If, as many believe and scholarship confirms, religion is particularly important in ethnic conflicts involving Muslims,1 how does this affect the nature of conflict in the Middle East?2

This is a simple question, but finding an accurate answer is not at all simple. It is complicated by two interrelated factors. First, due to differing perspectives and differing political agendas, interpretations of events in the region vary wildly, and accuracy is often lost. This problem is not limited to the study of the Middle East and, in fact, has been a central issue in political science since its inception. For instance, Karl Deutsch points out that comparative methodology, which consists of analyzing cases using introspection, intuition, and insight, while a powerful tool for analysis, is limited by our imaginations and preconceived notions.3 That is, a researcher who uses the method of familiarizing himself with as many facts as possible as well as with the insights of other scholars can easily be influenced by his own preconceptions.

This can be problematic because due to such preconceptions we often think we see relationships that we expect to see even if they do not exist, and we often fail to see relationships that do exist but that we never expected to exist or even imagined might exist. Applied to the issue at hand, we must ask ourselves if the perception that religion is particularly important in the Middle East and is the driving force behind many of the region’s conflicts is accurate, or merely a reflection of what we expect to see. To some extent, this is true of the study of religion and ethnic conflict in the Middle East.

Second, the issue of Islam’s role in generating conflict has become especially controversial since Samuel Huntington asserted that Islam has "bloody borders" and predicted that the dynamics of civilizational conflict in the post-Cold War era would reinforce and intensify this phenomenon.4 His analysis is part of a larger notion that conflicts are increasingly defined by a "clash of civilizations." Huntington holds three points: that post-Cold War conflict is mostly between world civilizations defined primarily by religion; that conflicts involving Islamic civilization will be particularly common and violent; and that Islamic civilization will be the greatest threat to Western civilization. While many, if not all, aspects of Huntington’s theory are controversial, his arguments concerning Islam have found acceptance among some policymakers. And that may, as some maintain, make Huntington’s analyses self-fulfilling.5 That is, if many Westerners expect Islam to be a threat and are disposed to perceive such a threat, whether it exists or not, then the expectations may wind up influencing policy.

Thus can political agendas, preconceptions, and popular academic theories obfuscate and perhaps even alter the role of religion in the Middle East. These complications make an accurate assessment of Islam’s role in Middle Eastern conflict all the more essential.

This study uses an empirical method to provide a perspective on the issue different from the comparative approach. While the empirical method has its limitations—primarily, it can analyze only what can be measured and measurements of social, political, and economic factors are often imperfect—it also has three advantages. It allows us to test and perhaps falsify theories. It makes clear what factors produce its results, and anyone analyzing the same data will get the same results. Also, it often produces surprising findings that would have never resulted from the comparative approach because no one would have looked for them.

This analysis proceeds in two stages. First, the results from previous empirical analyses are summarized. Second, we examine data on ethnic conflict to assess the extent and nature of the influence of religion in the Middle East.

Previous Findings

Empirical analyses, several by this author, have established that the relationship between religion and conflict in general can be summed up as follows:

Religious differences make conflict more likely and more intense. The more diverse a country’s religious population, the more violent its domestic conflicts tend to be.6 Similarly, when religious issues are important in ethnic conflicts, political, economic, and cultural discrimination, and rebellion all tend to increase.7 Religious differences also make international wars more likely.8

Religious issues influence the dynamics of conflicts. When religious issues are important in an ethnic conflict, religious institutions tend to facilitate unrest; but they tend to suppress conflict when religious issues are not important.9 Similarly, in countries where it is more legitimate to invoke religion in political discourse,10 ethnic conflicts generally focus more on religious issues as opposed to other political, cultural, and economic issues.11 Also, religious differences between combatants and religious affinities between minorities and those likely to intervene make foreign intervention in ethnic conflicts more probable.

Religion shapes discrimination against ethnic minorities. Religious and non-religious discrimination is more likely in states where it is legitimate to use religion in political discourse and in conflicts where religious issues are important. Religious discrimination is likely against minorities that are otherwise culturally similar to the majority group - that is, where religion is the chief differentiating characteristic between the majority and the minority. Perhaps connected, religious minorities seeking autonomy are more likely to suffer non-religious discrimination than ethnic minorities who are not distinguished by religion.12

Autocratic regimes are more likely than non-autocratic regimes to discriminate against religious minorities. Autocracies engage in higher average levels of discrimination against ethno-religious minorities than do democracies, but it is those regimes that are somewhere between full autocracies and full democracies (and that are known as semi-democracies), that discriminate the least. At the same time, regime track type seems not to be correlated with discrimination against ethnic minorities that are not religiously different from the majority group in their states.13

