Collage of book covers

In Review

A Double-Edged Dilemma

By David Little

With the collapse of the Cold War around 1990, the distinctions between and the connections among religion, justice, and peace became newsworthy in a critical, almost daily way. Nationalist conflicts in places like Bosnia, Northern Ireland, Sudan, Sri Lanka, and Israel/Palestine brought into sharp focus the alleged destructiveness of religion, both as a disturber of the peace and as a symbol of ethnic injustice. Of course, the underlying characteristics of these conflicts were not all that different before 1990, but not many people took notice because the focus was on the longtime contest between East and West.

Moreover, what was described as the momentous outburst of religious politics, dramatized earlier in 1979 by the anti-Soviet insurgency in Afghanistan and the Iranian Revolution, was contemporaneous with the hastening decline of the Soviet empire. The emergence of religious politics appeared, in turn, to bring religion and terrorism closer together, particularly but not exclusively in the Middle East, during and after the mid-1990s. As with nationalist clashes, these occurrences were believed to pose severe threats to justice and peace.

But not all the news was bad. Along with the negative reports, it was said that religion also played a constructive role in the events of the late 1980s and early 1990s by helping facilitate peaceful transitions from authoritarian, closed societies to relatively democratic, open ones. In instances like East Germany and Poland, church-related movements, resolutely committed to nonviolence and peaceful reconstruction, were credited with assisting in liberalizing post-Communist life there and in other parts of Eastern Europe.

Earlier transitions in other locations were similarly described. The Roman Catholic Church of the Philippines was lauded for leading a nonviolent coup against Ferdinand Marcos in 1976, and the Dutch Reformed Church of South Africa was likewise hailed for radically reversing its pro-apartheid policy in the early 1980s in favor of the idea of a racially inclusive constitutional democracy. These examples were still part of the Cold War, because the rigid authoritarianism rejected in both cases was defended as a bulwark against Communism.

The South African case is important in one other way. It famously exemplified a movement, claiming at least partial religious inspiration, to find innovative, unconventional ways of dealing with what is known as “transitional justice.” “Truth and reconciliation commissions,” such as the one established in South Africa after the fall of the apartheid regime, came to be proposed as a means of supplementing the conventional administration of justice in response to violations associated with authoritarian regimes. Though not uncontroversial, the movement attracted worldwide attention, and is credited with introducing terms like “forgiveness,” “reconciliation,” and “restorative justice”—previously relegated to theological schools and places of worship—into the political arena. As with reports of religiously inspired peaceful transitions to democracy, this example and its effects are frequently cited in support of the claim that religion, quite dramatically in the period surrounding the collapse of the Cold War, is not only a malevolent influence, but can also, under some circumstances, contribute to justice and peace.

These divergent perceptions of the relations of religion, justice, and peace have stimulated a vast and still expanding literature, reflecting diverse and sometimes contentious perspectives. Before sampling that literature, though, I must address some theory and terminology, since analysts are by no means agreed or consistent either on the meaning of basic terms or on the causal connections among them.


Some contemporary thinkers claim that the very idea of religion, as currently understood, is illusory.1 They argue that dividing the world up into “the religious” and “the secular” is an invention of Western scholars and colonial bureaucrats with an axe to grind. “Secular,” meaning what is modern, rational, and peaceful, is good, and “religious,” meaning antimodern, superstitious, and violent, is bad. They suggest that by allocating praise and blame on the basis of an inaccurate description, it serves, among other things, to mask and disregard the injustice and violence attributable to secular movements like nationalism and liberalism. Thus is the “myth of religious violence” traceable to an appalling misuse of language. By abandoning misleading conceptions of “the religious” and “the secular,” these thinkers believe we will be prompted to pursue more rigorously, because more empirically, the real-world complexity and variability of the causes of violence.

