Are Some Religions More Conflict-Prone Than Others?

, March 21, 2004

Jewish Political Studies Review 16:1-2 (Spring 2004)

 

This article focuses on the question of whether some religions are more conflict prone than others. There are several schools of thought on the topic, most of which focus on Islam. First some, like Samuel Huntington, argue that Islam is becoming increasingly violent in comparison to other religions. Second are those, like Daniel Pipes, who argue that some but not all Muslims are more violent. Finally, there are many who argue that no religion is more conflict-prone than others. This includes the modernization and secularization schools of thought which argue that religious conflict will decease because religion is becoming less important in the modern era.

However, there are others who argue that modernization has resulted in a resurgence of religion, causing an increase in religious conflict. This study tests these propositions using the State Failure dataset from the years 1965 to 2001. The results show that on an absolute level, Christians are involved in the most conflict, Muslims are involved in the most conflict in proportion to their population size, and the majority of these conflicts are intra-religious rather than inter-religious. Finally, the pattern of conflict does not conform to any of the theories in the literature. The absolute level of religious conflict followed the pattern of general conflict and increased steadily until the early 1990s and then dropped. However, religious conflict as a proportion of all conflict in any given year increased steadily throughout the period covered in this study, especially among Muslims.

 

Introduction

The argument that some religions are more prone to engaging in conflict than others is not a new one. For example, Max Weber argued that “religions with world accepting ideologies are more likely to take part in revolution than those with world rejecting ideologies.”1 More recently, especially since the events of 11 September 2001, many have posited that Islam is more conflict-prone than other religions. On the other hand, religions such as Buddhism – which include doctrinal pacifism – are seen as less prone to conflict. Nevertheless, nearly all of the evidence cited in such arguments is doctrinal, theoretical, or anecdotal. Accordingly, the purpose of this study is to assess these arguments systematically, using empirical data on domestic conflict from 1965 to 2001 from the State Failure (SF) dataset.

 

Theories of Religions and Conflict

This section is called “theories of religions and conflict” and not “theories of religion and conflict” because it does not assess how religion in general influences conflict, but rather assesses the debate over whether specific religious traditions are more conflict-prone. That is, the topic of discussion is specific religions and not the general phenomena of religion. This means that for the purposes of this study the term religion refers to religious identity – membership in a religious tradition. Accordingly, this discussion focuses on theories which posit that certain religions are either more or less conflict-prone than others, though it also examines theories which posit that all religions have similar propensities to violence.

Most of the recent discussion on the topic focuses on Islam. The debate over this topic can be divided into three camps: those who say that Muslims in general are more conflict-prone, those who argue that only certain Muslims are more conflict-prone, and those who argue that Muslims are no more conflict-prone than the followers of other religions. It is important to emphasize that this article takes no stand on this issue (other than the one based on the results of the empirical portion of this study) and tries to present the three sides of this debate as well as possible.

Samuel Huntington is perhaps the most well-known advocate of the argument that Islam is more conflict-prone than other religions.2 In his discussion of his “clash of civilizations” theory on the nature of conflict in the post-Cold War era, he argues that Islamic civilization will be more conflict-prone than any of the other religion-based civilizations he discusses. That is, Huntington divides the world into several civilizations, all but one of which are based at least in part on religion. He argues that most of the conflict in the post-Cold War era will be between these civilizations and that the Islamic civilization will be the most violent. It is important to note that all elements of his theory, especially the parts which deal with Islam, are highly controversial. However, this study focuses only on those aspects which deal with Islam.3

Huntington’s arguments regarding Islam are unambiguous. For example, he states that “the most violent fault lines [conflicts] are between Islam and its…neighbors.”4 He provides three reasons for this. First, while accepting modernity, Muslims reject Western culture and prefer to find the answers to their problems within Islam. One cause of this dynamic is that since World War II, most Islamic governments which were guided by Western ideologies like liberalism and socialism failed to successfully address the social problems of the day, increase standards of living, and provide political freedom and justice.5

This argument is not an uncommon one. Juergensmeyer6 makes the same argument with regard to religion in general in the Third World, positing that many states are returning to their indigenous religions due to the failure of governments guided by Western ideologies. Deeb,7 Piscatori,8 and Layachi and Haireche9 make similar arguments with respect to Islam.

Huntington’s second reason to believe that Islam is more conflict-prone is that Islam and the West, have historically mutually feared each other and rejected each other’s culture. Finally, he argues that this is exacerbated by doctrinal issues, especially the fact that the Islamic religion divides the world into those who follow Islam and those who do not.10

Huntington’s arguments concerning Islam and the West are, to say the least, not uncontroversial. However, the specific reasons for disagreeing with him are diverse and sometimes contradict each other. The arguments against Huntington can be divided into several categories. First, modernity and economic development are causing the Islamic world to become more secular. As a result, Islam has a decreasing impact on politics.11This argument reflects a wider secularization literature which posits that religion is becoming irrelevant in modern times. It is discussed in more detail below.

