Understanding violence and atrocities in Africa; A case study of Central African Republic (CAR) civil war (2004-present)

                      ‘The devil tried to divide us


On February 7th, 2019, the world watched in relief the signing of the 3rd peace agreement between the Central African Republic (CAR) government and 14different armed groups, currently controlling about 70% of the country, hoping it would allege violence. The situation remains fragile despite the Frenchs’ disarming operation Sangaris of the Seleka and the UN peacekeeping forces which are struggling to impose lasting peace. The habituation of violence in the CAR is making it harder for the conflict to cease. The Christian/Muslim dichotomy and the religious portrayal of the conflict has in reality sustained and nourished it.

There exists a vast and diverse volume of literature tackling causes and roots of civil wars, their consequences and implications, coming from a range of disciplines such as political science, sociology, anthropology and psychology. Approaches to understanding atrocities and violence in African civil wars gained momentum in the 1980’s with literature ranging from ideological to economic and political motivations such as greed, grievance and ancient hatreds (Kaplan, 1993; Collier & Hoeffler, 2004; Keen, 2012; Stewart, 2010; Kaufman, 2006) Recent literature focuses on African weak states and its implications, arguing that the warlord and militia groups are part of the African socio-political and normative structure(Isaacs-Martin, 2015; Tang, 2015). Political and economic development failure lies at the heart of the problem today, but long-term existing social practices also play a role in the perpetuation of atrocities among polarized groups (Lombard & Batianga-Kinzi, 2014).

This essay will critically assess the relationship between polarisation and the perpetration of violence in the case of the Central African Republic, on both an affective/psychology and structuralist/materialist account. This essay will argue that group-identity and the religious/ethnic divide between Christians and Muslims was not the main outbreaking cause of conflict, though tensions existed before the war, but rather a consequence of the evolution of it, in a certain given opportunistic time and space. Therefore, this essay will not focus on the causes of the civil war but rather will seek to explore both rationalistic and culturalist approaches as to how group polarization and group identity attenuate the perpetuation of atrocities and conflict in the CAR.

As a way to contribute to the already extensive and thorough body of literature on the top, this essay will also include some content analysis retrieved from interviews of victims of violence and atrocities in the CAR civil war found at the Visual History Archives database.

Civil war context

The Central African Republic is a landlocked country bordered by Chad, Sudan, the DRC, South Sudan and Cameroon, and has been subject to external influence since its independence in 1960 from the French. Little attention is paid to this small country of about 5 million inhabitants, despite the bloody civil war that is on-going. The CAR has seen coup d’état after coup d’état and political instability for two decades.

Francois Bozzizee, who took power in 2003, was finally ousted out by the Seleka armed group in 2013, leading to the leader of the Seleka coalition, Michel Djotodia, to proclaim himself president. By then, mass atrocities had been committed against civilians by the Seleka, most notably Christians in favour of Bozzizee. The Seleka (‘the alliance’), a coalition of Chadian, Sudanese and CAR soldiers from the north-east, rose in response to the oppression of the Muslims and in the aim of overthrowing Bozzizee. Djotodia ultimately failed to dissolve the Seleka armed group that got him into power and therefore put an end to the continuing violence between the former and Bozzizee supporters.

The anti-balaka consequently emerged as a self-defence group comprised of mostly Christians, aimed at ousting the Seleka from the country. Fighting, in reality, has been prevalent in the CAR between government and rebel forces for the past twenty years. The government has had a tendency to focus its socio-economic efforts in the capital city of Bangui and surrounding areas at the detriment of vast areas such as the north-east. After the Seleka took power in Bangui in December 2012, the anti-balaka took arms, and gained control of Bangui a year later in December 2013. Over the past decade, militarisation has amplifiedin the whole of the central African region that has further fostered a climate of insecurity, uncertainty and risk among the millions of people living there.