Although there are some things of value in these accounts, they elide, if not confuse, two quite distinct issues. The question of whether there is any proper use for the idea of religion is very different from whether religion and violence should be interdefined. While I agree that defining religion as inherently violent is unsupportable, I disagree that no good use can be found for the concept. In fact, once we better understand what the idea of religion is good for, the more we can appreciate why the pejorative reading is so misguided. We can also better appreciate, I believe, why opposing “the religious” to “the secular” or to the “liberal nation-state,” is similarly amiss. The correct conclusion is the rather unsurprising one that religion, properly identified and examined, may or may not cause violence; it all depends on the circumstances.

It is hard to do without the idea of religion because, once freed of ideological bias, there is important work for it to do related to the imperative to justify force. Force or violence,2 understood as the infliction of death, impairment, suffering, injury, or confinement, requires an extremely strong justification, wherever it occurs, for two principal reasons: the obvious adverse consequences that result from using it, and the strong temptation in human affairs to use it arbitrarily. What I propose to continue to call “religion” turns out to play a decisive role in distinguishing between justified and unjustified force or violence. I mean by “religion” a set of beliefs and practices organized around the idea of a sacred authority that is frequently associated, in a sustained and solemn way, with authorizing the regulation of force. “Sacred authority” refers to the belief in an ultimate source, taken to be “above and beyond” human determination, of comprehensive and sacrosanct belief and practice.3 Beliefs about the affinity between religion and the regulation of force turn out, again and again, to be highly ambivalent.4

It is impossible to read “religious” texts, like the Hebrew scriptures, the New Testament, the Qur’an, the Bhagavad Gita and other works in the Hindu corpus, and Buddhist literature, including uncanonical writings like the Vamsa literature of Sri Lanka, without observing the central focus on the dispositions, standards, and purposes according to which force is supposed to be administered. At the heart of all these traditions, in their different ways, is a preoccupation with justice and peace. Force is never an end in itself, never self-justifying. It is always at best instrumental toward something else. Consequently, to restrain force in accord with authoritative teaching is to specify certain rules of justice, and it is to do that with the objective of achieving what is at once the ideal form of restraint and the state of ultimate fulfillment, namely, peace.

If I am right, we can begin to understand why force is typically treated with deep ambivalence. On the one hand, it is so unavoidable an aspect of human experience that finding permissible ways of putting it to use becomes urgently important. Consequently, the office of political ruler, directly responsible for administering force, is regularly, if variously, afforded “religious legitimacy.” On the other hand, the inherently destructive character of force represents a continuing and irreducible threat to the objective of ultimate restraint or “lasting peace.” For that reason, there exists in most religions, whether Judaism, Christianity, Islam, Buddhism, or Hinduism, a distinction between “religious” and “political” authorities. However entangled the relationship, or diverse the expressions of it, there is consistent reluctance to collapse “spirit” and “sword.”

Turning, then, to the question of causation, scholars in my field have begun to look empirically at the relations among religion, justice, and peace. When we do, what appears is a complicated and not altogether satisfactory picture. Two areas that have attracted special attention in regard to the empirical study of religion and force are nationalist conflicts and terrorism.

As to nationalist conflicts, the results of several influential studies are in fact rather disappointing. Paul Collier, the World Bank economist, has famously argued, in his essay in Turbulent Peace: The Challenges of Managing International Conflict, that national civil wars are basically the product of “greed, not grievance,” presumably including ethnic and religious grievance. It is economic conditions, such as low national income connected with slow economic growth and accelerated population increase, together with the availability of unscrupulous entrepreneurs, that make countries susceptible to nationalist conflicts, he says: “Where rebellions happen to be financially viable, wars will occur.”5 Charges of cultural, economic, or political mistreatment by one or another party to a conflict are of no causal significance for the onset of violent conflict, except, Collier concedes, in multiethnic societies where one ethnic or ethnoreligious group constitutes between 45 and 90 percent of the population—”enough to give it control but not enough to make discrimination against a minority pointless.”

