Civil war can be defined as a violent conflict that pits states against one or more organized nonstate actors on their territory. This distinguishes civil wars from interstate conflict (where states fight other states), violent conflicts or riots not involving the state (sometimes labeled intercommunal conflicts), and state repression against individuals who cannot be considered an organized or cohesive group, including genocides or similar violence by nonstate actors, such as terrorism or violent crime. This entry first discusses the various kinds of civil war and reviews the key patterns and trends in such conflicts. It then presents theories on the causes of civil war and the current research agenda.
The above conceptual definition of civil war clearly encompasses many different forms of conflict. Some analysts distinguish between civil wars where insurgents seek territorial secession or autonomy and conflicts where insurgents aim for control of the central government. Conflicts over government control may involve insurgents originating from within the center or state apparatus, as in military coups, or challengers may come from the periphery, or outside the political establishment. Others separate between ethnic civil wars, where the insurgents and individuals in control of the central government have separate ethnic identities, and revolutionary conflicts, where insurgents aim for major social transformation. Colonial conflicts are sometimes singled out as a set distinct from civil wars on states' core territory. However, even if these suggested categories may be conceptually distinct types and could suggest different causes and dynamics, a given civil war will often combine several elements. For example, insurgencies may be both ethnic and ideologically based, and aims can shift over time from secession for a limited territory to controlling the entire state. Thus, most researchers do not impose such strict a priori distinctions between varieties to analyze these as separate types but instead study civil war as a common class of conflict.
Even though the modern state sometimes is defined by its alleged monopoly on violence, armed challenges to state authority are as old as states themselves. Despite numerous historical accounts of civil wars, there are few systematic data sources on civil conflict prior to 1945, since so little is known about the extent of conflicts outside the developed countries before this date. Figure 1 displays the number of ongoing conflicts (top) and new outbreaks (bottom) since 1945, using data from the Uppsala Conflict Data Program at the Department of Peace and Conflict Research, Uppsala University and Centre for the Study of Civil War at the International Peace Research Institute, Oslo (UCDP/PRIO), Armed Conflict Dataset. UCDP/PRIO classifies conflicts claiming more than 25 deaths in a year as interstate conflicts, extrasystemic (or colonial) conflicts, as well as internal conflicts, or as civil wars, including internationalized internal conflicts where other states fight on the side of the government. Figure 1 shows that while there are relatively few interstate wars since 1945, civil wars have been common. Whereas interstate conflicts tend to be short, civil wars often persist for a long time. Research also demonstrates that civil wars are less likely to be settled by formal agreements than interstate wars and much more likely to recur. Many observers saw the initial rise in new outbreaks of civil conflict after the end of the Cold War as evidence that the world would be more turbulent and violent after a period of stable deterrence between the superpowers. However, Figure 1 demonstrates a clear decline in civil war since this peak after the Cold War. The specific causes that may underlie this decline remain disputed, and the number of ongoing civil wars remains high in absolute terms.
Civil wars are generally less severe than interstate wars in terms of the direct battle deaths. However, civil wars have been more frequent and often persistent, and more than 90% of the recorded deaths in battle since the Cold War stem from civil wars. Further, war can have a substantial indirect impact on human welfare beyond the direct loss of life. Studies indicate that countries experiencing civil war see a pronounced fall in their gross domestic product and never recover their earlier growth trajectory. According to Paul Collier et al. (2003), “Civil war is development in reverse.” Civil wars also dispel trade and investment and leave large social legacies in unemployed former combatants and displaced individuals. The negative consequences of civil war are not limited only to the countries that experience them; studies find that neighboring countries also suffer a negative economic impact and may be more prone to violence themselves as a consequence of conflict among neighbors.
The literature on the causes of civil war is enormous, and it is impossible to provide an exhaustive review of the many arguments that have been presented here. Contributions have emphasized a variety of social, economic, and political factors. However, theories of civil war, in general, point either to specific motives for why people resort to violence or to specific opportunities that make violence more or less feasible or attractive. This entry examines theories of civil war by the specific clusters of factors emphasized, in an order roughly following the chronology of theory development and the specific events and cases that motivated these arguments.
