25) use Bayesian probability analysis to calculate the “risk” for Muslim and non-Muslim countries of being undemocratic. They conclude that “by 2004, a country which was undemocratic was more than twice as likely to be Muslim than one which was highly/moderately democratic. Using a complementary concept of risk, we were also able to show that by 2004, even though only a third of the world’s countries were Muslim, the chance of an undemocratic country being Muslim was 85 % of the chance of an undemocratic country being non-Muslim.”

Simple analyses similar to the one above have been carried out time and again yielding similar correlation and regression coefficients. However, such bivariate statistics, without any control variables, could easily result from spurious associations.

In a seminal and still influential work, Lipset (1959) argued that democracy was associated with economic and social development as well as an equalitarian value system. Since Lipset, it has been commonly accepted that economic development (or rather the more broadly defined concept of socioeconomic development) and democracy are correlated. Hence, one has to ask whether or not the high bivariate correlation coefficients (between democracy and religion) approach zero when one controls for development.

To test the spuriousness hypothesis, we run the above regression twice: once with a purely economic indicator (GDP/cap 2005) and once with a more widely accepted indicator of socioeconomic development, that is, the Human Development Index published by UNDP. The results are given in Table 14.1. When income is controlled for, not only is the religion effect still very highly significant (p < .000) but the beta coefficient – which allows a comparison of the relative magnitudes of independent variables – for the percent Muslim variable is more than twice greater than the income coefficient. And the two variables explain close to 30 % of the total variance in Effective Democracy Index. It is customary to use the logarithmic transformation of per capita income to smooth out the distribution and get rid of the outliers. Using log gnp/cap elevates the R squared to 0.41 and the beta coefficients for percent Muslim and income are about the same. The religion effect is still loud and clear when GDP/cap is replaced by HDI as control variable. Furthermore, the R squared jumps to 0.56. This finding corroborates Diamond (1992) who argued that HDI was a better predictor of democracy than per capita income.

Clearly, development, both economic and socioeconomic, is a significant predictor of the level of democracy in a society, but religion also plays a vital role over and above the development effect. This is a robust and hard-to-deny finding regardless of one’s preferences for indicators and measurement models. One may justifiably argue that these models are too simplistic to be conclusive and that they do not demonstrate causality. No doubt, this is true. But it is equally true that there is a story here that one would be well advised not to ignore.

Table 14.1 Regression of Effective Democracy Index on percent Muslim and socioeconomic development (standardized beta coefficients)

Up to this point we have treated the democratic system of government as a unified concept and have not analyzed any of its components separately. Given the myriad definitions of democracy, identifying its components would be complicated theoretical task indeed. However, there is no disagreement on the principle that democracy, above all, means equal rights for all citizens regardless of gender. It is also a fact that, as of yet, no country in the world has been able to achieve complete equality between its male and female citizens. Nevertheless, this is not a binary, yes-or-no variable, but rather an interval scale. Some societies, although not perfect, are much more advanced with respect to this all important aspect of democracy than others.

As comparative data compiled by various international organizations have shown, gender inequality is exceptionally high in Islamic societies. So much so that, many scholars have concluded that if there is one major area of divergence between Islamic and non-Islamic societies as well as Muslims and non-Muslims, it is related to gender equality and the status of women. Put differently, comparatively speaking, gender inequalities are greater in Islamic countries both at the institutional and individual values levels.

Comparing the values of Muslims with non-Muslims in an attempt to test the “clash of civilizations thesis,” Norris and Inglehart (2002: 260) conclude that the major cultural cleavage is not related to democratic values but rather in “social beliefs about gender equality and sexual liberalization. In this regard, the West is far more egalitarian and liberal than all other societies, particularly Islamic nations.” Esmer (2002) reaches a similar conclusion and underlines that a distinguishing characteristic of Islamic culture is its outlook on women (and sex). Fish (2011) explores the status of women in public life and popular attitudes towards gender equality and writes that “All the findings point in the same direction…. In none of the results is Islam associated with better status for females relative to males, and in most of the analyses it is clearly correlated with inferior conditions for females” (Fish 2011: 201). Nevertheless, he is cautious about making causal inferences from these findings and considers a number of alternative hypotheses that could explain the correlations.

