SciELO - Scientific Electronic Library Online

vol.23 issue70Rationality Economic and TerrorismLikelyhood of non-violent political action in societies fragmrnted by multiple conflicts: on the colombian case author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand



Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google


Análisis Político

Print version ISSN 0121-4705

anal.polit. vol.23 no.70 Bogotá Sept./Dec. 2010


Wath is the Link Between Religión and Violence? An Assement of the Literature

¿Cuál es el vínculo entre Religión y Violencia? Una Evolución de la literatura

Giovanni Mantilla

Giovanni Mantilla is a Ph.D. Candidate in the Department of Political Science at the University of Minnesota. He holds an M.A. fromthe same University, and B.A. degrees in Political Science and Languages and Sociocultural Studies from the Universidad de los Andes in Bogota, Colombia. University of Minnesota


This article critically reviews the main debates and literature produced by international relations and comparative politics scholars connecting religion to the eruption of violence. Centrally, it argues that, despite fierce fractionalization in other fields of study, academic work on the religion-violence nexus seems to be reaching an emerging consensus between rationalist and constructivist approaches, approximating what Katzenstein and Sil have recently dubbed "eclectic theorizing." It also surveys the virtues and weaknesses of current work, and suggests avenues for future research.

Keywords:Religion, Violence, Conflict, Constructivism, Rationalism.


Este artículo discute los debates centrales de la literatura académica sobre religión y violencia producida por los estudiosos de las relaciones internacionales y la política comparada. El argumento central es que, a pesar de que en otros campos de estudio las aproximaciones racionalistas y constructivistas se enfrentan como explicaciones alternativas, en los estudios sobre religión y violencia parece estar surgiendo un consenso que las combina, al estilo de lo que Katzenstein y Sil bautizaron recientemente "teorización ecléctica". El artículo destaca las virtudes y deficiencias del trabajo existente, y sugiere vías para investigaciones futuras.

Palabras clave: religión, violencia, conflicto, constructivismo, racionalismo


As Daniel Philpott and Eva Bellin argue in separate, newly published review articles, the study of global and comparative politics appears to have "found" religion only recently and sparsely (Philpott, 2009; Bellin, 2008.) A number of factors seem to account for this "delayed response" by social scientists—the most commonly mentioned ones being, on the one hand, the Cold War, whose end allowed for a shift the lenses of security scholars and political scientists in general from the ideological superpower rivalry toward non-ideological and non-international variants of conflict, and on the other, the pervasiveness of the secularism in assumption and methods(1). Infamously, within the academe, Samuel Huntington was one of the driving forces behind this move with his argument about an impending "clash of civilizations", which predicted that future conflicts would be waged by religiously-inspired cultural entities(2). Huntington's injunction, coupled with very visible events of religiously-inspired violence, notably the Al Qaeda attacks on the US on September 11, 2001, served to revive academic interest specifically on the link between religión and political violence (Huntington, 1997.) Now, almost twenty years after the controversial thèse civilisatrice was born, scholars from different disciplines, including a few political scientists, have produced important theoretical reflections and empirical findings in this subarea.

This literature review will attempt to present and assess the scholarly work that connects religión to violent conflict, placing a stronger emphasis on the main arguments and recent developments within Political Science(3). I will proceed in the following manner. First, to give the reader a sense of the dimension of the religion and conflict nexus in global politics, I concern myself with the few but important empirical studies that have tried to gauge the extent to which religion actually plays an increasing role in violent struggles around the world. Next, I turn the lenses toward the major theoretical arguments and approaches launched by scholars to understand this phenomenon. Though I aim to offer assessments and critiques throughout this essay, I will conclude with general remarks regarding the body of literature surveyed and with some ideas for future research. By the end I aim to have demonstrated that studies on the impact of religion on conflict represent a good example of what Peter Katzenstein and Rudra Sil (2008, 2010) recently dubbed "eclectic theorizing," insofar as the most persuasive analytical frames --indeed, the current "conventional wisdom"-- is constituted by arguments that combine rationalist and constructivist elements.

I. Establishing Correlations: What is the Impact of Religion on Conflict?

Until recently, claims regarding the alleged rise of religiously-based conflict were largely based on a few illustrative examples or on scattered evidence. Political scientists were often content to point to a few highly visible cases of religious groups engaged in violent conflict in places like Israel and Palestine, Lebanon, the Sudan or on US soil, but without substantive cross-case empirical assessments to sustain the generalizing tone of their arguments. Sociologists, for their part, were more systematic in building a body of comparative case studies to develop comprehensive frameworks for understanding fundamentalism (Appleby et al., 2004.) This important work notwithstanding, the actual dimension of the impact of religion on violent conflict, both domestic and international, remained unclear.

Recent work has begun to fill this gap. To begin with, in a series of quantitative studies, Jonathan Fox has devoted much effort to determine, among others, the extent to which religious conflicto has actually increased since the 1950s in absolute as well as in relative numbers-- that is, vis-à-vis non-religious conflicts (Fox, 2004a, 2004b.) Based on data from the State Failure project(4), which codes revolutionary, ethnic and genocidal wars from 1950 onward, Fox establishes that, contraHuntington, though religious conflicts have increased since the 1950, so have non-religious ones, dispelling the idea that the relative proportion of "wars over the divine" would outnumber civil conflicts. Furthermore, Fox's data suggests that both religious and non-religious conflicts have sharply decreased after the Cold War (at least until 1996, his cutoff point,) refuting Huntington's predictions.

Fox's findings also contradict arguments based on secularization theory, which predicted the general decline of religion's influence in the world: Though according to Fox's analysis religious conflict has started to drop in the early 1990s, the same trend is also seen in broader conflicto trends. As a result, the decline in conflict around the world cannot be readily attributed to a decline in the importance of religion. Interestingly however, Fox also finds that ethnic conflicts are more likely to be combined with religious motives, and that ethno-religious conflicts are slightly more intense (read: only slightly deadlier) but tend to last considerably longer than non-religious ethnic wars(5). Finally, Fox's work reveals that "by far the most common type of conflict is conflicto between two Christian groups," outnumbering conflicts between Muslim groups (2004a.) This finding is consistent with De Soysa and Nordås's work, which also attempts to show empirically that conflict between Christian groups is more common than between Muslims (De Soysa et al., 2007.) However, Fox's analysis also suggests that Islamic groups engage in more inter-religious conflict (though Christians seem to trail them by only a slight margin.) In all, these results appear to contradict or at least qualify Huntington's assertion regarding the conspicuous "bloodiness" of Islam.

