Introduction: Civil war and the international system
Civil wars have become the most frequent and destructive form of conflict in the international system. Between 1945 and 1999, 127 civil wars in 73 states resulted in 16.2 million battle deaths and more than 50 million displaced persons.Footnote 1 As we have seen in the cases of Lebanon, Bosnia, Kosovo, Liberia, Sudan and many others, civil conflicts generate enormous human suffering, along with economic and political devastation. Civil strife also creates important regional diffusion effects. Unrest in one country can pose threats to the stability of neighbouring countries, which ultimately might undermine regional and international security. Even after a conflict terminates, the negative repercussions of domestic conflict can continue to affect those in close proximity. In addition, problems posed by domestic strife also can lead to engagement of other states, international actors and organisations.
With their significant domestic and international implications, internal conflicts have attracted a great deal of attention from researchers and policymakers.Footnote 2 The good news is that the theoretical and empirical understanding of civil conflict, once called by Sambanis ‘the most poorly understood system failure in domestic political processes’, has progressed tremendously since the late 1990s.Footnote 3 Utilising qualitative case studies and quantitative data, scholars have identified a number of factors that make states more susceptible to civil wars and explored the causal mechanisms behind various relationships. Hence, today we are in a much better position to understand why such domestic conflicts emerge. Progress in the civil conflict literature, however, is by no means limited to determinants of onset. A great deal of research examines the conditions that influence the duration, severity, termination, reoccurrence and aftermath of civil conflict. Furthermore, progress in civil conflict research includes all stages of the research process: theorising, model development, collection of qualitative and quantitative data, measurement and testing. These developments have been critical not only for researchers but also for policymakers trying to locate and establish mechanisms to minimise the destructive impact of civil conflict. Prevention, management and termination of violence necessitate, first and foremost, a clear understanding of the complex dynamics of civil wars.
This article will review theoretical perspectives on the causes of domestic conflict and identify problems associated with the conceptual juxtaposition of opportunity versus grievance that predominates in the field. We do not intend to provide an overview of quantitative comparative analyses of civil conflict or distinguish between results that are more widely accepted and those results that are relatively less robust. Nor do we intend to delve into data and coding related issues. Although such efforts are beneficial for civil war research, they are beyond the scope of this article.Footnote 4 The next section presents the core ideas of opportunity and grievance, the basic categories of causal factors. The third section evaluates this heavily utilised dichotomy in the civil war literature and identifies some of the limitations associated with opportunity and grievance based explanations. Fourth, and finally, we provide suggestions for future research on civil conflicts.
Several themes emerge from reconsideration of how theorising and, in turn, hypothesis testing is carried out vis-à-vis civil wars. First, our review credits Paul Collier and Anke Hoeffler's theoretical dichotomy with starting a very important and lively debate. Their categories have played a highly constructive role in the evolution of the civil war literature. There is no doubt that grievance and opportunity are two critical aspects of the conflict process. However, like any other dichotomy or conceptual juxtaposition, while sharpening the debate and facilitating discussion, it also creates some difficulties in the process of understanding civil war. Conceiving of opportunity and grievance as two mutually exclusive and competing explanations can lead to reductionism and, in turn, limit our understanding of civil conflicts. Not all variables fall easily into either the opportunity or grievance categories for complex events such as civil wars. The existing dichotomy, moreover, encourages research that eschews the question of how collective action is achieved in some circumstances and not others. Similarly, opportunity and grievance based explanations emphasise the role of rebels rather than the state and strategic interaction between states and rebel groups. Most complex of all challenges to consider is the intertwining of factors tentatively identified as opportunity- or grievance-related when it comes to empirical exploration of cases. Theorising often is carried out most effectively, as will become apparent, when various factors are recognised as playing more than one role – generating a grievance at one stage, perhaps, but then an opportunity at another or vice versa.
Theoretical perspectives: grievance and opportunity
All countries host some citizens who are dissatisfied or frustrated with existing political, economic or social conditions. However, only in some instances does this latent unrest cross a certain threshold and transform itself into a massive armed conflict. Why do some states experience political violence in the form of civil wars while others do not? Which countries are more prone to violent domestic conflict? Scholars have investigated these challenging questions in an attempt to uncover the factors that best account for domestic political violence on a cross-national basis.
With regard to the roots of civil conflict, Collier and Hoeffler provide an influential scheme of analysis.Footnote 5 They have introduced a conceptual dichotomy to the literature: ‘greed’ vs. ‘grievance’. The greed argument refers to self-interested, even selfish, behaviour as the main motivation behind civil wars, whereas the grievance approach identifies discontent and frustration as the primary motivation for political action. Over the years, Collier and Hoeffler have modified the original conceptual division and shifted their emphasis from greed to a broader notion, namely ‘opportunity’, which refers to the factors that facilitate organised violence and influence the feasibility of action, over and beyond greed.Footnote 6 While greed can be an important source of motivation for a potential rebel and serve also as a facilitating factor, it fails to capture other dimensions of the opportunity structure that play critical roles in rebel calculations. Scholars studying civil war have used this dichotomy heavily.Footnote 7 It is commonplace for researchers to assign their key independent variables to one approach or the other: grievance or opportunity. Not surprisingly, there is a great deal of disagreement about which indicators should belong to what category and why.Footnote 8
Even though grievance and opportunity explanations are not exhaustive, they seem to be the main competitors in explaining collective political violence: ‘The history of the study of contentious politics since the 1960s may be summarised as a movement from culture to structure, from collective behaviour and relative deprivation to resource mobilisation to political process, or from Gurr (1970) to Tilly (1978).’Footnote 9 The main difference between the two approaches is the underlying explanatory variables they use to explain collective political violence. While the former emphasises discontent due to unjust deprivation as the primary motivation for political action, the latter focuses on mobilisation of group resources in response to changing political opportunities.Footnote 10 In other words, grievance theory assumes a direct and strong link between deprivation-induced discontent and political violence, while opportunity theory downplays the role of grievances and emphasises the importance of mobilisation and organisation of the rebel movement in the emergence of civil wars.
