13 Dervishes, moryaan and freedom fighters: Cycles of rébellion and the fragmentation of Somali society, 1900-2000 Jon Abbink There has been a state of near-permanent revolt in Somali society since 1991. This chapter offers a cultural analysis of patterns of political and military activity from the precolonial era through the Italian and British colonial period, state independence (1960-1991), to the present period of statelessness. The focus is on a comparison of éléments in the campaigns of revolt against the colonial states between 1900 and 1920 with those in the late Siyad Barre period (1988-1991) and the era of statelessness (1991 onwards). A transformation of ideas of revolt and violent action has occurred in which Somali notions of egalitarian social order, kinship and local leadership have taken on a different shape. This prevented the institutionalization of crosscutting alliances and the émergence of a wider political arena - except in certain régions such as Somaliland and Puntland. The cultural and social unity of Somali society has always been overestimated. Somali political culture is by nature centrifugal, preventing the institutionalization of a legitimate leadership at state level hut not at a regional clan level. Introduction The Somali remain an enigma. Fiercely attachée! to their independence and rebellious towards any perceived or real outside interférence m their country, their society appears deeply divided and chronically unstable, as exemphfied by the absence of a central state since 1991. Certainly, the decentralized and centrifugal Somali political culture, marked by localized if not fragmented authority structures, is one explanation. Their culture is rooted in a strong segmentary lineage and clan organization based on a predominantly pastoral way of life and guided by a remarkable genealogical self-consciousness' that has persisted through time. The phenomenon of group division based on instrumentalized clan identities, while not explaining everything, goes a long way in accounting for the perpétuai tensions in Somali socio-political organization, as the Somali themselves are the first to admit.2 At the same time, the diversity of patterns of economie ad'aptation and socio-cultural différence within Somali society should not be underestimated.3 Although more than 60% of Somalis are transhumant pastoralists, there is more sedentary settlement, hierarchy and class formation in the southern part of the country, while an urban trader/hustler class is increasing in size as well. In the riverine Digil-Rahanweyn area there is also more clan flexibility in that people have changed their clan affiliation, and many newcomers or migrants have been included in the clan system. In coastal cities there is a distinct sub-culture of traders and urban people of mixed origins. In addition, there are 'despised', low-status caste groups and other small minorities who have always been on the lower rungs of the social ladder and have been disproportionately hit by successive artned conflicts. In recent years, the non-urban and non-pastoral Somali, like the sedentary agricultural Digil-Rahanweyn peoples of the area between the Juba and Shebelle (see Map), the Bantu-speaking (or Jareer) groups,4 the Wa-Gosha and the occupational caste minorities (for example, 1 The gréât value attachée! to personal genealogical reckoning (abtirsiinyo) is proverbial in Somali culture. See for instance, I.M. Lewis, Blood and Bons. The Call of Kinship in Somali Society (Lawrenceville NJ, 1998), 19, 82, and I.M. Lewis, 'Doing Violence to Ethnography: À Response to Catherine Besteman's "Representing Violence and 'Othering' Somalia'", Cultural Anthropology, 13 (1998), 101. See also G. Prunier, 'Segmentante et Violence dans l'Espace Somali, 1840-1992', Cahiers d'Etudes Africaines, 146 (1997), 381. 2 Said S. Samatar, 'Somalia: The Politics of Poetry', Africa Report, 38 (1993) (Online version: www.anaserve.com/~mbali/samatars.htm); see also G. Hanley, Warriors and Strangers (London, 1971), passim. 3 Cf. L.V. Cassanelli, The Shaping of Somali Society: Reconstructing thé History ofa Pastoral People, 1600-1900 (Philadelphia, 1982); V. Luling, 'The Other Somali Minority Groups in Traditional Somali Society', in T. Labahn (ed.), Proceedings ofthe Second International Congress of Somali Studies, University of Hamburg, August 1-6 1983, vol. IV (Hamburg, 1984), 39-55. For a map of the Somali clan families, see www.lib.utexas.edu/maps/africa/somalia_ethnic92.jpg. F. Declich, '"Gendered Narratives", History and Identity: Two Centuries along the Juba River among the Zigula and Shanbara', History in Africa, 22 (1995), 93-122. See 4 -? 