I. Polarisation of ethnicity and religion: the enmeshment of group-identity in CAR and its link to atrocities

While research suggests that high levels of social, religious or ethnic polarizations fail to explain the outbreaks of civil wars and violence, (on the contrary, ethnic diversity is a deterrent rather than a cause of civil war) (Collier & Hoeffler, 2004), it nonetheless facilitates our understanding as to how and why it contributes to the perpetuation, variation and intensity of violence in the case of the CAR (Harkavy & Neuman, 2001; Esteban & Ray, 2008; Dutton, Boyanowsky, & Harris Bond, 2005). Polarisation entails a theory of deep divisions between groups and in some cases pre-existing social conditions than render violence more atrocious, but in the case of the CAR, cleavages both existed and were formed after the break of hostilities, while deep polarization didn’t necessarily precede the war and wasn’t necessarily what sparked the conflict, “war does not erase the social and cultural patterns existing in the places where it occurs, but rather transforms and adapts them (Lombard & Batianga-Kinzi, 2014).

The on-going civil war in the Central African Republic has been misleadingly described as a heinousreligious war between Christians and Muslims. Similarly to ethnic violence, religious and ethnic violence in Africais considered to be especially atrocious in its very essence, as opposed to conventional western ‘impersonal’ warfare, reducing “the cruel and brutal physical contact with the enemy.” (Trinquier, 1964). The mediatic focus on atrocities committed by the Seleka and anti-balaka towards each other and civilians in the following enactment of the civil war over the years has attributed a religious dimension to the conflict which isultimately not a cause but rather consequential of the ‘politicization of religion’ (Lombard & Batianga-Kinzi, 2014). What began as a politico-military alliance in the aim of overthrowing an established power turned into a war between two religions. It is far more complex than that, but what is certain is that polarisation, the sum of antagonisms between individuals belonging to a small number of groups that simultaneously display high internal homogeneity and high external heterogeneity(Kalyvas, Lange, & Bates, 2006; Esteban & Ray, 2008), has and continues to play a major role in the violence.The resulting categorisation of this conflict between two distinct entities aggravated and steered group-polarization and group/ethnic-identity dynamics, reducing the conflict to two visible entities (Knoope & Buchanan-Clarke, 2014).

Polarisation typifies the antagonisms, emotions and(sometimes) actions of a certain group and comes to characterize the identity of a group and the atrocity infringed on the out-group (Kalyvas, Lange, & Bates, 2006). This ‘identity shapes and structures violence and atrocities, “identity is what gets the blood boiling, what makes people do unspeakable things to their neighbours. It is the fuel used by agitators to set whole countries on fire. When the world is reduced to a battle between “us and them” … only mass murder will do, for “we” can only survive if they are slaughtered.” (BurumaIan, 2002) But group polarisation does not always lead to violence. In the case of the CAR, tensions did exist before the war between existing communities, but really intensified once the Seleka came into power, where ‘structural violence’ could play out.

This group-polarisation, Christian versus Muslims, led to further enmeshment of the phenomenon on a large scale all across the country. In 2013, the RJ (Revolution Justice), mostly composed of former Christian guard members of former president Patasse, and the 3J (Retour, Reclamation et Rehabilitation), aimed at protecting Muslim civilians, were founded in addition to the FPRC: Front Populaire pour la Renaissance de la Centrafrique) founded in 2014 (headed by Djotodia), and MNLC (Mouvement National pour la Liberation de Centrafrique) founded in 2017, composed mostly of Muslim fighters from CAR and Chad. This coping mechanism called “boundary-activation, occurs when a perceived threat between two identifiable groups (Seleka vs anti-Balaka), intensifies in-group and outgroup identity. Religion figured as the main constituent of the hegemonic frame for group polarisation in the violence.

Each of these groups and subsequent factions are composed of certain ethnic populations that has turned and twisted this politico-military conflict into an ethnic-religious one, with political factors mobilising social identity and polarisation as part of an overall strategy. Of course, political and historical factors additionally shape and influence the formation of groups and the selection of a target group (vice versa), and target groups are often designated based on historical relevance. As violence progresses, the number of perpetrators spreads and the selected target group increases the range of its membership acceptable for purging. (Dutton, Boyanowsky, & Harris Bond, 2005). Polarisation can therefore be considered as a cause and/or result of violence.