The problem with Collier’s argument, and with similar arguments by others,6 is that they do not prove what they set out to prove. Collier himself admits what turns out to be a huge and very consequential exception to his argument. For example, all of the ethnoreligious majorities in four cases reviewed in the recent Religion and Nationalism in Iraq: A Comparative Perspective—the Sinhala Buddhists in Sri Lanka, the Arabized-Islamized Northerners in Sudan, the Serbs in the former Yugoslavia, and the Shi’ite Arabs in Iraq—constitute between 45 and 90 percent of the relevant populations, and thus have reason to institute discriminatory policies against minority populations that, on Collier’s own terms, would significantly affect the causes of violent conflict. In other words, Collier’s thesis, by his own admission, has no explanatory power in regard to four very notable current nationalist conflicts!

Collier also admits that during the course of even unexceptional nationalist civil wars, grievances (presumably including those that are ethnic or religious), and not just greed, turn out to be potent influences on the direction and outcome of the war. He emphasizes the importance of a “sense of injustice,” and a perception of discrimination in legitimating an insurgency, and in rallying recruits to its cause, and he explicitly mentions that grievances, concerning things like equitable government provision for education and health, must be taken into account in any sustainable peace settlement.7 If that is so, then it is not simply economic factors that must be addressed in resolving national conflicts, as his thesis about the priority of greed over grievance would seem to imply. Even in regard to the conflicts to which Collier believes his thesis does apply, ethnic and religious grievances, undoubtedly along with other concerns, eventually become very salient when considered in relation to the overall course of a national conflict.

In Peoples Versus States: Minorities at Risk in the New Century, another scholar, Ted Robert Gurr, while assigning substantially more significance to grievance as a cause of national conflict than does Collier, nevertheless plays down the causal importance of religion in contrast to the ethnic factor. Yet some of his own data and arguments do not support this conclusion. In Peoples Versus States he writes, “The only conclusion drawn here [about the effects of group type on political action] is that religiously defined communal groups are not a relevant factor in a risk model of ethno-rebellion.”8 But 55 pages later, Gurr turns around and draws some large, ominous conclusions about the global risks of what he calls “ethno-rebellion” in a revealing comment about intolerant Islamist political leaders: “Probably the greatest threats to [peaceful ethno-national relations] come from predatory, hegemonic elites who use the state as an instrument to protect and promote the interests of their own people at the expense of others. These and other sources of communal warfare and repression remain in many corners of the world and will continue to cast up challenges to those who would contain ethnic violence.”9 As the reference to the Islamist leaders suggests, religious identity appears, on Gurr’s own accounting, to be relevant to the risks of nationalist conflict in regard both to minorities and to state structures.

If we were to take the conclusions of empirical studies like these at face value, we might be inclined to agree that the causes of violence in nationalist conflicts would be traceable either to “secular” economic or political factors, or to grievances that are minimally related to religion. The problem is that such studies, on close inspection, do not appear to prove any such thing. The evidence they provide points to a more complicated state of affairs, namely, that religion is part of the picture, at least as a proximate or related cause. In particular, grievances, including religious grievances, are by no means excluded as a possible reference point for legitimating, and rallying recruits for, one side or the other in a nationalist civil war.

Another recent empirical study of the role of religion in nationalist conflicts is more satisfactory, and therefore deserves sustained attention. In “Getting Religion? The Puzzling Case of Islam and Civil War,” Monica Toft explores why 34, or 80 percent, of 42 “religious civil wars” that occurred from 1940 to 2000 involved at least some belligerents on either the government or rebel side who identified with Islam. In only 21, or 50 percent, belligerents identified with Christianity, and in 7, or 16 percent, they identified with Hinduism. On Toft’s account, religion is considered central if the key issue is whether the state or a region of the state is to be ruled according to a specific religious tradition, and peripheral if belligerents identify with a specific tradition but do not fight over religious control of the state. Toft considers conflicts such as the one in the former Yugoslavia to be peripheral since the key issues of contention are, in her words, “national rather than religious.”10

Toft proposes a theory of “religious outbidding” to explain in general why religion might become central to a conflict, as it does in 25 of the 42 cases, thereby producing a set of notable results, and in particular why the belligerents in the large majority of those 25 cases identify with Islam. Conditions for religious outbidding obtain where either government or rebel elites feel threatened, the material and ideological resources needed to reduce the threat are obtainable by couching the conflict in religious terms, the society in question exhibits religious cleavages, and the government controls information. The conditions constitute a setting in which elites make “legitimacy bids” by competing to establish their religious credentials so as to attract the material and ideological support necessary to overcome the existing threat and secure their authority. The theory also predicts that the greater the incidence of religious outbidding, the more likely it is that religion will become a central rather than a peripheral issue, that such civil wars will be more destructive than other kinds, that they will yield higher numbers of combatant and noncombatant deaths, and that they will be less resolvable by negotiation.

