In the coming weeks, Reinventing Peace will feature a number of reflections on patterns of violence in Somalia that stemmed from our recent seminar on the topic. We kick off this series of memos with the feature below by Lidwien Kapteijns.
History of the Project: Stage One
This project started as research into Somali popular culture about Somali civil war violence. This study led to the insight that the most ‘prestigious’ and ‘legitimate’ mediations of civil war violence, that is to say, men’s words in genres that could be legitimately performed in simultaneously shared Somali public space, largely did and could not articulate who did what to whom in the civil war. The discovery of this aporia then led me to the more conventional historical tasks of analyzing, documenting, and contextualizing the civil war violence that marked the collapse of the Barre regime and the Somali state.
Analyzing, Documenting, and Contextualizing the Campaign of Clan Cleansing: Stage Two of the Project
My book is a study of the changing use of large-scale clan-based violence against civilians as a political instrument between 1978 and 1992 and argues that the clan cleansing campaign of 1991-1992 represented what I call a key shift that became the immediate cause of the collapse of the state. It acknowledges the relevance of earlier Somali history and presents the increasingly violent and divisive policies of the military regime of M.S. Barre (1969-1991) as crucial causal factors (Chapter Two). It also traces the War of the Militias, triggered in response to the clan cleansing, during which clan-based violence against ordinary people became normalized practice. However, the book’s major contribution lies in the documentation and interpretation of ‘the ruinous turn of 1991’, as well as, I hope, a conceptualization not just of why it may be necessary to speak truth about this past but also how this might be done without redrawing the very battle-lines of 1991 in a war of words and competitive memory-making today.
Why 1991 as key shift?
First, 1991 marked the first time that politico-military leaders used large-scale clan-based violence against civilians as a political instrument outside of the institutions of the state.
Second, 1991 was the first time that politico-military leaders did not only target civilians as victims of clan-based violence but also incited and organized civilians to become perpetrators of such violence. I see the moment at which leaders outside of the framework of the state tied their civilian followers to them by making them kill civilians of other clans, i.e. the purposeful incitement to and perpetration of communal violence, as a second aspect of 1991 as key shift. As the testimony of civilian survivors of the clan cleansing campaign clearly shows: they were hunted down intentionally because of their group identity and even by name by people who knew them well – precisely by people who knew them well. Meanwhile the top henchmen of the Barre regime whose clan backgrounds fitted the genealogical construct with which the USC associated itself were not just spared but welcomed into the political fold as heroes.
Third, 1991 marked the moment of an unexpected and abrupt reversal of the axis along which civil war violence (including the political use of large-scale clan-based violence against civilians) had occurred until now. Until this moment, the political dividing line had been between military dictatorship and opposition fronts, and large-scale violence against civilians had been meted out by the government against civilians associated because of their clan backgrounds with the armed opposition fronts. In 1991, the front that marched on Mogadishu and drove the dictator from the capital (as well as elements within the military regime itself), drew a new line – one based on clan. This meant that the opposition front that overthrew the military government in Mogadishu included and welcomed with open arms those die-hards of the regime who happened to be of their clan, while targeting for death and expulsion not only regime stalwarts but also tens of thousands of ordinary people who had themselves been the direct victims of the regime but who were now – simply because of their clan backgrounds – targeted for terror warfare and expulsion.
Fourth, Recent work in the fields of new genocide studies and the anthropology of violence have shown that silences, misrepresentations, and denials have been an integral part of acts and campaigns of genocide and ethnic cleansing – so much so, I argue, that they become part of the diagnostic of such violence and become of crucial importance to documenting such episodes and providing insight into their nature and contexts. In the book, I document such silences, denials, and purposeful distortions of 1991. I refer to accounts that simple skip the campaign of clan cleansing and go straight from the expulsion of Barre from the capital (January 26, 1991) to the war between USC-Caydiid and USC-Cali Mahdi April/November 1991). Stanley Cohen, in States of Denial: Knowing About Atrocities and Suffering, has a name for such denials that often accompanies genocides and speaks in this context of ‘literal,’ ‘factual,’ or ‘blatant’ denial, and even of ‘the classic cover-up’ (2001: 7, 138).