II. Group-polarisation and emotions

Conflicts along ethnic lines are often viewed as likely to turn exceptionally cruel and brutal. Psychology comes to play an important part in understanding group influence and behaviour. As polarisation facilitates cohesion within a group based on the common assumption’s members have of the group (ingroup) and outsider groups(outgroup), this in turn acts as a social facilitator, where de-individuation occurs, and groups perceive each other on different moral dimensions(Myers, 2011). The core paradigms of the group-think itself intensifies over time creating, in the case of the CAR, a volatile environment propitious to a cruel spiral of escalating violence and atrocities.

Emotions such as for example, grievance, vengeance, shame, guilt, resentment, hostility, jealousy, envy, embarrassment, and disgust surge amongst the specific environment of the group, “violence is inherently emotional (Lombard & Batianga-Kinzi, 2014).Group-think, group-identity and group-action are key to understanding how dehumanisation of the other relates to the intensity and nature the atrocities. The ‘reversal of morality’ (Staub, 1999), and therefore reverse in previously accepted norms, collapse and come to influence types of atrocities committed against the targeted other. This reversal of morality in intricately linked to the societal anomie (DiCristina, 2015) the CAR has been facing for the past twenty years.

The way group-identity dynamics is intricately linked to the atrocities committed is interesting in this case: polarisation allows for dehumanisation between groups to occur. The out-group or “other” is subject to a different moral system than to that of the in-group, resulting in moral splitting and moral exclusion (Staub, 1999). In other words, it’s all about belittling, demeaning and patronising, so as to completely dehumanise “the other”. While social groups are dehumanised or “otherised,” emotions intensify over time, which can sometimes lead to ‘continuum of destruction (Staub, 1999), where intergroup hostility goes from moderate to severeexterminatory aggression’ (Staub, 1999).

Over time, due to the number of internal and external factors that come into play, the conflict takes on certain dynamics and the focus on the ‘other’ can sometimes become stronger. Endogenous polarisation(Kaufman, 2006), in the case of the CAR, in addition to being a convenient and efficient tool for politicians to gain support, obliges the civilians not directly involved in the conflict to at some point identify themselves on an ethnic or/and religious basis as an instinctive and coping survival strategy towards threat and the fear for their own lives. Endogenous violence results from shifting prejudices within the dynamics of the conflict over time and the shaping of identities are constructed in relation to the wider socio-political normative structure. In fact, civil-war victims of the CAR seem to agree that nobody expected the seemingly political war to turn into a religious/ethnic one. Civilians have a tendency to describe this conflict in binary terms such as Christian versus Muslim, which shows just how in times of heightened violence, cultural practices facilitate categorisation, classification and therefore polarisation. Out of the 4 witnesses’ testimonies on the Visual History Archive, they all describe the violence as inherently between two binary groups and all refer to the anti-balaka and Seleka when asked to describe theconflict, as if they felt the need to position and situate themselves in all the chaos. Ironically enough, they also all evoke the fact that Christians and Muslims lived in peace prior to the conflict.

This type of endogenous polarisation (Kaufman, 2006), whereas the distribution of popular support during the war is an accurate reflection of (pre-war)cleavages, in times of high polarisation, the ‘carrying capacity’ of political actors reaches its maximum valueand, “many societies may experience high levels of group polarisation – a mass phenomenon where emotions such as resentment, fear and hostility between groups run high – but there may be little or no violence if there is no structural or material opportunity to act on these sentiments.” (McDoom, 2012) In the case of the CAR, the socio-politico structural environment has encouraged and promulgated polarisation, which coincides with research indicating that “the introduction of meaningful differences between groups in resources, status or power” alter the degree of group identification.(McDoom, 2012)

The ‘symbolic politics theory’ claims that three pre-conditions are necessary for ethnic/religious based conflict; (1) symbolic myths justifying hostility toward an ethnic group, (2) an opportunity to mobilize politically against a group and (3) ethnic fears (Kaufman, 2006). In the case of the CAR, while religious conflict did not characterize the pre-war times, tensions have existed between these different groups over history whilst thethird condition of ethnic fears has certainly spiralled the conflict between the two religious’ groups. Violence can produce/alter new fears, resentments, hostility and new ‘psychocultural narratives’ based on previous historical narratives can intensify conflict due to the strong emotional component of identity discourse (McDoom, 2012).