Grievance approaches
Grievance theory places a great deal of emphasis on psychological processes and factors with potential to create discontent among citizens. The basic premise of grievance-based approaches, including Gurr's relative deprivation theory, is that ‘justice-seeking behavior’ is at the origin of civil conflict and collective violence is a function of various forms of frustration, injustice and resentment experienced by individuals, which may be due to political, economic or both factors:Footnote 11
The primary source of the human capacity for violence appears to be the frustration aggression mechanism […] If frustrations are sufficiently prolonged or sharply felt, aggression is quite likely, if not certain, to occur […] The frustration aggression mechanism is in this sense analogous to the law of gravity: men who are frustrated have an innate disposition to do violence to its source in proportion to the intensity of frustration.Footnote 12
Relative deprivation theory asserts that individuals focus on expectations (relative deprivation) and compare themselves with others in society (structural inequality). These two variants of the relative deprivation argument entail different causal mechanisms but point to the same conclusion: both breed discontent.Footnote 13
The first mechanism highlights the importance of the difference and gap between aspirations of a group (that is, subjective value expectations) and its actual capabilities (that is, value capabilities) in catalysing insurgency.Footnote 14 A degree of inevitability exists in the relationship between deprivation and likelihood of collective violence. In other words, relative deprivation and postulated responses to it – namely anger, psychological strain, discontent, and grievances – are necessary conditions for civil conflict. Shared grievances foster a sense of group identity and generate a strong sense of in group/out group distinction over time, which in turn leads to politicisation and activation of discontent. Armed opposition against government is, therefore, a way to redress grievances and alter sources of discontent. Conflict also is expected to vary in magnitude according to the level of relative deprivation that people experience. As the perceived deprivation increases – relative to expectations – the risk of discontent and civil strife will increase.Footnote 15
Structural inequality, the second mechanism, is also important in understanding violence. When a substantial portion of the public does not receive what it perceives as a proper share in the allocation of scarce resources, anti-system frustrations tend to increase. Structural inequalities and collective disadvantages (economic and social) generate grievances among those at the bottom of society. If such conditions are handled poorly by the state, large-scale political violence can ensue.Footnote 16
Given that the state is responsible for keeping the balance between groups and preventing (or at least, managing) a security dilemma, scholars naturally investigate the impact of government policies on citizen angst and frustration. Some of the most prominent hypotheses from this theory therefore focus on deliberate political, economic or religious discrimination, systematic exclusion of minority groups from political power, limitation of religious and linguistic practices, political repression by the state, poverty and absolute shortages, inequitable division of economic resources, and human rights abuses. These policies foster a sense of injustice, frustration, and anger in society against an out-group (or the state) and people start organising to express their opposition to the government. In other words, the shared (common) identity brings people together and binds them. This makes the pursuit of common goals possible.
Although grievance theory is influenced heavily by psychological theories and emphasises the importance of relative deprivation, group disadvantages, and socially experienced frustration at the individual level, most empirical tests of the argument are conducted with cross-national, aggregate indicators. This is largely because grievance and injustice-driven explanations tend to build on subjective and elusive concepts like history, hatred, isolation, ethnicity, religion, exclusion and inequality, which are difficult to quantify. While earlier empirical investigations guided by grievance arguments focused mostly on the level of economic inequality (income, land) and deprivation,Footnote 17 later research turned to questions about repression, systematic discrimination (economic, political, gender related), human rights abuses and other non-economic inequalities (for example, limitation of political or religious rights, unequal education and health services).Footnote 18 A more comprehensive set of indicators became the norm as research on grievance progressed.
Despite being well theorised, grievance factors such as inequality and repression have not found systematic, cross-national support from research. Most of the influential large-N studies conclude that there is no obvious regularity in the interaction between inequality and civil conflict.Footnote 19 A handful of recent studies, however, offer some degree of support to the inequality-conflict linkage. For instance, Regan and Norton highlight the importance of political inequality for the risk of civil conflict and find that exclusion from political power, as well as limitations on political freedom, exhibits a strong conflict-inducing effect.Footnote 20 Caprioli's and Melander's results demonstrate that structural inequality and subordination of women are important predictors of civil conflicts.Footnote 21 Moreover, gender equality is associated with a pacifying effect on domestic conflict. Building on these results, Nel and Righarts find that violent civil conflict is more prevalent at intermediate levels of infant mortality, which is utilised as a proxy for economic inequality.Footnote 22
Similarly, testing the association of repression and domestic conflict produces rather weak support.Footnote 23 While a number of scholars report no statistically significant direct effect,Footnote 24 Regan and Norton's analysis reveals strong empirical evidence for the idea that a higher level of government repression is associated strongly with an increased likelihood of civil conflict.Footnote 25 Along similar lines, Thoms and Ron claim that human rights violations create deep grievances that can, under certain conditions, motivate collective violence and serve as conflict triggers.Footnote 26 Empirical studies also raise the possibility that there is an inverted U-shaped relationship between negative sanctions and political violence; the propensity for collective political violence is higher at intermediate levels of repression, since these regimes are oppressive, but not to the degree that it completely prevents mobilisation for collective action.Footnote 27 Overall, the empirical evidence on influence of repression on collective violence is neither consistent nor conclusive. Scholars are more confident about the negative impact of civil conflict on the levels of human rights abuse than the reverse.
