Where Terrorism Finds Support in the Muslim World
That May Depend on How You Define It - and Who Are the Targets
by Richard Wike, Pew Global Attitudes Project and Nilanthi Samaranayake, Pew Research Center for the People & the Press
What produces terrorists and what conditions allow them to multiply in number and power in the Muslim world? While many studies point to the important role public opinion plays in creating an environment in which terrorist groups can flourish, relatively few works have explored survey data to measure support for terrorism among general publics. Findings from the 2005 Pew Global Attitudes survey on attitudes toward suicide bombing and civilian attacks and other measures of support for terrorism offer some revealing perspectives on this question.1
Most notably, the survey finds that terrorism is not a monolithic concept–support for terrorist activity depends importantly on its type and on the location in which it occurs. For example, Moroccans overwhelmingly disapprove of suicide bombings against civilians, but, among respondents in the six predominantly Muslim countries surveyed, they are the most likely to see it as a justifiable tactic against Americans and other westerners in Iraq. Opinions about the United States, its attitudes in dealing with the larger world and the Iraq war are also powerful factors in shaping support for terrorism, as are perceptions that Islam is under threat. With the exception of gender, demographic differences, including income, explain little if anything about attitudes toward terrorism in the Muslim world, but country-specific differences are significant, suggesting the importance of local social, political and religious conditions.
These findings are generally though not entirely consistent with other studies of the origins and growth of Islamic terrorism. Much of the relevant literature, however, differs in its focus, concentrating instead on the motivations of terrorist organizations and their members. For example, groups may turn to suicide bombing when other strategies fail (Martha Crenshaw, 1998) or when they find themselves in competition for public support with other militant groups (Mia Bloom, 2005). Robert Pape (2003) finds that terrorism can be a “rational” strategy, pursued by groups, including secular groups, seeking territorial concessions from liberal democracies (2003). Several authors examine the link between political authoritarianism and terror. Alberto Abadie (2004) finds countries in transition from authoritarianism to democracy at a heightened risk for terrorist activities, while Gregory Gause (2005) argues that authoritarian regimes may be best equipped to stifle terrorism – he offers China as an example. Still others see support for terrorism driven in part by opposition to U.S. foreign policy. For instance, Scott Atran (2004) finds “no evidence that most people who support suicide actions hate Americans’ internal cultural freedoms, but rather every indication that they oppose U.S. foreign policies, particularly regarding the Middle East.”
Relatively few studies have addressed the public attitudes that allow terrorism to take root and grow in certain societies; those that have rely on earlier data than is provided by the 2005 Pew study. In their analysis of Lebanese Muslim attitudes, Simon Haddad and Hilal Khashan (2002) find that younger respondents and those who endorse political Islam are more likely than others to approve of the September 11 attacks. However, they find that income and education are unrelated to such opinions. Examining polling data from the Palestinian Center for Policy and Survey Research, Alan Krueger and Jitka Maleckova (2002) also conclude that, contrary to much conventional wisdom, poverty and low education are not key drivers of support for terrorism.
Similarly, in a recent study, Christine Fair and Bryan Shepherd (2006) analyze 2002 Pew Global Attitudes data and find that women, young people, computer users, those who believe Islam is under threat, and those who want religious leaders to play a larger role in politics are more likely to support suicide bombing and other attacks against civilians. Fair and Shepherd find that financial status is also a significant determinant — that the very poor are less, not more, likely to support such attacks.
What then do more recent data show?
Declining Support for Terrorism
Overall, the 2005 Pew Global Attitudes survey finds that support for terrorism has generally declined since 2002 in the six predominantly Muslim countries included in the study – Indonesia, Jordan, Lebanon, Morocco, Pakistan, and Turkey – although there are some variations across countries and survey items.
We will focus on results for three terrorism-related measures: attitudes about suicide bombing and other violence against civilians, views on suicide bombing carried out against Americans and other Westerners in Iraq, and opinions about Osama bin Laden. The first two measures were only asked of Muslim respondents. All respondents were asked their opinion of bin Laden; however, we will restrict our analysis to Muslim respondents.
The most basic measure of support for terrorism asked respondents the following question: “Some people think that suicide bombing and other forms of violence against civilian targets are justified in order to defend Islam from its enemies. Other people believe that, no matter what the reason, this kind of violence is never justified. Do you personally feel that this kind of violence is often justified to defend Islam, sometimes justified, rarely justified, or never justified?”