Second, most of the violence involving Muslims is by Islamic fundamentalists. However, Islamic fundamentalism is controversial even within Islam. Furthermore, these clashes are really clashes between Islamic fundamentalists and modernity, and such conflicts between fundamentalists and modernity exist in most religions, not just Islam. In addition, many argue that the real motivation for fundamentalist movements is economic and social, not religious.12 In fact, Islamic fundamentalism is more of a threat to the authoritarian regimes in Islamic states than it is to the West.13

Third, many clashes between Muslims and non-Muslims are not due to religious causes, but rather secular ones. These secular causes include economic, national, political, cultural, psychological, post-colonial, and strategic issues.14

Fourth, Islam and the rest of the non-West are embracing the West rather than rejecting it. Other civilizations aspire to be like the West,15 and the non-West actually fears the retreat of Western world leadership.16

Fifth, Islamic civilization is weak and divided. Most conflict involving Muslims is actually with other Muslims. Nationalism is a potent factor within the Islamic world, causing many divisions within it, and it is only one of the many issues which divide Muslims. Other divisions include those between Arabs and non-Arabs, between Sunni and Shi’i Muslims, and between the rich and the poor Muslims, both internationally and within states.17 The few quantitative studies which address this issue show that, the majority of conflicts involving Muslims are, in fact, with other Muslims.18

While this description of the debate over Huntington’s views on Islam is by no means complete, it is sufficient to highlight the arguments of those who believe that Islam in general is more conflict-prone than other religions, as well as those who argue that it is not.

Those who hold that some Muslims are more conflict-prone than members of other religions, in general argue that Islamic fundamentalists, often called Islamists, are the source of the higher levels of violence by Muslims and that other Muslims are no more conflict-prone than members of other religions. However, there is little agreement over the extent to which Islamists are more conflict-prone and the extent to which this interpretation of Islam is popular within the Islamic world.

Daniel Pipes is a prominent proponent of this type of argument. Pipes defines Islamism as “a political movement that takes the religion of Islam and turns it into the basis of a totalitarian ideology that shares much with prior versions, namely fascism and Marxism-Leninism.”19 This movement includes the majority of violent Islamic terrorist movements, including Al-Qaeda. Pipes believes that the Islamists are an influential minority within the Islamic world and that many of their actions are popular with broader segments of Muslims.

Abdullahi Ahmed An-Na’im argues that the extent of the radical movement within Islam is more limited. He believes that “a few militant and highly motivated gangsters – real criminals – are holding Muslim cultures and Muslim leadership hostage.”20 However, this movement is marginal in the larger picture. More specifically, “the notion of aggressive Jihad has become morally untenable…and the rise of the modern human rights movement has tumbled the moral foundations of segregation and discrimination against women and non-Muslims.”21

Others acknowledge the dual trends within Islam but do not take a stand about the level of popularity of Islamism. For example, Bernard Lewis cites the perceived “occupation” of Saudi Arabia by the Unites States and the Palestinian-Israeli conflict as being among the motivations for Islamist attacks against the West.22 However, he argues that this is an extreme interpretation of Jihad and that Islamic rules for behavior in war do not enjoin murder and terrorism or even consider the random slaughter of uninvolved bystanders. More formal examinations of the rules of war in Islamic texts come to similar conclusions.23 Esposito and Voll make a normative version of this argument, calling for Muslims to follow the example of Muslim leaders who interpret Islam in a more peaceful manner.24 Sachedina similarly argues that the Islamists are corrupting Islam.25

In addition to the arguments which focus specifically on Islam, there are several types of arguments regarding the propensity of certain religions to engage in conflict. One such argument posits that all religions have within them the potential to support both peace and violence. For example, in a discussion on terrorism, Rapoport argues that all religions have enormous potential for creating and directing violence and that fundamentalists can exploit the violent potential of a religion even when that religion is rarely perceived as having violent potential.26 Juergensmeyer argues similarly that the rhetoric of war is “prominent in modern religious vocabulary” and “virtually all cultural traditions are filled with martial metaphors.”27 Appleby calls this “the ambivalence of the sacred.”28 Paradoxically, religions support both peace and the sword. This is because most religions have complex traditions which include multiple interpretations of doctrine, some peaceful and some violent. Religious leaders can choose to emphasize either of these trends.29

This is true even of those religions commonly perceived as doctrinally peaceful. For example, a form of Buddhist doctrine supports the violence by the Buddhist majority in Sri Lanka against its Hindu, ethnic Tamil minority. They believe that Buddha charged them with preserving the true Buddhism – Therevada Buddhism – and gave the island of Lanka to the Sinhalese Buddhist majority to create a “citadel of pure Buddhism.” Accordingly, violence is justified in support of that goal.30 Another example of Buddhist justifications for violence is that used by Asahara, the Tokyo subway killer, who believes that since souls are transferred upon death, some people are better off dead, especially people living in evil situations that will cause damage to their karmas.31 While clearly many Buddhists dispute these interpretations, it demonstrates that even in a doctrinally peaceful religion, interpretations of doctrine that support violence are possible and, in fact, occur.