Toft advances a few reasons why a central role for religion changes the terms of a conflict as well as lowers its chances for being settled. She says that “religion and violence are generally associated” with religions like Judaism, Christianity, and Islam (and to a lesser extent Hinduism and Buddhism) because they share two features that increase “the likelihood that conflict between competing groups may escalate into violence.” One is a tendency to be uncompromising, and therefore to fight on, whatever the cost, when required to violate basic tenets of the faith; and the other is a readiness to discount physical as compared to supernatural survival, and thereby to accept self-sacrifice or martyrdom in the name of spiritual objectives. These characteristics make religious belligerents less susceptible to bargaining and deterrence of a “secular” sort, namely, to threats or offers based on physical and material benefits.11

For Toft, the fact that, at present, religious outbidding works better for Islamic elites than others, and is more frequently practiced by them, explains why the pertinent conditions of her theory—threatened elites, ability to attract local and outside support by religious appeal, and the existence of religious cleavages, as well as the predicted outcomes (resistance to negotiated settlement, increased destructiveness, greater lethality for combatants and noncombatants)—are mostly all empirically confirmed with respect to civil wars in which Islam is involved. She spends time showing why she thinks the case of Sudan is especially illustrative of the usefulness of her theory.

Toft mentions historical and doctrinal reasons why Islam is so important in her study. Unlike the Christian West, Islamic countries do not have a history of the secularization of political authority, including the separation of religion and state, first established by the Treaty of Westphalia of 1648. Western political institutions were transformed principally by “the extreme barbarity of the [Thirty Years] war and philosophical reaction to it—for example, the pervasiveness of the theme of ‘reason over faith’ during the Enlightenment.” That reaction “persuaded most survivors that a secular form of government and some sort of power-sharing arrangement should supersede the power of the formerly unfettered prince (whether an agent of religious authority or not). Worship would become a private matter distinct from the practice of ruling.”12 Toft’s idea of secularization, featuring as it does a notion of the imposed confinement and reduction of religious influence, opens the door to secular forms of conflict resolution, including bargaining and deterrence based exclusively on “this-worldly” appeals.

Moreover, she cites the effects of the colonial exploitation of several Islamic countries by the West, especially because of oil. Such activity, particularly when combined with the creation of the state of Israel in 1948, provoked both the rise of ideas of Arab and Islamic unity, and of Arab nationalism and Islamic fundamentalism. Also added to this combustible mixture, Toft suggests, is the Islamic doctrine of jihad, and what she calls an “obligation-to-defend-by-force component.” Presumably, the doctrine might be interpreted differently in different contexts, but when it is deployed against the troubled background of Western-Islamic relations, a violence-prone application is understandable.

There is much of interest in this account. Toft makes a strong case for the variable relationship of religion and the use of force, and for the negative effects religion may have on the conduct and resolution of conflict under certain conditions. She achieves that by means of a bold theory, applied with considerable care, as to why, when, where, and how religion tends to cause violence. She also gives some plausible reasons why, under present circumstances, Islam may be predominantly associated with religious civil wars.