As Table 1 illustrates, the share of the public that believes suicide bombing and other violence is justifiable varies considerably across countries, with Jordanian Muslims significantly more likely than others to support terrorist acts. Lebanon and Pakistan form a middle tier on this question, followed by Indonesia, Turkey, and Morocco, where solid majorities say these forms of violence are never justified. In five of the six countries, support for such attacks has dropped since the last time the question was asked, although the decline in Turkey is insignificant. The lone exception is Jordan, where support has actually increased 14 points since 2002.
The most dramatic drop in support for terrorism is seen in Morocco, a country that experienced a devastating terrorist attack in May 2003. Fully 79% of Moroccans surveyed in 2005 said that support for suicide bombing and violence against civilians was never justified–more than double the percentage (38%) who had expressed this view a year earlier.
A second question asked respondents specifically about suicide bombing in Iraq: “What about suicide bombing carried out against Americans and other Westerners in Iraq? Do you personally believe that this is justifiable or not justifiable?”
Interestingly, despite the overall decline in support for terrorist acts among its citizens, Morocco is the only country in which a majority says attacks on Americans and other westerners in Iraq are justified. Roughly half of Jordanian and Lebanese Muslims support such acts, while fewer than 30% of Muslims in Pakistan, Indonesia and Turkey agree. In all four countries where trends exist, support for suicide attacks in Iraq has declined, including a large, 21-point drop in Jordan.
Finally, respondents were asked how much confidence they have in Osama bin Laden to do the right thing in world affairs. The results show support for bin Laden has declined in four of the six countries. Jordan and Pakistan are the exceptions, with the percentage of Muslims who have a lot or some confidence in bin Laden rising five points among Jordanians and six points among Pakistanis.
Independence of Terrorism Measures
It is clear that across all three measures, support for terrorism has declined generally. However, it is also clear that levels of support vary across questions, suggesting that each measures a different facet of how people view terrorism.
This can be illustrated by examining the relationship between views about suicide bombing generally and suicide bombing specifically in Iraq. As Table 4 demonstrates, in some predominately Muslim countries a significant number of people who believe that suicide bombing and other attacks against civilians are at least sometimes justifiable still do not support suicide bombing against Westerners in Iraq. For example, in Turkey among respondents who say suicide bombing is rarely, sometimes, or often justified, a 49% plurality says that suicide bombing in Iraq is not justifiable. By contrast, in Morocco 81% and in Jordan 68% of those who say targeting civilians is at least sometimes justified also find it justifiable in Iraq.
Similarly, those who believe that suicide bombing and other attacks against civilians are at least sometimes justifiable do not necessarily have confidence in Osama bin Laden. Again, results vary significantly by country, with 71% of Jordanian Muslims who believe violence against civilians can be justified also having confidence in bin Laden, compared with only 5% of Turks.
Finally, the relationship between views about suicide bombing in Iraq and views of bin Laden also differ significantly among the six countries. For instance, 82% of Jordanian Muslims who think suicide bombing in Iraq against Westerners is justifiable also have a lot or some confidence in bin Laden. However, only 6% of Lebanese in the same category also have confidence in bin Laden.
Correlates of Support for Terrorism
As noted above, differences in opinions about terrorism have been linked not only to demographic variables, notably age and gender, but also to views about Islam, democracy, and the United States. Four sets of variables are used to explore whether these patterns are significant in the 2005 survey data.
- Demographic variables – these include gender, age, education, and income, as well as whether a respondent has a child under age 18 living in the household and whether the respondent regularly uses a computer. Since measures for education and income differ across countries, for the purposes of analysis respondents are characterized as low or high education, and as low, middle, or high income.
- Views about Islam – Both the academic literature and the popular press have emphasized links between terrorism and an extremist brand of Islam. Responses to three questions are used to explore any potential relationships between opinions on religion and terrorism. The first asks respondents whether their primary identity is as a Muslim or as a citizen of their country (Jordanian, Moroccan, etc.). The second asks how important it is that Islam plays a more influential role in the world than it does now. The third asks whether the respondent thinks there are any serious threats to Islam today.