330 Abbmk Midgan, Yibir, Tumaal), were mvolved m the civil war predommantly as victims and less as perpetrators 5 Civil war and disunity (1991-2001) have exacted a heavy toll among the Somah population m genera! After the failed war against Ethiopia m 1977-78,6 Somalia, under an increasmgly brutal dictatorship, crept towards oppressive clan-based rule,7 disunity and armed stnfe, with the northern régions becommg the prime victim This led to the death of tens of thousands of people, well over half a million refugees and displaced persons, a dramatic mcrease m clan-based conflict, malnutrition and famine, and economie and environmental crisis 8 As a result, the pohtical structure of a united Somalia unravelled in the 1980s during the regime of President Mohammed Siyad Barre and evaporated after the démise of his regime m January 1991 Smce then, the country has remamed vulnérable to famine, natural disaster and prédation by warlords Civil stnfe agam resulted m vast numbers of human casualties 9 also M Mohamed-Abdi, 'Les Bouleversements Induits de la Guerre Civile en Somalie Castes Marginales et Minorités', Autrepart (Cahiers des Sciences Humâmes N S), 15 (2000), 131-48 In the pre-colomal era, many Bantu-speakmg people were brought to Somalia as slaves 5 See A Haji Mukhtar, 'The Plight of thé Agro-Pastoral Society of Somalia', Review of Afneem Pohtical Economy, 23 (1996), 543-53, L V Cassanelli, 'Victims and Vulnérable Groups in Southern Somalia', Research Directorate, Immigration and Refugee Board Canada Occasional Paper (Ottawa, 1995), Mohamed-Abdi, 'Les Bouleversements', 132 6 See Gebru Tareke, 'The Ethiopia-Somaha War of 1977 Revisited', International Journal ofAfrican Histoncal Studies, 33 (2001), 635-67 7 By Marehan-Ogaden-Darod clan elites * 8 See Hussein M Adam, 'Somah Civil Wars', m T M Ah & R O Matthews (eds), Civil Wars m Africa Roots and Resolution (Montreal, 1999), 180-81, I M Lewis, 'UN Paperclips for Somalia', Horseed, online newspaper [www horseed com/jan/03/24/htm], 3, IRIN news message of 13 June 2002 on the massive destruction of remaimng Somah forests for charcoal for export to Saudi Arabia and the Gulf States See also www padngu gu se/EDCNews/Research/SomahaCharcoal html 9 For a good chronology of developments m Somalia over the past 41 years, see M A Perouse de Montclos, Interprétations d un Conflit Le Cas de la Somah (Pessac, 2001), 7-18 What is remarkable m this chronology is the high incidence of violent events and épisodes For a report on thé dévastation before 1991, see also www somahawatch org/archivefebOl/010202301 htm Dervishes moryaan andfreedom fighters 331 Erigavo _= OMALILAND ISSAQ ÄWARSANGELI Hargeisa / PUNTLÄNß National boundary Provincial boundary ""• Clan famîly boundary ?, : (approximate) o Map 13 l Somah clan famîly distribution WOkm '^ 333 332 Abbink Somali culture and politics There is by now a rather familiär account of the causes of the civil war and the havoc it has wreaked in post-1991 Somalia.10 Political analyses of the Somali drama tend to yield similar stories, referring to socio-political décline and internai division in the absence of a developed state tradition. Analysts only differ in the relative weight accorded either to certain political and historical factors or to the rôle of 'clan identities' used as a basis for political action or abused for personal or factional gain, igniting conflict. These points were reiterated, respectively, by those termed the 'traditionalist' and the 'transformationalist' scholars of Somali affaire.11 Those who aim to transcend thèse two polarized positions12 usually end up with a modified version of the second position, stressing elite private abuse. In my view, the core of the enigma of Somalia's violent décline is not only explained by an appeal to greedy elite compétition or class conflict at the level of state elites. This would not account for why competing factions and fighting groups have assumed such notable clan identities and also judge their opponents in the same way. In addition, the elite is divided and shows a highly shifting basis for the recruitment of supporters and private pursuits. The fact that Somali socio-political organization has lent itself to such monumental abuse and has actively worked against the émergence of any accepted central state authority or durable territorial or other alliances is telling of the peculiarities of and continuities in Somali society, and sets it apart from other societies. 