Most civil wars take place within relatively poorer societies, and many studies corroborate the link between development and income. Early contributions to the study of violence within societies tended to focus on economic deprivation and grievances as key motives. In particular, Ted Gurr highlighted inequality and how groups may resort to rebellion if they are dissatisfied with their current economic status relative to their aspirations. The literature on nationalist conflicts has emphasized how both relatively poorer and wealthier groups are likely to rebel against the center if they believe that they can do better under independence. Civil wars in Latin American countries were often interpreted within a framework focusing on economic grievances, in the form of either unequal land distribution or high income inequality. However, the empirical evidence linking individual income inequality and conflict is mixed. Older studies focusing on a broad range of political violence often found a positive effect of higher inequality, while newer studies of civil war more specifically tend to find little support. However, one should be cautious in making strong inferences from this, given the poor quality and coverage of cross-national data on individual income inequality.
More recent political economy studies of civil war tend to be very dismissive of the role of grievances. Some researchers have argued that grievances are ubiquitous and that it is more important to focus on variation in the opportunities for violence. Collier and Anke Hoeffler argue that low income makes it easier to mobilize insurgencies, since potential recruits have less to lose in forgone income from normal economic activities and wages for soldering will be much lower than in wealthier societies. James Fearon and David Laitin argue that civil war is primarily a problem of weak states, which in turn is largely determined by economic development. Researchers in this tradition also link mobilization to the role of individual incentives. Opportunities for insurgencies are better when participants can do well out of war, for example, through looting or rents derived from valuable natural resources. Empirical studies lend some support to link between natural resources and a higher risk of civil war. Civil wars in Africa are often taken to support these perspectives.
Political deprivation, such as colonial subordination or lack of political rights, provides another plausible motivation for resort to violence. Many conflicts after 1945 first emerged as groups sought to achieve independence for areas under colonial rule. The Indochina War and the Algerian War of Independence helped mobilize movements in other countries by showing how overwhelmingly more powerful colonial powers could be defeated through sustained violent campaigns. Although overseas colonies eventually received their independence, many ethnically distinct groups within contiguous empires such as Russia or Ethiopia see themselves in similar struggles of national liberation. Researchers have highlighted how violence may arise around such peripheral ethnic minority groups. The violent conflicts in the former Yugoslavia spurred renewed interest in ethnic conflict. Many researchers postulated strong parallels between the security dilemmas in an anarchic international system and relations between ethnic groups, where only full territorial partition could provide stable solutions.
Much of the subsequent empirical literature has emphasized that there is little evidence that ethnically more diverse countries are generally more prone to conflict. However, looking only at the number or relative size of ethnic groups disregards their political status and the extent to which ethnic groups are systematically excluded from political power or discriminated against by the state. Studies looking at ethnic exclusion find stronger evidence for a relationship to violence. Ethnically diverse countries are not necessarily more prone to conflict if they have inclusive institutions or grant autonomy rights, and control of the state or access to power does not always follow directly from the relative size of ethnic groups. Many ethnic civil wars see minority groups controlling political institutions and excluding larger groups from power, such as the Amhara in Ethiopia until their defeat by a coalition of other groups in 1991. Political, economic, and ethnic grievances are often linked. Studies that focus on inequalities that follow group lines rather than individual income inequalities find a stronger relationship to conflict.
Struggles for broader political rights in autocratic systems provide another context where violence may occur. Political democracy provides many avenues for actors to express dissent through nonviolent political means. Autocratic regimes typically deny citizens room for political activities and often resort to severe repression of protest, which in turn may motivate resort to arms. Protests against autocratic or exclusionary regimes have often turned violent, sometimes leading to sustained campaigns, as, for example, in South Africa. Claims for greater political rights and freedom are clearly important elements of the rhetoric and call for mobilization of many insurgent movements, even if these do not necessarily move to implement democratic institutions if successful in achieving power.