This situation where religions stimulate violence is, perhaps, most common when religion and nationalism are combined. This is the case in the two most well-known cases where Buddhists engage in violence, Sri Lanka and Tibet. Haynes argues that because Buddhism has no definitive scriptures it is more easily combined with nationalism.32

A second type of argument also posits that all religions have a similar propensity for violence. However, rather than postulating that all religions are prone to conflict, it posits that religion is becoming irrelevant in the realm of politics and therefore no religion will have any significant influence on conflict. This argument is known as the modernization theory in the political science literature and the secularization theory in the sociology literature. It posits that several trends inherent in the process of modernization are resulting in the demise of religion as an important political and social factor. Rational and scientific thought are replacing superstition as a basis for understanding the world and guiding society. This includes secular criteria for defining morality and correct behavior. For example, psychologists have replaced the clergy in defining what is considered sane, normal, and acceptable behavior, and rationalist-based concepts of human rights are taking the place of religion in defining what is moral. This is supported by mass education and increased literacy which give the masses direct access to information that was monopolized by the clergy in the past. Technology provides people with alternatives to religion. For example, contraception has undermined the traditional morality of abstinence. Urbanization is undermining the small traditional community in which religion thrived in the past.33

It is important to note that this body of theory has come under serious question. The facts that religious conflicts remain common and religious fundamentalism is present in most of the world contradict the notion that religion is becoming irrelevant. In fact, many would argue that modernity has caused the increase in religion’s relevance. That is, the modern threat to religion has caused a religious backlash against it.34 In a similar vein, many argue that fundamentalism is a modern reaction against modernity.35

These various trends in the literature can be distilled into several hypotheses. It is important to note that these hypotheses contradict each other as there is little agreement on the topic in the literature. The purpose of this study is to determine which of them are correct.

  • Hypothesis 1: Islamic groups are more violent than other groups.
  • Hypothesis 2: Groups of all religions engage in the same level of conflict.
  • Hypothesis 3: As all religions contain the potential for violence, there will be conflict which involves all religious groups.
  • Hypothesis 4: As religion is becoming irrelevant in modern times, religious conflict will decrease over time.
  • Hypothesis 5: As modernity is causing a religious backlash, religious conflict will increase over time.
  • Hypothesis 6: As the factors that cause religious conflict are stable over time, the extent of religious conflict will remain stable over time.

 

It is important to note that while a hypothesis predicting that some Muslims are more violent than other groups is warranted by the literature, there is no way to test this proposition with the SF data so this hypothesis was not included in the list above.

 

Research Design

This study uses the SF dataset to test the theories described above. The unit of analysis in the SF dataset is a year of violent conflict. Consequently, the dataset includes data on only those states which have undergone violent conflict for the periods of time when those conflicts were occurring. Thus, it includes a good representation of how many violent conflicts were occurring in any given year. The dataset includes conflicts from 1955 to 2001, but this study focuses on the data from 1965 to 2001 due to the low level of conflict in the first decade of data.

This study includes data from three sections of the SF dataset: revolutionary war, ethnic war, and mass killings. Revolutionary wars are defined as:

episodes of violent conflict between governments and politically organized groups (political challengers) that seek to overthrow the central government, to replace its leaders, or to seize power in one region. Conflicts must include substantial use of violence by one or both parties to qualify as wars.36

Ethnic wars are defined as: “episodes of violent conflict between governments and national, ethnic, religious, or other communal minorities (ethnic challengers) in which the challengers seek major changes in their status.”37 Mass killings are defined as:

the promotion, execution, and/or implied consent of sustained policies by governing elites or their agents – or in the case of civil war, either of the contending authorities – that result in the deaths of a substantial portion of a communal group or politicized non-communal group. In genocides the victimized groups are defined primarily in terms of their communal (ethnolinguistic, religious) characteristics. In politicides, by contrast, groups are defined primarily in terms of their political opposition to the regime and dominant groups. Geno/politicide is distinguished from state repression and terror. In cases of state terror authorities arrest, persecute or execute a few members of a group in ways designed to terrorize the majority of the group into passivity or acquiescence. In the case of geno/politicide authorities physically exterminate enough (not necessarily all) members of a target group so that it can no longer pose any conceivable threat to their rule or interests.38

The unit of analysis for the SF dataset is a conflict year. Each year during which a particular type of conflict occurred in a particular state is coded separately, including partial years in which the conflict began or ended.