How do humans go about feeling such strong emotions of those enumerated above, to go as far as committing atrocities against the other, and what coping mechanisms are necessary to do so? In the process of polarisation and therefore otherization, dehumanisation, classification and stereotyping of the ‘other’ out-group, symbolic representation(s) of each other are forged. Most importantly, the other becomes an existential threat to the other. In the case of the CAR, the anti-balaka emerged as a result of the Seleka rebel group killing Christians in the country, as a self-defence group aimed at protecting Christian communities and the other groups surfaced in reaction to the atrocities and therefore security threat, “the power of security threat lies in the power of group emotions, notably the core emotion of fear. Emotions activate psychological processes at the group level that lead to the polarisation of intergroup attitudes.” (McDoom, 2012)


According to research, emotions such as fear, threat and disgust come to be important factors in intragroup behaviour vis a vis the out-group, as a means of distancing themselves from the perceived threat (Dutton, Boyanowsky, & Harris Bond, 2005; Brubaker & Laitin, 1998; Esteban & Ray, 2008; Fearon & Laitin, 1996; Horowitz, 2001; Petersen, 2002), and emotional disinhibiting them. During the ‘otherization’ process, the sum of common or antagonistic sentiments and emotions amongst the individuals composing that group and psychological processes such as these come to facilitate violence. The “power of group emotions” such as fear, produce group violence, when material and structural opportunities permits it to be expressed as violence (McDoom, 2012), and violence in return may produce additional hostile emotions, which may or may not lead to violence itself, which were previously not necessarily deeply felt.Atrocities depend on a range of social and environment factors such as historical components and relations between intergroups, leaders’ behaviour, economic and social development, political and economic instability, the presence of threats (Fein, 1993). These factors all played a role in the case of the CAR and the polarisation between two identifiable groups.

The anti-balaka seem to have clearly emerged in response to the Seleka killing Christians and accuse the Seleka of a certain ‘otherogenic’ trait, of being a foreign group invading CAR from neighbouring Chad, which comes to constitute the premise of the ingroup’s outgroup negativity, increasing the likeliness of aggression (Taylor, 2006). Violent actions on behalf of the Seleka, which represents the minority of Muslims, on Christians, which constitute the majority of the population, polarises attitudes and behaviours of the majority (McDoom, 2012). The Seleka in turn emerged in response to their perceived oppression of the Muslims in the country and due to socio-economic inequalities. In any case, the emotion of anxiety and fear stemming from the power of insecurity and/or threat of extermination (Horowitz, 2001), seems to be inherently linked to the intensity of violence perpetuated. Both realistic and symbolic threats polarize groups and the more the threat, the more violent conflict will be. According to Louisa Lombard, social anthropologist having spent the past 10 years studying the civil war, this is precisely why it is so difficult for the atrocities to cease; this spiral of group polarisation can go on and intensify over large periods of time.

Consequently, the in-group cohesion based on in group bias comes to blur the distinction between actual armed fighting entities and civilians. The Seleka’s in group bias was such that all Christian and ethnic affiliated civilian outgroups were considered asthreatening, infectious and bound to be killed. The same goes for the anti-balaka, who emerged as a result of a perceived threat, and who in turn retaliated largely on ethnic Muslim civilians as a result of outgrouphomogenization. This tendency of generalising all Muslims, including civilians, as the target group, and vice versa for the Christians, reveals the generalising component of polarisation and how indispensable this process of extreme otherization is necessary for groups to commit atrocities on a humanistic level. A ‘social death’ must occur and that is done through historical psychocultural narratives and the use of discourse, like for example describing each other as potentially infectious and dangerous, “we must kill all Muslims.”