The World Economic Forum annually publishes a global “gender gap” report which ranks countries on various dimensions of gender equality.¹¹ We repeat the regression results reported in Tables 14.1, only replacing the Effective Democracy Index with the Gender Gap Index of the WEF as the dependent variable. (To facilitate comparison with the Effective Democracy Index, we use the 2006 scores for the Gender Gap Index as well.) The results given in Table 14.2 need no further interpretation. Proportion of Muslims in a country has a strong negative effect on gender equality net of the effects of both economic and social development. In fact, religion effects are greater than development effects which, at least in one case, is not even statistically significant (Table 14.2).

As noted, for consistency, we have used the Gender Gap scores for 2006. But the situation has not changed at all since then. The correlation between the 2006 and 2011 Gender Gap scores is almost perfect at 0.94. Of the 15 lowest ranking countries on the 2011 Gender Gap Index, 12 are Muslim-majority countries, the 3 exceptions being Cote d’Ivoire (28 % Muslim), Benin (18 % Muslim), and Nepal (4 % Muslim). If these indices (including the ones published by various UN agencies) have any validity, gender equality is a serious problem in Islamic societies which, in turn, translates into a democracy problem.

Table 14.2 Regression of Gender Gap Index on percent Muslim and socioeconomic development (standardized beta coefficients)

14.2.2 The Micro Level

A number of authors (Norris and Inglehart 2002; Rose 2002; Tessler 2003; Al-Braizat 2003; Inglehart 2007; Kotze and Garcia-Rivero 2009; Tausch 2009) have concluded that Muslims, as individuals, are just as democracy oriented as non-Muslims. One scholar has even reversed the “common wisdom” and, in a study of eight countries, argued that Muslim respondents espoused democratic values more than Christian (Eastern Orthodox, to be more precise) respondents at least in the countries included in the study (Hofmann 2004). Most, if not all, of these writers, however, have analyzed responses to questions that ask directly about attitudes towards democracy such as “how important is it for you to live under a democratic system of government?” or “would it be very good, etc. to have a democratic political system for governing this country?” It has been shown that responses to these and similar direct questions vary little from one society to another perhaps due to the fact that after what Huntington has called the “third wave,” democracy has become the “only game in town.” Whatever the underlying reason, it seems like one has to look beyond direct questions and rhetoric and try to unearth the existence or nonexistence of core democratic values in devout Muslim populations. This is not to say that the findings mentioned above need to be revised. It only points out the need for further analyses using questions that tap democratic values.

To gain further insight into the support for democracy at the individual level, we propose to define “the minimal democrat” with few core democratic values or attitudes. Taking the 2008 EVS Questionnaire as our starting point, our “minimal democrat” is the following individual. He/she:

• Thinks that being ruled by a democratic system of government would be very good

• Thinks that having a strong leader who does not have to bother with parliament or elections would be fairly bad or very bad

• Does not categorically reject the possibility of signing a petition even if he/she has not done so yet

• Has no objections to having persons of different race or color as his/her neighbor

• Does not believe that there is only one true religion (obviously one’s own)

• Does not think that men should have priority over women when jobs are scarce

It is easy and perhaps tempting to expand this list of six items, but we would like to keep the number to a bare minimum since we are after the “minimal democrat.” The dimensions of this six item list are obvious. These are:

• (a) Strong faith in democracy, parliament, and elections

• (b) Acceptance of conventional political participation

• (c) Tolerance for different races and religions

• (d) Gender equality

Using the latest round of EVS (2008–2009) and recoding the above questions as 1 (democrat) or 0 (not democrat), we obtain a simple additive index¹² that ranges between 0 and 6. We then try to assess the impact of religion on this index in 47 European countries.

A simple frequency distribution reveals that our index, intended to identify the “minimal democrat,” is rather optimistic. If this definition of the “minimal democrat” were accepted, one would have to concede that democratic individuals are a rare breed in Europe (Fig. 14.1). Indeed, only about 17 % of Europeans scored as “democrats” on all six questions. Again striking a pessimistic note, no less than a third of Europeans are on the nondemocrat side with respect to at least three of the six items.