10 See for instance, Adam, 'Somali Civil Wars'; 'Pérouse de Montclos, Interprétations, 19; Prunier, 'Segmentante et Violence'; M. Jama, 'Trajectoire du Pouvoir en Pays Somali', Cahiers d'Etudes Africaines 146 (1997), 403-27; Th. Vircoulon, 'La Crise Somalienne', Afrique Contemporaine, 177 (1995), 3-16. " See Abdi I. Samatar, 'Destruction of State and Society in Somalia: Beyond thé Tribal Convention', Journal of Modem African Studies, 30 (1992), 625-41. 'Traditionalists* are those seen to be putting (too much) emphasis on clan identities and lineage segmentation as thé persistent (and explanatory) basis of Somali politics, while the 'transformationalists' stress social inequality, personal abuse of resources, and exploitative élite or class rule as the motor of divisive politics and conflict. The 'traditionalists' reject thé epithet given to them by their opponents. 12 For instance, M. Jama, 'Sur la Violence en Somali: Genèse et Dynamique des Formations Armées', Politique Africaine, 47 (1992), 147-51; C. Besteman, 'Representing Violence and "Othering" Somalia', Cultural Anthropology, 11 (1996), 120-33. See also the recent PhD thesis by A. Mohamoud, 'State Collapse and PostConflict Developments in Africa: The Case of Somalia (1960-2001)', University of Amsterdam (2002). In this chapter, I do not repeat the conventional explanations of political crisis but instead reconsider the impact of some historical and cultural factors on Somali political and military rébellion and revolt across time. The ideas and practices of (armed) rébellion seem to have dominated Somali society in the past Century, and specifically in the past fifteen years, but the ultimate aims and benefits appear to have been lost along the way. In this respect, cultural premises and the manner in which they were made to interact with the political dynamics of Somalia (as determined by both interna! and external factors) need renewed attention. Cultural and social structural éléments are simply part of the longue durée structure of Somali history. Such a cultural analysis is, however, hampered by thé comparative lack of empirical field research done in Somalia in the past decade, and this chapter can only be a modest attempt towards an understanding of the topic. While conflicts and violence were common features of life in pré-colonial Somali society,13 the émergence of collective revolt can be located in the era of foreign colonization by the Italians and the British since the late 1890s. The response to this external domination was marked by violent ideological and armed rébellion. In the era of state independence (1960-1991) and right through thé period of statelessness (1991-présent) such a violent response surfaced again with renewed and much more destructive force. But it might be said that thé Somalis were not only in revolt against thé above-mentioned external forces but also against themselves, as is évident from persistent patterns of kin-group feuding, régional antagonism or rébellion against indigenous élite groups forming in thé urban centres and through trade networks. Most dramatic perhaps were thé revolts against the independent state that had become deeply suppressive and violent especially in the last decade (1980-1991) of Siyad Barre's rule. More historically and culturally informed explanations for the centrifugal forces that have prevented the Somali nation from successfully organizing its own state structure14 remain necessary. Indeed, the Somali expérience provides the most convincing argument to date that an 'ethnie group' (and one with illdefined boundaries at that)15 is not the best or even a necessary unit for the construction of a 'national state'. In particular, one could ask what the 13 Cassanelli, The Shaping of Somali Society; LM. Lewis, A Modern History of Somalia. Nation and State in the Hom of Africa (Boulder & London, 1988); G. Prunier, 'Segmentante et Violence'. 14 Cf. D. Laitin & Said S. Samatar, Somalia: Nation in Search ofa State (Boulder Co., 1988); G. Schlee, 'Régularités dans le Chaos: Traits Récurrents dans l'Organisation Politico-Religieuse en Militaire des Somalis', L'Homme, 161 (2002), 17-50. 