Interestingly in the CAR, people see the world as both possessing an occult, invisible side and a visible reality. Some are considered to take advantage of the dark occult world of the dead to empower themselves, which is ultimately seen as a threat. The anti-balaka for example wear animist accessories believed to shield them from bullets, death and occult ‘Seleka-Muslimforces, which furthermore strengthens in-group cohesion and outgroup negativity. The targeted group becomes both a moral and physical threat, endangering the opposing group, in which the sentiment of disgust plays an important role.

This perception of the world as having two worlds is directly linked to the atrocities committed in the case of the CAR: people believe that life continues after death and that the dead can continue to act in this occult world and avenge themselves. This may explainwhy the intensity and degree of violence is such: for the dead not to be able to avenge themselves, the corpse must be severely destroyed for it not to live on to the afterlife, “only by truly disaggregating a body – through desecration or tearing it limb from limb, or burning the corpse – can one finally extinguish the person’s ability to wreak vengeance from beyond the grave.” The belief in these two opposing worlds comes to play an important factor in the way these groups perpetuate atrocities, and reflects the way in which threats are handled in the CAR, as a ‘tool of management of threats’ (Lombard & Batianga-Kinzi, 2014), as a response to the anomie of society due to the lack of governmental presence in certain areas of the country, disorder and feeling of endemic insecurity, “violence has long been a central element of popularized modes of governance in the country” (Lombard & Batianga-Kinzi, 2014) and has become a popular mode of social control, way before the Seleka even came to power in 2012. Popular punishment and vengeance in the CAR over the past twenty years facilitated the perpetuation of atrocities at a given time, when conditions were fully ripe. These practices of violence that pre-date the war is a mode of action suited to the fractured solidarities of dislocated urban life”. (Bonhomme, 2012).


What is interesting in the case of the CAR civil war is that it embodies a case study capable of bridging the gap between the structural/materialist/instrumentalist and the culturalist/psychological debate on civil wars. These two seemingly contradictory accounts actually complement each other; while rationalist theories allowus to understand the socio-political context that permitsfor violence to be expressed, the affective/emotional theories allow us to understand the psychological concepts, processes and how emotions and dehumanisation influence forms and degrees of violence. Polarisation is useful in understanding how and why violence occurs, but also how highly polarised conflicts are, “susceptible to a loss of control, so that people end up doing things they later regret or do not understand (Lombard & Batianga-Kinzi, 2014). But polarisation has a tendency to generalise the conflict in terms of two groups, marginalising the micro-level analysis of it and tends to ignore the diffused aspect of violence. A diverse range of internal and external factors influence the way a conflict and subsequent violence expresses itself and as we have seen in the case of the CAR, there are many.

Social control mechanisms such as popular punishment and vengeance pre-dating the war also facilitated the perpetuation of violence. The gradual militarisation of the region and increasing insecurity and lack of governmental control has created a threatening environment propitious to high polarisation, though polarization does not always lead to violence, even if the structural opportunity presents itself. Fear in the case of the CAR however, did and still continues to affect the intensity of polarisation and intensity of violence, “among the strongest feelings people express are fears about physical attacks on their group, and on symbolic attacks on its identity… both fears involve feelings of vulnerability, denigration and humiliation that link past losses to present dangers (Ross, 2007). Ways in which people perceive threats and the way they handle them in a place like the CAR where government authority is weak, has enmeshed violence in everyday life as a means of dealing with threats and enforcing a sort of unwritten legal code. To prevent outbreaks in the future, the pervasive socio-economic and political insecurity amongst the population must allege. If not, high polarisation in some societies where structural conditions are ripe can lead to ‘high levels of hatred and compulsions to vengeance with the end result of brutality, massacres, and various forms of savagery(Harkavy & Neuman, 2001).


