![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
Are proponents of Islamic government in the Arab World more likely to support violent extremist groups like ISIS? Previous research indicates that the answer to this question is dependent on individuals’ beliefs about what constitutes proper Islamic government. In this study I theorize that individuals’ beliefs about the compatibility of Islamic government and democracy is a key predictor of support for armed groups like ISIS. Using 2017 survey data from six Arab countries I find that ‘nondemocratic Islamists’ – those who favor the implementation of Shari’a law and clerical rule but believe Islam to be incompatible with democratic principles – are significantly more likely to express support for ISIS. In contrast ‘democratic Islamists’ – those who view Islamic Government to be congruent with democratic rule – are not more likely to endorse ISIS goals or tactics. These findings underscore the conditional and complex nature of the relationship between Islamism and violent extremism.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1 It is important to note the complex legal origins and history of hudud laws and practices within the Islamic World. Contrary to popular understanding, specific hudud punishments are not enumerated in the Qur’an but rather should be understood to be the products of legal interpretation borrowing from Islamic, non-Islamic and pre-Islamic traditions. A thorough explanation of hudud and its complex origins and legal manifestations can be found in Jonathan Brown (2017) Stoning and hand cutting – understanding the Hudud and the Shariah in Islam. Yaqeen Institute. Available online at: https://yaqeeninstitute.org/jonathan-brown/stoning-and-hand-cutting-understanding-the-hudud-and-the-shariah-in-islam/#.Xff0GehKiUl.
2 It is important to also note that some survey research has failed to find a statistically significant link between attitudes toward democracy and support for political violence. See, for example, Ciftci et al. (Citation2017) and Shapiro and Fair (Citation2010).
3 Note, Roy’s ‘three camps’ of Muslim thought about democracy and Islam can be found across different sects of Islam, across the four different schools of Sunni jurisprudence and outside of the more puritanical Salafi interpretation of Islam and Islamic government. In the study, I broadly associate ‘democratic Islamists’ with the second two of Roy’s camps and ‘nondemocratic Islamists’ with the first camp.
4 The Arab Barometer surveys both Muslims and non-Muslims in Arab countries, though the majority (around 92 percent) of respondents in the Arab Barometer Fourth Wave reported their religious affiliation as ‘Muslim.’ For the purposes of this study, I include only Muslim respondents in the sample. This results in the exclusion of 638 non-Muslim respondents, out of a total sample of around 7,200, or around 8.9 percent of the total sample. The Arab Barometer survey asks respondents about their religion in question 1012. The question reads simply ‘Religion’ and gives respondents the choices of ‘Muslim,’ ‘Christian,’ ‘Other,’ ‘Jewish,’ ‘I don’t know’ and ‘Declined to answer.’ With the exception of the ‘Denomination’ question (question 1012a) asked only in Lebanon, the survey does not ask respondents which sect of Islam they belong to nor does it ask them about the major school of Islamic jurisprudence (Hanbali, Shafi’i, Ja’fari, etc.) They adhere to. The Arab Barometer contains no questions about whether the respondents adhere to Salafi (Wahabi) beliefs. This does not allow researchers to consider how the respondent interprets Islam in a nuanced way, nor does it allow researchers to consider the way that Islam and ‘Muslimness’ as identities are constructed. This is a limitation. Certain Muslim identities and interpretations – for example, more Salafi orientations – could affect support for ISIS, which is very much motivated by a Salafi interpretation of Islam. However, the Arab Barometer survey does not provide questions that would enable the investigation of such nuances.
5 The fourth wave of the Arab Barometer was also fielded in Egypt and among a sample of Syrian refugees. These are excluded from the study, however. Questions about ISIS were not asked of Egyptian respondents. I exclude Syrian refugees from the sample because I suspect their personal experiences fleeing ISIS violence affected their attitudes toward ISIS. However, their inclusion in the analysis does not change the core findings of the study. Results with Syrian refugees included are available from the author.
6 Data and codebook for Arab Barometer are available online at: http://www.arabbarometer.org/.
7 As such, the study provides only a ‘snapshot’ of the relationship between support for democratic and nondemocratic Islamic government and support for ISIS. Attitudes about and interpretation of Islam and Islamic government do, of course, change over time and are affected by major events facing the Muslim World. However, I am not able to investigate the impact of change in interpretation and attitudes, given the static nature of the survey data.
8 Of the four waves of Arab Barometer, only two have asked respondents about terrorism and political violence by non-state actors. The first wave, fielded in 2006, included questions regarding terrorist activity by Jihadi groups then fighting U.S. forces in Iraq. The second and third waves omitted all questions about terrorism.
9 Question 829. Responses: agree to a large extent; somewhat agree; somewhat disagree; disagree to a large extent.
10 Question 830. Responses: support to a large extent; somewhat support; somewhat do not support; do not support to a large extent.
11 Question 828. Responses: certainly represents true Islam; represents true Islam; does not represent true Islam; certainly does not represent true Islam.
12 Results of these tests are summarized in Appendix Table 1.
13 Respondents answering ‘agree to a large extent’ or ‘somewhat agree’ for question 829, ‘support to a large extent’ or ‘somewhat support’ for question 830 and ‘certainly represents true Islam’ or ‘represents true Islam’ for question 828 were coded with a ‘1’ indicating that they were ISIS supporters in the analysis.
14 Question 605. ‘Which of the following statements is the closest to your point of view? The laws of our country should be based … ’ Responses: ‘entirely on Shari’a;’ ‘mostly on Shari’a.’
15 Question 606.2. ‘Your country is better off if religious people (imams) hold public positions in the state.’
16 Question 607.1. ‘Democracy is a Western form of government that is not compatible with Islam.’ Responses: ‘strongly agree;’ ‘agree;’ ‘ disagree;’ ‘strongly disagree.’
17 Question 516a.
18 Question 1001. ‘Age.’
19 Question 1002. ‘Gender.’ Coded 1 for males.
20 Derived from question 1015. ‘What is the total monthly income for all household members? Less than or greater than … :’ Responses: 50,000 Algerian Dinars (question 1015aalg); 450 Jordanian Dinars (question 1015ajor); 500 US Dollars (question 1015aleb); 3,500 Moroccan Dirhams (question 1015amor); 3,000 Israeli New Shekels (question 1015apal); 500 Tunisian Dinars (question 1015atun).
21 Question 1005. ‘Are you … ?’ Response: Unemployed.
22 Questions 1003 and t1003. Responses: illiterate / no formal education; elementary; preparatory / basic; secondary; mid-level diploma / BA; MA and above.
23 Question 610.1. ‘Do you pray daily?’ Responses: always; most of the time; sometimes; rarely; never.
24 Calculated as the first-difference marginal effect of a one-unit increase in the independent variable while holding all other covariates in the model constant at their mean values.
25 P ≤ .132.
26 P = .362.
27 The substantive impact of gender is highest in model 2 (+2.2 percentage probability) and lowest (+2.0 percentage probability) in model 5.
28 Question 518.2. ‘To what extent [do] you think a parliamentary system in which only Islamist parties compete in parliamentary elections is … ’ Answers: not suitable at all; somewhat suitable; suitable; completely suitable.
29 Question 518.4. ‘To what extent do you think a system governed by Islamic law in which there are no political parties or elections is … ’ Answers: not suitable at all; somewhat suitable; suitable; completely suitable.