15 Schlee, 'Régularités', 18. 334 Abbmk mitigating or reinforcmg rôle of Somah Islam and cultural precepts related to customary law (or heer) have been in the articulation and perpétuation of conflict in Somalia. A theoretical approach to structure and agency would emphasize mat cultural organizing principles do not disappear under the force of political conjunctures but are brought to play a rôle in the power game.16 While attention will be paid to the intervening decades, two historical episodes stand central in this chapter: the upheaval and résistance led by a notable religiously inspired leader, Mohammed 'Abdulle Hassan17 (from 1899 to 1920, see below), and the period of'clan-faction' violence18 between 1991 and 2001.1 consider these two periods not to assert that they are similar in their way of revolt but to illustrate the continuity and change in the character of armed conflict and résistance in Somali society. In addition, I sketch the transformative impact of new forms of violent revolt on local society. Briefly, the dominant image in the literature is that of a 'Somali identity' in the political sense. The first period was a formative, nationalist one while the second was fragmenting. One of my theses in this chapter is that the mythical image of Mohammed 'Abdulle Hassan and his Dervish movement as 'nationalist' obscures the nature of Somali political identity. Although hè initially appealed to all Somalis in the name of a religious identity, to speak of his rebel movement as advancing a nationalist cause is somewhat anachronistic. The parallels between then and now are remarkable because of the effect of its violent tactics on Somali society. Somali politics has retained an undercurrent of cyclical conflict based on certain principles of kinshïp, ideology and (perceived) material group interests, and in this sense one could say that politics is 'absorbed' or reproduced by Somali culture. Somali cultural values, ideas of personhood and identity, achievement and self-image have changed much less over the past century than the vagaries of political crisis and conflict would suggest. In this sense, the dramatic démise of the Somali state in 1991 was not a sudden and unexpected event but the result of a longer, incrémental process of décline of a political structure that could never do what it was expected to do. I do not wish to succumb to any primordialist view of the force of clanship and segmentary identities as the basis of all politics, let alone as the explanation for recurring factional fighting and social chaos: there is no direct link between the existence of clan identities and their violent and divisive instrumentalization. But certainly it is a fact that the idiom of kin- and clanship, genealogy and segmentation forms the structural basis of compétition and political strife, aided by a territorial dimension of clans and clan families being concentrated in certain areas. As such, this idiom is an excellent vessel for manipulation by elite groups or classes that emerged in the colonial and postcolonial era, allowing for meaningful collectives to be formed as well as for flexible alliances.19 This is mainly because kin- and clanship and segmentation (i.e. situational positioning vis-à-vis others) provide the basic identity that people fall back upon and recognize as defming themselves. This is also true in modern conditions. No one with first-hand knowledge of Somalis and Somali society will underestimate this fact. In Hussein Adam's view,20 Somali political culture is pervaded not only by ideas of kinship and segmentation but also by an ethos of equality, pride and prestige between groups at any level of segmentation, jealously guarding public status vis-à-vis each other. The important point is that people want récognition and reciprocity, even if they are not prepared to give it to others. They also make claims to social justice. Following Hussein Adam, one could say that this characteristic attitude is politically relevant in Somali life, in that clan consciousness is mobilized in the struggle for prestige and justice. Thus, a political-economic view must be developed in dynamic linkage with a culturalist view to explain continuities and change in patterns of group identity and political agency that have shaped rébellion in Somalia. An understanding of what motivâtes the Somali people in socio-cultural terms would help in devising international assistance programmes and sounder political coopération with Somalia. 