For all the merits of Fox's quantitative tests, however, they also face serious shortcomings. First, Fox's use of the State Failure dataset (and until 1996,) which only covers domestic conflicts, severely limits the claims he makes, particularly since much of the recent political controversy (sparked by the Al Qaeda attacks) has focused on international inter-faith wars. Second, as he readily admits in his articles, these broad correlations do not prove causation. Absent in his work is a discussion of the theoretical mechanisms connecting religion to conflict, and as such, his findings are only inferentially suggestive. In other words, he lacks a theory of the religion and conflict nexus. Finally, and perhaps most damningly, Fox's conceptualization and measurement of religious conflict seems suspect, if not outright inadequate: His analysis conceptualizes religious conflict as that waged by groups belonging to/identifying with different religions. But does mere identification with a given religion necessarily make a conflict religious? The answer should be no, or at least not necessarily. Rather, it would seem more appropriate to conceptualize, and to count as religious, those conflicts in which religions plays either a key or relatively important role, and in which religious motives, causes or goals are professed and mobilized.

More recently, Monica Duffy Toft has also joined the debate regarding the impact of religion on war. In a number of important works, Duffy Toft has reported empirical support for the hypothesis that stresses the rise of religion's importance in conflict trends (Duffy Toft, 2006, 2007.) Armed with an original dataset, Duffy Toft finds that out of 133 civil wars that took place between 1940 and 2000, 42 were religiously-based. Additionally, in a helpful conceptual and empirical move (that addresses my earlier critique of Fox,) Duffy Toft codes wars in which religion was either a central or a peripheral component. By central Duffy Toft denotes wars in which combatants were fighting over whether a state or region should be ruled according to a specific religion. Peripheral refers to conflicts where combatants identified with a specific religious tradition and grouped themselves accordingly, but where the rule of a specific religious tradition could not be considered the object of their contention (Duffy Toft, 2006: 9.) With this in mind, she finds that out of the 42 cases, religion was central in 25 of those conflicts, and peripheral in 17.

In addition, Duffy Toft shows that there has indeed been an increase in the proportion of religious civil wars in the last half-century or so. She explains: "From the 1940s to the 1950s, the figure rose from 19 to 29 percent. The 1960s witnessed a modest decline to 21 percent, but the figure grew in the 1970s to 36 percent and continued to climb in the 1980s to 39 percent and into the 1990s to 43 percent. Religious civil wars also make up a disproportionate number of ongoing wars after 2000. There were 14 ongoing wars as of 2000. Of these, 7 (or 50 percent) were religious" (Duffy Toft, 2006: 9.) These findings appear to contradict Fox's assertion that the rise of religious conflict relative to other types of conflict was negligible.

As for the influence of particular religions on civil wars, again unlike Fox (2004a, 2004b) and De Soysa and Nordås (2007,) Duffy Toft finds that Islam was the most prevalent faith to engage in conflict: "One or both parties adhered to Islam in 81 percent of all religious civil wars (32 of 42 cases); Christianity was involved in 21 (50 percent); Hinduism in 6 cases (14 percent); and other religions in just a handful" (Duffy Toft, 2006: 12.) Her data reveals that Islam is also the most involved religion in intrafaith wars (with 9 out of 10 cases) though in interfaith conflicts, both Islam and Christianity seem equally well represented (25 cases or 78 percent for Islam, and 22 cases or 69 percent for Christianity.)

Interestingly, however, Duffy Toft suggests her data disproves Huntington's clash of civilizationsthesis, in that there has not been an increase in the rise of interreligious conflicts since the 1940s— rather, they have represented around one-fourth of total conflicts across the last six decades. Duffy Toft's research also refers to other important aspects of the religion and violent conflicto debate. She finds that religious wars outlast others by a two-year margin, and that those in which religion is "peripheral" last longer. In regards to the settlement of religious wars, unsurprisingly, she finds (like Fox) that those in which faith is "central" are twice as less likely to be resolved via a negotiated settlement, in comparison to those in which it is "peripheral." Her findings on war recurrence are similar to these, but in regards to deadliness, she finds that they are four times more intense.

To summarize, Duffy Toft finds that: "First, religious civil wars make up more than one-third of all civil wars fought from 1940 to 2000, and there is little sign that this trend will wane any time soon. Second, among the world's major religions, Islam was involved in just over 80 percent of these civil wars. Third, religious civil wars in some instances are more costly than nonreligious civil wars: they last longer, and when religion is central, they are more intractable and are deadlier to noncombatants than wars in which religion is peripheral" (Duffy Toft, 2006: 12.)

On balance, Duffy Toft's participation in the empirical debate has been helpful. Importantly, as noted, she addresses the shortcomings of Fox's work regarding both the conceptualization of religious conflict and the key differences between central and peripheral influence of religious motives and goals. Combined, though, both Fox and Duffy Toft empirically show that religión has in fact played an increasing role in the waging of war -a trend that shows no sign of decline-, and present revealing results regarding the relative endurance and deadliness of wars involving religious groups. In addition, both scholars' work confirms the prominent role of Islam vis-à-vis other faiths in both domestic and international conflict, which opens up a series of interesting puzzles and avenues for research.

The limits of this work are in part due to limits in the data that is currently available. As noted above, these studies place their focus on a relative short span of time, roughly since the mid-twentieth century. As illustrated throughout, though helpful for addressing current debates sparked by both secularization theories as well as by Huntington's argument, this body of empirical work could be fruitfully complemented with data that looks further back into history to assess the origins of the trends identified by these scholars. In addition, having established major trends and correlations that solidly connect religion to violent intrastate and interstate wars, future work should aim to discriminate between different types of conflict, ranging from smaller scale clashes (pogroms and riots, for example,) to communal violence and other expressions of violence, short of full-blown civil war. This line of work can be best pursued both quantitatively and qualitatively: On the one hand, Fox's and Duffy Toft's broad trends should be broken down with the use more fine-grained typology of violent outcomes in existing datasets, and qualitative case studies should be built (and extant ones could be harnessed) to identify the causal and constitutive mechanisms connecting religion to different types of conflict.

II. Why Is God Deadly? Theories of Religion and War

Empirical studies have proven that religion matters in the occurrence of war. But how and why does it matter? What factors influence its impact? This section reviews the major theoretical approaches to the impact of religion on violent conflict.