Regimes in Muslim states are more autocratic. One study finds that Muslim states are the most autocratic states in the world, based on a measure of the extent to which a state is a liberal democracy as well as on a separate measure of institutional democracy.14 However, this study finds no relationship between Islam and a third measure which focuses on political rights. Another study, which uses the measure for institutional democracy, finds these trends particularly pronounced in the Middle East.15

Taken together, these five results suggest that Middle Eastern conflicts should be particularly frequent and intense. One would expect religious factors to be particularly important and to exacerbate ethnic conflict in that region; religious discrimination to be high; and political behavior to be particularly influenced by religious differences between groups. This is especially true given the region’s high level of autocracy combined with the particular importance of religion in the region. Is this in fact the case?

Ethno-Religious Conflict in the Middle East

The Minorities at Risk (MAR) dataset, with information on 267 politically active ethnic minorities throughout the world, plus additional data collected by this author, provides insight into the nature of the Middle East’s ethno-religious conflicts.16

Three preliminary points: (1) The MAR dataset, developed by the MAR project headed by Ted R. Gurr, is particularly useful for our purposes. It documents all instances of ethnic conflict between 1945 and 1998 and contains a reasonably accurate list of all ethnic groups that are actively pursuing political agendas, violently or otherwise, or that suffer from high levels of discrimination. The dataset has been in existence since the mid-1980s and has received considerable attention and criticism which, over time, have led to a fairly accurate list of the groups that meet the stated criteria. Along with data collected separately on religion for use with the dataset, it constitutes one of the few aggregate-level datasets on religion and conflict that does more than simply measure whether the groups involved are of different religions.17

(2) For the purposes of this study, an ethnic minority is considered to be of a different religion than the majority group in that state if 80 percent of the population of the minority group is of a religion different from the majority group. Protestant and Catholic Christianity are considered separate religions, as are the Sunni and Shi‘i branches of Islam.

(3) MAR’s data refer only to conflicts involving ethnic minorities within states and not to conflicts within the same ethnic group. This means, for example, that the conflict between Egypt’s Coptic minority and Muslim majority is included but the same country’s conflict between the Islamist movement and the Egyptian state is not.

Are ethnic conflicts between religiously differentiated groups particularly prevalent in the Middle East? Yes: Table 1 shows that 54 percent of the politically significant ethnic minorities in the Middle East are also religious minorities. Only Asia scores higher and then only slightly so. In the rest of the world, ethnic minorities are considerably less often religious ones.

An examination of religious factors in ethno-religious conflicts reinforces this picture. Religious factors are considered to be present when any, and especially all, of the following conditions are fulfilled: religion is a significant issue in the conflict; religious discrimination exists; a minority group demands more religious rights; and religion is invoked in political discourse.

Comparing the results for the Middle East to other world regions and Muslim-majority states outside of the Middle East allows us to see something important: whether the Middle East is unique in the intensity of conflicts with religious factors - or whether such conflicts exist in other world regions or are common to all Muslim states. That, in turn, allows us to say whether the explanation for such conflicts is regionally based (unique to the Middle East) or religiously based (common to states with Muslim majorities).

It turns out that all four religious factors surface considerably more often in the Middle East than in other regions. From this, one can conclude that religion is indeed particularly important in the Middle East’s ethno-religious conflicts.

Comparisons involving non-Middle Eastern states with Muslim majorities are particularly interesting. In such states, all four religious factors occur more often in ethno-religious conflicts than in non-Middle Eastern states lacking Muslim majorities—but less often than in Middle Eastern states. Turning this around: ethnic conflicts in Middle Eastern states more often involve religious factors than similar conflicts in non-Middle Eastern states with Muslim majorities. The latter conflicts, in turn, more often involve religious factors than do similar conflicts in non-Middle Eastern states lacking Muslim majorities. Based on these findings, Islam provides a partial explanation for the unique importance of religion in Middle East conflicts but it cannot be considered a full explanation. Other factors must be at work.

To assess whether the Middle East’s ethno-religious conflicts are particularly violent or intense, Table 3 (see page 36) measures the presence of seven factors important in ethnic conflict: political discrimination, economic discrimination, cultural discrimination, repression, an expressed desire by a minority for autonomy, political demonstrations, and rebellion (terrorism, guerrilla warfare, or armed insurrection). Except for political discrimination, the Middle East scores near the world average on these measures. This permits us to conclude that the Middle East is unique in the importance that religion plays in its ethno-religious conflicts, but those conflicts are otherwise mostly similar to the ones occurring elsewhere.