Fig. 14.1

Distribution of the “minimal democrat index” in 47 European countries

Examining the relationship of the index to religion, we first note that those who do not claim to belong to a religion score higher on the index than those who identify with a religion or religious denomination (mean 4.41 vs. 3.94; data unweighted).

The EVS data allows us to compare the scores of four major denominations (for others the number of observations is insufficient). The means and the standard deviations for these four denominations are given in Table 14.3. Among the four religion groups, Protestants score the highest and Muslims the lowest. The difference between Protestants and Muslims is quite sizeable and the difference between Roman Catholics and Muslims is not negligible either.

Table 14.3 Minimal democrat index means for religions/denominations

To be certain that the difference between denominations observed in Table 14.3 are religion effects, one must hold constant possible confounding variables. Levels of income and education are the first such variables that come to mind. Thus, we run a regression with our minimal democrat index as dependent, religious denominations (dummy variables for Protestant, Catholic, and Muslim) as independent, and education (ISCED one digit coding) plus income (purchasing power parity in Euros) as control variables. The results are given in Table 14.4. With such a large sample size, it is very unusual not to get statistical significance; therefore, we do not comment on that. However, two things in Table 14.4 are worthy of our attention: (1) the signs of the coefficients and (2) the proportion of total variance explained. We observe that all four independent variables have positive effects on the minimal democrat index while the sign is negative only for Muslims. Furthermore, the R squared is quite respectable particularly taking into account that this is individual-level data.

Table 14.4 Regression of the “minimal democrat” index (data unweighted)

To sum up the analysis so far, one thing is clear. Both at the macro and the micro levels, the relationship between Islam and democracy is problematic – at least more problematic than other religions/denominations. Why? That question leads us to Part 2 of the chapter: the theoretical considerations that might be responsible for these findings.

14.3 Theoretical Arguments

There are two clearly demarcated and opposing camps with respect to the relationship between Islam as a faith (and its core teachings) and democracy. On one the side are those who claim that Islam is fully and entirely compatible with democracy. Not only is it compatible, they argue, but Islam is a call for democracy when its basic principles are considered – human rights, equal status of all human beings before Allah, rejection of racism, principles of justice, etc. Some go as far as asserting that the idea of democracy originated in Islam. Their evidence for this rather unsustainable claim is the “shura” principle that requires the leader of an Islamic polity to consult and seek the advice of believers before making important decisions. According to one vocal spokesman, “Shura constitutes one of the four cardinal principles in the Islamic perspective on socio-political organization. The other three are justice, equality and human dignity” (Sulaiman 1999; italics original).

The opposing camp argues that no other religion is as comprehensive and all-encompassing as Islam which leaves no room for civil society. In Islam, every aspect of human existence, it is argued, is governed by immutable rules – at least in principle. According to the proponents of this line of reasoning, Islam has no room for democracy as we know it.¹³

14.3.1 Democracy and Civil Society

Defined as the “…sphere of voluntary associations and of voluntary involvement at some distance to the state, to the market and to the intimate sphere of family” (Dekker and van den Broek 2005: 46), a vibrant civil society is regarded as a prerequisite for democracy because it ensures pluralism and limits state power. Indeed, civil society is a term with a positive connotation because it is assumed to be the antidote of absolutism and totalitarianism. As stated by Therborn (2002: 45) “it [civil society] has a number of attractive futures: a plurality of autonomous groups and associations; public institutions of culture and communication; privacy for individual choices and pursuits; a system of general rights demarcating plurality, privacy, and publicity from the state; perhaps also from the economy.”

In any geography, civil society does not come into existence abruptly or by decree, but it requires a hospitable cultural tradition. Put differently, the societal value map must be conducive to the well-functioning of civil society. In Gellner’s words, the central premise of civil society is “unconstrained and secular individual, unhampered by social or theological bonds, freely choosing his aims” (Gellner 1994 quoted in Sunar 2002: 9).

