Review Article

Internal War
Causes and Cures

Steven R. David

Roy Licklider, ed. Stopping the Killing: How Civil Wars End. New York: New York University Press, 1993, 354 pp.

Internal war is rapidly becoming a central concern for students of international relations, and with good reason. War has always been at the center of international relations, and internal war has become by far the most common form of armed conflict. Wars within countries are not a new phenomenon, but the intense interest in them is unprecedented. Also new are the approaches used to make internal war understandable. Rather than focusing on specific countries, much of the recent work draws upon insights from the field of international relations to develop generalizations for explaining domestic conflict. Those interested in the future course of world politics must address some critical questions, in particular, why there is an upsurge in interest in internal war at this time and whether using new tools we can expand our understanding of its causes and cures.

For much of the cold war internal war received relatively little scholarly attention, ¹ although there was a brief spate of interest in the 1960s that helped define the field. Scholars such as George Modelski, Harry Eckstein, and Samuel Huntington noted the importance of internal war in shaping the policies of the great powers (as seen in the Russian and Chinese Revolutions), the tendency of internal wars to drag in other states (Greece, the Congo), and the growing proportion of armed conflicts occurring within countries rather than between them. ² For the [End Page 552] most part, however, this early work was not developed further. Instead, with the state seen as the primary actor in international affairs and the Soviet threat looming, internal wars were consigned to those interested in country studies or counterinsurgency tactics.

The appearance of the books under review and a flood of other books and articles on the subject of internal war raises the question of why the sudden surge of interest at this time. Contrary to conventional wisdom, the number of internal wars has not increased markedly since the end of the cold war. Rather, the collapse of the Soviet Union has enhanced only slightly a trend toward more domestic conflicts that has been in evidence since the 1960s. ³ The origins of internal war have also not undergone any major transformation since the collapse of the Soviet Union. Most internal wars, whether dating from before or after 1989, stem from power struggles in which ethnic and religious groups seek to wrest control of the state from regimes they believe are oppressing them. Finally, the intensity of internal wars has not changed over the past decades. There is no evidence of an increase in the casualties or material devastation from internal conflicts since the end of the cold war. ⁴

The renewed attention to internal war can be explained by the following three reasons. First, the relative importance of internal war as a form of armed conflict has risen noticeably in recent years. With the end of the cold war, major armed conflicts have been almost exclusively domestic. While internal wars have made up over 80 percent of the wars and casualties since the end of World War II, ⁵ this preponderance has become even more striking since the end of the cold war. From 1989 to 1996 there have been ninety-six armed conflicts, of which only five have been between states. In 1993 and 1994 there were no interstate conflicts, and in 1995 there was only one (a minor skirmish between [End Page 553] Ecuador and Peru). ⁶ Those interested in contemporary warfare are left little choice but to focus on internal war. Moreover, unlike interstate wars, no one suggests that internal wars are becoming obsolete, cannot occur in democracies, or will diminish markedly in the post-cold war era. Instead, there is every expectation that internal conflicts will continue to erupt and fester throughout the world.

Second, interest in internal wars has increased because of where they are taking place. The collapse of the Soviet empire placed internal wars in full view of the developed world. Since the end of the cold war Europe has experienced greater growth than any other region in the number of internal conflicts. ⁷ The emergence of weak states in the former Soviet republics (such as Azerbeijan, Armenia, and Georgia), conflict in Russia itself (Chechnya), and most notably, the advent of civil war in what was Yugoslavia all occurred on the doorstep of the West. When internal war remained largely an affliction of weak, sub-Saharan African states, interest was not that great. Now that European states have showed that they too could fall victim to internal war (a victimization well documented by the media), those who saw internal war as a Third World phenomenon to be ignored suddenly found their interest in civil conflict piqued.

A third set of reasons for the revitalization of interest in internal war is that with the end of the cold war many scholars who had focused on the Soviet threat were forced to find a new interest. Thus, issues that used to concern students of security affairs such as the NATO-Warsaw Pact balance and the prospect of superpower nuclear war have essentially disappeared. And the problems that have taken their place--ethnic conflict, genocide, environmental degradation, human rights abuses--are often driven by internal war. These scholars have brought new ideas to explaining internal war, which, in turn, has drawn new attention to the field.

Given the importance of internal war, it is critical to determine its causes. It turns out, though, that they point in all directions. At times, internal war is the result of individual calculation; at other times group interests are the determining factor. Some internal wars are rational and purposeful, while others are emotional and nihilistic. Large-scale violence against the government comes about because it is seen as weak or because people no longer view its exercise of power as legitimate. Weak governments promote internal war by tempting would-be insurgents to challenge [End Page 554] the regime's authority. Strong governments promote internal war by encouraging the belief that the time is right to eliminate would-be insurgents. Stable international environments enable regimes to target threatening groups free of fear that outsiders will come to the assistance of those groups. Hostile international environments spur internal conflict as outside states back rebel groups in enemy countries. Adding to the confusion, the causes of internal war are so general that it is unclear how much they really explain. Any internal war can be made understandable after the fact by asserting that people were frustrated, or lost confidence in government, or were carried away by emotion. But these feelings or beliefs are always present and therefore may tell us little about why they sometimes produce internal war and sometimes do not. ⁸

The proposed cures for internal war are similarly contradictory. Outside involvement is seen as necessary to halt wars that belligerents cannot stop on their own, but it is also seen as making things worse by relieving the combatants of responsibility for settling their own disputes. Some proclaim that regional organizations are the key to ending civil strife, while others declare them incapable of producing peace. Cease-fires promote peace by enabling the belligerents to negotiate in an environment free from violence, but they can also prolong war by easing the pain endured by the fighters.

That there are so many possible causes and cures for internal war and that many of them contradict one another should not be surprising. For all its complexity, international war involves like units (states) operating under a given framework (anarchy). By contrast, the main actors in internal war range from government-directed modern armies to roving bands of youths. The contexts vary from ones of tight central control to situations of total absence of authority. Internal wars include such diverse conflicts as revolutions, insurrections, insurgencies, terrorist campaigns, and mass slaughters.

That there are differences between internal wars does not, however, preclude the possibility of being able to offer some generalizations about their origins and remedies. Even if there is not some single factor or theory to explain the origins and remedies of all internal conflicts (and, after all, there is no such theory to explain the far simpler problem of war between states), there may well be causes and cures common to many if not most of these conflicts.

The books under review, both edited volumes that apply new approaches [End Page 555] to the study of internal war, seek to provide some generalizations about the topic. Michael Brown's The International Dimension of Internal Conflict is perhaps the best comprehensive treatment of internal war. It brings together a mostly first-rate collection of essays on the causes, effects, and potential solutions to internal war. Roy Licklider's Stopping the Killing combines theoretical chapters with case studies of internal war to explore ways of halting domestic conflict. Despite their differences, these works share a belief that internal war is of critical importance not only for the countries in which they take place but also for the broader interests of the global community.