The following modifications were made to the data for this study. First, there are several cases where the SF dataset codes conflict by several groups against the state together as a single entry; this study separates them into separate cases.39 Second, many of the cases in the three categories overlap. For the tests performed on the entire dataset, the overlapping cases were removed from the study.40 As a result, 638 years of ethnic war, 233 years of mass killings, and 390 years of revolutionary war were coded. Taking overlapping cases into account, this totals 1,010 conflict years between 1965 and 2001.

Each of these years of conflict is coded for the purposes of this study as religious or not religious.41 A conflict is coded as religious if it meets one of the three following criteria: 1) it is between two groups which belong to different religions, 2) it is between two groups which belong to different denominations of the same religion, or 3) the descriptive material embedded within the SF data describes the conflict as religious. This latter category usually represents fundamentalist challenges to secular states.

In order to identify the groups involved in conflict, each side of a conflict was coded as Christian, Muslim, or “other.” While clearly there is considerable diversity within Christian and Muslim groups – not to mention the “other” category – this simplification was necessary in order for there to be enough cases in each category for meaningful comparison.

Each year of conflict was counted twice in many parts of the analysis, once for each of the two groups involved in the conflict. This means that a conflict between two Christian groups, for example, was counted twice: once for each Christian group. Similarly a conflict between a Muslim group and a Christian group would also be counted twice, once as a conflict involving Christians and once as a conflict involving Muslims. This allows us to account for the total participation of various religious groups in violent conflict. For the purposes of this study, this methodology is a participant-based analysis; counting each conflict once is referred to as a case-based analysis.

This study proceeds in several stages. First, we assess the extent of overall state failure on a yearly basis, using a participant-based analysis control for the specific religions of the groups involved. This allows us to assess the actual participation of each type of religion in conflict and test Hypotheses 1, 2, and 3. Second, we repeat the above analysis, controlling for the population size of each religion. This is because, for example, as there are less Muslims in the world than there are Christians, we would expect there to be less conflicts involving Muslims. That is, if there are less people who belong to a religion, we would expect the absolute amount of conflict in which members of this religion participate to be lower than the absolute amount of conflict in which members of a larger religion participate.42 This also tests Hypotheses 1, 2, and 3.

Third, we perform the same two tests as above, but only for religious conflicts. This allows us to examine the extent of religious conflict in the world and the extent to which each category of religion participates in religious conflict. While this primarily tests Hypotheses 4, 5, and 6, it is also relevant to Hypotheses 1, 2, and 3 with regard to religious conflict.

Finally, using a case-based analysis on a yearly basis, we examine the percentage of world conflict which is religious, both in general and for each category of religion used in this study. This provides a different perspective on the change in the amount of religious conflict over time. The previous analyses examine the absolute amount but this analysis examines the relative amount. This is because it is possible for the absolute amount of religious conflict to increase while the relative amount of religious conflict decreases if the absolute amount of non-religious conflict is increasing even faster than the absolute amount of religious conflict.

 

Data Analysis and Discussion

Table 1: All State Failures, 1965-2001

The analysis in Table 1 shows the absolute amount of conflict in which Christians, Muslims, and other groups engaged between 1965 and 2001. The results show that Christian groups engaged in the most conflict for nearly all of this period. This was true for every year except 1967-1970, 1994, 1997, and 1998. In the 1967-1970 period, “other” groups engaged in the same or higher levels of conflict than Christian groups, and in 1994, 1997, and 1998, Muslim groups engaged in the same or higher levels of conflict. Thus, in all but seven years of the thirty-seven-year period examined, Christian groups engaged in the highest level of conflict, but there is some evidence that Muslim groups became more violent in the 1990s. This table also shows that in general there was a steady rise in conflict until the early 1990s, followed by a drop.

Table 2: All State Failures, Controlling for Population Size, 1965-2001

(The values in this table represent the expected number of state failures if the denomination in question constituted the population of the entire world.)

Table 2 examines the same data as Table 1, but it controls for the proportion of the population that constitutes each category of group. That is, it presents the number of conflicts we would expect from each category if that category was the entire population of the world. The results are very different. Except for four years during the 1970s, this analysis shows Muslim groups to be the most violent in every year examined. It also shows that Christian groups are consistently involved in more conflict than groups in the “other” category. Thus, there is a clear hierarchy of violence by this measure. Muslim groups are the most violent, “other” groups the least violent, and Christian groups fall in between the two.