Major theoretical approaches to the religion-conflict nexus In a 2000 article, Andreas Hasenclever and Volker Rittberger review the tradition of scholarly work on religion and violence, slicing it up along three major approaches: primordialism, instrumentalism and constructivism (Hasenclever et al, 2000.) Below I present the main traits of each "camp," as highlighted by Hasenclever and Rittberger:

1) Primordialists, spearheaded by figures such as Huntington (though Gilles Kepel and Jeffrey Seul, among others, are also included here) argue that there are intrinsic elements within religioustraditions that propel the explosion of violence, and that such elements are so deeply ingrained that intolerance toward other religions will almost inevitably arise, ultimately leading to violent conflict. The scholarly emphasis here is usually placed on the so-called "religions of the book": Christianity, Islam and Judaism, though Huntington included a number of other "civilizations" in his list of warring cultures. Hasenclever and Rittberger explain that "according to the primordialist perspective, the reorganization of world politics will be accompanied by civil unrest and international wars. Domestically, non-Western civilizations will purge themselves from the remnants of the Cold War era. Religious militants face, and take up, the task of either ousting westernized political elites from power or converting their members into zealous believers, who recant secularism and devote themselves to building political institutions that protect and promote the religious traditions of their nations" (Hasenclever et al, 2000: 643.) This belligerent dynamic is also said to escalate internationally, thus leading to civilizational conflicts.

2) Instrumentalists, for their part, hold a less essentialist view of the relationship between religión and conflict. For them conflict between religions erupts not because of intrinsic doctrinal reasons but because religion offers a particularly viable resource for political leaders and elites to tap onto and mask a more fundamental interest: pursuing greater power and wealth. Conflicts that appear to be motivated by religious or doctrinal causes are rather fundamentally triggered by socioeconomic divergences and inequality, both within and between states. As such, they argue that the radicalization of religious groups "is more likely to occur in times of economic decay, social disintegration or state collapse" (Hasenclever et al, 2000: 645.) Domestically, socioeconomic grievances are said to spark social unrest among populations living in poverty, which leads them to find an "alternative orders" in religious communities to satisfy their need for welfare, security and recognition. Political leaders in these states recognize such unrest and, faced with an opportunity to increase their political capital, seize the opportunity to use religious arguments that effectively mobilize crowds.

Internationally, instrumentalist approaches posit that rational utility-maximizing leaders will resort to religious scripts to build alliances with co-religionist states when the economic and military incentives they offer are great. However, should other states that do not profess the same religión offer larger benefits than said co-religionists, rational leaders will have no qualms in prioritizing alliances with those states.

Put otherwise, for instrumentalists, religion is a resource to be tapped to reap political and economic gains, but is only one among many. As a result, though instrumentalists allocate religión some importance in precipitating mobilization and facilitating alliance-building by shrewd selfinterested leaders, at the end of the day religion is seen as a proxy variable for understanding the deeper motives driving violent conflict.

The study of suicide terrorism is a good example of the instrumentalist approach. In a widely influential study, Robert Pape argues that suicide terrorism follows a "strategic logic" that aims to extract short-term, usually territorial, gains from liberal democracies (Pape 2003, 2005.) Using crosscase quantitative data and case studies on a number of conflicts, including Sri Lanka, Lebanon, Israel-Palestine and the Kurdish conflict in Turkey, Pape contends that "suicide terrorism is on the rise" because terrorists have learned that "it pays." Importantly, Pape plays down the effect of religious motives, or in his words, "fanaticism" to explain suicide bombings. Though he recognizes that this conclusion, produced by scholarly work in the 1980s was "consistent with the data from that period," he also notes that later cases have confirmed that suicide terrorism is not limited to Islamic Fundamentalism (Pape, 2003: 343.)

Religious motives may be invoked, but at the core of terrorism lay the thirst for territorial and political gains. What is the strategic goal of terrorism, then? According to Pape, suicide bombers aim to coerce the opponent into making territorial concessions. As such, Pape argues, these tactics are not the result of "fanatical" or irrational fighters, but rather are carefully deployed, connected to nationalist claims, and selective in its targets. Stathis Kalyvas has made a similar argument about the massacres in Algeria (Kalyvas, 1999.)

3) Constructivism represents the third main approach to the impact of religion on conflict. For scholars writing in this theoretical vein, religion constitutes not a proxy or "mask" variable, but rather represents a powerful cognitive structure consisting "of shared understandings, expectations, and social knowledge, [which] provide[s] social actors with value-laden conceptions of the self and others, and consequently affect[s] their strategic choices" (Hasenclever et al, 2000: 648.) In addition, religion is neither intrinsically belligerent nor always readily accessible to the instrumental needs of greedy elites. Rather, religious doctrine offers a wide net of inter-subjective understandings and scripts over which interpretational battles are fought, and as such, attention should be placed on the contest of interpretational frames and strategies (and those who put them forth) for understanding both the peaceful and the violent impact of religion in politics.

Note, however, that constructivists do not disagree with instrumentalists on the fact that religion is susceptible of factoring into the interests of leaders, or on the idea that political entrepreneurship is needed to mobilize crowd support based on religious arguments. However, as Hasenclever and Rittberger note, constructivists do disagree with the instrumentalist tendency to see religion as a mask of deeper, material interests. Instead, they attempt to take doctrine more seriously and argue that it can, by virtue of the mobilization power granted by its own cognitive force, lead to violent outcomes.

Constructivists also disagree with instrumentalist approaches on the ease with which leaders (in the latter's viewpoint) galvanize societal support: for constructivists, political leaders must engage in processes of persuasive social legitimation via the interpretation and reinterpretation of religious doctrine to achieve mass mobilization, a task in which they usually face contending frames and interpretations that do not instigate violence. Scott Appleby's argument about the "ambivalence of the sacred", presented later in more detail, illustrates this line of research well: different interpretations and framing enterprises of religious doctrine may lead to war or to peace, to clash or to dialogue (Appleby, 1999.)

Seizing a Middle Ground: Eclectic Approaches Hasenclever and Rittberger's theoretical overview is extremely helpful in teasing out the differences between major social scientific approaches to the study of religion and violence. However, a closer reading of some of the most recent studies on religion and violence reveals that the above picture of clearly demarcated theoretical camps is a highly stylized one. Some of the same authors referenced above, particular Duffy Toft, Hasenclever and Rittberger, or Daniel Philpott, as well as sociological works such as those by Appleby and Marty, fuse instrumentalist and constructivist tools in their explanatory frameworks. In this subsection I aim to illustrate the "eclecticism" that characterizes the arguments of these authors, and which I argue is increasingly becoming the conventional wisdom in the subfield of religion and violent conflict.