16 For instance, it could be said that the failure (except in its humanitarian work) of the US-led 'Operation Restore Hope' in the early 1990s was based on a well-intentioned but monumental ignorance of the cultural substratum of Somali society and politics. 'Black Hawk Down', the recent film based on the 1993 armed confrontation m Mogadishu, does not suggest that the Amencans have learnt much since. Sec for a review, http://www.alternet.org/story.html?StoryID=l 2175 17 Hère I follow Abdi Sheik-Abdi, in his excellent book Divine Madness. Mohammed Abdulle Hassan (1856-1920) (London, 1993), 44, by rendering his name in the Somali not the Arabic spelling (which would be Muhammad ibn 'Abdullah Hassan). 18 I use 'clan-faction' mstead of the widely populär but somewhat misleading term 'warlord' violence. 19 This is because most Somali clans allow for filiation on a non-descent basis and have clientship or 'adoption' (sheegad) of people from other clans. 20 Hussein M. Adam, 'Militarism and Warlordism: Rethinking the Somali Political Expérience', African Studies Center, Boston University Working Paper no. 49 (Boston, 1993). Dervishes, moryaan andfreedomfighters 336 337 Abbmk Pre-colonial Somalis Revolt and résistance seem perennial thèmes in Somali history and are often tied to the segmentary nature of Somali society, dominated by patrilineal kinship links (toi waa tolaane - 'agnates are connected') as well äs reconstructed lineage and clan identifications21 of individuals and groups. While the system has a patrilineal bias, the utérine links in the paternal line as well as the links though the mother's clan or lineage are also of vital importance. Somalis traditionally neither knew a state authority nor recognized the primary binding force of an external power beyond the (sub-) clan. The clan/lineage and its nominal authority structure, based on elders and certain chiefs (boqor, ugaaz, garadf2 and buttressed by religieus leaders (wadaaddo, ma'allimiiri) and the structure of the Sufi orders (turuuq)^ formed the bedrock of society. Due to the nature of their mobile economy and social organization, Somalis could (and still can) largely afford to ignore the claims of state or other overarching institutions. An oft-cited comment is that by the Baganda companion of a British soldier travelling for the first time in the Somali area in 1910: 'Somalis, bwana, they no good; each man his own sultan'.24 Somalis still cultivate this self-image of personal pride and independence.25 It should also be emphasized that the Somali social order, while recognizing links through females, is deeply patriarchal, with public, religious and political activity strongly dominated by men. Married women are vocal, economically active and central figures in domestic life and maintain important links with their family and clan of birtb26 but husbands or male relatives should give their approval for what their womenfolk do. •J o 00 -Is . J3 S -'S >, s 1 o .- __0 -C **• o CO 21 Tol means patrilineal kinship linking and is situational as to the actual links stressed. Starting at the lowest level and then going up, Somalis are members of a qoys (family), a reer, njihb, a qolo and a qabiil (this is the most inclusive level, that of the six large clan families). Jilib and reer, and to a lesser degree qolo, can be mag (blood money)paying groups in the case of homicide. 22 These had no meanmgful executive power and cannot be considered on a par with the chiefs known elsewhere in Africa. 23 Ahmadiyya, Qadiriyya and Salihiyya. 24 Cited in Samatar, 'Somalia: The Politics of Poetry'. R.F. Burton's remarkable book First Footsteps in Africa (London, 1856) is replète with similar observations. 25 Gerald Hanley's memoir Warriors and Strangers, about his time as a British officer and admmistrator in Somaliland after 1941, offers a penetrating and delightful picture of this. 26 See Aman, with V.L. Barnes, Aman, the Story of a Somali Girl (London, 1997). 