Even though grievance approaches have limited empirical support, this type of theorising continues to attract attention and few venture so far as to dismiss such factors in explaining civil conflict. This is due largely to strong theoretical foundations and intuitive appeal.Footnote 28
Opportunity approaches
Significantly influenced by rational choice theory, the opportunity approach provides an alternative explanation for emergence of civil violence. This approach is premised on the idea that social movements are the product of the ‘calculus of risk, cost and incentive’.Footnote 29 The extent of collective political violence is determined by the political environment,Footnote 30 relative power and resources capability, opportunity structures and mobilisation processes. The decision of an individual to participate in a rebellion depends on potential costs and benefits from violent political action; presumably rational, would-be rebel soldiers will choose to join an insurgency only if the expected value is higher than that offered by the status quo.Footnote 31 Opportunity structures influence rebel expectations and, in turn, condition political behaviour and shape the pattern of contention.Footnote 32 Therefore, opportunity theorists do not perceive domestic violence as an emotional reaction to grievances but rather as a ‘rational response’ to circumstances. Participation in collective political violence is the product of the probability of victory and its consequences.
According to this theorising, social movements are triggered by incentives, but grievance, deprivation and anger are neither necessary nor sufficient conditions for violent behaviour; what matters is simply rational politics.Footnote 33 Collective action ultimately depends on the political environment and opportunities because these factors affect an insurgency's ability to mobilise resources.Footnote 34 Regardless of motive, for internal dissent to take place, rebel leadership needs an effective strategy, resources and people who are mobilised and ready to act collectively.Footnote 35 Therefore, the process of political mobilisation and conditions that render insurgency more feasible, attractive and less costly hold a central place in this approach.Footnote 36
In fact, as a set of theoretical and probabilistic explanations, grievances are likely to over-predict the onset of political violence, leading to a higher expectation than observed historically.Footnote 37 In attempting to explain this deficiency, Sambanis points out that if grievances can trigger collective action, the redressing of them becomes essentially a collective good whose provision presents typical collective action challenges.Footnote 38 In other words, in the process of transforming discontent into political violence, the most important challenge the insurgency needs to tackle is the free-rider problem. This is largely because potential rebels are aware of the fact that costs associated with participation are quite high and, more importantly, they can enjoy the benefits resulting from successful collective action, like regime change, even if they do not participate in an organised rebellion.Footnote 39 What this implies is that even when rational peasants have something to gain from therebellion, they will not choose to rebel unless the collective action problem is overcome.
To deal with this problem it is essential to ‘deflect some of the costs associated with participation in armed conflict’ and ‘change the potential payoff to the participants’.Footnote 40 This can be done by offering private side payments.Footnote 41 When leadership offers selective incentives – including economic (that is, material gains, direct payments, future benefits from looting), social (emotional or psychological gains) and physical benefits (protection from repressive and threatening actions of the government) only to participants, potential rebels are less likely to prefer abstention (Olson's idea of providing selective incentives). Dealing with defection is less of a challenge when rebel leadership controls extractable natural resources that can be used easily as an economic incentive to attract soldiers.Footnote 42
In sum, this approach perceives grievances and inequalities as ubiquitous, weak and relatively constant in impacting upon all societies. Hence it opposes the relative deprivation view that grievances are automatically politicised and that discontent will translate directly into an increase in the activities of rebel groups for collective action. According to this approach, the main motivating factor from the standpoint of a potential rebel is the expectation of reward and personal gain. The existence of grievances is no guarantee for mobilisation; the linkage between motivations and effective political action depends on the organisational capacity of groups and opportunity structures.
Compared to the grievance approach, opportunity theories have received stronger support from empirical studies.Footnote 43 One of the most important opportunity indicators in the existing literature is level of economic development. While previous research provides strong empirical support for the relationship between economic prosperity and civil peace, the reason behind such a linkage is far from clear. Scholars offer different explanations as to why wealthier states are at a lower risk of rebellion.Footnote 44
Scholars also have paid great attention to three other so-called opportunity indicators – natural resource endowment, demographic and geophysical characteristics of states – since they play important roles in shaping the context of rebellion. Countries with a larger population are expected to be at a much higher risk of civil unrest than countries with a smaller population. A large population increases the number of potential rebels who can be recruited by insurgents. In addition, central governments of countries with large populations face difficulties in controlling, managing and serving the population and meeting their demands. Rough terrain also is perceived to be an important opportunity-related factor. While plains and deserts offer little cover to insurgents, rough terrain, including mountains and forests, can provide natural sanctuaries to insurgency and limit the surveillanceand counter-insurgency activities of the government. Lastly, ‘loot’ is important for opportunity arguments because rebels need to have a continuous and adequate flow of revenue to sustain large-scale violence; natural resources can provide a source of financing for insurgents. Lootable natural resources also allow a rebel movement to provide selective benefits to soldiers in exchange for their continued support and loyalty and therefore alleviate the collective action problem. In addition, the possibility of future income though looting can motivate individual rebels and groups to take up arms and overthrow the government.
Evaluation of the current theoretical division: grievance versus opportunity
While scholars acknowledge both opportunity and grievance as important aspects of the conflict process, they continue to debate whether variables related to motivation or opportunity to fight are most important to understanding where and when civil war will take place. Although this conceptual distinction served a valuable purpose by sharpening the debate and bringing significant insight into the causes and conduct of civil conflict, perception of opportunity and grievance as competing explanations also poses difficulties in understanding highly complex events like civil conflicts.