Causes

Neorealism and Internal War

During the cold war, historians, sociologists, and country specialists dominated the study of internal war. As ethnic conflict seemingly exploded across the globe, there appeared a spate of new works exploring issues of identity, nationalism, social structure, and nation building. Benedict Anderson's, Imagined Communities, Donald Horowitz's Ethnic Groups in Conflict, Liah Greenfeld's Nationalism: Five Roads to Modernity, and Anthony Smith's National Identity are some of the best scholarly attempts to grapple with the question of why so many states were torn by ethnopolitical conflict. ⁹ Although these works differed in approach and in the questions they raised and the answers they proposed, they all shared a concern with one issue in particular: why belonging to the same state was insufficient to prevent groups from believing they had a right to attack one another and the government to which they ostensibly owed allegiance.

With the end of the cold war, a new literature has emerged that uses neorealism in an attempt to make internal war understandable. Although neorealists take diverse paths to understand internal war, all stress anarchy as the underlying or permissive cause of armed conflict within states, just as it is between states. This approach, some of which is reflected in the reviewed volumes, has not been received well by nonneorealists and other scholars who do not focus on international relations. Comparativists, sociologists, and most political scientists dismiss neorealism outright for ignoring the subtleties of the societies they [End Page 556] study, that is, such key factors as identity and social structure. ¹⁰ This kind of blanket repudiation is misguided. Neorealism poses a serious challenge to existing approaches to the study of internal war and deserves to be taken seriously, for a number of reasons.

First, neorealism is arguably the dominant approach to understanding the causes and cures of interstate conflict, and many argue that it should be applicable to internal war as well. In his chapter in Stopping the Killing Robert Wagner makes a strong case for using the insights of international relations theory to explain war within states just as it does war between states. He argues that it is impossible to distinguish between the international and domestic realms based on levels of violence or the presence or absence of effective government. Increasingly, the international domain is marked by declining conflict and seemingly enhanced conformity with the dictates of global institutions and norms. At the same time, large numbers of states are unable to impose order within their societies as civil conflict inexorably spreads. Perhaps, then, international relations theorists have things exactly backward: instead of a world of international anarchy and domestic order, it is more accurate to assume a world of international order and domestic anarchy. ¹¹ And for understanding violence and anarchy, there seems to be no more appropriate tool than neorealism.

Second, neorealism is worthy of consideration for understanding internal war if only because it promises so much. Like Alexander the Great cutting the Gordian knot, neorealism seems to have the potential to slice through the muddled complexities of the causes and cures of internal war with insights of startling simplicity. Thus, according to neorealism, once central authority collapses, a microcosm of the international system is replicated within the state; this, in turn, causes domestic groups inside a country to behave much as states do in the international system. As Barry Posen argues, under conditions of anarchy groups within states fear for their security in the same way that states do:

In areas such as the former Soviet Union and Yugoslavia, "sovereigns" have disappeared. They leave in their wake a host of groups--ethnic, religious, [End Page 557] cultural--of greater or lesser cohesion. These groups must pay attention to the first things that states have historically addressed--the problem of security-- even though many of these groups still lack many of the attributes of statehood. ¹²

A key concept of neorealism, the security dilemma, comes into play when the state can no longer protect groups within its borders. The security dilemma asserts that efforts to improve one's security in an environment of anarchy makes others feel less secure and thereby lessens security for all. ¹³ In the wake of governmental collapse groups that may have lived in harmony under a strong central government suddenly view each other with suspicion, interpreting every act as a threat. In Yugoslavia, for example, neorealists see the disintegration of central authority as leading directly to the warfare that has plagued the Balkans throughout the 1990s. It is tempting to point to interethnic hatreds, ancient feuds, and conflicts over identity to explain the wars within and among the former Yugoslav republics. But as Michael Ignatieff reminds us in Blood and Belonging, Serbs, Croats, and Muslims lived in relative harmony for most of the twentieth century. They had high rates of intermarriage, lived in the same villages, and attended many of the same schools. What turned friends and neighbors into crazed killers was the power vacuum that emerged in the wake of Tito's death. Without a central government to ensure order and guarantee personal security, people became afraid. They sought protection where they could find it, which turned out to be their ethnic brethren. As Ignatieff writes:

Once the Yugoslav Communist state began to spin apart into its constituent national particles, the key questions soon became: Will the local Croat policemen protect me if I am a Serb? Will I keep my job in the soap factory if my new boss is a Serb or a Muslim? The answer to these questions was no, because no state remained to enforce the old inter-ethnic bargain. ¹⁴

Ethnic antagonisms helped fuel the conflict, but they would not have led to war without the collapse of authority.

Neorealists contend that the implications of the security dilemma for the behavior of intrastate groups in an environment of anarchy are as predictable as they are for states in the international system. Thus, the security dilemma should be especially acute for groups isolated from [End Page 558] their ethnic brethren within states, just as it is for states that are similarly remote from potential protectors. When technology and topography favor the side seeking to go on the offense, the prospects for intrastate war rise as they would for war between states. Just as the logic of anarchy leads states to protect themselves in ways that may inadvertently lead to interstate warfare, so too the collapse of domestic authority leads groups within states to take measures to defend themselves that may lead to internal warfare, regardless of whether that was their original intention. ¹⁵

Neorealism also offers guidance in the form of two basic solutions on how to stop internal wars. First, since internal wars stem from domestic anarchy, what is needed is the restoration of a central authority capable of imposing order. In this view Thomas Hobbes was correct that the threat of internal war is ended by the establishment of a central government strong enough to take on all challenges. ¹⁶ Ideally, the state afflicted with internal war would establish its own strong government. Alternatively, outside states can step in and establish control, as Vietnam did in Cambodia and Syria did in Lebanon.

Another neorealist approach to ending internal wars is based on the balance of power, which deters interstate warfare and similarly, it is assumed, can be employed to prevent warfare within states. One means of doing this is to promote the breakup of the state into ethnically homogenous independent countries (assuming of course that ethnic hatreds are at the root of the conflict), each of which balances the other. Where balances cannot be achieved internally, outside states can step in and establish alliances to protect weaker powers. By facilitating peaceful "ethnic cleansing," the international community prevents the mass killings that so often accompany internal war and raises the hope of creating the same stable balance of power that has deterred conflict on the international level. ¹⁷ Thus, for example, John Mearsheimer and Stephen Van Evera urge an ethnically based partition for Bosnia, Croatia, and Serbia along with international support for Bosnia to make certain that it can withstand threats from its stronger neighbors. ¹⁸ Neorealists [End Page 559] also assert that outside states can end internal wars by intervening so as to tilt the local balance of power to one side or the other, thus enabling it to impose order. Since stalemates often encourage the prolongation of civil wars, making sure that one side wins ensures quick termination of the conflict. Richard Betts argues that the failure of the United States and its NATO allies to support one party over the others in Somalia and the Balkans helped prolong those civil wars. ¹⁹

History supports the neorealist contention that a major factor in internal war is a weak government unable to control unruly subjects. Thus, the spurt in internal wars in the 1960s coincided with the height of decolonization, at which time scores of fragile new states were struggling to impose order over groups that violently resisted their control. And similarly, the creation of new states in Europe in the wake of the collapse of the Soviet empire encouraged power struggles among groups whose aspirations for autonomy and independence had heretofore been kept in check. There is little question that the shift in the balance of power from the state to groups within the state is a major reason for the eruption of civil conflict. And in focusing on such occurrences, neorealism contributes to our understanding of internal war.