First, a few lines are necessary on what I mean by this "eclecticism." Social science research, including in Political Science, and within it specific subfields such as International Relations and Comparative Politics, has traditionally been divided along two broad meta-theoretical positions: rationalism and constructivism. As explained in the section above, rationalists focus on political actors and their strategic interests, usually said to be geared toward the maximization of their utility function (determined by a cost/benefit calculus). Though "thin" and "thick" variants of rationalism exist, the version that has usually gained more currency among scholars, perhaps following a market bias, has been associated rationalism with materialism and the attainment of wealth and power.

The other meta-theory of social action is constructivism, which, as explained earlier in reference to Hasenclever and Rittberger, takes actors' ideas and beliefs more seriously, and uses them not only to understand how said actors' "interests," strategies and identities are constituted and are subject to change over time via actors' interaction.

Scholarly practice has usually treated these two meta-theories as radically and mutually exclusive, and indeed many academic debates (and careers) have been built on a "dueling theories" approach to social science. However, recently, more scholars have become skeptical about the impermeability of the divide between these two camps, and have introduced into the debate the idea that rationalism and constructivism may be combined regardless of whether core philosophical and epistemological questions remain unresolved. As such, James Fearon and Alexander Wendt (2002) and more recently Katzenstein and Sil (2008, 2010) have promoted pursuing a "pragmatic" and "eclectic" route of research that combines the major insights from both approaches to produce more persuasive scholarly work. As Katzenstein and Sil explain: "Analytic eclecticism is distinguished by the fact that features of analyses in theories initially embedded in separate research traditions can be separated from their respective foundations, translated meaningfully, and recombined as part of an original permutation of concepts, methods, analytics, and empirics." (Katzenstein et al., 2008: 111.) This pragmatic turn to eclecticism is now hailed as the new mantra of the International Relations subfield, and key figures are trying to promote it as a sort of academic best practice.

Such enthusiasm, naturally, does not mean automatic uptake by the broader scholarly community, and as such, the turn to eclectic theorizing remains work in progress. In what follows, however, I want to argue that the subfield of religion and conflict reviewed here (if perhaps not the larger subfield of religion and politics) seems to have made this "eclectic" turn, insofar as many of the main theories currently held by major scholars incorporate insights and tools from both rationalism and constructivism. A distillation of the arguments of the scholarly camps discussed earlier helps to flesh out my point.

First, it must be said that primordialism has been widely discarded as a valid explanation of the religion and conflict nexus. This is the case, I would argue, for two possible reasons. One the one hand, it appears to be factually unsound. As the empirical work by Fox and Duffy Toft presented earlier show,(6) no single religion accounts for all religious conflict around the world, and in particular, Christians and Muslims share an important slice of the existing cases.(7) As a result, there is nothing intrinsic about Islam (or Christianity, or any other religion) that makes it more conflict prone, which, however, does not mean there may not be other factors (which Toft dubs "structural") that may explain Islam's larger share in both inter and intrafaith conflicts around the globe. On the other hand, it may simply be politically incorrect -indeed, politically dangerous- to single out any major religion as essentially and inevitably violence-prone.

Setting aside primordialism, three key factors are identified in the existing religion-and-violence literature: the strong sense of community in religious groups; religious doctrine; and the role of leadership in violent religious mobilization. These three factors correspond roughly to the variables central to constructivism and rationalism: identity, ideas, on the one hand, and strategic interaction on the other. Yet, as seen below, mono-causal arguments using any single one of these variables are rarely persuasive. Indeed, my survey of the literature suggests an emerging scholarly consensus on the idea that separately, though necessary, neither one of these traits is expected to be a sufficient cause of religious violence. As a result, rather than a rivalry between approaches, a combinatory approach may be the best route. Taking this eclectic perspective, different scholars have opted either to present mid-level hypotheses or to build broader interpretive frames to understand the religion and conflict nexus, and explain violent outcomes. Let us explore those arguments in turn.

How do religious group membership, religious doctrine and leadership combine and lead to conflict? In an important 2007 article, Daniel Philpott identifies a series of pathways (Philpott, 2007.) Based on constructivist insights, Philpott agrees that religion offers an ideal and powerful "cognitive" set of scripts through which mass support can be mobilized around religious identity. Group membership in religious communities, which usually rest on strong ties of brother- and sisterhood, may lead to in- and out-group dynamics that characterize social life generally and derive in violent clashes with believers of different faiths or against secularists. As Philpott argues: religion "shapes the identities and loyalties of warring communities – Serbs, Northern Irish Loyalists, or Buddhist Sinhalese" (Philpott, 2007: 518.) However, Philpott suggests that the actual doctrine of specific religions plays little role in this pathway—what matters more here is the use of religious language and symbols to "inflame the bellicosity of communities whose faith had come 'folk religion'—theologically desiccated, but rich in ritual, lore and ethnicity" (Philpott, 2007: 518.) For her part, as Duffy Toft (2006) notes, most major religions share a latent "defense of the religious community" component, that is, the idea that a believer's most important goal in life is to preserve his/her faith and community from threats or attacks by outsiders, which, if activated, may have the effect of radicalizing the behavior of co-religionists. A corollary of this is that a radicalized belief in the need to defend one's religion may lead believers to discount their fear of death, which would explain phenomena such as suicide terrorism.

For Philpott then, the above combination of factors (community and leadership, and doctrine, if only superficially) represents a first pathway of religiously-activated conflict: a strong feeling of attachment to a religious identity that is tapped onto by embattled political leaders who see it as a potent resource for mobilization. This pathway provides scholars such as Duffy Toft with the grounds to propose a more concrete theory to explain cases in when religion goes from peripheral to central in a conflict, which she dubs "religious outbidding" (inspired in Jack Snyder's work on nationalist wars.) As she explains: "The logic of [my] argument is that because the intangible benefits of religiously approved conduct outweigh the tangible costs of bargaining, combatants may refuse to bargain and continue to fight, especially because in religious traditions such as Christianity, Islam and Judaism, a conflict outcome is often considered a manifestation of God's will, and violent conflict itself is therefore transmuted into a test of religious faith" (Duffy Toft, 2006: 26.) Note that this explanation combines constitutive, that is, constructivist notions, as well as rationalist ones: once a strong religious identity has been constituted it is easier for threatened leaders to mobilize believers to fight –and die- for their faith. It must be noted that for scholars such as Duffy Toft, this approach (that I call eclectic) takes the wind from claims regarding the irrationality of religious violence. As she argues, "Religious actors are actually rational, but they base their utility calculations on intangible values" (Duffy Toft, 2006: 27.) Toft uses the cases of the First and Second Sudanese Civil Wars to test her theory, but more qualitative work beyond those cases remains to be done to ascertain, via process tracing and controlled comparisons, the causal link that her theory suggests.

