Some observers have argued that the Arab Spring was a fundamentally secular movement, arising as a rejection of not only the existing regimes, but the religious status quo as well. On the other extreme, some have suggested that the “Arab Spring” may in fact have been an “Islamist Spring,” in which individuals sympathetic to political Islam took to the streets in order to call for a greater role for religion in public life. Between these two poles lies a wide variety of accounts of exactly how religious the Arab Spring was. We set out to test competing assessments about the role of religion in Arab protest movements by using data from the Arab Barometer Project collected in Egypt and Tunisia shortly after the fall of their respective regimes. The findings are not what many might have expected. The link between religion and protest behavior is primarily found in personal piety and behavior rather than communal or mosque practice.
Mosque attenders were no more likely to participate in anti-regime protests than were non-attenders. In fact, frequent mosque attenders may have been slightly less likely to participate. While the differences between the most frequent attenders and non-attenders were not statistically significant, frequent attenders were about 25 percent less likely to participate in protest than were non-attenders. These findings contrast sharply with the “days of rage” narrative, which argues that a vast number of citizens who would not otherwise have protested chose to participate because of sermons delivered in mosques immediately before the protests began.
On the other hand, Koran readers were far more likely to engage in protest than non-readers. In both Egypt and Tunisia, frequent Koran readers were over three times as likely to report protesting against the regime than citizens who do not read the Koran. Protesters do appear to have been more religious than other citizens, but there is less evidence that religious institutions provided the crucial organizational advantages. If organizational resources were the driving force behind religious participation in the Arab Spring, then mosque attenders should have been substantially more likely to protest than non-attenders. Personal religious convictions and behaviors drove pious citizens into the squares, in contrast to the claim that mosques drove this participation.
What about the long-standing question of grievance versus opportunity in protest behavior? We find more evidence for grievance-related channels than opportunity. In both countries, Koran readers are significantly more likely to support democracy and to perceive unequal treatment from the government, both of which suggest goal-oriented mobilization rather than the mere presence of opportunity. Further, two of the most important variables related to opportunity – trust and efficacy – are actually negatively correlated with Koran reading, and sometimes significantly so. These “opportunity” mechanisms would lower the costs of participating in protest; if religious individuals were more likely to trust other citizens, then they would be less afraid to protest because they would be more confident that others would attend as well. Political efficacy would, on the other hand, raise the relative benefits of protest. If individuals believe that they can effect meaningful change, then protest should become a more attractive option. However, in each of these cases, Koran reading actually decreases perceived opportunities, and sometimes by a considerable margin. Frequent Koran-reading Egyptians were 23 percentage points less trusting than non-readers, and frequent readers in Tunisia were less than half as likely as non-readers to report high levels of efficacy. While there are many possible explanations for the (usually) negative relationship between Koran reading and perceived opportunities for protest, Koran reading seems to have inspired protest behavior through motivation rather than opportunity.
The role of Islamist ideologies also appears to have been fairly limited in these settings. Individuals who espoused Islamist political beliefs were slightly more likely to participate in protests, but these effects are nowhere near statistical significance and pale in comparison to the effect of Koran reading. In both countries, individuals advocating high levels of religious involvement in politics were about 34 percent more likely to report protesting compared to those opposing such involvement; the corresponding increase associated with Koran reading is over 300 percent.
The Arab Spring was not, at its core, a secular or anti-religious movement, and religion did not impede participation in anti-regime activities. In this setting, it can scarcely be claimed that religion “puts people to sleep” or makes them docile, uncritical citizens. In Tunisia and Egypt, the centers of the Arab Spring, engagement with religious texts is robustly related to engagement in anti-regime behavior. That organizational resources did not appear to increase citizens’ likelihood of participating is equally interesting. While religion does seem to have played a significant role in the Arab Spring, it seems to have motivated protest from within rather than simply bringing citizens from the mosque to the square.
Michael Hoffman is a doctoral candidate in the Department of Politics at Princeton University, studying comparative politics. Amaney Jamal is a professor of politics at Princeton University as well as director of the Bobst Center for Peace and Justice and the Workshop on Arab Political Development. She co-directs the Arab Barometer project.