Almost equally abundant are accounts that characterize what happened in 1991 as revenge killings exacted by the clans that had allegedly been victimized by the military regime from the clans that had purportedly been its supporters and beneficiaries. By uncritically accepting clan as the relevant category of analysis and the logic of revenge, these accounts imply that the victims were punished for wrongs they had actually committed. First, they sidestep any attribution of responsibility to politico-military leaders who mobilized and organized ordinary people to commit violence in the name of clan. Second, they conceal the fact that those targeted largely consisted of civilians who had as little benefited from the brutal regime as Somalis as a whole. Thus, like the clan cleansers, they paint whole clan groups with the brush of being supporters and beneficiaries of the regime, Third, like the inciters to clan cleansing, they paint whole clans with the brush of being killers and conceal the fact that even the groups in whose name people were incited to clan cleansing included not just inciters (the warlords) and perpetrators (clan-based militias, mooryaan, and some ordinary people), but also bystanders and even – as my book illustrates – rescuers and saviors. According to Cohen’s typology, denials of this kind fit the category of ‘interpretive denial,’ which ‘ranges from a genuine inability to grasp what the facts mean to others, to deeply cynical renaming to avoid moral censure or legal accountability’ for oneself or others (Cohen 2001: 9). Cohen also speaks of ‘implicatory denial’ and ‘denial of the victim,’ applicable to misrepresentations that do not deny what happened but find nothing wrong with that and imply that the victims somehow deserved what was done to them (2001: 7, 8, 61). This interpretation is at times also still going strong in the Somali context. Denials have been an integral part of the histories of genocide and ethnic cleansing everywhere. I argue that their existence and persistence in the Somali case are therefore not only a powerful diagnostic of the clan cleansing of 1991-1992 but also evidence of how unbewältigt this past continues to be.
Why the harsh and painful term of clan cleansing?
I use the term ‘clan cleansing’ in parallel with the usage of ‘ethnic cleansing’ in international law. In the context of international law, ‘ethnic cleansing’ has been defined by the United Nations Commission of Experts on the war in former Yugoslavia, as ‘rendering an area ethnically homogeneous by using force or intimidation to remove from a given area persons of another ethnic or religious group.’ The Commission described the means used in ethnic cleansing as including: ‘the mass killing of civilians, sexual assault, the bombardment of cities, the destruction of mosques and churches, the confiscation of property and similar measures to eliminate or dramatically reduce Muslim and Croat populations that lie within Serb-held territories’ (Bringa 2002: 204). This definition of ethnic cleansing applies, mutatis mutandis, to the violence of 1991, although the appropriate phrase in the Somali case would be a campaign or policy of clan (not ethnic) cleansing. A non-legal definition would emphasize that those who incite to this kind of terror warfare against civilians, in doing so, also destroy alternatives to their own power and authority. I do not assert that some victims of Somali civil war violence deserve more attention, sympathy, and so forth than others. However, when we conceptualize civil war violence, the turn to communal violence in the context outlined above deserves our attention, as does the question of whether public acknowledgement of the different kinds of civil war violence, including the so often denied and concealed clan cleansing, is crucial to peace and justice.
The question of intent
As Jacques Semelin (2003) has pointed out, it is always difficult for those who want to document past atrocities to ‘prove’ intent. Yet the intent to kill and expel – to conduct a campaign of clan cleansing – on the part of USC-Caydiid is a central part of the argument of my book. Intent becomes undeniable, I argue, because particular actions were taken and atrocities committed in a particular order and following particular patterns of organization, and because these atrocities were instigated and justified by particular discourses of incitement, namely the mythical constructions of history to which also many scholars, accepting these as facts, have contributed. I will briefly comment on aspects of the discursive elements, order of events, and patterns of clan-based, communal violence against civilians that argue for intent.
The discursive elements that argue for intent are analyzed in Chapters Three and Four and include the following: First, Caydiid’s rhetoric – especially repeated reference to the numerical insignificance of the Daarood and to the need to whip those who might survive into subject status (that of raaciye). Second, the broader anti-Daarood clan hate-narratives that helped incite and justify the clan cleansing (especially references to the Daarood as foreigners in the Somali territory from which they were to be cleansed (the accusation of allochthony) and to accusations of ‘one hundred years of domination’), together with the code words that evoked them and were used by USC and SNM leaders as well as ordinary people: faqash, haraadiga Siyaad (is raaciya; ha kala reebinina), badda ku dara; siliggaan geynaynaa, ninkii dhoof ku yimid …, and so forth. Third, the normalization of this discourse of hatred as evident, for example, from Radio Mogadishu but also from song, poetry, and doggerel (as cited in the book).