Philpott proposes a second pathway, which appears to take faith more seriously. In certain cases, he suggests, religious doctrine may not only help define the identities of the groups engaged in violence, but also their ulterior goals regarding political order. In what he dubs the "integrationist" nature of some religious causes, certain political theologies (and interpretations thereof) lead groups and their supporters to discriminate against other, often minority, groups.(8) This integrationist quality appears to have been shared by many conflict cases involving Islam. Duffy Toft seems to agree with Philpott in her analysis of interpretations surrounding the notion of jihad as a "structural-institutional" cause of Islam-related violence (Duffy Toft, 2007.) In addition, Toft also offers an additional historical-structural factor for the higher prevalence of Islam in recent conflicto trends, suggesting that in comparison to Christianity, Islam has not undergone a systematic process of differentiation between religion and state (Duffy Toft 2007: 21-24.) This analysis is shared by Philpott, who notes that many Muslim-majority states have adopted a statist-repressive approach to Islamic ideas and groups, initially forcing them into illegality but ultimately –in a sort of feedback loop- also prompting "revivalist" radicalized movements which aim to rule in accordance to their radical reading of Muslim doctrine. There are also cases (Iran being the foremost) in which such movements have succeed in capturing the state and pursue integrationist agendas, making them more "conflict-prone" (Philpott, 2007: 519.)

As a corollary to this second pathway, Philpott notes that in some conflicts, religious ends may combine with other strong cognitive structures such as ethnicity and nationalism. Such are the cases of the Sudan and Sri Lanka. Finally, Philpott also notes that "integrationist" political theologies sometimes also play a role in shaping the identities and goals of opposition groups. This combinatory approach, as is evident, also contains institutionalist elements to explain variance in violent and non-violent outcomes, not unlike the approach taken by Fearon and Laitin in the study of inter-ethnic (non-)cooperation (Fearon and Laitin, 1996.)

For their part, Hasenclever and Rittberger offer a similarly "eclectic" theory of religious mobilization leading to conflict. Like Duffy Toft, they focus on the choices of elites, seen as rational actors who calculate the costs and benefits of resorting to religious arguments in order to incite supporters toward violence. Importantly, elites' decisions are determined by the likelihood of mobilization success. Following a rationalist logic, they suggest that "controlling for the strength of the adversary, the prospects of success, in turn, are a function of at least two variables: (1) the mobilization of the rank and file and (2) the support that the group's goals and strategies enjoy within the broader society. As a result, "we should expect elite's choices to be affected by the degree of support they can muster for their cause and their strategies." (Hasenclever and Rittberger, 2000: 651.) What affects the likelihood that elites' mobilization strategies will succeed, and what determines public support for those goals and strategies? They point to four factors connected to religion:

- Because religious conflicts constitute struggles over deeply held values rather than economic interests or other types of social preferences, there is a greater likelihood that members will mobilize violently to defend their faith.

- Leaders making decisions to encourage violence among their followers need to consider whether their supporters will have a tendency to have self-sacrificing attitudes. Religious attachment, as mentioned earlier, has a tendency to increase believers' will to sacrifice to defend their cause.

- The fewer cooperative strategies appear available for attaining the group's goals, the greater the likelihood that elites will choose violent options. When peaceful options have been exhausted or there is a history of animosity between groups, leaders may find in religion a potent and effective catalyst for violent action.

- Leaders must weight the likelihood that their framing of the struggle in religious terms and the suggestion of violent conduct will be find echo in their rank-and-file. Leaders must then engage in persuasive framing of their cause "in terms that lend credibility to their claim that violence is unavoidable."

This model, like Toft's and Philpott's, combines a constructivist approach to the formation of interests and preferences to mobilize crowds toward conflict, with rationalist strategizing on the part of elite actors.

Finally, let me briefly discuss the work of sociologists Martin Marty and Scott Appleby. In an impressive series of volumes framed around a broader research project, Marty and Appleby, with the aid of dozens of contributors, including political scientist Gabriel Almond and historian Emmanuel Sivan, developed a collection of comparative historical case studies in order to construct an interpretive framework of fundamentalism, covering a diversity of religions and regions in the world. They define fundamentalism, essentially, as a "reactive and defensive [stance] toward the processes and consequences of secularization and modernization." (Appleby and Marty 2004: 405.) Fundamentalists are groups of believers in a particular faith who, concerned with "the erosion and displacement of true religion," react to it in a variety of ways, both violent and non-violent (Appleby and Marty 2004: 405.)

This ambitious project culminated with the 2004 volume entitled Fundamentalism Comprehended, which offers extremely rich insights regarding the diversity of fundamentalist groups and the multiplicity of factors that give rise to it in its various forms. Importantly, Almond, Sivan and Appleby suggest three set of causes for fundamentalism: (1) structural factors related to long term contextual conditions and changes; (2) contingent, chance factors; and (3) human factors of choice and leadership. Though the participants of this volume aim to arrive at a explanatory framework of fundamentalism, the extraordinary diversity and complexity of the subject (and the level of detail they obtain via a plethora of controlled case studies) lead them to produce not a parsimonious theory of fundamentalism but rather a list of factors that enable its appearance, consisting of no less than eleven structural causes, which they then couple with contingency and leadership.

Depending on the scholarly standards one embraces this is either an advantage or a hindrance: Is it better to have an astoundingly layered study containing broad typologies and strategies that cover multiple religions and regions or is it preferable to arrive at and propose concrete hypotheses or theoretical combinations that explain specific outcomes? Political scientists and sociologists of differing epistemological sensibilities will probably take diverse views here. In any case, from the description provided above it the reader can appreciate that there is a deal of overlap between the aspects considered by political scientists such as Duffy Toft, Philpott or Hasenclever and Rittberger, and sociologists as those involved in the Fundamentalism project, to the extent that they include elements from both doctrine and strategic behavior on the side of political leaders in their explanatory frameworks.