The Limitations of Neorealism

Despite its benefits, neorealism is nevertheless an inadequate approach for explaining internal war. It does well in explaining the dynamics of some internal wars once they erupt, but it is deficient in explaining the causes of internal wars and the cures that can be administered. Just as domestic politics is not international politics writ small, so too internal war is not simply a scaled-down version of international war. The key distinction is the role of anarchy. The explanations and predictions that flow from neorealism stem from its assumption that anarchy on the domestic level is roughly equivalent to anarchy on the international level.

That assumption is not correct, however. First, while neorealism correctly notes that domestic anarchy can be an underlying or permissive cause for some internal wars, it does not explain how anarchy was created in the first place. More often than not, where anarchy does emerge (for example, in Liberia and Somalia), it is taken as a given. But anarchy cannot simply be assumed in internal wars: rather, most states most of the time can ensure compliance. On the international level, by contrast, [End Page 560] there is general agreement that there is no overarching power to compel states to do what they do not want to do. International relations theorists may therefore ignore how anarchy comes about, but it is a central problem for students of internal war.

Moreover the greatest number of internal wars by far occur where governments continue to exercise some degree of control; that is, they do not take place in the environment of anarchy assumed by neorealism. The Russian attempts to suppress the Chechnya uprising and the civil war in Algeria are far more typical than are the Liberias and the Somalias, which have garnered so much attention in the post-cold war era. Indeed, in many internal wars neorealist logic is turned on it head, as it is the strengthening of central authority, rather than its weakening or collapse, that is the permissive cause of internal war. Domestic conflict originates in governments seeking to destroy would-be insurgents as well as insurgents challenging governments. Just as states do not embark on wars they believe they will lose, so it is with governments: they tolerate threats to their rule if they cannot eliminate them but make war on those challengers when they believe themselves strong enough to do so. Thus, for example, Ethiopia's internal war against Eritrea in the 1970s was far less intense when the Marxist government of Mengistu Haile Mariam was weak than when the Mengistu regime (armed with Soviet weapons) became strong. ²⁰ Where government rule remains strong and anarchy does not define the context of the fighting, then, neorealism contributes little to an understanding of internal war.

Apart from emphasizing anarchy, neorealism also stresses material factors and parsimony, at the expense of a vast array of other motivations--religious, ideological, and emotional--that fuel domestic conflict. Thus, since neorealism argues that individuals or groups will not make war on their governments if they think they will lose, how does it explain the many internal wars that are fought for decades (for example, in the Sudan) against seemingly impossible odds? The neorealist assumption that states are similar units that behave in a like manner is not helpful for understanding internal war, where it is precisely in the vast differences among states that one finds the key to explaining why internal wars occur in some countries but not in others. Finally, neorealism tells us little about the role of legitimacy of government. Clearly, [End Page 561] the occurrence of internal wars is related to whether the people believe in the government's right to rule. On this subject, neorealism is silent.

In sum, neorealism provides some help in understanding internal war, but it is far from a complete explanation. For those internal wars that take place in an environment of anarchy, neorealism's insights into the behavior of states in the international system can be usefully applied. Neorealism is also helpful in explaining that some internal wars--like wars between states--break out because there is no central authority to prevent them from doing so. Neorealism is not very helpful, however, in explaining how anarchy developed in the first place; nor is it helpful for cases where conditions of anarchy or near anarchy are not present. To respond to these critical issues it is necessary to consider approaches that examine internal war free from the constraints of neorealism.

Bad Leaders as a Cause of Internal War

If neorealism is an inadequate approach, are there other theories that can explain why wars break out in some states and not in others and predict where and when it will happen? There is no clear answer to this question in the literature. Indeed, the German philosopher and poet Hans Enzensberger cautions that it may be pointless to look for a rational cause at all. As he sees it, violence is an end in itself and civil wars are waged "about nothing at all." ²¹ Domestic conflicts will continue so long as there is no central power to stop them. Despite Enzensberger's contention, however, most states are not wracked by civil wars. The great majority of the industrialized states enjoy domestic peace, and even many countries in the Third World, especially in Asia and Latin America, are not threatened by internal war, at least for now. Enzensberger provides a useful qualification that internal wars can be fought not for material benefit (as the neorealists would have us believe) but for nothing more than the sheer pleasure of killing. Internal wars in places like Lebanon and Liberia support his contention that wars that may have begun for "rational" reasons can degenerate into mindless killing with a life of their own. But more generally, it is both unhelpful and misguided to assert that nihilistic impulses explain most civil conflicts.

A far better insight into the causes of internal war is presented in the book by Michael Brown, much of it by Brown himself. Brown begins by rebuking the "ancient hatreds" view of internal war, in which groups with long-standing grievances erupt into violence once a constraining [End Page 562] hand (for example, of the Soviet Union) falls from their shoulders. As Brown correctly points out, there are too many cases where hatreds did not lead to violent conflict (for example, the Czechs and the Slovaks, the Ukrainians and the Russians) to posit this as a major cause of internal war (pp. 12-13).

What then is at the root of civil conflict? It appears at first to be virtually everything. Thus, Brown identifies four broad sets of explanatory factors: structural (strength of the state, presence of ethnic minorities); political (the fairness of the political system, whether citizenship is ethnic or civic based); economic (health of the economy, stage of economic development); and cultural/perceptual (presence of discrimination against minorities, views groups have of themselves and others) (pp. 12-23). This list, while certainly accurate, suffers from two major drawbacks. First, it does not tell us anything new. It is hardly surprising the weak states that are plagued by ethnic animosity and face economic problems will be more prone to violence than states with a homogenous population, a strong government, and a healthy economy. Second, as Brown himself notes, while these factors may explain the underlying causes of internal war, they are less helpful in explaining what triggers domestic conflict. To a greater or lesser extent, these conditions exist everywhere. For those interested in a parsimonious explanation of why some states are afflicted by internal war and others not, these conditions disappoint.

Brown does much better when he advances his own explanation for internal war. He argues that while mass-level conditions help explain which countries are vulnerable to internal war, elite-level forces are much more important in explaining the proximate causes of domestic conflict. Internal wars happen not because one people hates another, but rather because of the rational and deliberate decisions of "bad" leaders. Heads of state make decisions that lead to war because they are more interested in staying in power than in preserving the peace for their citizens. In looking to identify what allows bad leaders to drive their people to internal war in some places but not in others, Brown distills the many underlying causes of internal war to the two most important: a strong sense of antagonistic group history and mounting economic problems. When these factors come together with bad leaders, internal war can be expected. Where even one is absent, the likelihood of internal war is dramatically less.