- Opinions about democracy – Two questions test these attitudes among respondents. The first asks whether democracy is a Western way of doing things that will not work in the respondent’s country or if democracy is not just for the West and would work in their country. The second asks respondents if they are more optimistic or more pessimistic these days that the Middle East will become more democratic.
- Attitudes toward the United States – In addition to a straightforward favorability question about the U.S., these measures include questions about: the extent to which the U.S. takes into account the interests of countries such as the respondent’s country when making international policy decisions; how worried, if at all, respondents are that the American military will become a threat to their country; whether the war in Iraq has made the world safer or more dangerous; and whether the U.S. government favors or opposes democracy in the respondent’s country.2
Comparison of levels of support for the three measures of terrorism against these four sets of variables reveals a number of associations. As seen in Table 6, across all three measures, men are generally more supportive of terrorism than are women. Meanwhile, individuals with children are less supportive of suicide bombing generally, but more supportive of bin Laden. Support for terrorism is also more common among persons who identify primarily as Muslim, those who believe it is important for Islam to play an influential role on the world stage, and those who believe Islam faces serious threats.
Whether or not an individual thinks democracy is solely a Western way appears to have only modest effects on support for terrorism (it should be noted that relatively few Muslims, ranging from 12% in Morocco to 38% in Turkey, believe democracy is solely a Western form of government). On the other hand, across all three measures, those who are pessimistic about the prospects for Middle East democracy have more favorable attitudes toward terrorism.
Views about the U.S. appear strongly associated with attitudes toward terrorism, with support for terrorism higher among people who have an unfavorable opinion of the U.S., those who believe American foreign policy does not consider the interests of countries like theirs, those who are concerned that the U.S. may pose a military threat to their country, and those who believe the U.S. opposes democracy in their country.
Still, the question remains whether many of these variables have independent strength in explaining attitudes toward terrorism or whether they are primarily proxies for other significant variables with which they themselves are correlated. To determine whether these associations remain significant once other factors are controlled for, we conducted two types of regressions3 including the variables described above as along with dummy variables to assess country specific effects.
As illustrated in Table 7, when other factors are controlled for, most demographic variables no longer show significant effects on opinions regarding suicide bombing and civilian attacks. However, gender remains significant in views about suicide bombing against Westerners in Iraq or confidence in bin Laden, with women less likely than men to support such bombing or the Al Qaeda leader. Income is also a significant determinant of support for bin Laden, with wealthier individuals holding a more negative view of the al Qaeda leader.
Two of the measured attitudes toward Islam also remain significant. The belief that it is important for Islam to play an influential role in the world is positively related to support for suicide bombing in Iraq and confidence in bin Laden. The perception that there are serious threats to Islam is positively associated with support for suicide bombing and other attacks against civilians, as well as suicide bombing against Westerners in Iraq. However, primarily identifying as a Muslim is not significantly related to any of the three dependent variables.
Variables measuring attitudes toward democracy show limited effects. The only instance in which either of the two democracy measures is significant is that people who believe democracy is not just a western way and can work in their country are less likely to support terrorist attacks against civilians.
By contrast, some attitudes toward the U.S. are strongly associated with views on terrorism. Support for terrorism is positively correlated with negative views of the U.S., a perception that the U.S. does not favor democracy in a respondent’s country, and a belief that the Iraq war has made the world more dangerous.
Finally, nearly all of the country indicators are significant, indicating that country specific factors have a great deal of influence on attitudes toward terrorism.4
The results show that the variables for Jordan and Lebanon are positively related to support for attacks against civilians, while the other three countries are negatively related to this measure. In the second model, with support for suicide bombing in Iraq as the dependent variable, variables for three countries — Morocco, Lebanon, and Jordan — are positively associated with approval of suicide attacks in Iraq. Meanwhile, the Turkey variable is negatively associated with support for suicide terrorism in Iraq. Finally, in the third model Morocco is the excluded category, and Pakistanis, Jordanians, and Indonesians are found to be more supportive of bin Laden, while Lebanese and Turkish Muslims are less likely to have confidence in bin Laden.
The findings suggest several general conclusions about public opinion regarding terrorism in these six predominantly Muslim countries. First, the 2005 poll finds support for terrorism on the decline, although there are a few exceptions to this pattern, and support remains rather high in some countries, notably Jordan. Previous research has shown that support tends to decline among publics after they have experienced attacks on their own soil, and future research will determine whether such a drop has occurred in Jordan following the November 2005 bombings in Amman.