It must be noted, however, that the Fundamentalist Comprehended volume does not zoom specifically into the subject of inquiry of this article regarding the impact of religion on violence. This is not t o say that the connection does not pervade the endeavor-- quite to the contrary. For instance, the volume does offer a helpful typology of fundamentalist groups and strategies according to the group's relationship to their immediate environment: world conqueror, world transformer, world creator and world renouncer. From these, the first two appear as the most prone to engage in violence. As Almond, Sivan and Appleby state: "The primary strategy of the world conqueror is to asume control of the structures of society which have given life to the enemy" (Almond et al., 2004: 428.) In the case of the "world transformer", these authors suggest that "a second means of abolishing the enemy is to reinterpret and influence the structures, institutions, laws and practices of a society, so that opposing fundamentalism my become more difficult, and so that conditions become more favorable for the conversion or marginalization of the enemy" (Almond et al., 2004: 428.) These categories are used throughout the book and connected by the authors to specific cases of religziously-inspired violence.

In his solitary work, however, Scott Appleby does tackle the question of religion and violence more directly. Despite the fact that, again, his approach is more interpretive than explanatory, he offers ideas that coincide with the models and pathways proposed by Philpott or Duffy Toft: community, doctrine and leadership. First, his very definition of fundamentalism contains these elements: "Fundamentalism… refers to a specifiable pattern of religious militancy by which selfstyled true believers attempt to arrest the erosion of religious identity, fortify the borders of the religious community, and create viable alternatives to secular structures and processes" (Appleby, 2000: 86.) Appleby goes further, and differentiating fundamentalism from extremism, he connects them to explain violent outcomes via the interpretational powers of group leaders. He says: "The scrupulous observance of the divine law can become an ideological and operational resource for extremists… The ability of religion to inspire ecstasy stands behind the distinctive logic of religious violence. As unpredictable and illogical as this violence may seem to outsiders, it falls within a pattern of ascetism leading to the ecstasy of self-sacrifice that runs as a continuous threat through most religions" (Appleby, 2000: 91.) In his analysis, as in the ones presented earlier by political scientists, religion and religious doctrine serves to construct the identities and interests of the believers (which as in those theories can also mix with other cognitive scripts such as nationalism or ethnicity,) but the presence of a leader that frames doctrine to mobilize militant member is crucial (Appleby explores the influence of Sayyid Qutb on fundamentalist Islam and his interpretation of the notion of jihad.) He falls short of suggesting an explicit model, but in my view, his insights here, as well as collectively with Marty and others, contain the same –constructivist and rationalistelements to constitute "eclectic theorizing" in the area of studies of fundamentalism.

III. Assessing the Literature

Thus far I have evaluated positively recent work put forth by both political scientists and sociologists combining rationalist and constructivist lenses. But what are some weaknesses of this scholarship? What remains to be done?

To begin with, for all the merits of the combinatory approach, one wonders whether existing theories take religious doctrine seriously enough. Faith here plays a distinctive role because it serves as particularly powerful "glue" for group identity, one colored with certain scripts (of selfsacrifice, for example) which then may make resorting to violence more likely. However, these scripts still need to be activated and formulated by leaders that mobilize crowds of believers, usually in the face of a threat (by secularists or non-believers) or for self-interested reasons (the pursuit of power and wealth.) Critical scholars such as Roxanne Euben have raised doubts about the functionalist tone of much work on religion, particularly on Islamic fundamentalism, which plays into secularization theory by framing religious motives as inherently reactive to modernity (Euben, 1999.) Though this critique applies more forcefully to instrumentalist arguments and particularly to theories propounding for the "irrationality" of religious motives, one wonders whether the "mobilization approach" put forth by most "eclectic" accounts is also guilty of this charge and thus, as Euben argues (citing Lawrence,) fail "to take account of the autonomous nature of the religious impulse" (Euben, 1999: 24.) Can religion be understood as anything other than providing fertile ground for the activation of masses toward particular objectives (violent or non-violent)? This is a major question that I do not intend to answer here, but one that future research in political theory and comparative politics should take up.

A critique of the mobilization approach that sustains major theories of religion and conflict can be waged from a more positivist standpoint. Though religious group identity and doctrine are present in the prevailing explanatory models of religion and violent conflict, it is ultimately the interpretative power and the strategic action of leaders (be they political elites or minority leaders) that ultimately triggers the conduct of believers. As a result, though not the sole causal factor in these models, leadership appears as the prime mover. For scholars seeking to come to causally parsimonious explanations, the role of leadership as a constant (rather as a variable) in existing models of religion and conflict may be problematic. This type of criticism has also been leveled at the broader literature on social movements. As Charles Tilly notes in his foreword to a volumen on Islamic Activism:

"As I see it, the apparatus of mobilizing structures, political opportunities, repertoires, frames and contentious action can never serve as more than an orienting device for the sorting of observations that investigators will then have to explain by other means. As a horde of critics have complained without much effect, no one has identified powerful empirical regularities or compelling causal models that account or all sorts of political opportunity structures, framing processes, or sequences of political mobilization. Regularities and explanatory principles operate at the level of mechanisms and processes, not at the level of these descriptive categories." (Tilly, 2004: xi.)

One way to address this problem may be to develop studies that analyze cases that present similar conditions but lead to different outcomes, that is, to compare cases of successful and failed violent religious mobilization. Another way to proceed is to take Duffy Toft's approach by proposing hypotheses about concrete sequences or causal chains connecting religion to violence and subjecting them to careful empirical testing across cases. Qualitative methods (process-tracing, controlled comparisons, etc.) may result most fruitful here. Furthermore, in addition specifying precise causal chains, scholars of religion and conflict should look to make clearer statements about the conditions under which their causal mechanisms and models lead to violent or nonviolent outcomes. Much can be drawn from existing work: As mentioned earlier, Appleby and Marty's volume offers an extremely extensive list of structural variables, including regime type, education levels, the role of mass communication, or differences in social structures and civil society density, that seem to matter in the mobilization of fundamentalist groups. This plethora of structural variables can be mined to produce more specific hypotheses and theories to see how well they apply to specific cases or sets thereof.