Based on the above results, Hypothesis 2 – that all groups engage in the same level of violence – can clearly be rejected as both tests show that this proposition is false. Similarly, Hypothesis 3 – that all groups engage in at least some violence – is supported by the evidence as this is true in both analyses. However, the results in Tables 1 and 2 are not consistent with regard to Hypothesis 1, which posits that Islamic groups are more violent than other groups. When examining the absolute level of conflict, Christian groups engage in the most conflict, but the level of conflict by Islamic groups began to approach that of Christian groups during the 1990s. When controlling for population size, Hypothesis 1 is clearly supported by the evidence. Thus, whether there is any truth to Hypothesis 1 depends on how one measures conflict. It is argued here that the proportional method is the most appropriate as we would clearly expect a larger group to engage in more violence simply due to increased opportunity. Once we account for this factor, Muslims clearly engage in more violence.

Table 3: Religious State Failures, 1965-2001

Table 3 examines the absolute extent of religious conflict in the world between 1965 and 2001. It shows that with the exception of four years during the 1970s, Muslim groups engaged in the most religious conflict. The analysis also shows a clear rise in religious conflict between the 1970s and early 1990s, followed by a drop. This pattern is most prominent among Muslim groups, but it also exists among Christian groups.

Table 4: Religious State Failures, Controlling for Population Size, 1965-2001

(The values in this table represent the expected number of state failures if the denomination in question constituted the population of the entire world.)

Table 4 examines the same data in Table 2, but controls for the proportion of the population that constitutes each group category. These results show that Muslims engaged in the highest level of religious conflict throughout the 1965 to 2001 period. Thus, the analyses in Tables 3 and 4 show that Hypothesis 1 is correct for religious conflict and Islamic groups engage in the most conflict. As is the case with the previous two tables, this analysis shows that Hypothesis 2 is false and supports Hypothesis 3.

Interestingly, this analysis does not support Hypotheses 4, 5, or 6. Hypothesis 4 predicts a decrease in religious conflict over time, which clearly does not occur. Hypothesis 6 predicts that the level of religious conflict will remain steady over time, which also does not occur. The evidence is closest to supporting Hypothesis 5, which predicts a rise in religious conflict over time. This was true as of the early 1990s, but after that religious conflict dropped. This may indicate that religious conflict is a cyclical phenomenon. In order to fully confirm this, data for a longer period of time than is currently available is necessary, preferably covering at least a century.

Another possible explanation for this is that religious conflict reflects the overall level of conflict which also rose throughout the early 1990s and then dropped. Accordingly, the analysis in Table 5 examines the proportion of conflict which is religious. That is, rather than focusing on the absolute level of conflict, it focuses on whether or not religious conflict is becoming more common relative to non-religious conflict. Overall, there is a clear rise in the proportion of conflicts which are religious. In the period 1965-1980, between 26.1 percent and 33.3 percent of all conflicts were religious in any given year. In the period 1981-1990, between 30.0 percent and 37.9 percent were religious in any given year. In the period 1991-2001, this rose to between 42.4 percent and 50 percent, with the 50 percent score occurring in 2001, the final year of this analysis. These results clearly support Hypothesis 5.

Table 5: Percentage of State Failures which are Religious, 1965-2001

The results in Table 5 also show that, more often than not, the group with the highest proportion of religious conflict is Muslim. In 1984, 1986, 1987, 1990, 1992, 1993, and 1994 the proportion of religious conflict by “other” groups was as high or higher than that of Muslim groups. By 2001, however, the proportion of Islamic conflict which was religious was almost 80 percent higher than that of groups in the “other” category. Thus, there was a surge of religious conflict by “other” groups beginning in the mid-1980s, but this surge began to weaken during the early to mid-1990s. Nevertheless, the proportion of conflict analyzed here by “other” groups which were religious at the end of the period is about as high as it was after the surge began in the mid-1980s, and higher than it was at any point before that. Thus, “other” groups are also participating in the overall rise in the proportion of world conflict which is religious.

Conflict involving Christian groups involves about the same proportion of religious conflict throughout the period analyzed here. The proportion of Christian conflict which is religious remains between 12.5 percent and 27.9 percent during the entire period examined here. Thus, the high point of Christian religious conflict is just barely higher than the low point for Muslim religious conflict (26.1 percent). This also means that the general rise in religious conflict as a proportion of all conflict does not apply to Christian groups.

 

Conclusions

This study set out to answer two basic questions with regard to domestic violent conflict. First, is any particular religious group more violent than others? Second, has there been any change in the level of religious conflict over time? The short answer to both questions is yes, but the more accurate answer is a bit more complicated.

If one examines only the absolute level of all domestic conflict, Christian groups appear to be consistently the most violent. Islamic groups were becoming more violent during the 1990s, but Muslim groups were less violent than Christian groups. However, if one takes into account the fact that there are about twice as many Christians in the world as there are Muslims, the proportional level of conflict by Muslims is clearly higher. The results for religious conflict are clearer, as Muslim groups engage in the most conflict overall in both absolute and proportional terms.