Second, terrorism is not a monolithic concept, and different facets of terrorism have different patterns of public support. Many individuals who say suicide bombing in defense of Islam may be justifiable do not support it in Iraq, and vice versa.5 For example, while Moroccans are the least supportive of suicide bombing when it is described in general terms, they are the most likely to approve of suicide bombing specifically in Iraq.
Third, demographic characteristics appear to have relatively small effects on attitudes towards terrorism, with the exception of gender. Contrary to Fair and Shepherd, we find that women are generally less likely to approve of terrorist acts and are less likely to hold favorable views of Osama bin Laden.
Fourth, views about Islam are linked, to some extent, to views about terrorism. In particular, and consistent with Fair and Shepherd, we find the perception that Islam is under threat is positively correlated with support for terrorism.
Next, we find that opinions of the United States and of American foreign policy are important determinants of attitudes towards terrorism. The perception that the U.S. acts unilaterally in international affairs, concerns about the American military becoming a threat, negative views of the Iraq war, the belief that the U.S. opposes democracy in the region, and a generally unfavorable view of America all drive pro-terrorism sentiments.
Finally, the multivariate analysis finds significant country-specific effects, suggesting that conditions giving rise to terror are greatly influenced by local political, social, and religious factors. Future studies should seek to shed more light on these country specific influences, as well as the factors that shape public opinion on terrorism across nations.
A longer version of the paper was presented at the annual conference of the American Association for Public Opinion Research, Montreal, Canada, May 18-21, 2006
Field dates for the survey, as well as the number of Muslims in each country sample are shown below.
|Indonesia||April 30-May 16, 200||N=970||Muslims|
|Jordan||May 3-24, 2005||N=967||Muslims|
|Lebanon||May 3-24, 2005||N=563||Muslims|
|Morocco||June 6-16, 2005||N=1000||General public (religion not asked)|
|Pakistan||May 2-24, 2005||N=1203||Muslims|
|Turkey||April 27-May 14, 2005||N=965||Muslims|
Abadie, Alberto. 2004. “Poverty, Political Freedom, and the Roots of Terrorism.” Kennedy School of Government Faculty Research Working Paper Series.
Atran, Scott. 2004. “Mishandling Suicide Terrorism.” The Washington Quarterly 27: 67-90.
Bloom, Mia. 2005. Dying to Kill: the Allure of Suicide Terror. New York: Columbia University Press.
Crenshaw, Martha. 1998. “The Logic of Terrorism: Terrorist Behavior as a Product of Strategic Choice.” In Walter Reich (ed.), Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind. Washington, DC: Woodrow Wilson Center Press.
Fair, C. Christine and Bryan Shepherd. 2006. “Who Supports Terrorism? Evidence from Fourteen Muslim Countries.” Studies in Conflict & Terrorism 29: 51-74.
Gause, F. Gregory III. 2005. “Can Democracy Stop Terrorism?” Foreign Affairs 84: 62-76.
Haddad, Simon and Hilal Khashan. 2002. “Islam and Terrorism: Lebanese Muslim Views on September 11.” Journal of Conflict Resolution 46: 812-828.
Krueger, Alan B. and Jitka Maleckova. 2002. “The Economics and the Education of Suicide Bombers: Does Poverty Cause Terrorism?” The New Republic June 24.
Pape, Robert A. 2003. “The Strategic Logic of Suicide Terrorism.” American Political Science Review 97: 343-361.
- See the 2005 Pew Global Attitudes Survey for more information, including sample sizes and exact field dates. The analyses in this report are based on Muslim respondents only, except for Morocco, where a religious preference question was not asked. However, given that Morocco's population is 99% Muslim, it is likely that nearly all of the Moroccan sample is Muslim. ↩
- The question was not asked in Turkey. ↩
- Standard OLS regression was employed for the questions regarding suicide bombing and other attacks against civilians and confidence in bin Laden and logistic regression for the question concerning whether suicide bombing against Americans and other Westerners in Iraq is justified. ↩
- To test the independent importance of the country variables, in each model the country closest to the median position on the attitude in question was excluded. As shown in Table 7, in the first and second models, Pakistan is the excluded country, and the other country dummy variables should be interpreted as in comparison to Pakistan. ↩
- This was a split form question asked of approximately half the sample. ↩