'S Pu oo g < Fieure 13. l Outline of the total Somah clan genealogy Mod fied on the basis of the chart in T. Lyons & A.I. Samatar, Somalia State ToUa^e, Multüateral Intervention and Stratèges for Political ReconstrucUon (Washington DC, 1996) 338 Abbmk Pré-colonial economie conditions in Somalia, a vast, semi-arid country with a widely dispersed population, never allowed state formation processes based on surplus extraction or power monopolization to take off because the main source of wealth was 'mobile capital' in thé form of livestock and local trade, neither of which is easily controlled or taxed by central authorities (like in the kingdoms of, for instance, Buganda, Ethiopia or Congo). Also, a state in Somalia could, almost by définition, only be a territorial, clan-based structure at the expansé of others, perhaps even at that of the Somali heer, the customary law based on principles of social contract. This was the case in the pre-colonial past as well as today. States formed by agreement or social contract, as in today's Somaliland under the Somali National Movement27 since 1991, remain beset by internai tension and are conditional upon careful respect for, and a balancing of, the rights and local interests of the common people, especially if they are of another, non-dominant clan (family).28 Nor was there any ideology or idea of nationalism in the pre-colonial period. ,There was a vague and decentred notion of cultural and religious identity of Somali-speaking groups but most of the clan and territorial units did not know each other and had little or no contact. Politics was localized and based on nested segmentation without central leaders. There was substantial diversity also in dialects, lifestyle and prestige between Somali groups, not only between the Samaale and Sab (see Figure 13.1) but also within these two gréât divisions and between them and the craft-worker groups and the Bantu speakers in the south.29 The only meaningful crosscutting loyalties were formed by the Sufi orders (turuuq) to which most Somalis (mâles and also many females)30 were affiliated, but even here the orders and their local settlements had membership of a dominant clan, and in times of conflict and feuding they did not prevent violent confrontation. Discussing violence in Somalia, Prunier31 bas described pre-colonial Somalia as a real 'culture of violence' and cites instances to mis effect. Violent stratégies were part and parcel of the harsh way of life in the country, and compétition was necessarily fierce. A forced reconstruction of an idyllic Somali social space where custom reigned and people settled their disputes in amicable rituals should therefore be rejected. Even the much reputed heer, or 27 Dominated by thé large Isaaq clan. For example, thé 'usa and Gadabursi in the Awdal area and the D'ulbahante in the east are minority clans under thé Isaaq. 29 Not to mention thé people of Hamar (Mogadishu), Baraawa, Kismaayu and other coastal towns. 30 See F. Declich, 'Sufi Expérience in Rural Somalia: A Focus on Women', Social Anthropology, 8 (2000), 295-318. 31 Prunier, 'Segmentante et Violence', 384. 28 customary contract law, was only valid within thé agnatic (clan) groups, not outside them, at least not if there was no prior agreement or patron-client relationship. What has to be kept in mind is that violence in Somalia - as in so many pre-colonial or pre-industrial societies - while fréquent, was not seen as controversial and problematic as it is today in thé global normative discourse on human rights. Violence as part of accepted relations between groups and thé threat of feuding were facts of life. The early travel literature (for example, R. Burton's First Footsteps in Africa, published in 1856) regularly mentioned such feuding and massacres. Hanley also described repeated incidents of Somalis robbing, killing and preying upon each other.32 On the coasts, they had a tradition of looting shipwrecks and also of piracy.33 The Somalis thus formed a loosely connected pre-colonial realm of dispersed pastoralists, coastal towns and Sultanates, related through ideas of common descent, clan fission and fusion, a violent pastoral economy, Islamic religious identification (since the thirteenth Century) and régional trade links. In this realm, thé European powers arriving in the 1880s, with their claims to take possession of coastal areas to protect sea-lanes and establish political control, were to pose a challenge. Italian and British interférence: Chimeras of governance and order As elsewhere in Africa, Western colonial expansion in the late nineteenth Century reshaped local Somali society and politics. Indeed, it led to the reactive forging of an incipient national identity, initially couched in a religious idiom.34 The British established a protectorate