Under what type of regimes do ethno-religious conflicts mostly take place—autocratic, semi-democratic, or democratic? Table 4 finds that 86 percent of Middle Eastern ethno-religious conflicts occur in autocratic states. This is a particularly strong finding considering that all of the 14 percent of cases of Middle Eastern ethno-religious conflict occurring in non-autocratic states occur in Israel.18 However, it is not a particularly surprising finding given that the Middle East is the world’s most autocratic region. Be that as it may, these results show that all Muslim majorities in the Middle East involved in ethno-religious conflicts, as documented by the MAR dataset, preside over autocratic states. As with the presence of religious factors, Islam can provide a partial explanation for this thorough-going autocracy in the Middle East, but not a full explanation. Non-Middle Eastern Muslim majorities involved in ethno-religious conflicts are not uniformly autocratic, although they do rule autocratic states more than twice as often as non-Muslim majorities (52 percent as compared to 20 percent). More importantly, these results imply that there may be a link between Islam’s, and particularly the Middle East’s, association with autocracy and the finding that in Islam and the Middle East religion tends to be particularly important.

This brings up a final question: do Islam and autocracy in the Middle East combine to make religion particularly important in the region’s ethno-religious conflicts? This requires a comparison of the Middle East with the non-Middle East. If the disproportionate importance of religion in Middle Eastern ethno-religious conflicts is due to a combination of Islam and autocracy, we would expect the religious factors to be similarly important outside of the Middle East in states that are both Muslim and autocratic.

To measure this, conflicts in countries outside the Middle East are broken down into five categories (with some overlap between the categories): all autocratic states; states that are both autocratic and Muslim; states that are autocratic but not Muslim; semi-democratic states; and democratic states. Table 5 shows ethnic conflicts in Muslim autocracies outside the Middle East include religious factors less often than within the Middle East. Ethnic conflicts in non-Muslim autocracies outside the Middle East include three of the four factors less often than do non-Middle Eastern Islamic autocracies. Most notably, outside of the Middle East, Muslim autocracies discriminate against religious minorities less often than do non-Muslim autocracies. These results rule out Islam as an explanation for the disproportionate importance of religion in Middle East ethno-religious conflicts. After all, if Islam resulted in the increased importance of religion, we would have expected to find that religion was considerably more important in non-Middle Eastern Islamic autocracies than in other non-Middle Eastern autocracies as measured by at least a majority of the variables. In fact, the reverse is true and the non-Islamic autocracies scored considerably higher on two of four variables and slightly higher on a third.

Autocracy is also ruled out as an explanation: if autocracy were the sole explanation for the disproportionate importance of religious factors in the Middle East, we would expect it to have a uniform impact outside of the Middle East, which is not the case.

What about the Middle East’s combination of Islam and autocracy? Well, outside of the Middle East, the combination is associated with results quite different from the same combination in the Middle East.


This study has both expected and surprising results.

Religion is more important in Middle Eastern ethnic conflicts than elsewhere. Religion is important in the ethnic conflicts of all Muslim states, and it is more important in the Middle East than in Muslim states outside the region. This means that while Islam may provide a partial explanation for the particular importance of religion in the region, it cannot provide a full explanation. One potential explanation for this is the historical importance of religion in the Middle East, a region that gave birth to three of the world’s major religions. On the other hand, this historical importance may also mean that whatever it is that makes religion particularly important in the Middle East is not a new phenomenon, and the findings presented here are simply the latest manifestation of an age-old phenomenon.

The Middle East is the most autocratic and least democratic region in the world. Muslim states outside of the Middle East are found to be more autocratic than other non-Middle Eastern states but less autocratic than Middle Eastern states. Again, Muslim states outside the Middle East are more often autocratic than their non-Muslim counterparts but considerably less often autocratic than those in the Middle East. Thus, Islam may provide no more than a partial explanation for the autocracy of the region.

In this case, history may provide an alternate explanation. Democracy and the liberal ideologies upon which it is based were developed in the West. Accordingly, it is not surprising that the West is the most democratic region of the world. Other regions particularly influenced by the West, such as the states of Latin America—which began as colonies of the West and whose inhabitants speak almost exclusively Western languages—also tend to be highly democratic, at least of late. Most of the former Soviet bloc is considered European but can be distinguished from the West in that it had limited historical exposure to central European experiences including the Reformation, Renaissance and Enlightenment.19 For nearly a century it followed another Western ideology, Marxism, and is now in the process of democratization. Other than the Middle East, the regions that are the most autocratic are Asia and Africa, regions that have retained much of their own cultures despite Western influences. Thus, the link between the Middle East and autocracy may be due more to cultural and historical momentum than anything else. Also, the link between Islam and autocracy may be due to the fact that most Islamic states are in the Middle East, Asia, and Africa, the regions that appear to have the highest levels of rejection of Western styles of government.