Thus, there is some support for Huntington’s contention that Muslims have “bloody borders.” However, the pattern of conflict is very different from the one Huntington describes. He predicts that this will be a new phenomenon in the post-Cold War era, which is not the case. Also, while he predicted a steady rise in the absolute level of conflict after the Cold War, the level of conflict dropped during the early 1990s, shortly after the end of the Cold War.

Huntington also predicted a rise in conflict between his religiously-defined civilizations, which seems consistent with the proportional rise of religious conflict. However, upon closer inspection much of this rise includes conflicts between fundamentalist rebels and secular governments which does not fit Huntington’s definition of civilizational conflict. When taking this into account, conflict between groups of different religions never reached higher than 40 percent of all conflicts in the post-Cold War era (1990 onward), and was often less than one-third of all conflict in this era.

With regard to the second major issue of this study – the pattern of overall religious conflict – it is clear that there are fluctuations over time in the level of religious conflict. However, the pattern of these fluctuations does not clearly conform to any of those described in the literature on the topic. On an absolute level, conflict – both religious and non-religious – began rising in the 1970s, peaked in the early 1990s, then dropped considerably. However, the proportion of conflict which is religious began to rise in the late 1980s and reached its highest point in 2001, the last year included in this study. In absolute terms, religious conflict thus appears to be cyclical. In proportional terms, however, it conforms to the predictions of those who argue that there will be a religious backlash against modernization, causing an increase in religious conflict. Thus, while the level of conflict overall appears to be cyclical, those conflicts that do occur seem to be becoming more religious in nature.

However, the extent to which this is true is different for different religious groups. For Muslims, the vast majority of domestic conflicts in the late 1990s and the new millennium are religious ones. Non-Abrahamic groups have also experienced a surge in religious conflict, but that surge, though continuing as of 2001, peaked in the early 1990s. Conflicts involving Christian groups, in contrast, have consistently been mostly non-religious.

In conclusion, this study shows that religion has undoubtedly not disappeared in the modern era, at least not in the form of religious conflict. There have been fluctuations in the level of religious conflict and some religious groups are more likely than others to participate in both conflict in general and religious conflict in particular. However, there is no doubt that all religious groups examined here have consistently participated in at least some religious conflict throughout the 1965 to 2001 period.

Thus, while religious conflict is a constant, the exact extent of religious conflict varies from group to group and over time. Unfortunately, as the SF data focuses mostly on the dependant variable of conflict, it does not provide sufficient information to examine why these variances occur. Furthermore, the fluctuations over time do not fully conform to any of the three trends predicted in the literature. Consequently, further research is needed in order to determine the causes of these periodic fluctuations in religious conflict.

*     *     *

Notes

1. Cited in David Kowalewski and Arthur L. Greil, “Religion as Opiate: Religion as Opiate in Comparative Perspective,” Journal of Church and State, 1990, p. 516.

2. Samuel P. Huntington, “The Clash of Civilizations?” Foreign Affairs, 72 (3), (1993): 22-49; Samuel P. Huntington, The Clash of Civilizations and the Remaking of the World Order (New York: Simon and Schuster, 1996).

3. For a further discussion of Huntington and his critics see Jonathan Fox, “Ethnic Minorities and the Clash of Civilizations: A Quantitative Analysis of Huntington’s Thesis,” British Journal of Political Science, 32 (3), (2002b): 415-434; Jonathan Fox, Religion, Civilization and Civil War since 1945: The Empirical Study (Lanham, MD: Lexington Books, 2004).

4. Huntington, The Clash of Civilizations and the Remaking of the World Order, p. 183.

5. Huntington, The Clash of Civilizations and the Remaking of the World Order, pp. 32, 109-120, 185, 209-218.

6. Mark Juergensmeyer, The New Cold War? (Berkeley: University of California, 1993).

7. Mary J. Deeb, “Militant Islam and the Politics of Redemption,” Annals, AAPSS, 524 (1992): 53-54.

8. James Piscatori, “Accounting for Islamic Fundamentalisms,” in Martin E. Marty and R. Scott Appleby, eds., Accounting for Fundamentalisms: The Dynamic Character of Movements (Chicago: University of Chicago Press, 1994), pp. 361-363.

9. Azzedine Layachi and Abdel-Kader Haireche, “National Development and Political Protest: Islamists in the Maghreb Countries,” Arab Studies Quarterly, 14 (2 & 3), (1992): 70.

10. Huntington, The Clash of Civilizations and the Remaking of the World Order, pp. 32, 109-120, 185, 209-218.

11. Faoud Ajami, “The Summoning” (a reply to Huntington’s “The Clash of Civilizations?”), Foreign Affairs, 72 (4), (1993): 2-9; Robert L. Bartley, “The Case for Optimism,” (a reply to Huntington’s “The Clash of Civilizations?”), Foreign Affairs, 72 (4), (1993): 15-18.