Our evaluation of the prevailing theoretical division is separated into two subsections. In the first section, we consider grievance and opportunity as separate categories. First, while grievance theories fail to explain how and when discontent leads to organised violence, opportunity theories fail to explain why certain groups take up arms but others with similar opportunities do not. Second, grievance approaches fail to account for other types of motivation behind insurgency. Third, even though the existing empirical evidence is much stronger, when it comes to opportunity-based explanations, the actual political mechanisms through which the opportunity variables operate still are unclear. Fourth, a number of variables do not fall cleanly into either the opportunity or grievance categories.
We argue in the second section that, even if we tried to consider both grievance and opportunity-based variables simultaneously, a number of critical issues – again both theoretical and empirical – would remain. First, these concepts make it difficult to explain cases where both factors appear to operate at different points in the conflict. Second, they fail to account for important cases in which neither opportunity nor grievance operated. And finally, the focus on grievance and opportunity places too much emphasis on rebels and fails to theorise adequately the role of the state and strategic interaction between states and rebel groups. Any approach to understanding civil war needs to capture the complex interaction between states and rebels around a variety of aspects of grievance and opportunity in a way that allows rebels to overcome collective action challenges. We close by presenting suggestions for future research that might move the field in this direction.
Considering grievance and opportunity as separate categories
First and foremost, theories that focus on underlying motivations offer a satisfactory explanation as to why: (1) people want to participate in collective political action; and (2) violence erupts. However, they fail to explain how individual feelings of grievance are transformed into collective action, how groups of people come together and organise themselves, which factors constrain or enable rebel movement and when discontented people rebel. Motivation forms the basis for action but is not a sufficient condition for organised internal conflict. Regardless of the nature and extent of resentment against the state, opposition obviously will not be able to realise its goals and pose a real challenge without soldiers, arms, resources and an efficient organisational structure. This explains why some targets of discrimination rebel, like Tamils in Sri Lanka and Kurds in Turkey, while others, such as Baluchis in Iran and Roma people in Europe, fail to do so.Footnote 45
Opportunity theories are similarly one-sided in their explanatory power. While they successfully account for the factors and context that make large-scale collective action more likely and sustainable, they do not address the role of public dissatisfaction and fail to explain why some groups take up arms but others with similar opportunities do not. It is unreasonable to expect people to take up arms against the state just because there is some opportunity for mobilisation and the likelihood of success is high. People need to have a strong motivation and belief in their cause to participate in a dangerous venture. They need a strong bond (that is, identity to mobilise) because it separates them from others and binds them together to pursue common goals.
Grievance approaches, in addition, are very focused and tend not account for other types of motivations behind insurgency. Grievance is defined by many as widely shared dissatisfaction among group members about their cultural, political and/or economic standing vis-à-vis a dominant group and hence it: (a) emphasises collective disadvantages; (b) identifies differences among people; and (c) designates injustice and deprivation as the main drivers of civil conflict.Footnote 46 While these are important sources of motivation, others also can fuel collective violence. Just as redressing grievances can be a major incentive for action, other motives – including, but not limited to, greed – can be the main incentives for rebellion. Income that will be acquired during the rebellion from looting – economic or political benefits anticipated from overthrowing the government – or benefits associated with capturing the state can motivate the insurgency.
Third, despite the fact that opportunity-based theories have garnered more empirical support so far than grievance-based explanations, they are beset collectively by significant variable specification problems. For example, Collier and Hoeffler find that natural resources, a variable they measure as the share of gross domestic product taken by primary commodity exports, is a significant factor in explaining civil war onset. Despite the empirical basis of the Collier and Hoeffler argument, important cases seem to provide disconfirming evidence either as false positives or false negatives. These cases suggest competing explanations as to why states with abundant natural resources are at a lower risk of rebellion.Footnote 47
Among case studies chosen and designed to test and refine the Collier and Hoeffler model,Footnote 48 on a number of occasions the model predicts a low probability of civil war when it does occur: these false negatives include Burundi, Nigeria between 1965–1969, Senegal, Mali, and Lebanon 1975–1990.Footnote 49 There are also a number of false positive cases where the model predicts a high probability of war but where the cases demonstrate no war (or the violence that does occur does not meet the threshold definition for civil war): Macedonia 1990–2001 and Nigeria 1985–1999.Footnote 50 Particularly in light of the false positive cases, one is left searching for an explanation as to why the critical conditions in the Collier-Hoeffler model seem to lead to civil war in some cases but not others. Snyder also seems to be motivated by a very similar question, pointing out that among the 42 countries that produced alluvial diamonds, tropical timber, and/or illicit drugs between 1960 and 1999, 24 of them (57 per cent) did not experience civil war.Footnote 51 Appendix A provides a summary of cases inconsistent with the Collier-Hoeffler model.