As for the chronological order of atrocities committed, this is the backbone of the narrative of Chapter Three, which includes events that become relevant to intent when one regards them as they succeeded each other in time. I want to isolate here just one episode of the clan cleansing campaign by USC-Caydiid at the end of February 1991. This surprise night attack on the residents of Gaalkacyo occurred three days after Radio Mogadishu had reported that an SSDF assembly, with representatives from different regions, had met with one of Cali Mahdi’s ministers and decided to accept the provisional administration’s invitation to attend a National Reconciliation Conference wherever and whenever it would be called.
There were therefore no Barre supporters from which this area needed to be ‘liberated’ by General Caydiid nor had there been for over two decades. Moreover, Gaalkacyo was not on the way to Kismaayo or to Gedo, where Barre was hiding out. Caydiid’s attack on Gaalkacyo had thus nothing to do with the war against forces loyal to Barre; it was a stage in the clan cleansing of Daarood Somalis from that large sweep of Somalia General Caydiid and his associates wanted to dominate. Oral accounts of the USC attack on Gaalkacyo in February 1991 express grief and indignation about all victims of the attack (which the group called Concerned Somalis in the diaspora estimated at 500 dead, 1000 wounded and 200 hostages), but especially dwell on the rounding up of the very elders who had promoted and were most likely to promote peace and reconciliation. Thus, clan-based violence against civilians was also, again, used by a warlord to eliminate any alternative to war and warlord domination.
Finally, relevant to the issue of intent are the eyewitness accounts of survivors on which I draw in the book. This is a source on which perpetrators (and those who associate themselves with perpetrators) almost always focus their denial and on which scholars, for complex reasons – from the time of the Armenian genocide under the Young Turks to the expulsion of the Palestinians in 1947-1948 and beyond – try so hard not to depend. The philosopher Marc Nichanian speaks in this context about ‘historiographic perversion,’ because survivors of the Armenian genocide have been and are forced to retell the stories of their experience of genocide time and again and again because their eyewitness accounts are never sufficient to those who demand proof. I have made limited use of such survivor accounts in the book but those accounts I included were to a large extent ‘sources-in-spite-of-themselves’ – accounts I had heard before I had even conceived of the book project. It was from hindsight that I realized that the pivotal point of these vignettes, their raison d’être, was the shocking moment of realization that the individual was targeted because of his or her clan identity and that USC leaders (such as a former Manifesto member and Xasan Cismaan Caato) were actually in charge of the lower-level perpetrators. One may add that these accounts prove that there was a pattern to the ways in which the clan cleansing was conducted, that USC leaders were trying to conceal it as it was happening, and that ordinary individuals of the clan on which the USC based itself were among the saviors.
In the book I outline the broad and complex context in which the clan cleansing campaign became possible such as the political and economic abuses of the Barre regime and its large-scale clan-based violence against civilians; the regime’s active undermining of state institutions; Barre’s refusal to step down and leave Mogadishu, and so forth. Moreover, without the context of war and the outbreak of armed fighting; the break-down of law and order; the histories of underlying regional, class, rural-urban, economic, political, and personal conflicts; opportunity with impunity, and so forth, the clan cleansing campaign might never have happened.
However, it did happen and, while elements of the campaign of clan cleansing had featured in earlier episodes of large-scale clan-based violence against civilians before it (under the military regime) and after it (during the War of the Militias), the combined features of campaign of the clan cleansing make it analytically distinct and mark it as a key shift in the use of large-scale clan-based civil war violence against civilians. These combined features include: its scale, especially its time-span, geographical scope, and numbers of people affected; its nature, that is to say the fact that it was communal violence, incited to and committed outside of the institutions of the state by perpetrators who included many civilians, often knew those they unexpectedly targeted for terror warfare and expulsion well, and intentionally sorted them out from individuals of other clan backgrounds (allowed to go free or join in in the campaign); the intent, both that explicit in the discursive triggers for the clan cleansing (the ‘mythical hate-narratives’ and the code words that stood in for them and served as rationales and justifications for the clan cleansing) and that implicit in the patterns of organization that characterized the campaign’s implementation; the incitement of Somali civilians that mobilized them at nearly the highest possible level of the clan template or genealogical construct, that of Daroodnimo versus Irirnimo, which leaves relatively few Somalis outside of its scope; the lack of acknowledgement of the clan cleansing campaign in shared Somali public space by the different iterations of the Somali political leadership since 1991, many of whom played a role in, stood silently by, or benefited from the clan cleansing; and the active concealment, distortion, and denial of the clan cleansing–from 1991 until today–in many journalists’ and scholarly accounts, reports by human rights organizations, political memoirs and autobiographies, and Somali website comments and editorials.