This is an important argument. It says that much of what scholars study is not at the heart of why internal wars occur. Factors such as the level of modernization in countries, the strength of governments, and [End Page 563] the prevailing culture are not as important in explaining the outbreak of internal war as are the decisions of individual leaders. Outside forces such as meddling states play a relatively minor role in bringing about civil conflict. If accurate, Brown's argument provides a parsimonious way of identifying which countries are most vulnerable to internal war; it also suggests ways of halting internal war.

There is much to suggest that Brown is at least on the right track in the factors that he isolates. No one can refute that antagonistic group histories have played a central role in internal conflicts. Algeria, Angola, Bosnia, Burundi, Liberia, Russia, Sri Lanka, and Sudan are just some of the countries where internal wars driven by groups who hated one another have occurred. Nor can one discount economic factors in explaining internal war. In virtually every case of internal war, the afflicted country faced serious and growing economic problems. Even more telling, countries that have escaped the horror of internal war have, for the most part, done well economically. Chapters in the Brown volume by Rachel Bronson on the Middle East and Trevor Findlay on Southeast Asia make a persuasive case that many states in these regions have been able to stave off internal war precisely because economic growth allowed potentially hostile groups to be co-opted.

If Brown's arguments on the importance of group relations and the state of the economy are persuasive, it is his assertion that domestic elites, particularly leaders, bring about internal wars that is most arresting. He is right to ascribe a central role to leaders, but he does not go far enough in exploring the theoretical justifications for doing so. The key question is why leaders--both of governments and of rebel groups--provoke internal wars. For most neorealists, heads of state act in the national interest to defend their country from threats posed by other states. However, much of the literature on the Third World has long argued that government leaders act not so much in the national interest as on behalf of their own personal interest--not so much to defend themselves against threats from other states as to protect themselves from domestic challengers. Remaining in power is a leader's paramount interest. Leaders not only seek the benefits of power (wealth, status), but they also wish to avoid the unpleasant consequences of being forcibly removed from power; in many states that includes the possibility of being killed. Similarly, leaders of insurgent groups will challenge the government to gain the benefits under its control, and in the interim they will act to maintain their own position of leadership out of concern for personal survival and individual aggrandizement. Internal war results, therefore, when leaders of states [End Page 564] or groups see its outbreak as a means of perpetuating themselves in power. ²²

There are many examples of leaders who provoke internal wars to guarantee their hold on power, for example, those who provoked the ethnic conflict between Hutus and Tutsis that led to genocide in Rwanda. Boris Yeltsin's shift from escalating the war with Chechnya to trying to reach a settlement with the insurgents reflected the changing impact of the conflict on his chances of remaining president. The persistence of the civil war in Angola can be explained at least in part by its providing rebel leader Jonas Savimbi with access to untold wealth from diamond mines controlled by insurgent forces. Warlords in Somalia and Liberia owe their position and prosperity to the perpetuation of internal wars they understandably have little desire to see ended.

In the Balkans there is no doubt that leaders of the former Yugoslavia, particularly Serbian head, Slobodan Milosevic, helped cause the fighting by inflaming ethnic nationalism. Although some neorealists hold that rabid nationalist appeals are made to mobilize support in order to better protect fledgling states from threats by other states, ²³ Brown's leadership-centered approach is the more persuasive argument: Milosevic (and others) stirred up ethnic conflict in order to realize their personal interest of remaining in power. The chief threats facing them come not from other countries but from non-Serbian groups within the state. Recognizing that he could not hold on to power in a multiethnic Yugoslavia with a Serbian minority, Milosevic deliberately fostered a racist nationalism that resulted in the replacement of most of Yugoslavia with a state that had a clear Serbian majority. ²⁴

Despite some promise, Brown's formulation has its problems. First, his assertion that the interests of leaders drive internal wars is not systematically tested against actual cases. Although he provides a chart showing that the overwhelming number of internal wars was triggered by elites (twenty-six out of thirty-four post-cold war cases), there is, with only a few exceptions, little evidence offered to support the assertion (p. 582). Moreover, Brown does not explain exactly what constitutes [End Page 565] an elite-level cause. All internal wars are products of both elite and mass behavior, but the difficult task of isolating one from the other is not adequately undertaken in this volume. The reader is left to wonder why, for example, the seemingly mass-based Kurdish revolt is described as produced by elites, while the Kashmir dispute, despite an important elite component, is seen as mass driven. Justifications for these and other classifications are mostly not provided.

Second, Brown chides scholars for focusing on mass-level factors to the exclusion of elites, who, he argues, are more important as the proximate cause of internal conflict. But why are leaders more important than other factors? Are their decisions really more important than the collapse of domestic authority, ethnic hatreds, economic problems, or repressive rule? Brown makes much of the fact that it is "bad" leaders that drive internal conflicts, but he assumes a latitude in decision making that he does not demonstrate they possess. In one sense, of course, the Milosevics of the world are "bad" in that they exacerbate hatreds to further their own ends. But can we really expect leaders to act in ways that would undermine their tenure in office? If there are examples of "good" leaders who willingly risk their position in power in order to prevent internal war, Brown does not mention them. The bad leaders model simply begs the question of when you get bad leaders and what does "bad" really mean when it may just indicate leaders who are rational and self-interested.

Brown alleges that societies experiencing economic growth are less likely to experience internal war. But he ignores arguments asserting that rising expectations that are produced by economic growth are often at the root of domestic conflict. ²⁵ Moreover, he does not adequately consider how problems of differential gains across social, ethnic, and religious groups can foster violence. Brown is very good at discussing how leaders incite groups to attack others in order to solidify their hold on power. But he does not pay attention to how differing domestic political structures either facilitate or exacerbate the mobilization of populations to serve the leaders' ends. ²⁶ Brown does a good job of explaining how the permissive conditions of economic problems and historic antagonisms lead to civil wars, but he is less successful at explaining the many more cases where they do not. [End Page 566]

In the final analysis, Brown's argument tantalizes but does not fully persuade. If indeed the proximate cause of most internal wars lies with leaders who have the latitude to choose other paths, and if these leaders are able to provoke conflicts only in states already afflicted with antagonistic groups and mounting economic problems, then our understanding of the causes of internal war have been advanced considerably. Since internal wars are so varied in their origins, however, efforts at generalization need to be greeted with some skepticism. Brown's promise to bring order to the chaos of the study of internal war remains just that--a promise. It is up to country specialists, historians, sociologists, and other scholars to determine through detailed case studies whether the parsimony of Brown's approach is achieved only at the expense of accuracy.

