Downloaded from Brill.Com09/24/2021 02:55:52PM Via Free Access

sociology of islam 5 (2017) 249-277

brill.com/soi

Contextualizing Events in Imagined Communities Pre- and Post-September 11th Egyptian Attitudes toward Non-Arabs and Jews

Sarah Shah Department of Sociology, University of Toronto, 725 Spadina Ave, Toronto, on M5S2J4 [email protected]

Gabriel A. Acevedo Department of Sociology, University of Texas at San Antonio, One utsa Circle, San Antonio, tx 78249

Andrea Liza Ruiz Department of Sociology and Criminology, Pennsylvania State University, 211 Oswald Tower, University Park, pa 16802

Abstract

The ongoing denigration of Arabs in the media, the Western democratic political shifts away from Muslim religious freedom, and increasing anxiety about Muslim radicalization prompt the question: How do Arabs respond to global events when the Muslim and Western worlds are perceived to clash? Our study draws on the theory of imagined communities to examine the extent to which exogenous world events influence attitudes towards out-group members in the Egyptian context. We apply a “pre-post” quasi-experimental design using the World Values Survey, and examine the influence the events of September 11th, 2001 had on Egyptian perceptions of Jews and non-Arabs. Results suggest that intolerance towards both Jews and ethnic minorities decreased after the attacks. Results also suggest a complex, dynamic association between religiosity and tolerance towards out-group members. We conclude by discussing the theoretical contributions of this paper by highlighting the significance of context and religion when framing imagined communities.

Keywords

Egypt – pan-Arabism – Islam – imagined communities – religious group identity – September 11th

250 Shah, Acevedo and Ruiz

Arab-Western relations are a major component of the global political stage. Huntington’s (1996) Clash of Civilizations thesis remains one of the seminal, yet highly controversial paradigms guiding the understanding of Arab-Western relations, postulating that conflict in contemporary societies would be less a function of material inequality and more based on ideological cleavages. In this, Huntington predicted eminent confrontations between the Arab world and the West. While this thesis has received substantial criticism for failing on both theoretical and empirical grounds (e.g., Acevedo 2008; Ahmed 2002; Chiozza 2002; Fox 2002; Henderson and Tucker 2001; Matlock 1999; Russett, Oneal, and Cox 2000; Tsygankov 2003), it nevertheless informs the often partial and partisan framing of Arab-Western relations, such that many continue to view the two cultures as inamicable. While actual political confrontations and alliances are much more complex, and beyond the scope of this paper, the public perception of conflict between the Arab and Western worlds is reflected in media representations of the Muslim Middle East. Media coverage, especially in the West, tends to portray Arabs and Muslims as violent, radical jihadists with little regard for human rights (Said 1977; Shaheen 2003). This conceptu- alization of Arabs is often reflected in the growing anti-Arab and Islamopho- bic political shifts in Western democracies away from the freedom of religious expression for Muslims. From the contentious and ongoing debates around Muslim women’s clothing in both France and Canada (see Levey 2009 for a review), to, what some have argued, are the racist motivations underlying Brexit (Green et al. 2016), to Trump’s victory based on both a populist and Islamo- phobic platform (Lichtblau 2016; Yan, Sgueglia, and Walker 2016), to several Western countries’ opposition to receiving Syrian refugees for fear of the entry of sleeper cells (Eglitis 2015; Koran 2016), Muslims and Arabs are increasingly seen as “a fifth column” in democratic societies (Cainkar 2002). Furthermore, the rise of Daesh and the increasing Western anxiety about the radicalization of Muslims prompts the question: How do Arabs respond to global events when the Muslim world is publically perceived to be in conflict with the West? This study focuses on Arab reactions to the events of September 11th, 2001, which sent shockwaves throughout the international community (Moaddel 2003). While several studies have examined the “post-9/11” experiences of Arab Americans and American Muslims (Cainkar 2002; Cainkar 2009; Detroit Arab American Study 2009:153–154; Panagopoulos 2006), limited work has focused on the Middle Eastern experience (see Moaddel and Abdul-Latif 2006 for an exception). Particularly of interest is Egypt, which some call the first modern Arab state (Henry, Amara, and Al-Tauqi 2003). Egypt has received sustained international attention following the events of its revolution and displacement of President Mubarak, instatement and displacement of Morsi, and current

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 251 leadership of el-Sisi. The significance of Egypt as an emerging democracy has been highlighted for decades before the recent revolutions (al-Mashat 1986; Kassem 2006). Since the revolutions, however, and as Egypt experiments with processes of democratic reform, critical attention has been directed at the nation’s gradual attainment of political stability (Azzam 2013; Bayat 2013). Conse- quently, the formation of national identity within this fledgling democracy is of vital interest to social scientists. For this study, we frame pan-Arabism as an imagined community in the Egyptian context, and, like Moaddel and Abdul-Latif (2006) we frame September 11th as a pivotal historic moment for the global community. Conse- quently, in keeping with the theoretical insights of the imagined communities perspective, we expect the events of 9/11 to have the effect of magnifying in- group ties at the expense of out-groups. In accordance with the contours of collective behavior described by theories of imagined communities and religious social identity, we posit that Egyptians will have a strengthened sense of inclusiveness with the larger Arab community, and therefore be less tolerant of out-group members. We employ data from the World Values Survey, which was collected in Egypt before (2000) and immediately after (2002) the events of September 11th, allowing for a quasi-experimental design. This enables us to capture post-9/11 outcomes of pan-Arab nationalism in the Egyptian context, and provide empirically informed conclusions regarding Arab reactions to global events. We find that intolerance towards both people of other races as well as Jews decreased significantly after 9/11. This effect was moderated by education, such that those with higher levels of education were more likely to be more tolerant post-9/11. Religiosity plays a dynamic and complex role, such that both God salience and religious service attendance modestly correlate with less tolerance for people of other races, and religious service attendance modestly correlates with less tolerance for Jews. However, God salience and Muslim religious affiliation strongly correlate with more tolerance for Jews. Before proceeding to the empirical findings of our study, we discuss the imagined communities perspective, which frames the processes of nationalist identity formation.

Imagined Communities

Anderson (2006) proposed that modern nations might be conceptualized as “imagined communities” in which nationalist identities foster enhanced levels of social integration among individuals who share similar characteristics. These communities are “imagined” because they do not rest on concrete relationships

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

252 Shah, Acevedo and Ruiz between individuals, as the traditional definition of community would assume (Grudz et al. 2011:1296), but rather on an intangible, abstract sense that others are collectively identified in a larger group. Anderson (2006:6) states that members of “even the smallest nation will never know most of their fellow- members, meet them, or even hear of them, yet in the minds of each lives the image of their communion.” While traditional use of imagined communities points to nations, more recent scholarship demonstrates how the theory can be applied to non-national communities (e.g. Grudz et al. 2011; Powers 2011). Hage (2005), however, has cautioned against the use of imagined communities beyond political spaces. While an imagined pan-Arab community is beyond a national context, it is within a political space, satisfying this reservation. The pan-Arab community, which is composed of the Arab-speaking countries of the Middle East region, has been regarded as a unique “supranational territory” in the formation and perpetuation of identity and community (Culcasi 2011). It is important to note that although the communities are imagined, they also are collective lived experiences, embodied by the members (McDougall 2011; Wien 2011). Such national communities, according to Anderson (2006), succeeded the formation of modern states in Europe because of the rise of vernacular print and literacy. Literary advances made the spread of informa- tion through the common language possible. It is for this reason, as well as the implications for increased social cohesion when individuals are able to com- municate via common mediums, that language is considered focal to the theory of imagined communities (Tiryakian 2011:1396). The centrality of language, likewise, has been cited in pan-Arab movements and in Arab nationalism. The role of language in forming a sense of community is antecedent to the birth of modern pan-Arabism. For over a millennium, Arabs referred to those who could not speak Arabic properly—if at all—as Aʻjamī. The word Aʻjamī liter- ally translates to “people unable to speak properly,” but referring to another as A‘jam also implies that they are mute or dumb. Aʻjamī is an Arab-ethnocentric term, and, as such, has a very negative connotation in denoting non-Arabic speakers as inferior, both culturally and ethnically (Esposito 2003:12). Thus, the cultural marker of language in the Arab context has historically and contem- porarily created a sense of community, which acts as a litmus test for who does and—more importantly—does not belong. Today, the use of Modern Standard Arabic in unifying the Middle East (Pintak 2009) has been underscored as es- sential in the formation of an Arab imagined community. Additionally, the role of religion cannot be overlooked in the formation and fortification of the pan-Arab identity. For example, the majority of Middle Eastern Arabs identify with religious, nationalist, and ethnic identities, which in turn strengthen identifying with pan-Arabism. In a study on public opinion,

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 253

99 percent of Egyptians consider themselves religious (as opposed to secular) and 79 percent identify with the Arab-Muslim world (Moaddel and Abdul-Latif 2006). Although there are ongoing tensions between pan-Arabism and Arab nationalism (Henry et al. 2003), religion continues to be used as a cohesive agent for both national and supranational unity. Take, for example, Pintak’s (2009:91) illustration of Cairo in 2006:

Cheering crowds in downtown Cairo made up of secular liberals and members of the conservative Sunni Muslim Brotherhood, waving—side- by-side—photographs of Hassan Nasrallah, leader of Lebanon’s Shi‘ite Islamist movement Hizbullah, and Gamal ‘Abd al-Nasser, the icon of secular pan-Arabism.

