Islamic Conservatism and Support for Religious Freedom
The literature on the Muslim world has found that although Muslims have a high support for democracy, that same support has been lower when it comes to democratic virtues such as tolerance. It has been unclear, however, what specifically about the Muslim world that contributes to this paradox. I offer methodological and theoretical contributions to this scholarly discussion. Methodologically, I employ Bayesian item response theory (IRT) that improves the previous studies by relaxing the requirement that all units of analysis answer an identical set of questions. I present a picture of conservatism levels in 26 countries and offer an insight on the relative importance of social issues among Muslims across the world. On the theoretical level, I provide evidence that it is Islamic conservatism that leads to the tolerance deficit. This conservatism is related to the perception of Islam superiority and different from being pious or traditional.
Although early research argued that Islam is incompatible with democracy (e.g., Kedourie 1992; Lipset 1994, 5; Huntington 1997), contemporary studies have provided evidence for the opposite. Support for democracy is high among Muslims and in Muslim-majority countries (e.g., Esposito and Mogahed 2007; Jamal and Tessler 2008; Mujani and Liddle 2007, 2009; Norris and Inglehart 2002; Rose 2002; Tessler 2002, 2003). Muslims, just like non-Muslims, prefer a democratic leader to a strong leader and democracy to authoritarianism.
Stating that one supports democracy, however, is not the same as saying one submits to democratic values, among the most important being religious tolerance. In addition to (or perhaps more than) elections, democracy needs tolerance—a willingness of the people to put up with ideas, groups, or beliefs they disagree with (Gibson 2010, Linz and Stepan 1996). It is this aspect of democracy that studies start to show the Muslim world is lacking (e.g., Gu and Bomhoff 2012; Spierings 2014). Muslims support democracy, but are less willing to tolerate offensive acts, especially if the acts are hostile to religion (Djupe and Calfano 2012).
In part because this line of research is new, little is known about the factors behind this relatively low level of tolerance. I aim to advance the literature by focusing on religious tolerance and by offering methodological and theoretical contributions. On the methodological level, I employ a modeling technique of survey response that is based on Bayesian item response theory (IRT). The method offers three advantages relative to the past studies. First, instead of treating observed survey responses as given, the method models them as influenced by an unobserved latent variable “Islamic conservatism.” This variable efficiently summarizes attitudes toward multiple issues and makes a comparison between individuals or between aggregation categories easier. Second, the approach acknowledges the empirical fact that not all survey questions are created equal. Some may be better than the others in separating the most and least conservative respondents. Questions may also differ in their likelihoods to elicit affirmative responses from the respondents. Third, insofar as it is possible to create a common space and connect respondents based on the questions they answer, the method allows for a cross-countries comparison even if the respondents do not answer an identical set of questions.
On the theoretical level, I argue that Islamic conservatism is the main drive behind the relatively low tolerance level among Muslims in the Muslim world. Islamic conservatism is different from social conservatism in that the former pertains to issues specifics to Islam whereas the latter relates to general social issues debated by Muslims and non-Muslims alike, such as abortion, euthanasia, and alcohol. I provide individual-level evidence that Islamic conservatism predicts support for religious freedom above and beyond the effects of economic condition and level of piety. By differentiating Islamic and general conservatism, I argue that when it comes to respecting rights of religious minorities in Muslim-majority countries whether Muslims have become more liberal in the Western sense (such as by supporting abortion or same-sex marriage) matters less than whether Muslims submit to a religious ethnocentrism.
I divide the remainder of this article into six sections. In the first section, I review the literature on democracy and tolerance and discuss how in the Muslim world support for democratic values has been lower than the support for democracy as a system. The second section presents the argument for the need to differentiate Islamic and general social conservatisms. A section on Bayesian item response theory then follows. It is not meant to be technical, but rather to provide the reader with a broad understanding of the method. In the fourth and fifth sections, I outline the analytical procedures and present the results. I show how levels of Islamic conservatism vary across countries and provide evidence for the predictive power of the construct on the support for religious freedom. Lastly, I offer discussions of the results and highlight avenues for future research.