Despite any apparent correlation between opportunity variables and civil war onset, the actual political mechanisms remain unclear. Cases exist where a country has some traits that would put it at risk for civil war but unexpectedly lacks other seemingly critical elements. Algeria, for example, is a mixed case: high primary commodity export levels, which make it high risk for civil war onset, but relatively low income from its diasporas, which place it at low risk.Footnote 52 What, then, explains the decade of civil war in Algeria that began in 1992? It is by no means obvious. The case of Bosnia is similar. It had relatively low primary commodity exports in 1990–1992, a fact that makes its violence inconsistent with the Collier-Hoeffler model; however, this case does appear consistent with a number of other opportunity-related variables, such as geographic dispersion and mountainous terrain.Footnote 53
Natural resources and other primary commodities help to explain the onset of several civil wars, but they do so in ways different from each other and from the logic of the Collier-Hoeffler model. For example, in the Democratic Republic of Congo, natural resources do seem to have been a significant determinant of civil wars; however, geographic concentration and resulting unequal distribution of those resources among ethnic groups appear to have mattered more than the state's dependency on their export.Footnote 54 In Indonesia, the effects of primary commodities do not appear to manifest themselves in easier financing for rebels as Collier and Hoeffler would predict, but instead in a set of grievances over resource distribution, a larger military and a lack of government credibility.Footnote 55
Fourth, the dichotomous grievance-opportunity approach does not lend clarity to the role of explanatory factors that cannot be singularly classified. Some variables could be capturing aspects of grievance, opportunity, or both. Many factors, indeed, can be incorporated in (or, attributed to) both approaches, depending on their impact. Corruption is a prime example. Pervasive corruption in government, indicating the misuse of public office for private gain, allows certain groups to have access to services while others are being excluded, sometimes completely. Differential treatment leads to inequality that, in turn, creates a gap between ordinary citizens and state institutions and decreases trust as well as legitimacy of the political system. In addition to fuelling grievances, pervasive corruption plays a major role in providing the necessary opportunity structure for insurgency. Its existence signals to opposing groups that the state is incapable of performing its duties. Besides, corruption can promote insurgency through its hindering impact on long-run economic stability and prosperity. Countries with high levels of corruption suffer from lack of investment because private economic agents prefer stable countries in which doing business is easy. The devastating impact of weak institutions on the productivity and functioning of national economies makes it easier for insurgents to recruit rebels in the quest for political change beyond constitutional boundaries.Footnote 56
Among many that can be attributed to both approaches (depending on their impact), three other factors are economic development, repression and regime type. Poor economic conditions enhance mobilisation potential of the aggrieved communities. These conditions can decrease the opportunity cost of insurgency and make it more attractive for potential rebels. As the opportunity cost of participation decreases, obtaining supporters becomes a less daunting task for the rebel leadership. Weak economic conditions also can heighten the level of absolute deprivation and economic grievances, especially when disparities among groups, regions and individuals are widespread. Last, but not least, poor economic performance can limit the ability of the state to buy off opposition and take costly reforms to mitigate grievances.Footnote 57
Repression is similar to economic development in the sense that its impact can be attributed to both approaches, depending on circumstances. It is simply a strategic choice that governments make to create or maintain political quiescence and a climate of fear. Repressive acts by the state could include a wide variety of coercive behaviours and often induce intensely hostile psychological reactions.Footnote 58 Gurr states that the threat of sanctions is ‘equivalent to the concept of anticipated deprivation, the innate emotional response to both is anger’.Footnote 59 Moral distress and the search for justice are expected to increase the level of participation in insurgency.Footnote 60 In these circumstances, according to Lichbach, ‘the apathetic become politicized, the reformers become radicalized, and the revolutionaries redouble their efforts’.Footnote 61 High levels of state repression also can mitigate the free-riding problem that an insurgency faces by making defection an extremely costly option for potential rebel soldiers. Therefore, to avoid state-led abuse, potential soldiers might choose to offer their services to the insurgency. Repressive activities and tight social control, however, also can play a vital role in limiting an insurgency's ability to mobilise supporters by increasing costs. Impeding the ability of groups to challenge the government and mobilise resources like people, money and guns significantly increases the cost of collective action.
Regime type also can go both ways when it comes to insurgency-related effects. It commonly is asserted that democratic political structures are less likely to foster civil wars because they are more inclusive, constrained, egalitarian, tolerant, accountable and responsive to the wishes of the people. Citizens in fully democratic regimes tend to have more freedom, access to power and resources than those in dictatorships. More importantly, individuals can make use of available legitimate channels to express grievances in peaceful ways, like voting, bargaining and negotiation rather than resorting to violence. In many ways, the availability of formal institutions and effective channels to express dissent puts the brakes on large-scale violence. These characteristics of democratic representation can decrease significantly the grievances of citizens. Regime type also significantly influences the opportunity structure for rebellion. In democracies, people who want to articulate their demands enjoy a broad range of options. Opportunity structures are open for rebellion, but also other kinds of collective action. In dictatorships and closed regimes, even when there is widespread discontent, people do not have access to effective, peaceful mechanisms to articulate their opinion. Closed political systems inhibit emergence of collective political action because the repressive nature of the regime increases the costs associated with peaceful and public, as well as violent and clandestine, means of expression.
Similar arguments can be offered for the impact of education, an important tool in improving social welfare. Lack of educational opportunities or discrimination against certain segments of the population can create resentment against the state that, in turn, incites violence. By investing in education, the state provides a very valuable service that otherwise would not be available to some elements of society. It is important to note that the poorest and most vulnerable segments of society depend heavily on government and cannot replace public services with private ones. Therefore, by investing in an economically and socially productive area, namely education – rather than rent-seeking activities – governments can enhance their ties with citizens and the power base of their regime. Higher levels of education also can increase the opportunity cost of participation in an insurgency and therefore make the recruitment process more difficult.Footnote 62
Considering grievance and opportunity simultaneously
Even taken together, the categories of opportunity and grievance often fail to capture the empirical complexities associated with the way that rebel groups overcome their particular collective action challenges. Three other comments are necessary in this regard: the dichotomous approach to studying civil war: 1) makes it difficult to explain cases where both factors appear to operate at different points in the conflict; 2) fails to account for important cases in which neither opportunity nor grievance operated; and 3) does not theorise adequately the role of the state and strategic interaction between states and rebel groups.