There is, however, a major shortcoming. Toft’s theory, which draws on the work of the political scientist Jack Snyder, a prominent scholar of “democratization and nationalist conflict,”13 is virtually silent on one of the key features of Snyder’s theory. That is the institutional setting in which nationalistic outbidding takes place. Snyder’s work focuses on the conditions under which societies in transition from authoritarianism to democracy restrain or encourage resort to force or violence. Violence is restrained, according to him, by means of “thick versions” of liberal or constitutional democracy, which make up what Snyder calls, “civic nationalism.” It consists of an ample set of preconditions for “a stable, productive, peaceful society.” These are “a certain degree of wealth, the development of a knowledgeable citizenry, the support of powerful elites, and the establishment of a whole panoply of institutions to insure the rule of law and civic rights.” Snyder concludes, “The findings of this book suggest that only thickly embedded liberal polities are well insulated from the risk of developing belligerent, reckless forms of nationalism in the course of democratization.”14

Conversely, Snyder notes, “democratization in weakly institutionalized settings often plays into the hands of nationalist demagogues and swaggering populists.”15 Of special worry in this regard is what Snyder calls, “ethnic nationalism.” Here, appeals are made to “common culture, language, religion, shared [history], and/or the myth of shared kinship,” and they are typically employed “to include or exclude members from the national group.” It is in these circumstances that exclusionary outbidding, potentially leading to violent conflict, is most likely to succeed, since the institutions for restraining and diffusing the impact of such appeals are weak or absent.

Toft should have paid more attention to the institutional setting in which outbidding succeeds, and even developed it as one of the conditions of her theory. The fact, according to Snyder, that outbidding leads to violence in relation to ethnic nationalism but not civic nationalism is surely of great empirical significance. But there is a still more important implication for our purposes: religion should be reexamined very carefully in relation to both types of nationalism, so as better to understand the reasons religions give for resorting or not resorting to the use of force.


Toft’s account of the role of religion in the Western tradition is one-sided in regard to civil nationalism. The origins of the “whole panoply of institutions to insure the rule of law and civic rights,” including equal freedom of conscience, protection of the rights of property and political participation, a sharpened difference between religious and political authority, not to mention “secular” rules for the national and international regulation of force, and, in some circles, a strong emphasis on nonviolence, go back well before the Enlightenment to Western Christianity of the medieval period and the Protestant Reformation.

Thanks to recent scholarship,16 it is possible to detect the beginnings of a particular kind of politically oriented popular consciousness in medieval Catholicism, and especially the Conciliar movement, which at its height in the fourteenth and fifteenth centuries challenged papal authority as unrepresentative, as well as in seventeenth-century liberal Calvinism found in England and Colonial America. Both movements consisted of a new sense of popular awareness and empowerment, a new concern for political participation, and a new emphasis on national identity.

At the heart of this development was the concept of “natural rights,” which finds its first expression as early as the twelfth century. The belief that, regardless of religious, ethnic, socioeconomic, political, or gender identity, individuals as such (or “by nature”) possess a legitimately enforceable moral title empowering them to condemn, and, if possible, to resist, arbitrary coercion in regard to the exercise of conscience, political participation, control of property, and economic activity would play a huge role in the rise of civic nationalism. This belief engendered an invigorated sense of popular authority according to which social institutions, particularly in regard to church and government, ought to be reorganized and administered.

Of special importance, this new vision expressed itself in “national” arrangements, which obligated newly constituted territorial states to be responsive to the “people” or “nation.” The “civic” form of nationalism, as described by Snyder, first arose in seventeenth-century England and Colonial America where ardent Puritans like John Milton, the Levellers, and Roger Williams designed an influential social and political ideal—certainly “secular” in character, so long as the term is carefully understood—that was based on the equal freedom of all citizens, the right of political participation and the freedom of conscience, the separation of church and state, the division of executive and legislative powers, due process protections, and regulations against arbitrary government interference in trade and business and in the confiscation of property. The image was to have a decisive impact upon John Locke and on the influential vision of civic nationalism that is associated with his name.17 At the heart of this image is the idea that the proper jurisdiction of government extends primarily to material and temporal interests.

The salient point of this history is that civic nationalism, which provides standards of national and international justice that, empirically, mitigate or restrain the use of force in important ways, together with a doctrine of nonviolence, is partially the product of religious influence. Historically, religious people can and have promoted peace by sometimes agreeing out of conviction either to give up altogether or to severely restrict the use of force in advancing their mission. There is, incidentally, important evidence that these tendencies are not exclusive to Western Christianity. The recent volume Peacemakers in Action, which I edited for Cambridge University Press, recounts the stories of 16 Jewish, Christian, and Muslim peace activists from around the world and supplies dramatic contemporary examples of the same devotion to national and international standards of civic justice, including a strong preference for nonviolence and a distaste for religion that is coerced or imposed.