I do not argue that the divide that opened up as a result of the clan cleansing campaign of 1991-1992 – one that lined up with the opposing genealogical constructs of Daaroodnimo and Irirnimo and Hawiyenimo – is immutable or the most relevant to all levels of conflict in Somalia. However, I argue that this divide continues to underlie current popular mindsets and political contestations about the state in Somalia because of its nature as a key shift (explained above) and immediate cause of state collapse, and the fact that it remains publicly largely unacknowledged and has been and continues to be actively denied by the majority of those who associate themselves with the perpetrators.
How to Bring the History of the Clan Cleansing Campaign into the Present: Stage Three of the Project
How might we go about thinking and talking ‘truth to Somali history’ in a context of a total lack of public acknowledgement of what happened and with the hope of making things better rather than worse? In the book I speak about the three principles I have adopted in bringing this account of the past into the present (Chapter Four). These are: (1) reject false categories of analysis and do not attribute single agency to groups/clans; (2) reject the mythical hate-narratives that provided the rationales for large-scale violence against civilians in Somalia in this era, and (3) make clanship matter and not matter at the same time.
Most of the participants in this symposium have contributed to an analysis of the many causes of the civil war. I have tried to outline those causes but the book’s major contribution may lie in its analysis of what I (after Lieberman 2006) ‘mythical clan hate-narratives.’ I would like to make the following points about this: In my work I have tried to outline a history of the changing uses of clan as a political instrument (Kapteijns 2013 and 2010b), for I do not deny or underestimate the power of clan as a political tool and template, and as a dimension of popular mindsets, networks, and group identities. However, I try to insist on the need to never take the concept of clan for granted and to always analyze its specific workings in their diachronic and synchronic contexts. Thus, while I insist that clans did not kill but that people killed in the name of clan, I also emphasize (after Mamdani 2002) that we must explain why so many people flocked to clan banners in the Somali civil war. In this context, clan-based violence against civilians itself is a powerful motivator, but the neopatrimonial state that purposefully divided and ruled through the manipulation of clan as mindset (Compagnon 1995) represented a crucial stage.
I have come to believe that much of what we think we know about Somali history is the result of purposeful and powerful political spin. What empirical historical realities do concepts such as ‘Majeerteen dominance’ in the era of the civilian administrations, ‘MOD’ during the Bare regime, ‘one hundred years of Daarood domination’ during the clan cleansing really represent if one refuses to attribute single agency to clans? I am not arguing that they are without content; just that we have not studied them.
This brings us to neopatrimonialism and the favoritism of the state (civilian and military) towards its clients. As Compagnon wrote, Barre did not rule for ‘the’ Mareexaan but through them (1995). We know that many Mareexaan benefited from the Barre regime but even more did not. What do we really know about clan-based favoritism? Have we not neglected studying it in any detail because we accepted the principle of collective clan punishment and clan-based political spin, which has also remained largely unexamined? And should one and, given the available sources, can one hope to distinguish between the historical fact and fiction on which mythical clan hate-narratives draw?
When I think about truth and justice, I would like to see a number of research projects go forward, two of which I will mention here:
1. A historical project that examines and analyzes aspects of the mythical group hate-narratives and their precursors and tries to get a handle on fact and fiction. I consider of crucial importance, as I said before, research about what political and economic patronage as an instrument of clan-based divide-and-rule under the Barre regime actually meant. But we may also need to examine the history of the armed fronts in more detail (and is that feasible given the stakes and the secrecy?). If we do not step away from accepting discursive or interpretive ‘collective clan punishment’ as substitute for research and documentation, then we can never speak truth to history and disown and debunk the very stories that facilitated and accelerated large-scale clan-based violence against civilians.
2. A biographical project that documents the lives and political acts of those who played central roles in recent Somali history, whether in the Barre regime, the armed fronts, the clan cleansing campaign and War of the Militias, and large-scale violence against civilians since then. I believe that this should focus on those who are guilty of human rights violations, war crimes and crimes against humanity and are still playing (or aspiring to play) a role in national and regional politics. It should also focus on those who have written about the period of Somali history in which they were major actors but concealed their own involvement, whether sincerely or criminally, temporarily or long-term.