As previously discussed, the utility of exclusive analytical categories – like opportunity and grievance – is questionable since they seem to be inextricably intertwined and complement each other, especially in poor country settings and during different points of an ongoing conflict.Footnote 63 In fact, recent research emphasises the difficulties this dichotomous approach has in explaining changes in conflict after the initial outbreak of violence.Footnote 64 The case of Sierra Leone provides an illustration of the oft-intertwined nature of these concepts. In this conflict, grievances matched the importance of greed and opportunities.Footnote 65 After 24 years of mismanagement, manipulation, corruption, abuse, suppression and exploitation, civil war in Sierra Leone was initiated by a small rebel group, the Revolutionary United Front (RUF), in 1991.Footnote 66 RUF had many grievances towards the corrupt and ethnically oriented political regime of the All People's Congress Party (APC) and had the goal of overthrowing the government and re-establishing multiparty democracy.
Externally encouraged liberalisation attempts in the 1970s and 1980s affected the conflict, because privatisation led to increasing inflation, devaluation, significant cuts in wages of state employees, private oligopolies and rampant corruption. The decline in public services not only made rebel recruiting easier but also generated strong resentment in society. As revenues continued to fall, the infrastructure deteriorated, smuggling escalated and the army failed to provide security for the people.Footnote 67 In addition, widespread discrimination and corruption decreased citizen trust in government and the legitimacy of the state was questioned. As corruption and inequality increased, anger and fear mounted in society. It eventually led to violence.Footnote 68
Controlling the diamonds became one of the biggest motivations of the RUF, as the state failed to fulfil its duties. The lack of state capabilities, mismanagement and prevailing corruption made the exploitation of vast natural resources especially attractive for young people who were largely unemployed, poor, vulnerable and uneducated. During the dictatorial rule of the APC, conditions became very favourable for manipulation and mobilisation of such marginalised youth into organised violence. Insurgents’ control over the diamond fields became of special importance for the RUF not just because of greed. Diamonds provided certain opportunities for them – money for buying arms, recruiting new people and sustaining the violence. Diamonds also played a critical role in the involvement of Charles Taylor, former president of Liberia, in this civil war. With the aim of destabilising the country and gaining access to the natural resources of Sierra Leone (mostly diamonds), he allegedly supported the RUF with personnel, weapons, ammunition and military training in exchange for diamonds. The rebel forces encouraged by Taylor launched armed attacks within the territory of Sierra Leone and terrorised the civilian population. This example shows not only why during times of civil unrest, diamonds can emerge as the key to power, but also how that can complicate the conflict by encouraging external actors to get involved both before and after the initial outbreak of violence.Footnote 69
In addition, conflict can start as a political, grievance-driven rebellion, but economic opportunity considerations can eventually prevail over the original aspirations. Malone and Nitzschke note ‘gemstones or drugs became a prominent source of rebel-financing in the grievance-driven conflicts in Burma, Cambodia, Afghanistan, Columbia, Peru and the Democratic Republic of the Congo (DRC) only after fighting there had broken out’.Footnote 70 In the case of the DRC, grievances against both the Mobutu and Kabila regimes played a huge role in the initial stages of the civil war. However, once the fighting surfaced, greed rather than grievances motivated rebel groups, kept the movement going and determined its intensity. Capturing the natural resource fields and rents provided rebels and warlords with a great incentive to continue fighting.Footnote 71 In other words, even when economic incentives might not be the primary cause of the conflict, they still can interact with other motivations and produce a security dilemma.
It is worth noting that patrimonial politics, in which political, economic and military power overlap with each other, is the norm in much of Africa. Consider Zaire, for example, where patrimonial politics (that is, elites using clientalistic value systems to sustain their rule and political support) gave free rein to the enrichment of certain individuals and groups (including the president and his associates in government). This process, in turn, generates political decay and contributes to state failure.Footnote 72
Second, the long-lasting armed conflict in Uganda shows that while grievance and opportunity may explain a small part of the conflict, they cannot account for the main drivers of strife. While looting of villages and trucks carrying valuable supplies financed the insurgency, capturing the state to acquire access to valuable natural resources did not provide the main motivation behind the war. Grievances originating from exclusion and underdevelopment played a role at the beginning of the insurgency; however, as the conflict progressed, these factors lost their importance, largely replaced by fear of ethnic extinction, spiritual redemption and existential motivations.Footnote 73 Since Museveni formed the National Resistance Army (NRA), the Acholi people and members of the Ugandan National Liberation Army forces have lived in fear of reprisals for their actions in the past. The leadership of the Lord's Resistance Army (LRA), especially Alice Lekwena and Joseph Kony, became convinced that the Acholi people consisted of sinners on the brink of extinction. Despair led them to pursue a solution in redemption and spiritual escape. Therefore, to offer a satisfactory explanation for the actions of Kony, and especially the question of why Kony killed, tortured or abducted his own people, one needs to understand the notion of ‘redemption through internal ethnic cleansing’.Footnote 74 Spiritual redemption represented an internal mechanism rather than ‘a function of an external grievance’.Footnote 75 While grievances and economic issues played minor roles, in this extreme example, the pursuit of power and survival become key motivations for armed groups. Continuation of violence turned into an end in itself.Footnote 76
Another factor that carries a great deal of importance for both theories is the role of the state as a participant in the outbreak of intrastate conflict.Footnote 77 However, neither side of the dichotomous approach adequately captures – theoretically or empirically – the role that the state plays in the emergence of internal violence. While scholars generally agree that weak states are at a higher risk of insurgency, they tend to emphasise the opportunity-enhancing nature of state weakness and neglect its role in generating grievances.Footnote 78 Most existing treatments of state capacity in the civil war literature seem to follow Charles Tilly's understanding of government capacity, which can be summarised as ‘the extent to which governmental agents control resources, activities, and populations within the [state's] territory’.Footnote 79 These studies tend to emphasise the ability of the state to deter violent challenges to its authority and therefore focus on the coercive (and to some extent, administrative) capacity of the state against insurgents. In this understanding, state power is associated with the degree to which it can exert control over citizens. Weak states – the ones without sufficient bureaucratic penetration into the society and institutions of coercion like an efficient police force and military – are expected to provide conditions conducive for challengers to recruit non-combatants to their forces and pursue goals through violence.Footnote 80