As to ethnic nationalism, Toft makes a mistake, it seems to me, in too readily concluding that the conflict in the former Yugoslavia is “religiously peripheral” since the central issues under contention are, in her words, “national rather than religious.” Her claim ignores the powerful way in which “the national” or “ethnic” are, under some circumstances, deeply intertwined with “the religious,” and in which religious sensibilities, symbols, and images infuse the understanding of “nation” or “people.” By raising this subject, we confront, of course, the other, darker, side of religion’s political influence.

Another recent volume that I had a hand in editing, Religion and Nationalism in Iraq: A Comparative Perspective, helps, I believe, to clarify these matters. The book compares the current situation in Iraq to three other cases of “ethnoreligious nationalism,” Bosnia, Sri Lanka, and Sudan, and concludes that in all four cases religion is, unfortunately, more than peripheral in regard to the causes of violence. That does not mean that these conflicts concern only religion, or that the outbidding of government or insurgent leaders does not at times invoke “secular” political and economic interests and “temporal” or physical security. Whatever else it is, nationalism is inescapably about this-worldly matters, about gaining political and economic control over a given territory. At the same time, nationalism is also about the process and ideals according to which a “people” or “nation” gains and administers that control, and that is where religion, or “ethnoreligion,” often comes in. As Religion and Nationalism in Iraq makes clear, religion frequently has deep ties to ethnicity, and ethnicity, in turn, affects the understanding of “peoplehood” or “nationhood” that underlies nationalist campaigns. In a word, nations are ethnic groups in pursuit of political sovereignty.

The idea of ethnicity can, it is true, mean different things. It may designate membership in a group determined by inherited characteristics (genealogical lineage, language, manners and mores, etc.) that are not necessarily religious. At the same time, ethnicity may also and often does assume distinctly religious connotations. In that case, groups are distinguished and evaluated depending on their adherence to a preferred religion. In combination with religion, then, genealogy, language, inherited manners and mores, and other ethnic characteristics take on a sacred coloring reflected in recurring terms like “chosen people.”

It is the tendency of many ethnic groups in various cultural contexts to authenticate themselves religiously that calls attention to the potentially “ethnoreligious” character of nationalism. That tendency also explains why, in particular cases, it is artificial to try to distinguish too sharply between religious and nonreligious ethnic attributes. Though ethnic groups do not necessarily need a religious reason to promote and protect their identity in pursuit of political sovereignty, a claim of religious legitimacy is likely to strengthen and intensify such a campaign.

The close connection between what Snyder calls “ethnic nationalism” and resort to force depends on the role ethnicity plays in a nationalist campaign, and how it is understood. In fledgling or transitional democracies where “thick versions” of liberal or constitutional democracy constitutive of civic nationalism are lacking or enfeebled, ethnic loyalty readily gravitates toward “ethnocractic” government, or a system of ethnic favoritism and exclusion. As Snyder argues, it is that sort of political arrangement which, in turn, heightens the prospects for violent conflict. If a strong dose of religious support for such an arrangement is added, as it is in the cases of Bosnia, Sri Lanka, Sudan, and Iraq, the potential for resort to force is all the greater, and precisely for the reasons Monica Toft gives in her article. Incidentally, one virtue of stressing the importance of religion in settings like these is to prompt attention to the reasons religious leaders give for justifying ethnocratic government, the conceptions of justice and peace they hold in supporting one group over others, which Religion and Nationalism in Iraq explores.

It is, in fact, in religiously colored ethnocratic settings like these where ethnoreligious outbidding works so well. Three of the four cases illustrate this point perfectly, as they are all “transitional” societies in the process of moving away from authoritarianism or colonial domination and trying to become independent democracies; Sudan is in some ways an exception.