Bonhomme, J. (2012). The dangers of anonymity: Withcraft, rumor and modernity in Africa. Hau: Online Journal od Ethnographic Theory, 2(2), 205-233.

Brubaker, R., & Laitin, D. D. (1998). Ethnic and Nationalist Violence. Annual Review of Sociology, 24, 423-452.

Collier, P., & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford Economic Papers, 56, 563-595.

DiCristina, B. (2015). Durkheim’s theory of anomie and crime: A clarification and elaboration. Australian & New Zealand Journal of Criminology, 49(3), 311-331.

Dutton, D. G., Boyanowsky, E. O., & Harris Bond, M. (2005). Extreme mass homicide: From military massacre to genocide. Aggression and Violent Behaviour, 10, 437-473.

Elbadawi, I., & Sambanis, N. (2000). Why Are There So Many Civil Wars in Africa? Understanding and Preventing Violent Conflict. Journal of African Economies, 1-24.

Esteban, J., & Ray, D. (2008). Polarization, Fractionalization and Conflict. Journal of Peace Research, 45, 163-177.

Fearon, J. D., & Laitin, D. D. (1996). Explaining Interethnic Cooperation. American Political Science Review, 90(4), 715-735.

Fein, H. (1993). Genocide: a sociological perspective .London: Sage Publications.

Harkavy, R., & Neuman, S. (2001). Strategies, Operations Tactics in Low-Intensity Conflict. In Warfare and the Third World (pp. 189-232). New York: Palgrave Macmillan.

Horowitz, D. L. (2001). The Deadly Ethnic Riot. California: University of California Press.

Isaacs-Martin, W. (2015). The Motivations of Warlords and the Role of Militias in the Central African Republic. Conflict Trends(4), 1-4.

Kalyvas, S. N., Lange, P., & Bates, R. (2006). Logic of violence in civil war. Cambridge: Cambridge University Press.

Kaplan, R. (1993). Balkan Ghosts: A Journey through History.New York: Random House.

Kaufman, S. J. (2006). Symbolic Politics or Rational Choice? Testing Theories of Extreme Ethnic Violence. International Security, 30(4), 45-86.

Keen, D. (2012). Greed and Grievance in Civil War. International Affairs, 88(4), 757-777.

Knoope, P., & Buchanan-Clarke, S. (2014). Central African Republic: A Conflict Misunderstood. Retrieved from Clingendael : Netherlands Institute for International Relations: https://www.clingendael.org/sites/default/files/pdfs/CAR_Report.pdf

Lombard, L., & Batianga-Kinzi, S. (2014). Violence, Popular Punishement and War in the Central African Republic. African Affairs, 52-71.

McDoom, O. S. (2012). The Psychology of Threat in Intergroup Conflict: Emotions, Rationality, and Opportunity in the Rwandan Genocide. International Security, 37(2), 119-155.

Myers, D. G. (2011). MYERS’ PSYCHOLOGY FOR AP. New York: Worth Publishers.

Petersen, R. D. (2002). Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe. Cambridge: Cambridge University Press.

Richard, J. (2002). Violent International Conflict and the African Statr: Towards a Frameowkr of Analysis. Journal of Contemporary African Studies, 20(1), 29-52.

Ross, H. M. (2007). Cultural Contestation in Ethnic Conflict.Cambridge: Cambridge University Press.

Staub, E. (1999). The origins and prevention of genocide, mass killing, and other collective violence. Peace and Conflict: Journal of Peace Psychology, 5(4), 303-336.

Stewart, F. (2010). Horizontal Inequalities as a Cause of Conflict, A Review of CRISE Findings. Oxford: Centre for Research on Inequality, Human Security and Ethnicity.

Tang, S. (2015). The Onset of Ethnic War: A General Theory. American Sociological Association, 33(3), 256-279.

Taylor, K. E. (2006). Intergroup atrocities in war: a neuroscientific perspective. Medicine, Conflict and Survival, 22(3), 203-244.

Trinquier, R. (1964). A French View of Counterinsurgency.London: Pall Mall LLM