Cures

Whatever the causes of internal war, there is little doubt that domestic conflicts will persist for the foreseeable future. Consequently, a great deal of attention has been paid to coping with internal wars. Whether they focus on the afflicted states themselves or on outsiders, various recommendations have been put forth to manage or end domestic conflicts. For the most part, the results have been disappointing. To a greater extent than even with international war, effective solutions for internal war have not materialized. Insofar as the literature on internal war (including the books under review) has made a contribution, it lies in accurately portraying the dismal prospects of ending violent civil conflict anytime soon. ²⁷

Nowhere are the problems inherent in halting internal wars more apparent than in Stopping the Killing, the edited volume by Roy Licklider. Using case studies and theoretical analyses, Licklider and his fellow authors tried to produce lessons to assist scholars and policymakers find ways to stop civil wars in such a way that they do not begin again. Each of the authors was asked to consider such factors as the nature of the issues underlying the conflict, the internal politics of each side, the military balance in the field, the role played by third parties, and the [End Page 567] inclusiveness of the society that surfaced after the settlement. Useful guidelines for "stopping the killing" were expected to emerge from these focused comparisons.

Despite the various authors' attempts to grapple with the problem of halting civil war, no consensus emerges on what to do, and the usefulness of such lessons as are put forth is questionable. Thus, William Zartman's argument that settlements require a "hurting stalemate" in which both sides see no benefit in continuing the struggle does not take into account the view (of Richard Betts and others) that the way to end an internal war is not through deadlock but rather through one side having enough of an edge to impose a settlement (pp. 24-25). Licklider attempts to reconcile various suggestions for ending internal war in the last chapter, but it is a forced, unpersuasive effort. In the final analysis, the greatest value of Stopping the Killing is in leaving us with the probably accurate if pessimistic conclusion that the vast differences in the origins and nature of civil conflict make it difficult, perhaps even impossible, to achieve overarching solutions; they are inevitably either too general to be useful or too specific to apply to most or many cases.

The empirical record bears out the intractability of internal conflicts highlighted by Stopping the Killing. Negotiated settlements of civil wars are much less common than negotiated settlements of international wars: between 1940 and 1990 only 20 percent of civil wars were resolved through negotiations, as compared with 55 percent of interstate wars. ²⁸ Even when a settlement is reached, there is no guarantee it will last. Data collected by Licklider (not presented in Stopping the Killing) demonstrate that negotiated settlements of civil wars are far less likely to endure than are the results of military victories. Renewal of conflict occurred in fully half of the negotiated settlements, but in only 15 percent of the military settlements. ²⁹ From 1991 to 1994, as John Stedman notes, negotiated settlements of civil wars in Cambodia, Liberia, Angola, and Rwanda fell apart, resulting in renewed warfare (p. 363). Should one draw the conclusion, then, that military victories are the preferred way of ending internal conflicts? Perhaps not, for as Licklider shows, military victories in civil wars are more likely than are negotiated settlements to lead to genocide, as was seen in Rwanda and Cambodia. Whether it is worthwhile to increase the probability of mass murder to produce an enduring settlement is a question not easily answered. ³⁰ [End Page 568]

Why are solutions to internal wars so much more difficult to reach than are solutions to interstate wars? As Licklider notes, in an interstate war both sides can retreat to their own territory. In an internal war members of the opposing sides must live together in the same country, often with the loser being forced to disarm. Further, negotiated settlements of internal wars leave in place groups with the power to challenge the new government's authority whereas military victories over states eliminate potential rivals. Finally, the absence of credible institutions to ensure security, the fear of living alongside an avowed enemy, and the high cost of misplaced trust all impede efforts to bring internal wars to an end (Licklider, 4-5).

Before turning to possible solutions to internal war, an important qualification is in order. The volumes under review and most of the literature in the field are premised on the assumption that armed domestic conflict is an unfortunate development to be avoided at all costs. There is little recognition that the greatest threat to most of the world's people comes not from internal war but from their own leaders, and that violent resistance is often the only way that individuals and groups can free themselves from the murderous tyranny of their governments. According to R. J. Rummel, over 169 million people have been killed by their own governments in this century alone, a figure that dwarfs the 38.5 million killed in international and civil wars during the same period. ³¹ Nevertheless, rarely is internal war seen in a positive light, as a means of removing intolerable regimes or providing autonomy for repressed peoples. Instead, the focus is on how states can better defeat domestic challenges. Threats to great power interests and to regional stability, as well as the concomitant humanitarian disasters, explain efforts to end internal wars. But this does not justify the assumption that the creation of strong governments capable of ending internal conflict is always the most appropriate goal.

Neorealist Cures

The difficulty of finding solutions for internal wars is that much more problematic when the causes are inadequately identified. This is the case with neorealism, which emphasizes the need to establish a strong central authority as a means of preventing internal wars or stopping them once they have started. But this "cure" raises more questions than [End Page 569] it answers. First, as discussed above, the establishment of a strong central authority can just as easily provoke internal war as bring about its end. Second, even where a strong central government might be the solution to domestic conflict, how does one establish it? Neorealism provides no directions for building a government powerful enough to overcome problems such as arbitrary borders inherited from colonialism, ethnic strife, the lack of effective institutions to channel conflict, and the absence of a competent bureaucracy.

This is not to suggest that the development of strong governments is impossible. It happened in Western Europe, but only after centuries of civil wars waged over religious disputes, ethnic conflicts, factional fights, and power struggles. Therefore, the prospect of newly emerging post-World War II countries developing in a matter of decades into strong states in which internal war is more or less obviated as a possibility is highly unlikely. All the more so, as Mohammed Ayoob argues, because new states also confront crushing poverty, demands for mass participation, and interference by other powers. ³² The lack of interstate war as a stimulus further inhibits the creation of strong states. As Charles Tilly explains, the states of Western Europe evolved into strong cohesive units only after several centuries of brutal international conflict. ³³ Today, states do not need to become stronger to deal with external threats because the international community guarantees their survival. ³⁴ Instead of the strength-enhancing effects of interstate warfare, contemporary states are more likely to experience internal war, which weakens the state by interfering with taxation and undermining government control, thus laying the foundation for future domestic strife. ³⁵

The neorealist recommendation that states establish strong governments to deter internal war ignores the fact that the very act of creating a powerful central authority often requires internal war. Much of the literature on internal war assumes that armed civil conflict is a pathology [End Page 570] that can be overcome. But this view overlooks the argument that internal war is a natural result of governments extending their power over unruly groups who resist centralized control. Until states accumulate enough power to impose order, it is to be expected that groups within the state will seek to preserve their autonomy by resisting (at times violently) state control. Internal war is especially likely in countries that begin the process of state building with modest amounts of power relative to the groups they seek to govern. Whether the focus is on the European civil wars of the sixteenth and seventeenth centuries or today's conflicts in Central Europe and the Third World, the process of redistributing power from the periphery is often a bloody, long-term affair. ³⁶ In the meantime, for those interested in speeding up the process of establishing a strong central authority, it is necessary to open the "black box" of the state to examine its unique ethnic, cultural, and economic context. The approach of comparative politics is far better suited to this task than is neorealism.