In the former Yugoslavia, the point explains the amazing effectiveness of Slobodan Miloševíc’s shameless appeals to the tradition of “Christoslavism,” a set of beliefs built around the death of a Serb prince who was killed in a battle against the Ottoman Turks in 1389 CE, and was thereafter imagined as a Christ figure taken to be the rallying cry, some 600 years later, of the “Serb nation” and its right to territorial expansion. It also explains the continuing efficacy of appeals by political and religious leaders in Sri Lanka, since the time of independence and before, to Sinhala Buddhist themes in support of special privileges for the Sinhala majority. Again, it explains the enormous saliency of references to Shi’a and Sunni Islam in post-Saddam Iraqi politics. As in Bosnia, religious affiliation in Iraq has become a key marker of national identity, at least among the Arabs. While in Sudan the religious bids of the Khartoum regime have no doubt had an important effect on rallying support for the Islamic campaign in that country—especially since President Gaafar al-Nimeiry “got religion” in 1983—the government, after the military coup of 1989, has remained strictly authoritarian, and therefore does not qualify as “transitional,” in Snyder’s sense.

It should be added, of course, that the ambivalence of religion is still at work even in cases like this. While the predominant influence is in favor of ethnic nationalism, there always remain dissenting religious voices which seek in various ways to mitigate policies of exclusiveness and discrimination so evident in these cases.


There are good reasons to retain the notion of religion, and one of them is the close association of many religions, conventionally so called—like Judaism, Christianity, Islam, Hinduism, Buddhism, and Confucianism—to the task of justifying force. When that becomes the focus, it is clear why religion is not always violent. Religions like these are characteristically, if variously, ambivalent toward force. They typically accept the need for force, though they seek to regulate it according to standards of justice, that are, in turn, aimed ideally at the perfect form of just restraint, or true peace. Certain segments of some of these religions go even farther and renounce altogether the human use of force in the name of peace. On the other hand, religious support also appears frequently in defense of ethnic favoritism.

The ambivalence of religion toward the use of force is my major conclusion of reflections on the efforts by social scientists to study these matters empirically. Many accounts dismiss prematurely the role of religion, as in the work of Paul Collier and Ted Robert Gurr, or they oversimplify its role, inclining to a misleading distinction between “the religious” and “the secular,” as though those categories were unalterably opposed to each other in the achievement of justice and peace. Monica Toft appears to hold that religious advocacy, particularly when politically resonant, leads to violence, whereas a secular frame of mind, understood as a reaction against religion, fosters peace.

Dilating on the work of Jack Snyder, I have asserted that the picture is more complicated. Accepting Snyder’s empirical conclusions regarding the correlation of civic nationalism with peace, and of ethnic nationalism with violence, I think it is reasonable to claim that religion can “go both ways.” There is strong historical and contemporary evidence that certain segments of Western Christianity, not to mention like-minded thinking in other traditions, have favored and encouraged “secular” national and international patterns of justice and peace associated with civic nationalism. Conversely, there is also much historical and contemporary evidence that other segments of Western Christianity, as well as of traditions like Eastern Christianity, Islam, and Buddhism, have continued to give aid and comfort to ethnic nationalism, with all of the violent consequences that are so apparent in places like Bosnia, Iraq, Sri Lanka, and Sudan.

Thus, both nationalism and liberalism are supported by and frequently entangled with religion, and nationalism is strongly correlated with violence to the extent religion appears on the scene promoting policies of discrimination and exclusion. Moreover, there is a serious question of just how consonant U.S. Iraq policy is to the standards of national and international justice identified with civic nationalism. Might not a plausible case be made that U.S. policy has substantially defied existing constitutional guarantees and international legal obligations that are to an important degree rooted in seventeenth-century religious advocacy, and that continue up to the present to be undergirded, along with other opinions of course, by fervent religious commitment?

Might it not also be argued that a political context that is dominated by a fear of looming terrorist threats to national security, and that is therefore disposed to sacrifice national and international standards of justice, is a most congenial environment in which religious outbidding, such as the president and his supporters have engaged in with some success, is able to rally support for a policy that appears to encourage not civic, but ethnic nationalism?