In addition, as mentioned in the first section of this article, scholars should aim to distinguish between types of violent outcomes in terms of their scale, duration, "bloodiness," among others. For example, much current work focuses –unsurprisingly- on fundamentalist terrorist groups, but wide controversy remains in regards to the relevance and the specific effect of religion. Pape argues that terrorists are strategic, rational actors, and that even when it does appears to be relevant, religion is usually proxy variable masking territorial gains. However, other authors such as Fawas Gerges, Quintan Wiktorowicz and Simon Haddad question this thesis from different perspectives, which suggests that a more careful consideration of religious doctrine as constitutive is needed for a fuller understanding of suicide terrorism (Gerges, 2005; Haddad et al. 2002; Wiktorowicz, 2003.) Indeed, one wonders to what extent Pape's own argument should necessarily exclude the use of constructivist insights. One way of combining approaches is to analytically separate the "strategic" (i.e. rational) and the "constitutive" (i.e. constructivist) role of religion as they apply in mobilization dynamic involved in these models. Indeed, one could hypothesize that in some cases, the leaders that mobilize groups to pursue terrorist goals are themselves the strategic actors with clear shortterm utility-maximizing goals, but that believers themselves, the mobilized, are constituted by the cognitive power of the interpretations of religious doctrine promoted by their leaders, and as such, are subject to constructivist mechanisms. Along these lines, using survey research, Haddad and Khashan have suggested that support of terrorism in areas such as Lebanon or Palestine is strongly linked to an attachment to political Islam (Haddad et al., 2002.)

The relationship between rationalism and constructivism in the study of suicide terrorism could also work in a more fundamental way. For example, rather than separating the motives of leaders and believers, Wiktorowicz and others have argued that for some of these religious groups, including Al Qaeda, religion can act as the criterion for "utility-maximization."(9) That is, rather than material, territorial or economic goals, what extremist groups pursue is precisely the exaltation of their religious belief-based mandate, with the purpose of furthering the strength of their community vis-à-vis other competing worldviews and orders (Wiktorowicz et al., 2006.) This is a controversial approach to rationality, because some may argue that separating materialism from rationalism may prevent scholars from meaningfully differentiating between "ideational" and "material" motives, but the epistemological jury is still out, which means the debate should be carried further.

In addition, Pape's argument about the fact that suicide terrorism is not limited to Islamic fundamentalists, while relevant perhaps for general theory of terrorism, does not mean the link between religion and terrorism is not worthy of exploration in and of itself. That is, while it is obvious that terrorism is not strictly religiously-based across the board, for scholars in this subfield, understanding the cases, motives and dynamics of the cases where religion does seem to play a role, is still a goal worth pursuing.(10) Finally, connections between Pape's argument regarding the central role of nationalism raises doubts as to whether, in many of the explored cases, nationalism and religiosity, along with ethnicity, combine to produce violent outcomes, as noted for example, by Daniel Philpott.

This brings me to a final issue to which I have admittedly not devoted much space in this literatura review: What is the relationship between religion, ethnicity and nationalism as motivations or factors accounting for violent conflict? Given the prevalence of conflicts that combine these "cognitive schemas" (as seen in Fox's quantitative trends, but also in the work of Duffy Toft, Philpott and even Pape,) how fruitful is it to try to separate them analytically? Can this type of delinking be made persuasively at all? As presented by Ashtoush Varshney is his review of the subfield of ethnicity and ethnic conflict, similar approaches (primordialist, instrumentalist, constructivist and institutionalist) have been put forth by scholars in that subarea, in part because in many accounts religion has been taken as a proxy variable for ethnicity, or acting in combination with it (Varshney, 2007.) Yet, as this literature review attests, a small body of work has surfaced that has "taken religion seriously" as a potentially independent factor.

How can progress on this front be made then? In cases where ethno-religious motives combine, privileging one schema while "ignoring" or displacing the other seems like the least desirable option, and framing them as "competing" variables appears equally unpersuasive.(11) One option is to take a more inductive approach -again via the use of single or controlled comparison of case studies- to elucidate the ways in which elements from both interact or not to produce different outcomes, to identify the conditions under which one schema becomes more relevant than the other in the explosion of violent outcomes.

In sum, as has become evident in this essay, the study of religion and conflict is burgeoning in the social sciences, with scholars from different disciplines participating in the debate. My argument throughout has been that, at least within Political Science, there appears to be a theoretical convergence toward the combination of rationalist and constructivist approaches. On balance, however, it is as of yet unclear whether the existing work on the religion and conflict constitutes a carefully delineated research program.(12) However, as I have aimed to illustrate, solid bases have been established. Certain theories, particularly primordialism, seem to have been widely discarded. What is needed now is perhaps a more specific, consensual agenda to substantiate and qualify broad trends, and this seems to be the direction current research is taking. In the process, though, scholars working in this area would be well advised to stay in dialogue with their peers studying a range of related topics, from the formation of democratic political parties by Islamic groups, to the moderation of or the uptake of non-violence by fundamentalist movements across religions.


1. Ashutosh Varshney documents a similar story for the study of ethnicity and ethnic conflict in comparative politics. See Varshney, 2007.

2.Although Huntington's was perhaps the most mainstream and well-known –because of the controversy that ensued- academic work linking religion to conflict, he was not the first scholar to raise the importance of or to explore that link. See also Juergensmeyer, 1993 and Barber, 1995.

3. For broader overviews regarding the study of religion and politics, see the cite review pieces by Bellin, 2008 and Philpott, 2009.

4. See

5. The complications and implications of this overlap will be considered later in this article.

6. The comparative case studies built by sociologists also share this trait of exploring a wide variety of religions.

7. And any historically-minded scholar will make sure to remember, alongside the recent surge of Islam-driven violence, the perfidy of the Christian Crusades.

8. Philpott defines "political theology" as the set of ideas that a religious body holds about legitimate political authority (2007: 511.)

9. Duffy Toft, in the work reviewed here, takes this position.

10. Though not explicitly engaged with debates about religion and terrorism, recent work by Max Abrahms has argued, contra Pape and others, that what drives terrorism is not a "strategic" utility-maximizing logic, but rather is explained by the maximization of "solidarity ties." Though still recent and underdeveloped, this type of organizational approach to terrorism may easily be integrated with other arguments about the force of religious community and identity ties and the conditions under which they may lead to violent outcomes. Marty and Appleby, in the conclusions to their edited volume, highlight the importance of organizational factors for the actions of fundamentalist groups. This line of work merits further attention also because organizations traits and dynamics have become key explanatory variables for explaining patterns of combatant violence during civil war. See Abrahms 2006, 2008, and Weinstein, 2007 on civil war.