If states cannot establish strong governments on their own, there is always the possibility of outside powers intervening to impose order. Syria's intervention into Lebanon demonstrates what a determined state can accomplish in ending or at least mitigating internal conflict in another country. But as neorealists are quick to point out, most states are not prepared to intervene to stop internal wars in other states. The same international anarchy that permits war between states also permits war within states. At least in the international realm, war making is constrained by the actions of other states that perceive their interests to be threatened. In the domestic sphere, however, there are few conflicts that threaten the important concerns of states powerful enough to affect their outcome. As a result, internal wars are far more numerous and last longer than their international counterparts.

It has not always been the case that internal wars posed little or no threat to the key interests of great powers. During the cold war the superpowers found their strategic interests endangered by several internal wars, hence the American involvement in Korea, Vietnam, Lebanon (1958 and 1983), Nicaragua, and El Salvador; and the Soviet involvement in China, Korea, Vietnam, and Afghanistan, among others. Following the cold war, however, not one of the domestic conflicts has threatened the tangible strategic and economic interests of great powers. Either their impact is largely domestic and humanitarian (Somalia), or they spill over to similarly weak states (for example, as Sudan's civil war [End Page 571] affected Ethiopia) that cannot impose order. That governments often brutally suppress their own people is cause for compassion but, according to most neorealists, does not justify involvement by major powers.

For those few instances of internal war in which outside states become involved, the cures offered by neorealism raise more questions than they answer. Creating ethnically homogenous states in balance with one another begs the questions of what to do with minorities left behind and how to classify those of mixed parentage. Moreover, organizing states according to ethnic criteria creates an international norm that may legitimize and encourage heretofore quiescent minorities to seek statehood and thus contribute to worldwide instability. The faith of neorealists in the peace-inducing effects of the balance of power seems naïve, given the millennia of conflicts that have occurred under this system. Breaking up states into mutually hostile entities may enhance internal peace only at the expense of increased international war. ³⁷ As for giving one side the preponderance of power, this is easier said than done. Civil wars are messy affairs in which picking winners and losers is especially difficult. Even where a clear winner emerges, all too often the victor turns its wrath on the losers with genocidal ferocity. Not many states, especially those in the Western democratic world, would choose to become party to such an outcome.

Most damagingly, as John Stedman's indictment of international action makes clear, the efforts of outside states (and international institutions) to halt internal wars often wind up prolonging them. They do so by transforming wars that would have ended quickly, with one side winning, into protracted stalemates. In 1990 Charles Taylor's National People's Liberation Front was at the point of victory in Liberia. Only the intervention of Nigeria and other West African states under the authority of the Economic Community of West African States (ECOWAS) prevented Taylor from assuming control. As a result, the Liberian civil war stalemated and has continued for six more years, killing some 150,000 people (p. 252). Internationally sponsored cease-fires can also hurt more than they help. Because internal wars often end when it becomes clear that the costs of fighting have become unbearable for the leadership and the fighters, the establishment of cease-fires may delay a settlement by reducing the pain of the conflict. Most internal wars end without a cease-fire precisely because the threat of continued fighting is too much for one side to bear. As Stedman points out, in Zimbabwe, [End Page 572] Angola, Mozambique, Nicaragua, and El Salvador, negotiations to end the conflict took place in the absence of any cease-fire (p. 353). Moreover, cease-fires that have been established with outside help tend to break down (as was the case in Cambodia, Liberia, and Angola), since enforcing them frequently depends on outsiders who are not willing to pay the price to impose order. Even humanitarian assistance can prolong wars when it serves the fighters and not the innocent civilians, as happened in Bosnia, Rwanda, and Sudan. Before getting involved in the internal wars of others, much less promoting the breakup of states, outside actors might consider taking a version of the Hippocratic oath promising at least not to make things worse.

Leadership-Focused Cures

If the neorealists do not offer much help in solving the problem of internal war, what assistance can one find in Brown's view that bad leaders are the cause of most domestic conflicts? Two broad solutions to internal war follow from his argument. The first is to concentrate on coercing the leaders themselves. The second is to make potential followers less susceptible to the blandishments of their wayward rulers. While either approach may work in specific circumstances, neither offers a general guideline for halting internal conflicts.

Brown himself is understandably uncomfortable when addressing the question of what to do about internal wars driven by bad leaders. As he writes:

Many internal conflicts are triggered by self-obsessed leaders who will do anything to get and keep power. They often incite ethnic violence of the most horrific kind for their own political ends. If the international community is serious about preventing internal conflicts, it needs to think more carefully about the kinds of political behavior that should be proscribed and the kinds of actions it will be willing to take to steer or seize control of domestic political debates. These are extremely difficult problems, both intellectually and politically, but they have to be confronted if international actors are to prevent the deadliest internal conflicts. (p. 614)

Brown, however, does not confront these problems in any detail, for understandable reasons. Using force against leaders of sovereign states is one of the most sensitive issues in the international arena. Backing coups, undermining regimes, and even assassination are policies that are rarely discussed and even more rarely carried out effectively. The criticism of the American Central Intelligence Agency for its efforts to topple or otherwise intimidate foreign governments during the 1960s [End Page 573] and 1970s and the more recent failure to overthrow Iraq's Saddam Hussein illustrate the enormous problems of a leadership-centered approach. ³⁸ Brown mentions in passing the possibility of international military forces seizing bad leaders and placing them before international tribunals, but the tremendous difficulties encountered in attempting to capture and try war criminals in the Balkans raises questions about the feasibility of this strategy. ³⁹

If leaders cannot be gotten at directly, there is always the possibility of impeding their ability to mobilize support. Here the recommendations are on more familiar ground and conform to suggestions offered by many neorealists. Thus, John Mearsheimer and Stephen Van Evera emphasize the need to teach "honest" history in schools, thus dampening the flames of hypernationalism. ⁴⁰ Brown talks of launching "international information campaigns" and jamming radio and television stations to counteract or block harmful propaganda (p. 614). Even those not in the neorealist camp support efforts to make people less susceptible to the blandishments of their leaders. Jack Snyder and Karen Ballentine emphasize that unconditional freedom of speech can be manipulated by unscrupulous leaders to serve nefarious ends. They urge international patience and understanding for states wracked by internal conflict that seek to impose temporary restraints on speech, at least until democratic institutions have had time to mature. ⁴¹

Few will object to the goal of providing sources of information apart from that of the leadership. The feasibility and effectiveness of such an approach, however, is questionable. It is difficult to imagine that countries threatened by internal war will allow outsiders to come in to revise their textbooks and determine the classroom curriculum. Alternative sources of information are already being sent throughout the world both via traditional means (radio, television) and through newer technologies [End Page 574] such as the internet. Whether they will have a peace-inducing impact is another matter. Exposure to international media has done precious little to stifle the flames of hypernationalism in the Balkans and elsewhere. We simply do not know under what conditions, if at all, the dissemination of information will act to constrain the ability of "bad" leaders to garner support.