Pintak continues this discussion by criticizing the supposed incompatibility of simultaneous identities of Islamism and Arab nationalism—in other words, he critiques the perception of the secular and religious divide in Arab nationalism. Instead, he argues that religious and ethnic identities are complexly inter- twined to form pan-Arabism, an imagined community. The interweaving of religion and ethnic identities is exemplified by the religious concept of jihad, which, in a post-imperialist era, is used to repel colonial and neocolonial influences. Theologically the word jihad is often associated with a general struggle or striving, including the necessary commitment re- quired to live a pious and devout life. While jihad is often translated as “holy war” in the West, it is understood as a form of resistance in the Arab world. For example, scholars have noted the sharp increase of nationalist and religious sentiment in response to threats of external influences, which often refer to jihad to expel threatening imperial forces (Tiryakian 2011). Because pan-Arabism was shaped by the historical legacy of imperialism, colonialism, and resistance against Western conquests (Said 1977:331–332), a culture of inclusion and exclusion arose, which is typical of community trajectories (Smith 1995). Within this imagined community were Arabs; outside, non-Arabs. As mentioned above, language, religion, and ethnicity played crucial roles in syner- gistically framing the contours of this community. However, communities do not exist in vacuums; as such, external forces significantly influence community formation. When threatened by outside groups, the identity of imagined community members strengthens and grow more salient. For example, Egyp- tians, like members of other Middle East nations, identified more strongly with pan-Arabism following conflicts between a majority Muslim state (Leba- non) and an out-group nation (Israel) (Pintak 2009). In terms of more recent events influencing pan-Arabism, none has been as pivotal in redefining and

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

254 Shah, Acevedo and Ruiz

strengthening community identity as September 11th (Pintak 2009). While previous research has demonstrated the salience of September 11th as an event in changing public attitudes about secular democracy and gender equality (Moaddel and Abdul-Latif 2006), changes in attitudes about groups of people as well as religious modifiers of those attitudes remain unstudied.

Egypt in a Post-9/11 World The Middle East has been considerably shaped by over a century of complex and generally tense relations with Western powers; a history that Arabs rarely forget, but which Americans rarely remember (Farber 2007). Given this legacy of political, cultural, and economic imperialism, animosity tends to increase when members of the pan-Arab community feel threatened by out- group members. For instance, Egyptians bore greater respect for the u.s. when the American government refrained from interfering with the Middle East. However, the goodwill Egyptians held for America has been undermined by relations between the u.s. and the Middle East over the past century, especially in the form of “demonstrations of American economic and financial muscle” (Sherbiny 2005:832) as well as the continued, unyielding support of the u.s. for Israel—what some have termed the perception of the “American led ‘Jewish Crusade alliance’” by Muslim political extremists (Moaddel and Abdul-Latif 2006). The public critique of the West in general, and America in particular, has been exacerbated by religious influences within Egypt, where there is a strong relationship between conservative forms of Islam and resentment towards globalization. For fundamentalists, globalization is perceived as a prob- lematic aspect of modern culture, which imposes “foreign” norms and values, which, in turn, disrupt the pan-Arab culture (Najjar 2005). Past interactions and ongoing relations, such as the u.s. intervention in the 1967 Six Day War and the 1973 October War, sustained support for Israel, instatement of auto- cratic governments, and backing of military coup d’états to overthrow existing governments, have laid the grounds for both religious and secular Arabs to resent American influence in the region. More recently, in the events following the tragedy of September 11th, Egyptians were very critical of the re- taliatory actions taken by the United States (Pew Research Center 2012). Given this context, we expect that after 9/11, members of the pan-Arab community in the Egyptian context will have significantly less favorable attitudes towards out-group members, including both persons of non-Arab races as well as Jews. There are, in fact, no significant Jewish populations in Egypt. The majority of Egyptian Jews, who lacked citizenship due to discrimination by the Egyptian government in the previous century (Shamir 1987), were pressured to flee Egypt in the mid 20th century after the formation of Israel. The emigrating Jews who

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 255 could afford to relocate to the Americas or Europe did so; the remainder went to Israel (Beinin 1998). Thus, in the absence of Jewish neighbors, Egyptians conflate identities of “Jews” with “Israelis” and “Zionists,” particularly after occurrences of Arab-Israeli conflicts (al-Mashat 1986). Even before the mass emigration of Jews from Egypt during the mid 20th century, when Jewish communists living in Egypt were critical of and rejected Zionism, Egyptians tended to conflate “Jews” with “Israelis,” “Zionists,” and “communists,” despite the ideological incompatibility of the latter two (al-Mashat 1986; Beinin 1998; Ginat 2011). Thus, because of the tendency to collectively exclude non-members of the pan-Arab community after experiences of regional conflict, we expect the events of 9/11 to exacerbate unfavorable views towards these out-group members. This leads us to consider social identity theory as a framework to assess the role of religious factors in shaping views towards perceived out-groups.

Social Identity and Outgroup Perceptions Conceptually rooted in the work of Henri Tajfel, with origins in the classic work of Mead and Cooley, the social identity perspective has been applied by social scientists seeking to explain prejudice and intergroup conflict— integrating ideas like social categorization, ethnocentrism, and intergroup relations (Berger 1966; Berger and Luckmann 1967; Tajfel 1974; Turner et al. 1987). According to this approach, social identity is a type of awareness of self-concept that shapes feelings of belonging and offers in-group valuation (Tajfel 1972:292). At the heart of social identity is the idea of social categorization (Turner et al. 1987). Social categorization describes attributes that, in essence, capture similarities and differences within and between groups, while also defining the type of behaviors that are acceptable to the in-group. Being categorized as a group member identifies a person’s normative behaviors that are based largely on appropriate group norms and expectations. This depersonalization of the individual and emphasis on the group fosters group solidarity as well as in-group trust and liking (Hogg 2004). Religion is a system of ideas used by individuals to interpret the world around them, which in turn shapes self- concept, modes of behavior, and moral guidelines by which to live, while in- voking notions of the sacred and divine to derive personal meaning in one’s life (Kimball 2002; Mol 1976). The significance of religious principles, when compared to other normative standards, stems from religion’s broad scope in addressing core cosmological questions related to existence, the afterlife, meaning, and morality. In this way, religious ideologies often serve as a master identity that provides ideological and behavioral guidelines as well as moral codes and principles which encourage understanding, kindness, compassion and love (Hogg 2004; Hogg 2006). On the other hand, these standards may also

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

256 Shah, Acevedo and Ruiz facilitate a framework by which “righteous” behaviors are rewarded, while jus- tifying the punishment of those who are “immoral” (Silberman 2005). There is, however, a consequence for those with strong religious identification and in-group preferences. Heightened religious categorization fosters ideas of superiority in one’s own worldview and an inflexible intolerance of moral contradictions held by members of out-groups. Such ethnocentric ideologies have been linked to forms of extremism (e.g. intolerance, domination, and violence), whereby outsiders are viewed as heretics (Haslam 2006; Haslam et al. 2008; Silberman, Higgins , and Dweck 2005). Group-identification through religion has been viewed as a form of coping that may, in some instances, alleviate feelings of doubt, acting as a powerful solace for people in times of collective uncertainty (McGregor et al. 2008), including after significant events like 9/11. Therefore, we expect a positive association between religiosity and intolerance for members of out-groups. Further, we anticipate that more frequent religious service attendance and higher God salience will exacerbate in-group preferences, and thus, less favorable attitudes for out-group members. Higher frequency of religious service attendance would reinforce group solidarity, while higher God salience would indicate the importance individuals place on religion in their self-identification. The points raised thus far lead us to generate several key expectations. Stated in a series of formal hypotheses, we expect the below outcomes following the events of September 11th:

H1: Levels of intolerance towards persons of other races will increase after 9/11. H2a: Increased beliefs in the importance of God (God salience) will be associated with greater levels of intolerance towards persons of other races. H2b: Increased attendance of religious services will be associated with greater levels of intolerance towards persons of other races. H3: Levels of intolerance towards Jews will increase after 9/11. H4a: Increased beliefs in the importance of God (God salience) will be associated with greater levels of intolerance towards Jews. H4b: Increased attendance at religious services will be associated with greater levels of intolerance towards Jews.