The level of ethnic conflict in the Middle East is about average. This comes as a surprise, given that religion is disproportionately important in the Middle East and that the region is disproportionately autocratic. It may be due to a feeling among the region’s inhabitants that religion and autocracy are normal, at least within the Middle Eastern context, and therefore do not deserve any special response. Perhaps expectations in the Middle East are lower. In the rest of the world, participants in ethnic conflicts are more sensitive to the presence of religious issues and more likely to respond to them than in the Middle East. To be specific, perhaps the Shi‘i minority in Saudi Arabia, the Copts in Egypt, and the Christian and Baha’i minorities in Iran understand that religious discrimination by the autocratic governments of the region is par for the course. Because of this, even though they suffer from high levels of religious discrimination, they engaged in no protest or rebellion in 1998 (the most recent year for which data is currently available). By contrast, in eastern Europe, where religion was suppressed by communist regimes until the 1990s, groups that suffered from religious discrimination tended to react more forcibly in 1998: the Turks in Bulgaria engaged in large demonstrations; the Abkhazians in Georgia rebelled against the state; the Chechnians in Russia have been continuously rebelling against the state; and the Albanians in Kosovo both demonstrated and rebelled.

Islam is not an explanation for the Middle East’s uniqueness. As expected, religion is particularly important in the Middle East, and the region is the most autocratic in the world. Yet Islam cannot fully explain these findings; and the disproportionate importance of religion and the presence of autocracy in the region do not lead to the increased levels of ethnic conflict one would expect.

These findings show that the obvious explanations for phenomena are not always the correct ones. It is easy to assume that the prevalence of religious conflict in the Middle East is due to the region’s Islamic and autocratic character. It is also easy to assume that the region’s high concentration of autocracy is due to the region’s Islamic character. Yet neither of these assumptions appear to be correct. Furthermore, the findings of this study show that, except for the finding that religion is particularly important in the region, ethnic conflicts which take place in the Middle East are not considerably different from similar conflicts elsewhere.

In sum, ethno-religious conflict in the Middle East is unique but not in the way many believe. Yes, religion is disproportionately important in the region’s ethnic conflicts and the region is the most autocratic in the world. But neither of these findings is explained by Islam. Furthermore, most Middle Eastern ethnic conflicts are otherwise similar to ethnic conflicts elsewhere.


These findings have several implications. First, they contradict Huntington’s notion of Islam’s "bloody borders," for ethno-religious conflict in the Middle East between Muslims and non-Muslims is not more violent than other ethnic conflicts. And Islam is not the cause of differences between ethnic conflicts in the Middle East and those taking place elsewhere.20 Despite this, Huntington’s theory remains popular, and if it prevails, it can become a self-fulfilling prophecy in which Western fears of Islam cause threats to be seen where they do not exist and small threats to be exaggerated into large ones. This can lead to reactions that will provoke the very conflicts and threats that at first were only imagined.

That the issues involved in the Middle East’s ethnic conflicts are particularly religious does not bode well for conflict resolution, for such conflicts are among the most long-lasting, violent, and difficult to settle. One possible avenue may be to apply the growing literature on religious-based conflict resolution, which advocates using those aspects of religious traditions that emphasize compromise and accommodation to prevent, settle, and mitigate conflict.21