Recommendations for Future Study

Although the neorealists and Brown have contributed valuable insights into the causes and cures of internal war, much more work needs to be done. As a first step, less emphasis should be laid on searching for generalizations that purport to explain all internal wars. It is simply too diverse a phenomenon to lend itself to parsimonious theory. Better to concentrate on differentiating internal wars by type and seeing what kinds of contingent generalizations can be produced. Categorizing internal wars by origin is a good first start. Chaim Kaufmann's work is a good example of this kind of effort. ⁴² He distinguishes between internal wars based on ethnic strife (Rwanda, the Balkans) and those fought over ideological issues (Cambodia) and makes the case that outside military intervention can succeed in settling the former type of conflict but not the latter. Other categories can include internal wars whose origins are mass based and those that are elite driven, wars fought for territory as opposed to those over political control, and wars in which outsiders played a major role and those where they did not. Making these distinctions will be difficult and scholars will have to be scrupulous in applying objective criteria over the range of cases. What is lost in parsimony will be made up, one hopes, in better explanations for the very different conflicts that constitute internal war.

The likely impact of protracted internal war on world stability is another area that has received little attention and warrants additional consideration. Russia and China, for example, are far more vulnerable to internal wars than they are to outside attack. More generally, additional research is needed on unintended threats to global stability. Whereas traditionally, states worried about being attacked by other states, now they need to worry about the consequences of internal wars spreading beyond the borders of the countries in which they originated. The threat of a Soviet nuclear strike becomes one of "loose nukes" in a chaotic Russia, fear of an oil embargo is replaced by destruction of oil [End Page 575] facilities in a Saudi Arabia wracked by internal conflict, concern for a Warsaw Pact invasion of Western Europe is transformed into concern for an invasion of refugees fleeing civil strife. Much thought is needed about the impact of these and other "nondeterrable" threats to global stability.

Finally, the role and efficacy of international norms in spelling out the responsibilities of states that are not directly affected by internal war needs more attention. The great majority of internal wars do not threaten the security or economic well-being of great powers. But just as the post-cold war world has created space for a focus on internal conflicts, perhaps it will also lay the groundwork for the construction of a postwar interest in stability and regard for human rights that had been mostly ignored in the face of superpower confrontation. How such a norm might emerge and what its impact might be on halting or reducing the spread of internal war are key items for scholars to ponder in the coming years.

Making sense of internal war will be a critical task of the twenty-first century. Neorealism offers some insights into its causes and cures but ultimately falls short, since it is, after all, a theory designed to explain interstate affairs. The literature on internal war takes us further in explaining why such wars occur and what can be done about them, but the field is still very much in its infancy. The task confronting both international relations theorists and specialists in internal wars will be to determine which combination of causes and cures explains the greatest number of cases. The books under review have made a start, especially in their focus on the role of domestic elites, but it is only a beginning. As the post-cold war era unfolds, one would welcome an expanded attention to internal wars that matches the kind of attention that interstate conflict has attracted in the past. Whether that attention will yield useful generalizations on internal war remains to be seen.

Steven R. David is Professor of Political Science and Director of the International Studies Program at The Johns Hopkins University. He is currently working on a book examining the impact of internal war on global security.

Notes

1. Drawing from the Brown volume, I define "internal war" as a violent dispute whose origins can be traced primarily to domestic rather than to systemic factors, and one in which armed violence occurs primarily within the borders of a single state. The terms internal war, civil war, and domestic conflict are used interchangeably throughout this essay.

2. Eckstein, ed., Internal War: Problems and Approaches (Glencoe, Ill.: Free Press, 1964); Modelski, "International Settlement of Internal War," in James N. Rosenau, ed., International Aspects of Civil Strife (Princeton: Princeton University Press, 1964); and Huntington, "Patterns of Violence in World Politics," in Huntington, ed., Changing Patterns of Military Politics (New York: Free Press, 1962).

3. There are several internal war data sets that demonstrate the steady increase in their number (and the absence of any surge after the cold war). They include Peter Wallensteen and Margareta Sollenberg, "The End of International War? Armed Conflict, 1989-1995," Journal of Peace Research (August 1996); Ted Robert Gurr, Minorities at Risk: A Global View of Ethnopolitical Conflict (Washington, D.C.: United States Institute of Peace, 1993); Roy Licklider, "The Consequences of Negotiated Settlements in Civil Wars, 1945-1993," American Political Science Review 89 (September 1995), 688-89; and Michael E. Brown, "Introduction," in Brown, ed., The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996), 4-7.

4. The continuity of the nature of internal wars from the cold war period to the post-cold war era is best expressed by Ted Robert Gurr, "Peoples against States: Ethnopolitical Conflict and the Changing World System," International Studies Quarterly 38 (September 1994).

5. This figure is calculated from Ruth Leger Sivard, World Military and Social Expenditures, 1996 (Washington, D.C.: World Priorities, 1996), 18-19.

6. Wallensteen and Sollenberg (fn. 3), 353.

7. Margareta Sollenberg, ed., States in Armed Conflict, 1994, Report no. 39 (Uppsala: Department of Peace and Conflict Research, Uppsala University, 1995), 9.

8. Two excellent surveys of the causes of internal war are James B. Rule Theories of Civil Violence (Berkeley: University of California Press, 1988); and Ted Robert Gurr, ed., Handbook of Political Conflict (New York: Free Press, 1980).

9. Anderson, Imagined Communities: Reflections on the Origins and Spread of Nationalism (London: Verso, 1983); Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985); Greenfeld, Nationalism: Five Roads to Modernity (Cambridge: Harvard University Press, 1992); and Smith, National Identity (London: Penguin, 1991).

10. For an explicit criticism of the neorealist approach for its failure to take into account social structures (in this case, regarding the amount of military power that can be generated by states from different cultures), see Stephen Peter Rosen, "Military Effectiveness: Why Society Matters," International Security 19 (Spring 1996).

11. The reversal of second and third images is found in the work of several scholars focusing on unstable states. See, for example, Robert Jackson, Quasi-States: Sovereignty, International Relations and the Third World (Cambridge: Cambridge University Press, 1990); Barry Buzan, People, States and Fear: The National Security Problem in International Relations (Chapel Hill: University of North Carolina Press, 1983); Mohammed Ayoob, "Security in the Third World: The Worm about to Turn," International Affairs 60 (Winter 1983-84).

12. Barry Posen, "The Security Dilemma and Ethnic Conflict," in Michael E. Brown, ed., Ethnic Conflict and International Security (Princeton: Princeton University Press, 1993), 104.

13. The security dilemma comes from Robert Jervis, "Cooperation under the Security Dilemma," World Politics 30 (January 1978).

14. Ignatieff, Blood and Belonging: Journeys into the New Nationalism (New York: Farrar, Straus and Giroux, 1993), 42.

15. Posen (fn. 12), 103-11. For more on security dilemmas in internal wars, see also Chaim Kaufmann, "Possible and Impossible Solutions to Ethnic Civil Wars," International Security 20 (Spring 1996), 147-51.