Methods

For the purposes of this study, we utilize a subset of the World Values Sur- vey (wvs 2009), focusing on two waves of survey data administered in

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 257

Egypt: the year prior to (2000 [pre-9/11] N = 3,000) and after (2002 [post-9/11] N = 1,000) the 9/11 attacks on the United States. Data collection for the two waves was conducted under the direction of the emac Research and Train- ing Center in Shams University, Cairo, Egypt. Both waves were conducted using self-administered and telephone interviews, and were based on a strati- fied probability sampling design with a systematic random selection. Neither wave includes a weight variable; thus, all analyses below are estimated using unweighted data. While a multi-stage sampling design was used to allow for generalization to the larger Egyptian population, there was no oversampling utilized; thus, the data do not contain a weight variable. Moaddel and Abdul- Latif (2006) previously used this data in a similar design to study changes in public attitudes about secular democracy and gender equality pre- and post- 9/11; this study aims to extend that research by analyzing changes in attitudes about out-groups as well as assessing effects of religiosity on those attitudes.

Analytical Strategy Quasi-experimental design can examine causality in phenomenon wherein random assignment is not possible. This research design allows for non- random assignment while retaining the traditional form of experimentation of a comparison group and multiple assignments per treatment. The problem with randomized experiments is the inability to control for various validity threats, leading to unsound generalizability of study relationships and wa- vering statistical inferences (Cook and Campbell 1986; Fisher 1925). Research exploring diverse topics ranging from family planning to organized war has utilized quasi-experimental design for its simplistic, yet rigorous approach (Leon and Cuesta 1993; Pelowski 1971). While some non-experimental research designs appear practical superficially, they might distort variable relationships. Other times, more complex designs overcomplicate an assessment of parsimo- nious relationships. Not only does quasi-experimental design allow for valid methods to address these concerns, it also provides avenues to resolve additional methodological issues. For example, Romanelli and Tushan (1986) employed a quasi-experimental design in their comparative-longitudinal study examining the effect of differing contextual conditions on varying patterns of change in organizational activity. Since neither randomization nor control groups could be established, the researchers formed “quasi-experimental groupings”—a basic comparability of units—to examine those dimensions which influence a subject’s response to stimuli (Romanelli and Tushan 1986:610). This allowed stimuli effects to be parsed out from pre-treatment elements. In a similar design to our analysis, Legewie (2013) examined European perceptions of immigrants as a

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

258 Shah, Acevedo and Ruiz

consequence of terrorism, arguing that under certain conditions, events shape attitudes and perceptions of out-group members. Using a quasi-experimental design, this study found that the attack on Bali significantly exacerbated anti- immigrant sentiments (Legewie 2013). In another study with a similar design to our analysis, Inglehart, Moaddel, and Tessler (2006) analyzed the impact of insecurity on xenophobia and in-group solidarity in Iraq. Also using a quasi- experiment design, the researchers found that insecurity significantly exacerbated xenophobia and in-group preference. Given both the validity and value of using quasi-experimental design in measuring attitudinal consequences of events as well as the advantages of a quasi-experimental design over alterna- tive designs, we find this analytic method compelling as a means of assessing our key hypotheses.

Dependent Variables The wvs includes items intended to gauge tolerance for individuals in specific social categories. The prompt reads as follows: “Could you please sort out any [from a series of categories] that you would not like to have as neighbors?” In accordance with the exclusivity element of the theory of imagined communities, we are primarily interested in the level of intolerance for non-Arab persons that are included in the wvs data for Egypt. These groups also contain sufficient variation in responses to warrant analysis. Thus, the first group we analyze are, “People of a different race,” and secondly, “Jews.” The dichotomous outcomes were “Not mentioned” (=0)—implying tolerance—and “Mentioned” (=1)—signifying intolerance. This design allows us to estimate models using binary logistic regression and present odds ratios for intolerant views.

Key Predictors In order to assess the effects of 9/11 on levels of intolerance, we dummy code survey wave, allowing us to compare public opinion after 9/11 (=0) to pre-9/11 levels (=1, omitted).1 This strategy accounts for variation in opinion as a function of a major world event occurring just prior to and after the data collection stage, so that 9/11 approximates a treatment effect in quasi-experimental design. Furthermore, our study also explores possible correlates with religiosity, net of socioeconomic and demographic controls. Our first measure of religiosity is a measure of God-salience, which asks, “How important is God in your life?” A five-point ordinal scale captures

1 Analysis of survey data was carried out using SAS 9.3, which omits the highest value category.

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 259

responses that range from “Not at all important” (=0) to “Very important” (=4). Due to the low number of Egyptians reporting low levels of God salience, we recode the original 10-level scale and collapse the lowest levels of salience to create our five-point scale. The measure of religious service attendance asks, “Apart from weddings, funerals and christenings, about how often do you at- tend religious services these days?” We retain and reverse code the original eight-item ordinal scale such that it ranges from “Never or practically never” (=0) to “More than once a week” (=7). We had initially included an interaction term between gender and religious service attendance, since masjids in Egypt may be male dominated. This term was not found to be significant, possibly reflecting a gendered difference in kind, not degree or frequency, of women’s religious service attendance in Egypt (see, for example, Mahmood 2004). Our final religiosity indicator takes the denominational affiliation codes for Egypt to generate a country specific, dummy-coded measure that allows us to distin- guish between Egyptian Christians (=0) and Muslims (=1, omitted).

Control Variables Our study includes a series of controls for socioeconomic and demographic factors. Level of education is operationalized by an ordinal measure that asks, “What is the highest educational level you have attained?” Possible responses range from, “No formal education” (=0), through “University with degree/High- er education – upper-level tertiary” (=9). Household income is ascertained by a wvs prompt that reads, “Here is a scale of incomes. We would like to know in what group your household is, counting all wages, salaries, pensions and other incomes that come in.” We use the income variable as recoded into three intervals by original investigators but add an additional flagged dummy for “not asked/refused” (=0) to retain cases. Remaining categories are coded as follows: “Low income” (=0), “Middle income” (=1), and “High income” (=2, omitted). wvs data for Egypt include a measure of subjective social class, which reads, “People sometimes describe themselves as belonging to the working class, the middle class, or the upper or lower class. Would you describe your- self as belonging to the,” “Lower class” (=0), “Middle class” (=1) or “Upper class” (=2, omitted). For missing cases, we again created a flagged dummy for the “don’t know/refused” category (=0). Our models also include three distinct demographic controls: gender, marital status, and age. Gender is dummy coded with male (=0) and female (=1, omitted). Likewise, marital status is dummy- coded with married (=0) and not currently married (=1, omitted). Finally, age is dummy coded, comparing younger respondents (“15–29” = 0) to their older counterparts (“over 29” = 1, omitted).

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

260 Shah, Acevedo and Ruiz

Results

Before moving to results of logistic regression models, we point out several key characteristics of the descriptive statistics shown in Table 1. First, we can see by the data that intolerance is greater for persons of “other races” (32 percent) compared to “Jews” (9 percent). Next, we note the high levels of religiosity in Egypt as reflected by measures of central tendency of our two indicators of religiousness. God salience exhibits a mean score of 3.66 on a five-point scale (midpoint = 3), reflecting a skewed distribution for this measure. Likewise, the mean score for religious service attendance is 2.53 on a six-point scale (midpoint = 2.5), which reflects a slightly skewed distribution. Conversely, levels of education are generally low among respondents in our study sample, with a mean score of 3.36 (midpoint = 3.5). Finally, while income shows a bimodal distribution, with the highest percentages found in the low and high income categories, the measure for subjective social class seems to follow a relatively normal distribution. We limit our discussion of descriptive statistics to these key features of the data and move on to the results from multivariate models. The next tables present results of nested models that are designed to assess the impact of relevant predictors on measures of intolerance, net of statistical controls, towards people of other races (Table 2) and Jews (Table 3). For both tables, Model 1 includes only the effect of survey wave (e.g., pre- or post-9/11), Model 2 introduces religious predictors, while Models 3 and 4 add socioeconomic and demographic controls, respectively. We include an interaction term (education by wave) shown as Model 5 of Table 2. In an independent-samples t-test, we compared levels of education between pre- and post-9/11 data waves, and found that education levels were significantly higher in the latter wave (not shown; t-test available upon request). We therefore opted to include an interaction between education (centered) and survey wave to explore how this sociodemographic shift impacted tolerance levels towards out-group members. Additional measures of demographic shifts were inductively tested as interactions, but failed to reach statistical significance, and, as such, are not presented in our results.