The Middle East’s remaining the world’s most autocratic place means that the post-Cold War trend of democratization has hardly reached it. While this may change, there is no reason to believe that that will happen any time soon—as symbolized by the fact that the region’s newest political entity, the Palestinian Authority, has been vigorously repressing dissension. In all likelihood, the region’s ethnic conflicts will continue to flourish in an autocratic setting and the peaceful political avenues available to ethnic groups in democratic settings do not have a bright future in the Middle East. Opposition activity is likely to be violent. Political discrimination against minorities will probably continue.
Jonathan Fox, a lecturer in political studies at Bar Ilan University, is the author of Ethnoreligious Conflict in the Late 20th Century (Lexington Books, forthcoming).
1 Jonathan Fox, "Is Islam More Conflict Prone than Other Religions? A Cross-Sectional Study of Ethnoreligious Conflict," Nationalism and Ethnic Politics, Summer 2000, pp. 1-23.
2 In this study, the Middle East includes the Arabic-speaking countries (excluding Sudan), Turkey, Israel, and Iran.
3Karl W. Deutsch, "On Nationalism, World Regions, and the Nature of the West," Mobilization, Center-Periphery Structures, and Nation Building, ed. Per Torsvik (Oslo: Universitesforlaget, 1981), p. 51-93.
4 Samuel P. Huntington, "The Clash of Civilizations?" Foreign Affairs, 72 (1993): 22-49; idem, The Clash of Civilizations and the Remaking of the World Order (New York: Simon and Schuster, 1996).
5 Pierre Hassner, "Morally Objectionable, Politically Dangerous," The National Interest, Winter 1997, pp. 63-69; William Pfaff, "The Reality of Human Affairs," World Policy Journal, 14 (1997): 89-96; Tony Smith, "Dangerous Conjecture," Foreign Affairs, 76 (1997): 163-164; Stephen N. Walt, "Building Up New Bogeymen," Foreign Policy, 106 (1997): 177-189.
6 Rudolph J. Rummel, "Is Collective Violence Correlated with Social Pluralism?" Journal of Peace Research, 34 (2): 163-175.
7 Jonathan Fox, "The Effects of Religious Discrimination on Ethnic Protest and Rebellion," Journal of Conflict Studies, Fall 2000, pp. 16-43.
8 Errol A. Henderson, "The Democratic Peace through the Lens of Culture, 1820-1989," International Studies Quarterly, Sept. 1998, pp. 461-484; idem, "Culture or Contiguity: Ethnic Conflict, the Similarity States, and the Onset of War, 1820-1989," Journal of Conflict Resolution, Oct. 1997, pp. 649-668.
9 Jonathan Fox, "Do Religious Institutions Support Violence or the Status Quo?" Studies in Conflict and Terrorism 22 (1999): 119-139.
10 The legitimacy of the use of religion in political discourse is measured based on the variable which measures four aspects of religious involvement in politics in 1994-95: a history of religious involvement in politics; whether religious leaders are using religious rhetoric to mobilize the population; whether religion is offered as a solution to non-religious problems; and whether the state has an official religion. At least two of these factors must be present for religion to be considered legitimate in political discourse.
11 Jonathan Fox, "The Influence of Religious Legitimacy on Grievance Formation by Ethnoreligious Minorities," Journal of Peace Research, 36 (1999): 289-307.
12 Jonathan Fox: "The Salience of Religious Issues in Ethnic Conflicts: A Large-N Study," Nationalism and Ethnic Politics, Autumn 1997, pp. 1-19.
13 Jonathan Fox, "Religious Causes of Ethnic Discrimination," International Studies Quarterly, Sept. 2000, pp. 423-450.
14 Manus I. Midlarsky, "Democracy and Islam: Implications for Civilizational Conflict and the Democratic Peace," International Studies Quarterly, Sept. 1998, pp. 458-511.
15 Keith Jaggers and Ted R. Gurr, "Tracking Democracy's Third Wave with the Polity III Data," Journal of Peace Research, 32 (1995): 469-482.
16 Minorities at Risk (MAR) dataset is available at www.bsos.umd.edu/cidcm/mar. The MAR dataset currently includes 275 cases but eight of them, added after the religion data was collected, are excluded from this study.
17 For more information on the MAR dataset and the general findings on ethnic conflict based upon it, see Ted R. Gurr: Minorities At Risk (Washington, D.C.: United States Institute of Peace, 1993); idem., Peoples Versus States: Minorities at Risk in the New Century, (Washington D.C.: United States Institute of Peace Press, 2000).
18 Turkey is also a non-autocratic state but the MAR dataset includes no ethno-religious minorities in that country, thereby ignoring the Alevis, a large Shi‘i minority. On them, see Tord Olsson, et al. eds., Alevi Identity: Cultural, Religious and Social Perspectives (Istanbul: Swedish Research Institute in Istanbul, 1998). The Kurds are Sunni Muslims and so not an ethno-religious minority.
19 Huntington, The Clash of Civilizations, pp. 45-46.
20 Testing Huntington’s theories with regard to Islam more directly, I came up with similar results; see Jonathan Fox, "Two Civilizations and Ethnic Conflict: Islam and the West," Journal of Peace Research, July 2001, pp. 459-472.
21 See, for example, Marc Gopin, Between Eden and Armageddon: The Future of World Religions, Violence, and Peacemaking (Oxford: Oxford University Press, 2000); and R. Scott Appleby, The Ambivalence of the Sacred: Religion, Violence, and Reconciliation (Lanham, VA: Rowman and Littlefield, 2000).