16. Thomas Hobbes, Behemoth or the Long Parliament, ed. Ferdinand Tonnies, 2d ed. (London: Frank Cass, 1969); idem, Leviathan, ed. C. B. McPherson (New York: Penguin, 1968).

17. The creation of ethnically homogenous states is supported by Michael Lind, "In Defense of Liberal Nationalism," Foreign Affairs 73 (May-June 1994). For countries torn by ethnic strife, Chaim Kaufmann supports the separation of ethnic groups into defensible enclaves--though not necessarily into separate states. See Kaufmann (fn. 15), 161-65.

18. Mearsheimer and Van Evera, "When Peace Means War," New Republic, December 18, 1995.

19. The value of creating an imbalance of power rather than letting a war stalemate is persuasively argued by Betts, "The Delusion of Impartial Intervention," Foreign Affairs 72 (November-December 1994).

20. The intensification of the Ethiopian war against Eritrea after the strengthening of the Mengistu regime is discussed in virtually all accounts of the conflict. See, for example, Colin Legum and Bill Lee, The Horn of Africa in Continuing Crisis (New York: Africana, 1979); Marina Ottaway, Soviet and American Influence in the Horn of Africa (New York: Praeger, 1983); and Paul B. Henze, Russians and the Horn: Opportunism and the Long View, European-American Institute for Security Research, eai Papers, no. 5 (Marina del Ray, Calif.: European-American Institute, 1983).

21. Enzensberger, Civil War (London: Granta, 1994), 30.

22. For a fuller treatment of this view, see Steven R. David, "Explaining Third World Alignment," World Politics 43 (January 1991).

23. Barry R. Posen makes the argument that states engender nationalism in order to enhance their ability to defend themselves from external threats; see Posen, "Nationalism, the Mass Army and Military Power," International Security 18 (Fall 1993).

24. Many have made the case that the war in the Balkans stems from Milosevic's desire to remain in power. See V. P. Gangon Jr., "Ethnic Nationalism and International Conflict: The Case of Serbia," International Security 19 (Winter 1994-95); Alex Djilas, "A Profile of Slobodan Milosevic," Foreign Affairs 72 (Summer 1993); and Misha Glenny, The Fall of Yugoslavia: The Third Balkan War (New York: Penguin, 1992).

25. See, for example, James Davies, "Toward a Theory of Revolution," American Sociological Review 27 (February 1962); idem, "The J-Curve of Rising and Declining Satisfaction as a Cause of Revolution and Rebellion," in Ted Gurr and Hugh Davis, eds., Violence in America: Historical and Comparative Perspectives (Beverly Hills, Calif.: Sage, 1979).

26. The best argument for lack of institutionalization as a cause of domestic unrest is made by Samuel Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968).

27. There is a large and growing literature on how to end internal wars, mostly focusing on negotiations. See, for example, William Zartman, ed., Negotiating Internal Conflicts (Columbia: University of South Carolina Press, 1993); idem, Negotiating an End to Civil Wars (Washington, D.C.: Brookings Institution, 1995); Paul R. Pillar, Negotiating Peace: War Termination as a Bargaining Process (Princeton: Princeton University Press, 1983); Richard A. Falk, The International Law of Civil War (Baltimore: Johns Hopkins University Press, 1971); Stephen John Stedman, Peacemaking in Civil War (Boulder, Colo.: Lynne Rienner, 1991); and Barbara F. Walter, "The Resolution of Civil Wars: Why Incumbents and Insurgents Fail to Negotiate" (Ph.D. thesis, University of Chicago, 1994).

28. Pillar (fn. 27), 25.

29. Licklider (fn. 3), 681.

30. Ibid., 687. The pessimism expressed by many regarding negotiated settlements is not completely unqualified. Stedman and Licklider both acknowledge that under the right conditions (for example, when the outside powers are willing to impose order), negotiated settlements can work. Moreover, there have been too few post-cold war cases to allow for any definitive conclusions regarding the efficacy of negotiations for ending civil conflict.

31. Rummel, Death by Government (New Brunswick, N.J.: Transaction, 1994), chap. 1.

32. Ayoob, "The Security Problematic of the Third World," World Politics 43 (January 1991). Ayoob focused on the difficulties encountered by Third World leaders in trying to undertake state building and nation building simultaneously, but his pessimistic outlook holds for new states outside the Third World as well.

33. Tilly, "Reflections on the History of European State Making," in Tilly, ed., The Formation of National States in Western Europe (Princeton: Princeton University Press, 1975), 42.

34. This is a principal argument of Robert Jackson (fn. 11).

35. Jeffrey Herbst makes these points with regard to Africa. See Herbst, "War and the State in Africa," International Security 14 (Spring 1990).

In contrast to Herbst, Bruce Porter argues that civil wars can strengthen states, as seen by the European civil wars in the fifteenth and sixteenth centuries. The key, however, was the existence of a functioning bureaucracy that persisted throughout the conflict. Many of the states afflicted by internal war today have no such bureaucracy. Porter, War and the Rise of the State: The Military Foundations of Modern Politics (New York: Free Press, 1994).

36. Youssef Cohen, Brian Brown, A. F. K. Organski, "The Paradoxical Nature of State Making: The Violent Creation of Order," American Political Science Review 75 (December 1981), 901-10.

37. For criticisms of partition as a solution to internal war, see Radha Kumar, "Partition's Troubled History," Foreign Affairs 76 (January-February 1997); and Gidon Gottlieb, "Nations without States," Foreign Affairs 73 (May-June 1994).

38. For an exhaustive treatment of some of the problems the United States has had in attempting to intimidate (or remove) foreign leaders, see U.S. Congress, Senate, Hearings before the Select Committee to Study Governmental Operations with Respect to Intelligence Activities, 94th Cong., 1st Sess., vol. 7, "Covert Action" (Washington, D.C.: GPO, 1975); and Alleged Assassination Plots Involving Foreign Leaders, November 20, 1975, Senate Intelligence Committee, Select Committee to Study Government Operations with Respect to Intelligence Activities, 94th Cong., 1st Sess., Senate Report 94-465, 13098-8. On Saddam Hussein, see Yossi Melman, "Why the Plot to Kill Hussein Failed," Los Angeles Times, July 21, 1991, 1-2.

39. On problems of bringing war criminals to justice, see Theodor Meron, "Answering for War Crimes," Foreign Affairs 76 (January-February 1997), 2-9.

40. Van Evera, "Hypotheses on Nationalism and War," International Security 18 (Spring 1994), 37; Mearsheimer, "Back to the Future: Instability in Europe after the Cold War," International Security 15 (Summer 1990), 56.

41. Snyder and Ballentine, "Nationalism and the Marketplace of Ideas," International Security 21 (Fall 1996), 37-40.

42. Kaufmann (fn. 15), 136-75.

[an error occurred while processing this directive]