Much of civil war research has focused on the role of political institutions in structuring the opportunities for violence rather than the potential accommodative effect of greater democracy. Many have argued that although autocratic institutions provide fewer avenues for nonviolent political activities and protest, autocratic regimes are often sufficiently repressive to successfully deter dissent. By contrast, anocracies, or regimes combining autocratic and democratic features, are seen as the most prone to see violent conflict by combining the lack of political freedom motivating violence with the sufficient opportunities afforded by a less repressive regime. Many studies have found evidence for such an inverted U-shaped relationship between democracy and civil war. However, there is no consensus on the underlying mechanisms (e.g., reforms in autocracies may be a response to conflict potential rather than a prior cause). Some hold these relationships to be in part artifacts resulting from the definition of empirical democracy measures.
Most of the theories discussed above emphasize structural factors that rarely change or change only slowly over time. Such persistent structural features do not provide clear explanations for why civil wars break out at specific times and not others. Research on social movements suggests that certain events can create “political opportunity structures” that afford groups better prospects for extracting concessions from the state or center. This may include demonstrations of state weakness, conflict between elites, or events that make it easier for groups to mobilize, for example, by bringing groups together or indicating focal points for organizing protests. The concept of opportunity structures has so far not had much direct impact on studies on civil conflict, but there are many existing arguments and findings in civil war research that can be interpreted within this framework. Regime change and other signals of weakened state authority can increase the perceived chances of success or extracting concessions from a government. Studies have shown that economic crises and natural disasters can increase the risk of conflict. This is consistent with the idea that crises and emergencies can help provide a setting for rallying protest against the government. For example, the 1973 earthquake in Nicaragua, and the massive corruption and lack of subsequent reconstruction, generated widespread disillusionment and helped a long-standing Marxist insurgency dramatically increase recruitment.
Most research on civil war has assumed that since civil wars are “internal” conflicts rather than conflicts between states, their main causes must also be domestic or located within state boundaries. However, factors outside individual countries can play an important role in the outbreak of conflicts, as well as how they evolve. Many actors in civil wars are not necessarily confined within individual countries. Ethnic groups often span international boundaries, and transnational kin frequently participate in or provide support for insurgencies in other states. The status of international borders generates different constraints and opportunities for governments and rebels. Borders are, in a technical sense, just lines in the sand and are often not difficult to cross from a purely military perspective. However, the fact that borders formally delineate state sovereignty makes it more difficult for governments to violate the sovereignty of other countries, while such constraints are less relevant for rebels. Governments risk retaliation from neighboring countries from territorial incursions and face difficulties in targeting transnational support. This in turn means that rebels can have a logistic advantage in operating out of extraterritorial bases, and transnational rebel movements can be more difficult for governments to deter or defeat. The presence of conflict in another state can help facilitate violent mobilization, either through emulation of successful rebellions or through the direct imports of arms and combatants. Finally, civil wars are often closely linked to interstate war. Poor relations between states may motivate governments to support insurgencies in rival countries, and civil wars may in turn promote military conflict between states, for example, as a result of conflict over border violations, alleged support for insurgents, or conflicts over the externalities generated by conflicts. Western Africa in the 1990s, for example, provides many examples of governments supporting insurgencies in neighboring countries and retaliating against alleged support and border violations. The fact that conflicts are not necessarily limited to single countries and may involve participation by other states in various forms demonstrates how a strict dichotomy between civil and interstate wars often may be untenable.