Intolerance towards Neighbors of “Other Races” Table 2 shows results of binary logistic regression models predicting odds of greater intolerance, operationalized here by identifying people of “other races” as undesirable neighbors. Survey wave is shown to be a robust and significant predictor of decreasing the odds of bearing intolerant attitudes towards persons of other races as a consequence of the events of 9/11. Thus, contrary to our

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 261

Table 1 Sample characteristics: Egypt 2000 and 2001 World Values Survey (N = 4,000).

N Percent Mean sd Min Max

Dependent variables: Not as Neighbors Different race mentioned 1,296 32 – – – – Jews mentioned 371 09 – – – – Main predictor Post 9/11 1,000 25 – – – – Pre 9/11 3,000 75 – – – – Religion Variables Christian 243 06 – – – – Muslim 3,755 94 – – – – God salience 4,000 – 3.66 0.83 0 4 Religious service attendance 4,000 – 2.53 1.90 0 5 Control Variables Education 3,998 – 3.36 2.74 0 8 Income (Not Asked) 380 10 – – – – Low income 1,205 30 – – – – Middle income 936 23 – – – – High income 1,479 37 – – – – Subjective Social Class (dk/Refused) 111 02 – – – – Lower class 1,190 30 – – – – Middle class 1,687 42 – – – – Upper class 1,012 25 – – – – Male 2,027 51 – – – – Female 1,973 49 – – – – Married 2,660 67 – – – – Marital Status (others) 1,973 34 – – – – Age (15–29) 1,365 34 – – – – Age (over 29) 2,631 66 – – – – first hypothesis, we find that across models, September 11th had the effect of lowering odds of holding intolerant views towards people of other races, leading us to reject our first hypothesis (H1). Specifically, in Model 4, the full model that includes all socioeconomic and demographic controls, when compared to pre-9/11 views, respondents at the time immediately following 9/11 have a 25% lower odds of responding that they would not be willing to have persons

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

262 Shah, Acevedo and Ruiz

Table 2 Estimated odds ratios predicting intolerance towards neighbors of “a different race”.

Model 1 Model 2 Model 3 Model 4 Model 5

Main Predictor Variable Post 9/11 (omitted = Pre) 0.72*** 0.71*** 0.75*** 0.75*** 0.75*** Religion Variables God Salience 1.11** 1.10* 1.09* 1.10* Religious service attendance 1.03† 1.03† 1.03† 1.03† Christian (omitted = Muslim) 1.03 1.08 1.10 1.11 Control Variables Education 0.97* 0.97* 0.99 Income (omitted = high income) 1.09 1.09 1.09 “dk/Not asked” Low income 1.18† 1.18† 1.17† Middle income 0.96 0.96 0.95 Social class(omitted = higher) 0.72 0.73 0.76 “dk/Refused” Lower class 1.42** 1.46*** 1.47*** Middle class 1.27* 1.28** 1.27* Male (omitted=female) 0.86* 0.86* Married (omitted = other) 1.00 1.00 Age 15–29 (omitted = over 29) 1.00 0.99 Education × Post 9/11 0.91** Intercept 0.52*** 0.31*** 0.28*** 0.30*** 0.28***

Likelihood ratio χ- square 17.47*** 27.56*** 73.70*** 78.18*** 87.40*** −2 log likelihood 5021.31 5006.70 4954.49 4950.01 4940.79 Df 1 4 11 14 15 N 4,000 3,998 3,994 3,994 3,994

†p < .10. *p < .05. **p < .01. ***p < .001. of other races as neighbors (OR = 0.75, p < .001). In other words, Egyptian respondents report lower levels of intolerance for persons of other races after, compared to before, September 11. Turning to measures of religiosity, we find general support for our religion hypotheses (H2) in the form of a modest positive association between religiousness and heightened feelings of intolerance towards persons of other

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 263 races. First, with each unit increase in God salience (e.g. “How important is God in your life?”), there is an approximately 9% higher odds of intolerance towards persons of other races (Model 4, OR = 1.09, p < .05), lending support to H2a. Religious service attendance has a similar, albeit weaker, impact on intolerance towards neighbors of other races. Each unit increase in religious service attendance is associated with a 3% higher odds of not wanting to have persons of other races as neighbors (Model 4, OR = 1.03, p < .10), lending partial support for H2b. We next examine several noteworthy effects of key control variables. While level of education and income have moderate effects when predicting intolerance towards people of other races, subjective social class has the most pronounced effect. First, each unit increase in education is associated with an approximately 3% lower odds of intolerance towards neighbors of other races (Model 4, OR = 0.97, p < .05). Next, compared to high income respondents, individuals who report low income status have an approximately 18% higher odds of holding intolerant views, net of other factors, but these effects are marginally significant (Model 4, OR = 1.18, p < .10). However, subjective social class seems to be significantly and palpably correlated with intolerant views. Compared to those reporting “upper class” status, Egyptians who self-report as belonging to the “lower class” or “middle class” have greater odds of intolerance towards neighbors of other races (46%, p < .001; 28%, p < .01, respectively). Finally, and surprisingly, men have lower odds (14%) of reporting intolerance than women (Model 4, OR = 0.86, p < .05). For this study, we also tested the possible moderating effects between the pre- and post-9/11 survey wave and educational attainment (Model 5, Table 2). The positive and significant interaction effect between level of education and survey wave indicates that the positive association between the post-9/11 wave and tolerance towards neighbors of other races is moderated by increased education. While 9/11 had the effect of lowering intolerance towards neighbors of other races, this effect is magnified for Egyptians with higher levels of education. We discuss the implications of this finding in greater detail below after reviewing our findings on intolerance towards Jewish neighbors.

Intolerance towards Jewish Neighbors We now turn to Table 3, which shows the models estimating intolerance towards Jewish neighbors. What immediately stands out is the even more robust, sizable, and significant post-9/11 effect on changing attitudes towards Jewish neighbors when compared to the previous results for intolerance towards neighbors of other races. Ancillary analysis using bivariate probit models

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

264 Shah, Acevedo and Ruiz

Table 3 Estimated odds ratios predicting intolerance towards Jewish neighbors.

Model 1 Model 2 Model 3 Model 4

Main Predictor Variable Post 9/11 (omitted = Pre 9/11) 0.01*** 0.01*** 0.02*** 0.02*** Religion Variables God Salience 0.84** 0.86** 0.86*** Religious service attendance 1.06* 1.06* 1.06 Christian (omitted = Muslim) 2.41*** 2.55*** 2.58*** Control Variables Education 1.02 1.03 Income (omitted = high income) 1.45* 1.44† “dk/Not asked” Low income 0.88 0.88 Middle income 1.08 1.08 Social class (omitted = higher) “dk/ 4.55*** 4.59*** Refused” Lower class 2.49*** 2.53*** Middle class 1.40* 1.40* Male (omitted = female) 0.94 Married (omitted = other) 0.95 Age 15–29 (omitted = over 29) 1.00 Intercept 0.14*** 0.21*** 0.10*** 0.10***

Likelihood ratio χ- square 204.82*** 235.99*** 292.75*** 293.29*** −2 log likelihood 2266.01 2234.44 2176.91 2176.37 Df 1 4 11 14 N 4,000 3,998 3,994 3,994

†p < .10. *p < .05. **p < .01. ***p < .001.

confirms a stronger effect on intolerance towards Jews than neighbors of other races (results available upon request). As with our first hypothesis (H1), we reject our third hypothesis on the grounds that across models, 9/11 had the effect of drastically lowering odds of holding intolerant views towards having Jews as neighbors. Specifically, in Model 4, when compared to pre-9/11 attitudes, respondents at the time immediately following 9/11 have a 98% lower odds of responding that they would not be willing to have Jewish persons as neighbors, net of controls (OR = 0.02, p < .001).