11.The work of Stuart Kaufman is an excellent example of an "eclectic" approach to ethnic conflict that may be connected to these reflections on religion and violence. See Kaufman, 2001.

12. Though collaboration between major proponents, such as Philpott and Duffy Toft, has started to take place. See Philpott et al, forthcoming.


1. ABRAHMS, Max. "What Terrorists Really Want: Motives and Counterterrorist Strategy,"International Security 32:4 (spring 2008) 78-105.        [ Links ]

2. ABRAHMS, "Why Terrorism Does Not Work," International Security 31:2 (fall 2006) 42-78.        [ Links ]

3. ALMOND, Gabriel, Emmanuel SIVAN and Scott APPLEBY, 2004. "Explaining Fundamentalisms"in Fundamentalism Comprehended, Scott Appleby and Martin Marty (eds.) Chicago: The University of Chicago Press.        [ Links ]

4. APPLEBY, Scott. 2000. The Ambivalence of the Sacred: Religion, Violence and Reconciliation. New York: Rowman and Littlefield.        [ Links ]

5. APPLEBY, Scott and Martin MARTY, eds. 2004. Fundamentalism Comprehended. Chicago: The University of Chicago Press.        [ Links ]

6. BARBER, Benjamin. 1995. Jihad vs. McWorld. New York: Times Books.        [ Links ]

7. BELLIN, Eva. "Faith in Politics," World Politics 60 ( January 2008) 315-47.        [ Links ]

8. De SOYSA, Indra and Ragnhild NORDAS. 2007. "Islam's Bloody Innards? Religion and Political Terror?" International Studies Quarterly 51: 927–943.M        [ Links ]

9. DUFFY TOFT, Monica. 2006. "Religion, Civil War, and International Order," Discussion Paper 2006-03 Cambridge, Mass.: Belfer Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University.        [ Links ]

10. DUFFY TOFT, Monica. 2007. "Getting Religion: The Puzzling Case of Islam and Civil War," International Security 31, 4: 97-131.        [ Links ]

11. EUBEN, Roxanne. 1999. Enemy in the Mirror. Princeton: Princeton University Press.        [ Links ]

12. FEARON, James and Alexander WENDT. 2002. "Rationalism v. Constructivism: A Skeptical View," in Handbook of International Relations, Walter Carlsnaes, Thomas Risse and Beth Simmons (ed.) New York: Sage.         [ Links ]

13. FOX, Jonathan, 2004a, "Religion and State Failure: An Examination of the Extent and Magnitude of Religious Conflict from 1950 to 1996," International Political Science Review 25: 1, 55-76.        [ Links ]

14. FOX, Jonathan, 2004b. "Are Some Religions More Conflict-Prone than Others?" Jewish Political Studies Review 16: 1-2.        [ Links ]

15. GERGES, Fawad. 2005. The Far Enemy: Why Jihad Went Global. Cambridge: Cambridge University Press.         [ Links ]

16. HADDAD, Simon and Hilal KHASHAN. 2002. "A Comparative Study of Lebanese and Palestinian Perceptions of Suicide Bombings," International Journal of Comparative Sociology, 45: 5, 337-363.        [ Links ]

17. HASENCLEVER, Andreas, and Volker RITTBERGER. 2000. "Does Religion Make a Difference? Theoretical Approaches to the Impact of Faith on Political Conflict" Millenium - Journal of International Studies 29: 3, 641-674.        [ Links ]

18. HUNTINGTON, Samuel. 1997. The Clash of Civilizations and the Remaking of World Order: New York: Touchstone.        [ Links ]

19. JUERGENSMEYER, Mark. 1993. The New Cold War? Religious Nationalism Confronts the Secular State. Berkeley: University of California Press.        [ Links ]

20. KALYVAS, Stathis. 1999. "The Logic of Massacres in Algeria," Rationality and Society 11: 3, 243- 285.        [ Links ]

21. KATZENSTEIN, Peter and Rudra SIL. 2008. "Eclectic Theorizing in the Study and Practice of International Relations," in The Oxford Handbook of International Relations, Christian Reus-Smit and Duncan Snidal (eds.) Oxford: Oxford University Press.        [ Links ]

22. KATZENSTEIN, Peter and Rudra SIL, 2010. "Analytic Eclecticism in World Politics: Reconfiguring Problems and Mechanisms across Research Traditions" Perspectives on Politics 8:2, 411-431.        [ Links ]

23. KAUFMAN, Stuart. 2001. Modern Hatreds: The Symbolic Politics of Ethnic War. Ithaca: Cornell UP.        [ Links ]

24. LAITIN, David and James FEARON, "Explaining Interethnic Cooperation," American Political Science Review 90:4, 715-735.        [ Links ]

25. PAPE, Robert. 2003. "The Strategic Logic of Suicide Terrorism," American Political Science Review 97: 3, 343-361.        [ Links ]

26. PAPE, Robert. 2005. Dying to Win: The Strategic Logic of Suicide Terrorism. New York: Random House.        [ Links ]

27. PHILPOTT, Daniel. 2007. "Explaining the Political Ambivalence of Religion," American Political Science Review 101, 3: 505-525.        [ Links ]

28. PHILPOTT, Daniel. 2009. Has the Study of Global Politics Found Religion? Annual Review of Political Science 12:183-202.        [ Links ]

29. PHILPOTT, Daniel, Timothy S. SHAH and Monica DUFFY TOFT. 2011. God's Century. New York: W.W. Norton & Company. Forthcoming.        [ Links ]

30. TILLY, Charles. 2003. "Foreword" in Islamic Activism: A Social Movement Theory Approach, Wiktorowicz, Quintan (ed.) Bloomington: University of Indiana Press.        [ Links ]

31. VARSHNEY, Ashutosh. 2007. "Ethnicity and Ethnic Conflict" in The Oxford Handbook of Comparative Politics, Carles Boix and Susan Stokes (eds.) Oxford: Oxford University Press.        [ Links ]

32. WEINSTEIN, Jeremy. 2007. Inside Rebellion: The Politics of Insurgent Violence. Cambridge: Cambridge University Press.        [ Links ]

33. WIKTOROWICZ, Quintan, ed. 2003. Islamic Activism: A Social Movement Theory Approach.        [ Links ]

34. Bloomington: University of Indiana Press.        [ Links ]

35. WIKTOROWICZ, Quintan and Karl KALTENTHALER, "The Rationality of Radical Islam," Political Science Quarterly 121: 2, 295-319.        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License