When I think of social reconstruction and moral repair, I do not think that a centralized formal Truth Commission or other centralized forms of retributive or restorative justice are suitable for Somalia. However, a range of projects of historical documentation and commemoration are an integral part of speaking truth to history. They are, in any case, already under way, whether academically trained researchers participate in them or not.
Somalis have, of course, publicly engaged with the violence of the civil war for a long time, and this engagement has taken many forms, from poetry and fiction to academic and journalistic analysis and website commentaries. All these renderings of civil war violence are mediations of this violence; that is to say that they represent and interpret aspects of this violence as they also attempt to intervene in it and shape it; the memory-making about the past in which such mediations engage is often also an imagining of the future (Kapteijns 2010a).
My book, which draws on these mediations, as well as many other Somali and non-Somali sources, also represents a particular mediation of the violence of 1991-1992, namely one using the conventions and interpretive tools of the academic field of history. It is based on the premise that the truth about 1991 exists and can and should be recovered, and that public acknowledgement of the large-scale, clan-based violence against civilians, war crimes, and other gross violations of human rights is necessary for social reconstruction and moral repair.
However, in line with the scholarship about truth and post-conflict reconstruction, I accept what Eltringham concludes in his study about Rwanda: ‘that the past is a contested place and that different interpretations of it should be explored (rather than dismissed) because they reveal what actors hold to be current disparities’ (2004:148, my emphasis). This means that, when it comes to making peace, the narratives of all parties may have to be represented at the table. However, without an acknowledgement that not all narratives are equally valid about the past, such peace-making efforts may not constitute moral repair.
Many Somalis are stuck in the narratives of their own victimization and thus unable to engage (let alone publicly acknowledge or accept some form of responsibility for) what was done if not by them as individuals then in the name of their clan. I believe that these truths must become part of critical memory work that engages in the present the moral freight of what happened in the past. I also believe that that work can best be done by Somalis, gradually, in many different ways, forms, and spaces.
Bringa, Tone. 2002. “Averted Gaze: Genocide in Bosnia-Herzegovina, 1992-1995.” In Annihilating Difference: The Anthropology of genocide, ed. Alexander Laban Hinton, 194-225. Berkeley: University of California Press.
Cohen, Stanley. 2001. States of Denial: Knowing About Atrocities and Suffering, Cambridge: Polity Press.
Compagnon, Daniel. 1995. “Ressources politiques, régulation autoritaire et domination personnelle en Somalie: Le régime de Siyaad Barre (1969-1991).” Ph.D. dissertation, Political Science, Université de Pau et des Pays de l’Adour.
Eltringham, Nigel. 2004. Accounting for Horror: Post-Genocide Debates in Rwanda.London: Pluto Press.
Kapteijns, Lidwien. 2010a. “Making Memories of Mogadishu in Somali Poetry about the Civil War.” In Mediations of Violence in Africa: Fashioning New Futures from Contested Pasts, ed. Lidwien Kapteijns and Annemiek Richters, 25-74. Leiden: Brill.
Kapteijns, Lidwien. 2010b. “I. M. Lewis and Somali Clanship: A Critique.” Northeast African Studies n.s., 1, 1: 1-25.
Lieberman, Ben. 2006. “Nationalist Narratives, Violence Between Neighbours and Ethnic Cleansing in Bosnia-Herçegovina: A Case of Cognitive Dissonance.” Journal of Genocide Research 8, 3: 295-309.
Mamdani, Mahmood. 2002. When Victims Become Killers: Colonialism, Nativism, and the Genocide in Rwanda. Princeton, NJ: Princeton University Press.
Nichanian, Marc. The Historiographic Perversion, translated by and Gil Anidjar, New York, NY: Columbia University Press, 2009).
Semelin, Jacques. 2003. “Toward a Vocabulary of Massacre and Genocide.” Journal of Genocide Research 5, 2: 193-10.
Tagsadvocacy Africa African Union arms trade atrocities AU book review Bosnia conflict data corruption Democratic Republic of Congo Drugs Egypt elections Eritrea Ethiopia famine foreign policy gender genocide human rights memorial intervention Iraq justice Libya Mali mediation memorialization new wars peace political marketplace Re-Framing the Debate Research Somalia South Africa South Sudan Sudan Syria trafficking Uganda UN Unlearning violence US Youth Zenawi