Although most research focuses on accounting for the original outbreak of civil war, there has also been growing interests in understanding the prospects for conflict termination and why some conflicts are so persistent. Many researchers argued that there must be some symmetry between the original causes of civil war and the factors that lead to their eventual termination. From this perspective, efforts to foster conflict settlement should focus on addressing the issues that gave rise to the conflict. For example, civil wars would be likely to end earlier in more developed countries, where the opportunity costs are higher and states are stronger. Events such as political reform and increasing accommodation could help promote the end of conflicts. However, other researchers argue that the factors that make conflicts endure can be quite different from those influencing the initial outbreak of conflict. While interstate wars tend to be relatively short and typically end quickly in some kind of formal settlement once the relative strength of the actors becomes clear, civil wars often persist for a long time, as in the case of Southern Sudan or Angola. A number of explanations have been proposed to account for this paradox. Some emphasize how civil wars tend to be low-intensity conflicts fought in the periphery, where the features that facilitate insurgency make it difficult for governments to conclusively defeat rebels. Others point to how insurgents that can do well during war through looting and control over valuable resources may have little interest in seeking an end to conflict. Other research emphasizes how conflict termination involves severe problems of credible commitment. Even if it may be relatively easy to reach consensus on the terms of an agreement in principle, carrying through with an agreement may be difficult since parties have incentives to make promises that they later renege on. For rebels, laying down arms is a risky choice, since they will be left less able to defend themselves. Moreover, it can be difficult for governments to ascertain whether rebels will uphold their end of agreements, for example, through decommissioning or controlling violent breakaway factions in the aftermath of agreements. Research on credible commitment problems after civil war argues that the success of agreements often hinges on whether third parties can serve as external enforcers. New research on peacekeeping and conflict management suggests that involvement by international organizations such as the United Nations (UN) can increase the prospects for settling conflict and preventing wars. However, many interventions in civil conflicts have clearly been less than fully successful. There is little consensus on how best the UN may contribute or to what extent the characteristics of conflicts or initial prospects for settlements determine whether external actors are willing to be involved in the first place.
Civil war has gone from being a peripheral area to a major focus in the study of conflict. Moreover, the field shows clear positive synergies from the interaction between theory development and research design. Older studies of civil war were often quite descriptive, focused on single cases, and frequently reluctant to consider general theories of civil war. The renewed interest in civil wars in the 1990s recognized the need to take into account nonconflict cases and turned to cross-national studies, comparing the various characteristics believed to make countries more prone to conflict across countries and its relationship to civil war. However, recent research has highlighted how the proxy measures used in cross-country comparative studies are often quite far removed from the key theoretical concepts. Moreover, looking at states and countries at large often downplays the nonstate actors in civil wars and ignores the often considerable variation within countries. Conflicts tend to be localized and often involve actors and regions that are markedly different from national averages, as, for example, the Chechen conflict in Russia. Recent contributions, accompanied by many new data developments, often turn to greater disaggregation within countries and a more specific focus on actors to more carefully evaluate theories of civil war. A number of studies have looked at the specific conflictual events within individual countries over time to better understand the microlevel interactions between actors in conflict. Some studies have looked at dyads of peripheral groups and the center to take into account how the attributes of nonstate actors may influence the risk of conflict. Other studies have looked at the characteristics of specific rebel groups to better understand capacity or opportunities for conflict or used individual-level data on participants in insurgencies to examine motivations. Some researchers consider geographically disaggregated data, either on the characteristics of the places where conflict occurs or using smaller units within countries, for a more detailed resolution of how social, economic, and political factors may be related to conflict. These studies indicate that the characteristics of conflict zones tend to be quite different from national aggregates and averages. Moreover, the conclusions from country-level studies on who participates in insurgencies and their motivations are often not supported by more direct evidence. Such innovations in the study of civil conflict attest to the vibrancy of this research area. Moreover, they underscore the important relationship research design and the conclusions that we reach, and how theory must inform the former. Although existing research has clearly generated many useful insights about civil war, this review illustrates the considerable changes in the conventional wisdom on civil war over time. Advances in theory development and research are likely to continue to spur further changes.
Authoritarian Regimes, Conflicts, Democratization, Dissatisfaction, Political, Ethnicity, Inequality, Economic, Mobilization, Political, Nationalism, Nonstate Actors, Peace, Revolution, Secession, State Failure, Violence, War and Peace, Warlords
Related Credo Articles
Even though civil wars by definition take place within national borders, they can become global problems when they become a serious threat to...
Different social sciences have adopted different perspectives in the study of conflict. Most economists have approached this issue using a...
Civil war can be defined as a violent conflict that pits states against one or more organized nonstate actors on their territory. This...