In any case, we can be grateful to all the authors mentioned here for provoking us to face the deep complexities of the relations among religion, justice, and peace.


From Voting to Violence: Democratization and Nationalist Conflict, by Jack Snyder. W. W. Norton & Company, 320 pages, $25.

Peoples Versus States: Minorities at Risk in the New Century, by Ted Robert Gurr. United States Institute of Peace Press, 448 pages, $30.

Turbulent Peace: The Challenges of Managing International Conflict, edited by Chester A. Crocker, Fen Osler Hampson, and Pamela R. Aall. United States Institute of Peace Press, 936 pages, $31.50.

“Getting Religion? The Puzzling Case of Islam and Civil War,” by Monica Duffy Toft, in International Security 31, no. 4 (Spring 2007).

Religion and Nationalism in Iraq: A Comparative Perspective, edited by David Little and Donald K. Swearer. HDS/CSWR, 213 pages, $22.50.


  1. The article “Does Religion Cause Violence?” by William Cavanaugh, published in the Spring/Summer 2007 Harvard Divinity Bulletin, represents this perspective.
  2. The term “violence” itself is fraught with ambiguity. It is sometimes defined as “illegitimate force,” or a form of force that is illegal and/or unjust—as something that is, so to speak, “out of control.” On the other hand, violence is sometimes understood as synonymous with force. I use the term in the second sense, as synonymous with force. That means violence, like force, may be regarded as legitimate or illegitimate.
  3. I am sympathetic to William P. Alston’s flexible definition of religion, according to which various defining characteristics are understood to apply in different combinations to different religions. See his Philosophy of Language (Prentice-Hall, 1964), 8.
  4. This point underscores the importance of Scott Appleby’s framing in his well-known book, The Ambivalence of the Sacred: Religion, Violence, and Reconciliation (Rowman & Littlefield, 2000). Those religions that are conventionally so designated—Judaism, Christianity, Islam, Hinduism, Buddhism, and Confucianism— do exhibit such ambivalence.
  5. Paul Collier, “Economic Causes of Civil Conflict and Their Implications for Policy,” in Turbulent Peace: The Challenges of Managing International Conflict, ed. Chester A. Crocker, Fen Osler Hampson, and Pamela R. Aall (United States Institute of Peace Press, 2001), 154.
  6. See, for example, James D. Fearon and David D. Laitin, “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97, no. 1 (February 2003): 75–90, which emphasizes political over economic deficiencies as causes of national civil wars.
  7. Collier, “Economic Causes of Civil Conflict,” 153.
  8. Peoples Versus States: Minorities at Risk in the New Century (United States Institute of Peace Press, 2000), 232. See also Ted Robert Gurr et al., Minorities at Risk: A Global View of Ethnopolitical Conflict (United States Institute of Peace Press, 1993).
  9. Gurr, Peoples Versus States, 287.
  10. Monica Duffy Toft, “Getting Religion? The Puzzling Case of Islam and Civil War,” International Security 31, no. 4 (Spring 2007): 97, note 1.
  11. Ibid., 99, 100, 101.
  12. Ibid., 108.
  13. Jack Snyder, From Voting to Violence: Democratization and Nationalist Conflict (W. W. Norton, 2000). Toft refers to Snyder’s book on p. 102, note 23. See also Edward D. Mansfield and Jack Snyder, Electing to Fight: Why Emerging Democracies Go to War (MIT Press, 2005).
  14. Ibid., 316–317, 318.
  15. Ibid., 319.
  16. I summarize and draw upon notable recent work on the rise of nationalism in my unfinished volume, “Protestantism and the Problems of World Order.”
  17. See Anthony Marx, Faith in Nation: Exclusionary Origins of Nationalism (Oxford University Press, 2003).

David Little is Professor of the Practice in Religion, Ethnicity, and International Conflict at HDS and Faculty Associate at the Weatherhead Center for International Affairs.

Please follow our Commentary Guidelines when engaging in discussion on this site.