From Gellner’s perspective, then, we have two prerequisites for a healthy civil society: secularism and individualism. Ozdalga (2002: 74–75) agrees that “secularization is a necessary condition for the development of civil society. Individuals who are not able to act independently from the community of believers cannot become the building stones of intermediary organizations on which civil society is built.” There is a vast literature on the association between civil society and democracy which is hardly contested by anyone. Within the context of Europe and using EVS data, Dekker et al. (2003: 220) provide empirical evidence for “The assumed positive relationships between civil society and political democracy” both at the macro and micro levels. We would go one step further and argue that the relationship is not just assumed but demonstrated as well with reliable data. Blaydes and Lo’s (2012: 112) conclusion that “democracy occurs when regime liberalizers who prefer democracy to a narrowed dictatorship interact with a civil society that will honor democratic principles” may be debatable, but the emphasis on a “civil society that will honor democratic principles” has been widely accepted.

Recognizing crucial importance of civil society for a sustainable democracy, the next question that needs to be asked is the conduciveness of Islam to civil society. If Islam is not compatible with or at least not hospitable to civil society, this may provide an explanation for the empirical findings observed both at the macro and micro levels and summarized above. If, on the other hand, Islam does not constitute a barrier to civil society neither in theory nor in practice, we will have to look for an explanation elsewhere.

14.3.2 Islam and Civil Society

According to Gellner, let alone being conducive to it, Islam is the rival of civil society just like Marxism and Fascism. The reason for this unequivocal stance is Gellner’s assessment of Islam as a “secularization resistant” religion (Gellner 1992, 1994).

If we recall Gellner’s reasoning that secularization is one of the two main requisites of civil society and if Islam does not allow secularization, it follows that civil society cannot flourish in an Islamic society. Prophet Mohammad was not only a religious and spiritual leader but, at the same time, a political and military leader and the ultimate distributor of justice. This is why, it is argued, Islam cannot secularize, that is, exist outside the realm of politics. According to Huntington (1984: 208), in Islam “no distinction exists between religion and politics or between the spiritual and the secular….”

At a more general level, the major reasoning for the assumed incompatibility of Islam with civil society is based on the all-comprehensive nature of the Islamic faith. It would not be unfair to say that it regulates all aspects of social and political existence. Islam establishes principles in areas ranging from civil law to criminal law and from state formation to economic institutions. And in many cases, these principles are not just general and abstract codes but rather finely detailed and concrete obligations on the part of believers. This all-encompassing nature of the Islamic religion, it is said, leaves no room for civil society.

Gellner is not alone in regarding Islam as a rival of civil society. Mazlish (2005), for instance, sees global Islam, along with global America and global civil society, as contenders for future global society although he predicts that Islam will not be a winner in this competition.

To recapitulate the line of reasoning we have been summarizing:

• (a) Democracy needs civil society.

• (b) Civil society needs secularization and individualization.

• (c) Islam, in principle, allows neither secularization nor individualization mainly because it does not confine itself to the spiritual.

• (d) Therefore, Islam is not conducive to democracy.

To be sure, those who disagree with the line of reasoning represented here by Gellner would be in distinguished company as well. We have already mentioned a number of scholars who emphasize that Islam is not different from any other world religion with respect to its compatibility with democracy.

Once again, we need to look at the evidence in order to take a position concerning the association between religion and civil society. This is not going to be easy if only because there is little agreement on the precise meaning of the term civil society (Anheimer 2005). Nevertheless, an attempt in that direction is not completely futile. The Quality of Government Institute at the University of Gothenburg has made available some indicators of civil society for a large number of countries. We use two of these measures¹⁴: (1) civil society organizations working on corruption and civil society organizations per million population. We regress these two indicators on the proportion of Muslims as the independent and UNDP/HDI as the control variable. The results are given in Tables 14.5 and 14.6. We see that, with both indicators, the proportion of Muslims has a statistically significant and negative effects net of the effects of HDI (which in one case is negative and nonsignificant). Taking into account the low precision (and possibly low validity) of our civil society indicators, we can only tentatively conclude that Islam decreases the ubiquity and effectiveness of civil society.