12. Many, including Marty and Appleby, echo the argument that the rise in fundamentalism is occurring internationally and is, at least in part, a reaction to modernity. Martin E. Marty and R. Scott Appleby, eds., Fundamentalisms and the State: Remaking Politics, Economies and Militance (Chicago: University of Chicago Press, 1991); Martin E. Marty and R. Scott Appleby, eds., Fundamentalisms and Society: Reclaiming the Sciences, the Family, and Education (Chicago: University of Chicago Press, 1993); Marty and Appleby, eds., Accounting for Fundamentalisms: The Dynamic Character of Movement; Bartley, “The Case for Optimism”; Graham E. Fuller and Ian O. Lesser, A Sense of Siege: The Geopolitics of Islam and the West (Boulder, CO: Westview, 1995); Shirleen T. Hunter, The Future of Islam and the West: Clash of Civilizations or Peaceful Coexistence? (Westport, CT: Praeger with the Center for Strategic and International Studies, Washington, DC, 1998).

13. John L. Esposito, The Islamic Threat: Myth or Reality? (Oxford: Oxford University Press, 1995); Fred Halliday, Islam and the Myth of Confrontation: Religion and Politics in the Middle East (New York: St. Martin’s Press, 1996).

14. Esposito, The Islamic Threat: Myth or Reality? ; Fuller and Lesser, A Sense of Siege: The Geopolitics of Islam and the West; Mahmood Monshipouri, “The West’s Modern Encounter with Islam: From Discourse to Reality,” Journal of Church and State, 40 (1), (1998): 25-56.

15. Jeane J. Kirkpatrick et al., “The Modernizing Imperative,” Foreign Affairs, 72 (4), (1993): 22-26.

16. Kishore Mahbubani, “The Dangers of Decadence,” Foreign Affairs, 72 (4), (1993): 10-14.

17. Hunter, The Future of Islam and the West: Clash of Civilizations or Peaceful Coexistence?; Mahbubani, “The Dangers of Decadence”; Brian Beedham, “The New Geopolitics: A Fading Hell,” Economist, 31 July 1999: s10; Zerougui A. Kader, “The Clash of Civilizations and the Remaking of World Order,” Arab Studies Quarterly, 20 (1), (1998): 89-92; Monshipouri, “The West’s Modern Encounter with Islam: From Discourse to Reality.”

18. Jonathan Fox, “Islam and the West: The Influence of Two Civilizations on Ethnic Conflict,” Journal of Peace Research, 38 (4), (July, 2001): 459-472; Jonathan Fox, “State Failure and the Clash of Civilizations: An Examination of the Magnitude and Extent of Domestic Civilizational Conflict from 1950 to 1996,” Australian Journal of Political Science, 38 (2), (2003): 195-213; Fox, Religion, Civilization and Civil War since 1945: The Empirical Study.

19. Daniel Pipes, “A New Round of Anger and Humiliation: Islam after 9/11,” in Wladyslaw Pleszczynski, ed., Our Brave New World: Essays on the Impact of September 11 (Stanford: Hoover Institution Press, 2002), pp. 41-61; Daniel Pipes and Mimi Stillman, “The United States Government: Patron of Islam?” Middle East Quarterly, 2002.

20. Abdullahi Ahmed An-Na’im, “The Islamic Counter-Reformation,” New Perspectives Quarterly, 19 (1), (2002): 31.

21. An-Na’im, “The Islamic Counter-Reformation,” p. 33.

22. Bernard Lewis, “License to Kill: Usama bin Ladin’s Declaration of Jihad,” Foreign Affairs, 77 (6), (1998): 14-19.

23. Khaled A. El Fadal, “Muslims and Accessible Jurisprudence in Liberal Democracies: A Response to Edward B. Foley’s Jurisprudence and Theology,” Fordham Law Review, 66 (4), (1998): 1227-1231; Sohail H. Hashimi, “Saving and Taking Life in War: Three Modern Views,” The Muslim World, 89 (2), (1999): 158-180.

24. John L. Esposito and John O. Voll, “Islam and the West: Muslim Voices of Dialogue,” Millennium, 29 (3), (2000): 613-639.

25. Abdulaziz Sachedina, “Religion and Global Affairs: Islamic Religion and Political Order,” SAIS Review, 18 (2), (1998): 60-61.

26. David C. Rapoport, “Comparing Fundamentalist Movements and Groups,” in Marty and Appleby, eds., Fundamentalisms and the State: Remaking Politics, Economies and Militance, pp. 429-461.