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 265

Next, we turn to measures of religiosity, which indicate much stronger and pronounced effects on views towards Jewish neighbors than was the case for persons of other races. First, for each unit increase in God salience (e.g. “How important is God in your life?”) there is an approximately 14% lower odds of intolerance towards Jewish persons, net of other factors (Model 4, OR = 0.86, p < .001). Once again, we find evidence contrary to our initial expectations, leading us to reject H4a. Interestingly, religious service attendance has a reversed, albeit weaker and less robust, impact on intolerance towards Jew- ish neighbors. In Model 3, each unit increase in mosque attendance is associated with a 6% higher odds of not wanting to have Jewish neighbors (OR = 1.06, p < .05), but the effect is mediated by sociodemographic controls (Model 4, OR = 1.06, ns). Thus, H4b is only partly supported by the data. We discuss the implications of these finding below but first note the key effects of control variables. Unlike the effects of several control variables on views towards neighbors of other races mentioned above, only subjective social class stands out as a notable predictor of intolerance towards Jews. The overall effect indicates that lower social class status is palpably related to heightened intolerance towards Jewish neighbors. When compared to self-identified members of the Egyptian “upper class,” individuals who identify as “lower class” have 153% higher odds (Model 4, OR = 2.53, p < .001) while members of the “middle class” report 40% higher odds (Model 4, OR = 1.40, p < .05) of holding intolerant views towards Jewish neighbors, net of statistical controls. Interestingly, in terms of religious affiliation, identifying as Christian has the strongest and most robust effect when predicting intolerance towards Jews. Across models, when compared to Muslims in Egypt, Christian affiliation significantly increases the odds that Egyptians will hold intolerant views towards Jews as neighbors. Estimates shown in Model 4 suggest that Christians in Egypt have over one and a half times higher odds of holding intolerant views towards Jews when compared to Egyptian Muslims, net of all statistical controls (OR = 2.58, p < .001). We discuss the implications of this finding below, placing our results in both an explanatory and theoretical context.

Discussion

Our study leverages a subsample of the World Values Survey and utilizes a quasi-experimental design where the events of 9/11 serve as a treatment, allowing us to assess attitudinal variation prior to (2000) and immediately after (2002) the September 11th attacks. We consider the implications of Anderson’s

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

266 Shah, Acevedo and Ruiz

(2006) theory of imagined communities for in-group/out-group processes in the context of pan-Arabism. Contrary to expectations, our results strongly suggest that there was greater tolerance for out-group members in Egypt post- 9/11. Parallel to these findings, researchers have presented results that indicate attitudinal change in favor of secular democracy and gender equality in post- 9/11 Egypt (Moaddel and Abdul-Latif 2006). Contrary to both Huntington’s (1996) thesis and assumptions about Arab/ Muslim intolerance for others, we find a heightened climate of global conflict fosters greater tolerance for out-group members in Egypt. After 9/11, we find that Egyptians were more likely to be tolerant of neighbors who are either persons of other races or Jewish. We also find significant but inconsistent religiosity effects on intolerance. Our discussion will frame these results within both explanatory and theoretical contexts. We first address plausible social, economic, and religious mechanisms in Egypt that might explain our central findings. We then connect these points to the theoretical frame guiding our study and assess the merits and areas for expansion in the imagined communities perspective. We close by discussing study limitations and avenues of future research. A starting point is consideration of the overall effect of 9/11 reported here, which indicates a substantial decrease in levels of intolerance for neighbors of non-Arab races. In other words, Egyptians were substantially and significantly more accepting of out-group members immediately after the events of 9/11 than prior to the attacks. This response warrants detailed consideration. Like Moaddel and Abdul-Latif (2006), we surmise that the 9/11 attacks—through the global media dissemination of the incident—impacted public opinion in multiple ways. On the one hand, our findings suggest that 9/11 might have led to feelings of condolence for the victims of the attack, which may foster a greater sense of integration among persons experiencing a heightened sense of vulnerability and threat (Schwarzer, Bowler, and Cone 2014). In this sense, Egyptians may have turned to feelings of heightened tolerance as a coping mechanism (Jennings and McLean 2013). It is likely that the global collective trauma post-9/11 fostered an emotional climate amongst Egyptians conducive to higher levels of national integration that undermined pre-existing sentiments of exclusion directed towards members of out-groups (see also Rime, Paez, Basabe, and Martinez 2010). For this discussion, we depart from a monocausal interpretation that the sole factor in lowering intolerance for out-groups among Egyptians relates to the events of September 11th. Instead, with a focus on domestic issues pressing Egypt, we find compelling evidence that economic circumstances situated a climate within which the events of 9/11 could foster greater and more progres- sive tolerance, as opposed to reactionary intolerance for out-groups. Previous

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 267

research has indicated that while times of duress may lead to increased intolerance and prejudice aimed at members of out-groups (Doebler 2014; Inglehart et al. 2006; Schafer and Shaw 2009), tragic collective events may also have the unintended effect of fostering social cohesion and altruistic behaviors (Morgan, Wisneski, and Skitka 2011). It is likely that our results largely reflect the latter proposition, which was possible in Egypt because of economic tensions facing the nation. We move to a brief consideration of these economic trends in Egypt prior to and after 9/11. Between 2000 and 2002, the Egyptian economy underwent a significant downturn as evidenced by several economic and financial indicators. These include decreases in Gross Domestic Product (gdp), Gross National Savings, Gross Domestic Investments, and public financing, along with increases in Total External Debt Stock (edt) and domestic debt (African Development Bank 2006). Additionally, despite significant increases in education, unemploy- ment rates continued to increase between 2000 and 2002. Egyptians may have felt a sense of greater vulnerability for their own safety after the attacks—both physical and fiscal—while simultaneously fostering an empathetic response to the suffering of others. This would translate to lower levels of intolerance for out-group members in Egyptian society. In the past, when facing severe economic issues, Egyptians have not only called for greater rights and freedom but also demonstrated an urge to shift towards democracy. And, in times of sociopolitical and economic unrest, Egyptians have also manifested a desire to make peace with Israel (see, for example, al-Mashat 1986). Our findings also indicate distinct patterns related to religious factors as correlates of intolerance. As shown above, God salience and religious service attendance both have moderate and positive associations with intolerance to neighbors of other races. In other words, stronger God salience and increased frequency of religious service attendance both increase the likelihood of Egyptians holding intolerant views of non-Arab persons, while religious affiliation (Christian as opposed to Muslim) has no effect. However, the same religious factors show mixed effects on views towards Jewish neighbors, such that while God salience relates to more tolerance for Jewish neighbors, frequency of religious service attendance as well as Christian religious affiliation relates to less intolerance. We treat each of these items individually with emphasis on previous scholarship that has explored links between religious factors and perceptions of intolerance. God salience and religious service attendance variables are both related to increased levels of racial intolerance. As previous studies have shown, religion may influence racial attitudes in distinct ways (Doebler 2014). In some instances, religion has been shown to foster tolerance and a greater acceptance of members of other racial/ethnic groups (Kunovich and Hodson 1999; Shen

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

268 Shah, Acevedo and Ruiz et al. 2013). However, there are reasons to suspect that religiosity may likewise promote intolerant attitudes towards members of out-groups (Johnson et al. 2011; Scheepers, Gijsberts, and Hello 2002). Religion may invoke strong feelings of association and fellowship with fellow congregants, a faith-based imagined community. The commitment to this community may, in turn, translate to suspicion and mistrust—and therefore, exclusion—aimed at out-group members (Scheepers et al. 2002). In the context of Egyptian society, where religious membership and racial identification are often concomitant categories of political identification with pan-Arabism, exclusion of out-group members (racial minorities) may be amplified by religiosity. As previous studies have shown, religiousness may serve to amplify ethnocentric feelings against out- groups, subsequently exacerbating levels of racial or ethnic intolerance (Hall, Matz and Wood 2010). But if this is the case, what might explain the opposite effect when consid- ering the impact of God salience on tolerance for Jewish neighbors? As noted above, while religious service attendance seems to have a positive but negligi- ble, and ultimately mediated, effect on intolerance, God salience and Christian affiliation each have sizable and significant effects on attitudes towards Jewish neighbors. We will, again, treat these outcomes individually, first addressing the effects of God salience. Muslims with higher levels of God salience likely adhere to pluralistic Quranic teachings to a greater extent than those reporting lower levels of God salience. A reoccurring phrase in the Quran mentions both Jews and Christians as monotheistic “people of the book” (e.g. the Bible) and therefore merit respect and rank alongside Muslim believers (see, for example, Quran 2:62 and 5:69). Thus, the religious pluralism encouraged by these verses may translate into greater tolerance for Jewish neighbors—but not necessarily neighbors of other racial backgrounds, whose monotheistic status is unknown. Next, we consider the highly significant religious affiliation effect show- ing much higher levels of intolerance towards Jewish neighbors on the part of Egyptian Christians when compared to Muslims—contrary to Orientalist expectations of Muslims as more intolerant. This finding is not surprising for the Egyptian context: Coptic Christians, the predominant Christian group in Egypt, are highly religious and identify strongly with their own faith-based community (Scott 2010:169). Given the strong association we found between religiosity and intolerance for individuals of other backgrounds, the higher concentration of religiosity among Egypt’s Christian community positions them to be less tolerant of Jews than their Muslim counterparts. Furthermor e, the ongoing Israeli occupation of West Jerusalem and the seizure of Bethle- hem as well as the subpar living conditions of Palestinian Christians may impact the general attitudes towards Jews among Egyptian Coptic Christians. The