Weak states, in addition to providing opportunity, can generate citizen frustration that motivates insurgency. These processes are not mutually exclusive, and they can easily take place simultaneously. Levi defines a strong state as follows: (1) representative and accountable to its people; and (2) effective, which means capable of protecting the population from violence and ‘supplying other public goods that the populace needs and desires’.Footnote 81 She claims that a state's ability to secure cooperation and compliance from its citizens is critical for preventing the emergence of violent predation. Weak states encounter serious difficulties in maintaining ‘quasi-voluntary’ complianceFootnote 82 of a broad segment of the population and in co-opting political opposition and decreasing incentives for organising a rebellion. Credible commitment is a key ability that distinguishes strong states from weak ones; inept states that ‘impose too much repression and not enough redistribution’ will generate grievances, lack legitimacy and, in turn, have difficulty in maintaining civil peace.Footnote 83
Adding the state into the opportunity versus grievance framework, however, presents other challenges. Recent approaches to the study of civil war focus on how particulars of strategic interaction between states and rebels might lead to peace or conflict. Governments and rebels base their actions on perceptions of what the other party did in the past and what it is likely to do in the future. This brings into play the possibility that factors such as misperception and credibility are important determinants of intrastate conflict, along with both parties’ relative signalling abilities. Governments and rebels tend to think more about the future than current theories might anticipate. They are careful to build into their strategic calculations the probable outcome and expected duration of the current conflict,Footnote 84 as well as the anticipated risks and costs of future challenges.Footnote 85 Overall, the relative strength of both parties has important consequences for signalling their respective abilities to commit and build ongoing reputations in bargaining.Footnote 86
Signalling by respective parties to the conflict (or crisis) is recognised as an important component of this strategic interaction.Footnote 87 Signals sent by either party provide the other side with information about relative capabilities or interests. Such information can influence strategic decision-making. In the literature on civil wars, several studies focus on motivations underpinning particular ethnic groups seeking self-determination. These motivations include the extent to which the group feels relatively disadvantaged within a country compared to other ethnic groups,Footnote 88 the degree to which it feels relatively incapable of achieving its particular economic goals,Footnote 89 and/or the extent to which it perceives the ruling government's central authority to be in decline.Footnote 90 In addition, governments may be more or less willing to fight against the insurgency on the basis of relative capabilities as well. The government's choice to fight the insurgency (as opposed to ignoring it) is likely based on the economic, political and/or psychological value of the territory in question.Footnote 91
Conclusions and recommendations
Civil wars have unfortunate implications both for the states embroiled in them and neighbouring countries. Even after a conflict terminates, its negative repercussions continue to affect people and nations in close proximity. Therefore, understanding the determinants of civil conflict is of growing theoretical and policymaking importance. For policymakers who want to reduce the risk of domestic violence by devising prevention tools and bring a successful end to unrest by applying the right mechanism, it is critical to understand the origins of conflicts. The good news is that empirical literature on civil conflicts reveals significant advances, especially since the late 1990s. As a result, our theoretical and empirical understanding of civil conflicts has progressed tremendously.
Throughout this process, Collier and Hoeffler's theoretical dichotomy has played a critical and highly constructive role. They not only set the stage for the future theoretical and empirical development in the field by offering a conceptual division, but also started a very important and lively debate. Over the years, scholars have extended the debate by pointing out advantages and limitations of the dichotomy and by testing new relationships and analysing the implications of results. Some scholars find the categories of opportunity and grievance very helpful for purposes of comparison and even declare victory for one perspective. Others suggest that that the distinction is a ‘crude and unhelpful juxtaposition’ and ‘the conceptual division between greed and grievance is not in fact terribly useful, either in explaining the motivation or persistence of civil wars’.Footnote 92 In other words, the dichotomy is rejected by some because they see it as requiring the choice of one side over the other and leading to reductionist applications.
We argue that the juxtaposition definitely has sharpened the debate and brought significant insight into the determinants and conduct of civil conflict. However, we also identify limitations of the dichotomy and issues associated with the perception of opportunity and grievance as competing explanations: ‘Separating the explanations of civil war into “greed versus grievance” has imposed an unnecessarily limiting dichotomy on what is, in reality, a highly diverse, complex set of incentive and opportunity structures that vary across time and location’.Footnote 93 Note also that ‘protest and rebellion are consequences of complex interactions among collective experience, normative commitments, contention for power, and strategic assessments about how best to promote individual and collective interests’.Footnote 94
Furthermore, we argue that motivation and opportunity should be perceived as jointly necessary conditions rather than alternative explanations that are mutually exclusive. To understand the complex dynamics behind civil wars, it is essential to account for the full set of underlying motivations at different levels. On the one hand, without incentives it would be impossible to overcome the collective action problem and unite people toward such dangerous ends. On the other hand, the causes of resentment, like social isolation or discrimination, are not assumed to hit everyone with equal force, so deprivation will not always lead to violence or civil war. The nature and the magnitude of frustration and availability of various means determine the response. For groups to pose a serious threat to the state, they need to have resources, internal or external support and a strong sense of unity. This is why a synthesis is essential and reductionist approaches cannot account for the complex dynamics of civil wars. We need a model of civil strife that incorporates motivations (sources of discontent), identities, mobilisation and the process of collective action and accounts for the interplay of these aspects at various levels. Therefore, instead of speaking in ‘either-or’ terms, the priority should be to synthesise the insights gained from both approaches and develop an integrated, comprehensive approach that can account for complex conflict processes.