27. Mark Juergensmeyer, “Sacrifice and Cosmic War,” Terrorism and Political Violence, 3 (3), (1991), p. 106.

28. Appleby, The Ambivalence of the Sacred: Religion, Violence, and Reconciliation.

29. Similar arguments are made by Marc Gopin, Between Eden and Armageddon: The Future of World Religions, Violence, and Peacemaking (Oxford: Oxford University Press, 2000); Marc Gopin, Holy War, Holy Peace: How Religion Can Bring Peace to the Middle East (New York: Oxford University Press, 2002); Ben Mollov, Religion and Conflict Resolution (Israel: Yuval Press, 2003); Mohammed Abu-Nimer, “Conflict Resolution, Culture, and Religion: Toward a Training Model of Interreligious Peacebuilding,” Journal of Peace Research, 38 (6), (2001): 685-704.

30. James Manor, “Organizational Weakness and the Rise of Sinhalese Buddhist Extremism,” in Marty and Appleby, eds., Accounting for Fundamentalisms: The Dynamic Character of Movements, pp. 770-784.

31. Mark Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence (Berkeley: University of California Press, 2000), p. 114.

32. Jeff Haynes, “Religion, Secularization, and Politics: A Postmodern Conspectus,” Third World Quarterly, 18 (4), (1997): 721.

33. For a more detailed discussion of this body of theory and its critics see Jonathan Fox, Ethnoreligious Conflict in the Late 20th Century: A General Theory (Lanham, MD: Lexington Books, 2002).

34. Emile Sahliyeh, ed., Religious Resurgence and Politics in the Contemporary World (New York: State University of New York Press, 1990); Jeff Haynes, Religion in Third World Politics (Boulder, CO: Lynne Rienner Publishers, 1994), p. 7.

35. Marty and Appleby, Fundamentalisms and the State: Remaking Politics, Economies and Militance; Marty and Appleby, Fundamentalisms and Society: Reclaiming the Sciences, the Family, and Education; Marty and Appleby, Accounting for Fundamentalisms: The Dynamic Character of Movements.

36. Ted R. Gurr, Barbara Harff, and Monty G. Marshall, “Internal Wars and Failures of Governance, 1954-1996,” www.cidcm.umd. edu/inscr/stfail, 19 May 1997.

37. Gurr, Harff, and Marshall, “Internal Wars and Failures of Governance, 1954-1996.”

38. Gurr, Harff, and Marshall, “Internal Wars and Failures of Governance, 1954-1996.”

39. There are episodes of conflict which were broken up. First, the joint Serb and Croat war against the Bosnian government from 1992 to 1995 was broken up into two separate cases, one for Serbs and one for Croats; four conflict years were added as a result. Second, the Abkhaz and South Ossetian rebellion from 1991 to 1993 in Georgia was broken up into two separate cases, one for Abkhaz and one for South Ossetians; three conflict years were added as a result. Third, the Shi’i and Kurdish rebellions in Iraq from 1991 to 1998 were broken up into two cases; eight conflict years were added as a result.

40. There are 34 conflict years where all three categories overlap. In addition, there are 97 conflict years of mass killings which overlap with ethnic wars, 65 conflict years of revolutionary wars which overlap with ethnic wars, and 21 conflict years of revolutionary wars which overlap with mass killings.

41. This variable is not included in the original SF dataset and was developed by the author of this study.

42. According to David B. Barret, George T. Kurian, and Todd M. Johnson, World Christian Encyclopedia: A Comparative Survey of Churches and Religions in the Modern World, 2nd ed. (Oxford: Oxford University Press, 2001), Muslims constituted 15.0 percent of the world population in 1970, 18.3 percent in 1990, and 18.9 percent in 1995. Christians were 33.5 percent, 33.2 percent, and 33.1 percent, respectively. It is possible to use these population statistics to construct an analysis controlling the population size of the world’s religions. The number of groups rebelling in a particular category was divided by the proportion of the world’s population that religion constituted (i.e., in the 1970-1974 period, the number of rebelling Muslim groups was divided by .183 and the number of rebelling Christian groups by .335).

*     *     *

JONATHAN FOX received his Ph.D. in government and politics from the University of Maryland in 1997. He is currently a senior lecturer in the Political Studies Department of Bar Ilan University and a fellow in the Begin-Sadat Center for Strategic Studies. He specializes in the influence of religion on politics, which he examines using both quantitative and qualitative methodologies including analyses of the impact of religion on domestic conflict, terrorism, international intervention, and international relations as well as separation of religion and state. He has published numerous articles on this topic as well as his recent books: Bringing Religion into International Relations (New York: Palgrave-MacMillan, 2004) and Religion, Civilization, and Civil WarK/i> (Lanham, MD: Lexington Books, 2004).

Jonathan Fox

Jonathan Fox received his Ph.D. in government and politics from the University of Maryland in 1997. He is currently a senior lecturer in the Political Studies Department of Bar Ilan University and a fellow in the Begin-Sadat Center for Strategic Studies. He specializes in the influence of religion on politics.