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 269 pilgrimage to Palestine, now banned by Coptic Christian leaders because of the Israeli occupation, is an important religious practice for Coptic Christians (but not Muslims); thus, we can understand the higher levels of intolerance for Jews generally, and Israel specifically, by Egyptian Christians when compared to Muslims. We also find evidence of relationships between socioeconomic factors and levels of intolerance. Higher levels of education and income are associated with greater tolerance towards neighbors of other races. On the other hand, neither education nor income seems to impact views of Jewish neighbors in any significant way. What does stand out is the substantial and consistent relationship between subjective social class and intolerance. For both sets of potential neighbors, when compared to Egyptians who self-identify as upper class, identification with the lower and middle classes is a robust predictor of intolerance towards both neighbors of other races and Jewish neighbors, suggesting a relationship between lower levels of socioeconomic status and greater intolerance. This general finding resonates with previous studies that have shown links between lower social class status and attitudes against out-groups (for a review of the literature, see Carvacho et al. 2013). On the one hand, as with our findings, social class categorizations have been shown to work in divergent ways that are often distinct from other measures of socioeconomic status (Weeks and Lupfer 2004). In our case, it is subjective social class that has the great- est impact on the odds of intolerance. Persons categorizing themselves as members of lower income or disadvantaged groups may often live in highly segregated areas where biases are reinforced by structural factors that sepa- rate persons based on ascribed statuses (Coenders and Scheepers 2003). This pattern of forced segregation may translate to increased racial intolerance for members of out-groups, particularly those concentrated in higher status oc- cupations and neighborhoods. It is also possible that perceived social class may lead members of disadvantaged groups to feel greater levels of existential threat from members of out-groups (Quillian 1995). This line of inquiry may offer fruitful avenues of future scholarship. As with other studies that make use of cross-sectional data, we are aware of several limitations present in our study. We discuss these shortcomings and present avenues of future research. First, the study design approximates a quasi-experimental design where a major world event serves as the treatment between two time periods in a specific context. While we are careful to include an extensive array of control variables as part of this study, assessing causality is always a concern when using data that typically allows for correlational (not causal) claims. As with other quasi-experimental designs, it is difficult to

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

270 Shah, Acevedo and Ruiz offer an absolute causal assessment, since findings could be the result of other potential factors. Future studies drawing from panel study designs could examine the impact of significant historical, sociopolitical, and economic events surrounding panel waves, while accounting for individual-level factors. Such an approach would allow researchers to assess more direct linkages between exogenous events situated within specific contexts and relative changes in social attitudes. There are also limitations related to the data used for this study. First, the wvs is limited in the number of religiosity measures available in Egypt. A more extensive array of religious items would allow for measurement of distinct—and Muslim-appropriate—dimensions of religiousness (e.g. see González 2011). Similarly, the wvs measures of intolerance are framed around neighborhood context and are not general measures designed to capture intolerance. It is possible that asking specifically about neighbors may be interpreted in a very specific context that might not accurately capture feelings of intolerance in a more general sense. Furthermore, single-item measures lack the reliability of multiple- item indicators of both religiosity and intolerant attitudes. Future studies could incorporate multiple item scales and indexes that might capture the intended constructs with greater specificity than the currently available data. Finally, intolerance as an attitudinal construct is a complex and multifac- eted concept that we can only tap into here at the level of cross-sectional data. Particularly in light of our research design, which explores the potential impact of a large-scale exogenous factor, we cannot claim with certainty that alterna- tive explanations have been ruled out. It is possible that there are additional factors impacting sudden changes in intolerance not captured by quantitative data. The formation and interpretation of meaning structures in light of major historical events would benefit greatly from qualitative approaches, which are more suited to properly inform interpretive processes that individuals en- gage in when coping with large-scale social disruptions, including economic crisis as well as military attacks, natural disasters, and epidemics, among other events. Through in-depth interviews, it may be possible to better contextualize and specify associations between such events and processes by which imagined in-group/out-group identities are magnified, minimized, or dynamically dealt with. Despite these limitations, our study contributes to several areas of interest. First, we examine intolerance against out-group members as a function of traumatic, exogenous events within the specific context of an imagined community. Our study indicates substantial effects following a major world event within a specific context and its impact on views of out-group members. As such, our findings present implications for Anderson’s (2006) theory of

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 271 imagined communities. Following the contours of the theory, in particular applying the concepts of inclusion and exclusion of out-groups, we hypoth- esized that tolerance for out-group members would significantly decline after the events of 9/11. However, we found the opposite effect, and argue this effect is most likely the outcome of the specific context in which Egyptians were situated. As noted above, the Egyptian economy declined between 2000 and 2002, and economic deterioration has been historically linked to E gyptians’ increased tolerance for members of out-groups. We therefore argue that the theory of imagined communities must be contextualized to the specific political, social, and economic history of the community under analysis. Attention must be paid to unique cultural and sociodemographic shifts. In addition to the aforementioned financial circumstances pressing Egypt, we found significant effects for education moderating tolerance from pre- to post-9/11, such that education magnified the increase of tolerance in the latter wave. Just as Weber (2009:275) emphasized the significance of cultural context, the “switch- men” of social trajectories, we reiterate the necessity of taking an imagined community’s sociopolitical, economic, and historic context into consideration as these factors situate responses to and evaluations of both domestic and foreign events, including those related to September 11th. Furthermore, we present our results and implications of those findings within the context of a predominantly Muslim nation. Consequently, we place our work within the scant yet bourgeoning scholarship in the Sociology of Islam. We then offer quantitative, empirical evidence that informs the imagined communities literature in the Arab context—work that is often associated with interpretive, qualitative approaches. Finally, we add to methodological discussions on the use of quasi-experimental designs when historical events serve as the treatment variable. By applying such methodological approaches to understanding the multiple dynamics that influence social outcomes among understudied populations, we have opened an avenue for future scholarship.

References

Acevedo, Gabriel A. (2008). Islamic fatalism and the clash of civilizations: an appraisal of a contentious and dubious theory. Social Forces, 86(4): 1711–1752. doi: https://doi .org/10.1353/sof.0.0033. African Development Bank. (2006). Egypt: country strategy paper update. Regional Department North 1. Retrieved 2 February 2017 (www.afdb.org/en/documents/ document/2000-2002-egypt-country-strategy-paper-12986/).

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

272 Shah, Acevedo and Ruiz

Ahmed, Akbar. (2002). Ibn Khaldun’s understanding of civilizations and the dilemmas of Islam and the west today. Middle East Journal, 56(1): 20–45. al-Mashat, Abdul-Monem. (1986). The Egyptian-Israeli settlement. Journal of Arab Affairs , 5(1): 81–110. Anderson, Benedict. (2006). Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York: Verso. Azzam, Maha. (2013). Egypt’s democratic experiment: challenges to a positive trajec- tory. Insight Turkey, 15(2): 157–169. Bayat, Asef. (2013). Egypt and its unsettled Islamism. In Bayat, Asef (ed.), Post-Islamism: The Changing Faces of Political Islam, pp. 185–239. New York: Oxford University Press. doi:10.1093/acprof:oso/9780199766062.003.0007. Beinin, Joel. (1998). The Dispersion of Egyptian Jewry: Culture, Politics, and the Forma- tion of a Modern Diaspora. Berkeley: University of California Press. doi: 10.1017/ S0364009402400116. Berger, Peter L. (1966). Identity as a problem in the sociology of knowledge. European Journal of Sociology, 7: 105–115. Berger, Peter L. and Thomas Luckmann. (1967). The Social Construction of Reality; A Treatise in the Sociology of Knowledge. Harmondsworth, England: Penguin. Cainkar, Louise A. (2002). No longer invisible: Arab and Muslim exclusion after September 11. Middle East Report, 224: 22–29. doi: 10.2307/1559419. Cainkar, Louise A. (2009). Homeland Insecurity: The Arab American and Muslim American Experience After 9/11. New York: Russell Sage Foundation Press. Carvacho, Héctor, Andreas Zick, Andrés Haye, Roberto González, Jorge Manzi, Caroline Kocik, and Melanie Bertl. (2013). On the relation between social class and prejudice: the roles of education, income, and ideological attitudes. European Journal of Social Psychology, 43(4): 272–285. doi: https://doi.org/10.1002/ejsp.1961. Chiozza, Giacomo. (2002). Is there a clash of civilizations? Evidence from patterns of international conflict involvement, 1946–97. Journal of Peace Research, 39(6): 711–734. doi: 10.1177/0022343302039006004/. Coenders, Marcel and Peer Scheepers. (2003). The effect of education on nationalism and ethnic exclusionism: an international comparison. Political Psychology, 24(2): 313–343. doi: 10.1111/0162-895X.00330. Cook, Thomas D. and Donald T. Campbell. (1986). The causal assumptions of quasi – experimental practice: the origins of quasi-experimental practice. Synthese, 68(1): 141–180. doi: 10.1007/BF00413970. Culcasi, Karen. (2011). Cartographies of supranationalism: creating and silencing ter- ritories in the ‘Arab homeland.’ Political Geography, 30(8): 417–428. doi: 10.1016/ j.polgeo.2011.08.003. Detroit Arab American Study Team. (2009). Citizenship and Crisis: Arab Detroit after 9/11. New York: Russell Sage Foundation.