Table 14.5 Regression of civil society (corruption) on percent Muslim and UNDP/HDI

Table 14.6 Regression of civil society organization/million population on percent Muslim and UNDP/HDI

14.4 Islam and Civil Society in Europe

Finally, we test the relationship between civil society and Islam within the European context making use of the EVS data. The question we will address is whether or not there is a correlation between religious affiliation and the attitudes and values that are conducive to civil society.

Our indicators will be the two premises of civil society proposed by Gellner (secularism and individualism) in addition to membership in voluntary organizations that form the essence of civil society. Furthermore, we will look at data related to interpersonal trust and political participation – two important correlates of civil society (Dekker et al. 2003). To sum up, then, we will be comparing Muslims and non-Muslims in Europe with respect to:

• (a) Premises of civil society (Gellner)

• (b) Major correlates of civil society (Dekker et al. 2003)

• (c) Essence of civil society

The survey questions we will use to tap these dimensions are the following:

Secularism

Question 1:

Politicians who do not believe in God are unfit for public office (agree/disagree).

Question 2:

Religious leaders should not influence government decisions (agree/disagree).

Individualism

Question 1:

Regardless of what the qualities and faults of one’s parents are, one must always love and respect them, or one does not have the duty to respect and love parents who have not earned it by their behavior and attitudes.

Question 2:

Please use this scale to indicate how much freedom of choice and control you feel you have over the way your life turns out (a 10-point scale with higher scores indicating more freedom).

Interpersonal trust

Question:

Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?

Political participation

Question:

Signing a petition (have done/might do/would never do under any circumstances).

Membership in voluntary organizations

• Mean number of membership in 15 different types of voluntary organizations.

Table 14.7 summarizes responses to all seven indicators according to religious denomination. Admittedly, this is a very elementary analysis that is only intended to serve as a first step. It needs to be expanded to include controls for socioeconomic and demographic variables as well as tests for nation effects which we do not include in this chapter due to space limitations. However, with these caveats, we can say that on the premises, essence, and correlates of civil society, European adherents of Islam lag far behind Protestants and behind Roman Catholics as well. Members of the Orthodox Church in Europe come closest to Muslims – a finding that calls for further investigation.

Table 14.7 Behavior and values related to civil society (data unweighted)

14.5 Conclusions

We start with the rather commonplace empirical observation that regardless of which international index of democracy, freedom, or good governance one uses, Muslim-majority countries fare worse than non-Muslim nations at similar levels of economic and social development. Using Alexander and Welzel’s “Effective Democracy Index,” we once again demonstrate this rather strong negative association. This being the case at the macro level, the next obvious question – and one that many social scientists have tackled – is whether or not Muslims, as individuals, are less democratic. In other words, is there a significant difference between Muslims and adherents of other religions with respect to the core values related to democracy? Unlike the macro-level data, there is much disagreement on the nature of the relationship between individual Muslims’ value systems and democracy.

Using the EVS data, we compare the values of Muslims and Christians (Roman Catholic, Protestant, and Orthodox) in Europe on the “Minimal Democrat Index” composed of six core democratic values. The differences are significant even after controlling for education and income. However, as we have noted more than once, many scholars working on global, regional, or country data disagree with this outcome. Regardless of that debate (which is not a trivial issue, to be sure), we have to ask why is it that Muslim-majority countries are less hospitable to democracy.

One possible explanation, among others, is the incompatibility of Islam with a vibrant civil society which, in turn, is a prerequisite for a healthy and sustainable democracy. We test this hypothesis at two levels: (a) using two global civil society indices and (b) using indicators of the premises, essence, and correlates of civil society in Europe. Islam effect is still observable particularly in analyses with the global civil society indices as the dependent variable. With respect to the civil society indicators at the individual level in Europe, however, Orthodox Christians and Muslims have rather similar attitudes and values. Overall, the civil society hypothesis does not seem to be one that can be discarded easily, although further analyses both in Europe and in other regions of the World are needed.