We also argue most emphatically that association of a concept, especially a multifaceted one like state capacity, with one theory – either opportunity or grievance – is not helpful. Like other factors noted above, state capacity can induce or inhibit violence through a number of different mechanisms. It is possible that a factor can contribute to the opportunity structure and be a source of motivation at the same time. Instead of trying to attribute these variables to a certain theory and declaring ‘victory’ over its rivals, we need to account for direct, indirect and interactive impact on the risk of civil conflict. This points to the necessity of examining underlying causal mechanisms behind various associations though carefully crafted case studies. Furthermore, categorisation of variables as political, demographic, institutional, social and geophysical might be useful because these designations are more nuanced and do not assume a connection to a particular theory. A scheme of organisation, in sum, should facilitate, not hinder, research that establishes linkages to both sides of the ‘Great Divide’ over grievances and opportunity.
We argue, in addition, that grievance should be replaced with the more encompassing concept of willingness, which is associated closely with all types of motivations behind collective action, if the dichotomy is going to be utilised further. Credited earlier to Starr and various associates,Footnote 95 the concepts of opportunity and willingness, taken together, more thoroughly combine motivations for collective action leading to civil war. Opportunity replaced greed some time ago. The suggestion now is to make an analogous replacement because grievance is not the sum total of willingness. Grievance as a term is associated with a specific type of motivation, which is closely linked to notions of identity, disadvantage and discrimination. It is perceived to exclude other types of motivations, including pure material benefits, the pursuit of power and survival as in the case of Uganda, etc. Discussion of specific factors in the preceding review should make it clear that opportunity and willingness are more appropriate to describe the full set of factors identified in research so far.
Along with underlying motivations and opportunity structures, mobilisation potential, which depends on characteristics of the group – size and level of mobilisation, salience of shared identity and cohesion – can play major roles in generating civil conflict and should not be neglected: ‘it is difficult to motivate groups to fight one another without historical grievances even when valuable resource rents are at stake’.Footnote 96 When activated by collective grievances, a community's mobilisation potential can help to overcome collective action problems. This, in turn, determines the resilience and overall strength of collective action.Footnote 97 Gurr asserts that ‘[i]f peoples’ grievances and group identity are both weak, there is little prospect for mobilization by any political entrepreneurs in response to any external threat or opportunity.’Footnote 98 However, existence of a strong group identity and deep-seated grievances, as in the case of black South Africans and Shi'i and Kurds in Iraq, ‘provides highly combustible material that fuels spontaneous action whenever external control weakens. The combination animates powerful political movements and sustained conflict whenever it can be organized and focused by group leaders who give plausible expression to minority peoples’ grievances and aspirations.’Footnote 99 All of these factors explain differences among the responses of various groups that are discriminated against and suffer from repression in various parts of the world.
Beyond these general observations, we offer more specific recommendations for future research. It makes sense to enumerate these ideas within respective methodological traditions, since the field generally is organised in that way and the practice is likely to continue.
With regard to quantitative studies, the priority should be multi-equation modelling. Separate equations should be developed for mobilisation and escalation to civil war. The first equation is intended to answer the ‘why?’ question, while the second would focus on ‘when?’ regarding civil strife. Numerous models should be estimated to discover which factors are more relevant to willingness (that is, what had been labelled as grievance), opportunity (what had been designated as greed), or both. These models would be expected to incorporate existing knowledge about functional form among the linkages, such as the curvilinear connection involving repression and violence. Modelling should begin with variables that already have demonstrated staying power, but not to the exclusion of potentially valuable new ideas. One example is the experience of natural disaster, a factor that might impact upon both opportunity and willingness. Research on natural disasters and domestic conflict is in its infancy, but preliminary results are encouraging and further work would make a welcome addition to the modelling enterprise on civil war.Footnote 100
Our approach to civil war is consistent with greater systematic recognition of multiple theoretical pathways for certain variables as they impact upon the likelihood of violence. Thus we recommend more game-theoretic work on strategic interaction between and among governments and rebels. New work should focus on equilibrium analysis in sequential games with varying information conditions among governments, rebels and third parties. This research can build on existing game matrices that produce equilibria under the assumption of complete information among players.
With regard to qualitative research, Appendix A creates a natural agenda for future work. The cases listed in Appendix A are those that cannot be explained within the framework of Collier and Hoeffler. Additional factors should be explored within this set of challenging cases. Qualitative research also can focus quite naturally on anomalies that emerge from ongoing quantitative research.
This review of the research enterprise on civil war has attempted to identify the basic characteristics of work so far in order to derive insights for future efforts. The framework of opportunity and willingness seems like the best option for future theorising about the multiple pathways toward civil war. Within that context, case studies, data based research and strategic analysis all can play a significant role in producing greater understanding of civil wars as the most lethal events in the contemporary world.
Appendix A. Disconfirming cases and the Collier-Hoeffler ModelFootnote 116