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 273

Doebler, Stephanie. (2014). Relationships between religion and intolerance towards Muslims and immigrants in Europe: a multilevel analysis. Review of Religious Re- search, 56(1): 61–86. doi: 10.1007/s13644-013-0126-1. Eglitis, Aaron. (2015). Paris terror unites East Europe against Merkel’s refugee plan. Bloomberg. Retrieved 21 January 2017 (www.bloomberg.com/news/articles/ 2015-11-17/paris-terror-uniting-east-europe-against-merkel-s-refugee-plan). Esposito, John L. (2003). Oxford Dictionary of Islam. New York: Oxford University Press. doi: 10.1093/acref/9780195125580.001.0001. Farber, David. (2007). What They Think of us: International Perceptions of the United States Since 9/11. nj: Princeton University Press. doi: http://www.jstor.org/stable/j .ctt7t34g. Fisher, Ronald A. (1925). Statistical Methods for Research Workers. London: Oliver and Boyd. Fox, Jonathan. (2002). Ethnic minorities and the clash of civilizations: a quantitative analysis of Huntington’s thesis. British Journal of Political Science, 32(3): 415–434. doi: 10.1017/S0007123402000170. Ginat, Rami. (2011). A History of Egyptian Communism: Jews and their Compatriots in the Quest of Revolution. Boulder, co: Lynne Rienner. González, Alessandra L. (2011). Measuring religiosity in a majority Muslim context: gender, religious salience, and religious experience among Kuwaiti college students – a research note. Journal for the Scientific Study of Religion, 50(2): 339–350. doi: 10.1111/j.1468-5906.2011.01571.x. Green, Sarah, Chris Gregory, Madeleine Reeves, Jane K. Cowan, Olga Demetriou, Insa Koch, Michael Carrithers, Ruben Andersson, Andre Gingrich, Sharon Macdonald, Salih Can Açiksöz, Umut Yildirim, Thomas Hylland Eriksen, Cris Shore, Douglas R. Holmes, Michael Herzfeld, Marilyn Strathern, Casper Bruun Jensen, Keir Martin, Dimitris Dalakoglou, Georgos Poulimenakos, Stef Jansen, Čarna Brkovič, Thomas M. Wilson, Niko Besnier, Daniel Guinness, Mark Hann, Pamela Ballinger, Dace Dz enovska. (2016). Brexit referendum: first reactions from anthropology. Social Anthropology, 24(4): 478–502. doi: 10.1111/ 1469-8676.12331. Grudz, Anatoily, Barry Wellman and Yuri Takhteyev. (2011). Imagining Twitter as an imagined community. American Behavioral Scientist, 55(10): 1294–1318. doi: https:// doi.org/10.1177/0002764211409378. Hage, Ghassan. (2005). A not so multi-sited ethnography of a not so imagined community. Anthropological Theory, 5(4): 463–475. doi: https://doi.org/10.1177/ 1463499605059232. Hall, Deborah L., David C. Matz, and Wendy Wood. (2010). Why don’t we practice what we preach? A meta-analytic review of religious racism. Personality and Social Psy- chology Review, 14(1): 126–139. doi: https://doi.org/10.1177/1088868309352179.

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

274 Shah, Acevedo and Ruiz

Haslam, Nick. (2006). Dehumanization: an integrative review. Personality and Social Psychology Review, 10(3): 252–264. doi: https://doi.org/10.1207/s15327957pspr1003_4. Haslam, Nick, Stephen Loughnan, Yoshihisa Kashima, and Paul Bain. (2008). Attrib- uting and denying humanness to others. European Review of Social Psychology, 19: 55–85. doi: 10.1080/10463280801981645. Henderson, Errol A. and Richard Tucker. (2001). Clear and present strangers: the clash of civilizations and international conflict. International Studies Quarterly, 45(2): 317–338. doi: 10.1111/0020-8833.00193. Henry, Ian P., Mahfoud Amara, and Mansour Al-Tauqi. (2003). Sport, Arab nationalism and the Pan-Arab games. International Review for the Sociology of Sport, 38(3): 295–310. doi: https://doi.org/10.1177/10126902030383003. Hogg, M.A. (2004). Uncertainty and extremism: identification with high entitativity groups under conditions of uncertainty. In Yzerbyt, Vincent, Charles M. Judd, and Olivier Corneille (eds.), The Psychology of Group Perception: Perceived Vari- ability, Entitativity, and Essentialism, pp. 401–418. New York: Psychology Press. doi: 10.4324/9780203644973. Hogg, M.A. (2006). Social identity theory. In Burke, Peter J. (ed.), Contemporary Social Psychological Theories, pp. 111–136. Palo Alto, ca: Stanford University Press. Inglehart, Ronald, Mansoor Moaddel, and Mark Tessler. (2006). Xenophobia and in- group solidarity in Iraq: a natural experiment on the impact of insecurity. Perspec- tives on Politics, 4(3): 495–505. doi: https://doi.org/10.1017/S1537592706060324. Jennings, Lauren E. and Kate C. McLean. (2013). Storying away self-doubt: can narra- tives dispel threats to the self? Journal of Research in Personality, 47(4): 317–329. doi: 10.1016/j.jrp.2013.02.006. Johnson, Megan K., Wade C. Rowatt, Lucy M. Barnard-Brak, Julie A. Patock-Peckham, Jordan P. LaBouff, and Robert D. Carlisle. (2011). A mediational analysis of the role of right-wing authoritarianism and religious fundamentalism in the religiosity- prejudice link. Personality & Individual Differences, 50(6): 851–856. doi: 10.1016/j .paid.2011.01.010. Kassem, Maye. (2006). Egypt: politics in the new millennium. unisci Discussion Papers , 12: 121–134. Kimball, Charles. (2002). When Religion Becomes Evil: Five Warning Signs. San Fran- cisco: Harper Collins. Koran, Laura. (2016). Will refugees bring Europe’s terror woes to us homeland? cnn. Retrieved 21 January 2017 (cnn.com/2016/09/15/politics/syrian-refugees-isis -screenings). Kunovich, Robert M. and Randy Hodson. (1999). Conflict, religious identity, and ethnic intolerance in Croatia. Social Forces, 78(2): 643–668. Legewie, Joscha. (2013). Terrorist events and attitudes toward immigrants: a natural experiment. American Journal of Sociology, 118(5): 1199–1245. doi: https://doi .org/10.1086/669605.

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 275

Leon, Federico R. and Agustín Cuesta. (1993). The need for quasi-experimental meth- odology to evaluate pricing effects. Studies in Family Planning, 24(6): 375–381. doi: 10.2307/2939247. Levey, Geoffrey B. (2009). Review article: what is living and what is dead in multicultur- alism. Ethnicities, (9)1: 75–93. doi: https://doi.org/10.1177/1468796808099905. Lichtblau, Eric. (2016). Hate crimes against American Muslims most since post-9/11 era. The New York Times. Retrieved 21 January 2017 (www.nytimes.com/2016/09/18/us/ politics/hate-crimes-american-muslims-rise.html). Mahmood, Saba. (2004). Politics of Piety: The Islamic Revival and the Feminist Subject. Princeton, nj: Princeton University Press. Matlock, Jack F., Jr. (1999). Can civilizations clash? Proceedings of the American Philo- sophical Society, 143(3): 428–439. McDougall, James. (2011). Dream of exile, promise of home: language, education, and Arabism in Algeria. International Journal of Middle East Studies, 43(2): 251–270. doi: 10.1017/S0020743811000055. McGregor, Ian, Reeshma Haji, Kyle A. Nash, and Rimma Teper. (2008). Religious Zeal and the uncertain self. Basic and Applied Social Psychology, 30(2): 183–188. doi: http://dx.doi.org/10.1080/01973530802209251. Moaddel, Mansoor. (2003). 2002 keynote address: September 11, global exigencies, and sociology: a new challenge. Michigan Sociological Review, 17(Fall): 1–5. Moaddel, Mansoor and Hamid Abdul-Latif. (2006). Events and value change: the impact of September 11, 2001 on the worldviews of Egyptians and Moroccans. In Moaddel, Mansoor (ed.), Values and Perceptions of the Islamic and Middle Eastern Publics, pp. 249–297. New York, Palgrave Macmillan us. doi: 10.1057/ 9780230603332_10. Mol, H. (1976). Identity and the Sacred: A Sketch for a New Social-Scientific Theory of Religion. New York: Free Press. Morgan, G. Scott, Daniel C. Wisneski, and Linda J. Skitka. (2011). The expulsion from Disneyland: the social psychological impact of 9/11. American Psychologist, 66(6): 447–454. doi: 10.1037/a0024772. Najjar, Fauzi. (2005). The Arabs, Islam and globalization. Middle East Policy, 12(3): 91–106. doi: 10.1111/j.1061-1924.2005.00215.x. Panagopoulos, Costas. (2006). The polls-trends: Arab and Muslim Americans and Islam in the aftermath of 9/11. The Public Opinion Quarterly, 70(4): 608–624. doi: 10.1093/poq/nfl029. Pelowski, Allan. (1971). On the use of a quasi-experimental design in the study of international organization and war. Journal of Peace Research, 8(3/4): 279–285. doi: https://doi.org/10.1177/002234337100800308. Pew Research Center. (2012). Egyptians have sharply negative attitudes towards the u.s. Retrieved 2 February 2017 (www.pewresearch.org/daily-number/egyptians-have -sharply-negative-attitudes-towards-the-u-s/).

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access

276 Shah, Acevedo and Ruiz

Pintak, Lawrence. (2009). Border guards of the ‘imagined’ Watan: Arab journalists and the new Arab consciousness. The Middle East Journal, 63(2): 191–212. doi: 10.1353/ mej.0.0047. Powers, Jillian L. (2011). Reimaging the imagined community: homeland tourism and the role of place. American Behavioral Scientist, 55(10): 1362–1378. doi: https://doi .org/10.1177/0002764211409380. Quillian, Lincoln. (1995). Prejudice as response to perceived group threat: population composition and anti-immigrant and racial prejudice in Europe. American Socio- logical Review, 60(4): 586–611. doi: http://dx.doi.org/10.2307/2096296. Rimé, Bernard, Darío Paez, Nekane Basabe, and Francisco Martinez. (2010). Social sharing of emotion, post-traumatic growth, and emotional climate: follow-up of Spanish citizen’s response to the collective trauma of March 11th terrorist attacks in Madrid. European Journal of Social Psychology, 40(6): 1029–1045. doi: 10.1002/ ejsp.700. Romanelli, Elaine and Michael L. Tushman. (1986). Inertia, environments, and stra- tegic choice: a quasi-experimental design for comparative-longitudinal research. Management Science, 32(5): 608–621. doi: http://dx.doi.org/10.1287/mnsc.32.5.608. Russett, Bruce M., John R. Oneal, Michaelene Cox. (2000). Clash of civilizations, or realism and liberalism déjà vu? Some evidence. Journal of Peace Research, 37(5): 583–608. doi: https://doi.org/10.1177/0022343300037005003. Said, Edward W. (1977). Orientalism. New York: Vintage Books. Schafer, Chelsea E. and Greg M. Shaw. (2009). Trends-tolerance in the United States. Public Opinion Quarterly, 73(2): 404–431. doi: https://doi.org/10.1093/poq/nfp022. Scheepers, Peer, Mèrove Gijsberts, and Evelyn Hello. (2002). Religiosity and prejudice against ethnic minorities in Europe: cross-national tests on a controversial relationship. Review of Religious Research, 43(3): 242–265. doi: http://dx.doi .org/10.2307/3512331. Schwarzer, Ralf, Rosemarie M. Bowler, and James E. Cone. (2014). Social integration buffers stress in New York police after the 9/11 terrorist attack. Anxiety Stress and Coping, 27(1): 18–26. doi: http://dx.doi.org/10.1080/10615806.2013.806652. Scott, Rachel M. (2010). The Challenge of Political Islam: Non-Muslims and the Egyptian State. Stanford, ca: Stanford University Press. Shaheen, Jack G. (2003). Reel bad Arabs: how Hollywood vilifies a people. The Annals of the American Academy of Political and Social Science, 588(1): 171–193. doi: https:// doi.org/10.1177/0002716203588001011. Shen, Megan Johnson, Megan C. Haggard, Daniel C. Strassburger, and Wade C. R owatt. (2013). Testing the love thy neighbor hypothesis: religiosity’s association with positive attitudes toward ethnic/racial and value-violating out-groups. Psychology of Religion and Spirituality, 5(4): 294–303. doi: 10.1037/a0033648. Sherbiny, Naiem A. (2005). Can Russia and Germany save the Middle East? Russia in Global Affairs, 3(1): 167–175.

sociologyDownloaded of islam from 5 Brill.com09/24/2021 (2017) 249-277 02:55:52PM via free access

Contextualizing Events in Imagined Communities 277

Shamir, Shimon. (1987). The evolution of Egyptian nationality laws and their applica- tion to the Jews in the monarchy period. In Shamir, Shimon (ed.), The Jews of Egypt, A Mediterranean Society in Modern Times, pp. 33–67. Boulder: Westview Press. Silberman, Israela. (2005). Religion as a meaning system: implications for the new millennium. Journal of Social Issues, 61(4): 641–663. doi: 10.1111/j.1540-4560.2005.00425.x. Silberman, Israela, E. Tory Higgins, and Carol S. Dweck. (2005). Religion and world change: violence and terrorism versus peace. Journal of Social Issues, 61(4): 761–784. doi: 10.1111/j.1540-4560.2005.00431.x. Smith, Anthony D. (1995). Nations and Nationalism in a Global Era. Cambridge, ma: Polity Press, Blackwell. Tajfel, Henri. (1972). Social categorization. In Moscovici, Serge (ed.), Introduction à la psychologie sociale, pp. 272–302. Paris: Larousse. Tajfel, Henri. (1974). Intergroup behaviour, social comparison and social change. Unpublished Katz-Newcomb lectures, University of Michigan, Ann Arbor. Tiryakian, Edward A. (2011). The missing religious factor in imagined communities. American Behavioral Scientist, 55(10): 1395–1414. doi: https://doi.org/10.1177/00027 64211409563. Tsygankov, Andrei. (2003). The irony of western ideas in a multicultural world: Russians’ intellectual engagement with the ‘end of history’ & ‘clash of civilizations.’ Interna- tional Studies Review, 5(1): 53–76. doi: 10.1111/1521-9488.501003. Turner, John C., Michael A. Hogg, Penelope J. Oakes, Stephen D. Reicher, and Margaret S. Wetherell. (1987). Rediscovering The Social Group: A Self-Categorization Theory. Oxford, England: Blackwell. Weber, Max. (2009). The social psychology of world religions. Gerth, Hans H. and C. Wright Mills (ed.) From Max Weber: Essays in Sociology, pp. 267–301. New York: Routledge. Weeks, Mathew and Michael B. Lupfer. (2004). Complicating race: the relationship between prejudice, race, and social class categorizations. Personality and Social Psychology Bulletin, 30(8): 972–984. doi: https://doi.org/10.1177/0146167204264751. Wien, Peter. (2011). The long and intricate funeral of Yasin Al-Hashimi: pan-Arabism, civil religion, and popular nationalism in Damascus, 1937. International Journal of Middle East Studies, 43(2): 271–292. doi: 10.1017/S0020743811000067. World Value Survey. (2009). World Values Survey 1-981–2008 official aggregate v.20090902, 2009. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: asep/jds Data Archive, Madrid, Spain. Yan, Holly, Kristina Sgueglia, and Kylie Walker. (2016). ‘Make America white again’: hate speech and crimes post-election. cnn. Retrieved 21 January 2017 (www.cnn .com/2016/11/10/us/post-election-hate-crimes-and-fears-trnd/).

sociology of islam 5 (2017) 249-277 Downloaded from Brill.com09/24/2021 02:55:52PM via free access