in 1885 in the north, the Italians their présence in 1889 in the south, and both carved out their domains of influence in subséquent years. The French, after concluding treaties with local chiefs, had, in 1884, secured what is now Djibouti as a vital port on the Red Sea route. While the British originally had no intention of occupying the hinterland beyond the coastal strip, the Italians - established in southern Somalia with its fertile inter-riverine plains - sought to control the interior and make it productive and to settle substantial numbers of Italian agrarian 32 Hanley, 33 'Warriors and Stangers', 80,98,101. Cf. L. Kapteijns & J. Spaulding, 'From Slaves to Coolies: Two Documents from the Nineteenth-Century Somali Coast', Sudanic Africa, 3 (1992), 2-3. 34 Cf. E.R. Turton, 'Somali Resistance to Colonial Rule and the Development of Somali Political Activity in Kenya, 1893-1960', Journal o/African History, 13 (1972), 119-43; Said S. Samatar, Oral Poetry and Somali Nationalism. The Case ofSayyid Mahammad 'Abdüle Hasan (Cambridge, 1982). Dervishes, moryaan and freedom fighters 340 Abbmk immigrants there.35 Italy's greater aim was to ultimately conquer the fertile highlands of Ethiopie.36 The purpose of the British and French was not to substantially colonize and transform indigenous society. Even more than in other African régions their attitude to the Hörn was one of 'indirect rule', taking advantage of trade and maintaining public order. But even this modest form of political control proved to be a challenge on the Somali coast. From the start, the colonizers did not intend to alter the religieus character of the local population: their primary motive was political and they hoped for consent from or co-optation of the local leaders. The colonizers did not pursue or even allow a missionary Christian agenda, as was asserted by Somali religieus opponents. The colonial state was more of a secular administrative structure, with an often-remarkable fascination with and admiration for the Somali's pride and demeanour.37 Colonialism certainly brought new forms of power, economie control and submission in the Somali régions. These challenged the local economy of land use and pasture, as weil as the decentralized political system of Somali clan groups and their kin-ordered social structures.38 Stirrings of résistance were not unusual in the early colonial period. One important anticolonial party was that of the Somali slave owners.39 The nature of the colonial enterprise in the Hora was thus mixed. France, Britain and, to a lesser extent, Italy were primarily there to compete with each other and with Ethiopia (that resented European intrusion and defeated Italy in the 1896 Battle of Adwa) for sea and trade routes. In the case of the British and the French, there was no effort to root out customary law and local leadership in Somalia. The Italians, however, came to see Somalia as a settler colony, the productive resources and political structures of which were to be utilized. Their policies (for example of creating 'state domains') were a threat to local Somali forms of collective tenure of land and pasture.40 But these policies came to 35 As in Eritrea, the northern coastal part of Ethiopia, which was made a colony in 1890. Somalia was an Italian protectorate until 1905 when it was also made a colony. 36 M. Guadagni, 'Colonial Origins of the Public Domain in Southern Somalia (18921912)', Journal of African Law, 22(1978),!. 37 Cf. A. Jardine, The Mad Mullah of Somahland (London, 1923), 34, 319; Hanley, 'Warriors and Strangers', 20, 49, 107, 124-27, 149; also LM. Lewis, 'Unravelling a 'Flawed' History', Journal of African History, 41 (2000), 522. These social and cultural relationships of colonizer and colonized, while prominent in recent anthropological and historical debates, have not been well studied m the case of Somalia. 38 See Guadagni, 'Colonial Origins', 5-8, 12, on the Italian land laws in Somalia and the frequent opposition of the local population. This résistance was independent of the Dervish movement and based on people's direct material interests. 39 Cf. Cassanelli, The Shapmg ofSomah Society, 231. 40 Guadagni, 'Colonial Origins', 25. fruition mainly after the Dervish revolt had fizzled out, and m southern areas where the Dervishes were much less active. Incipient modernity, Dervish revolt and violent transformation The colonial venture inevitably was one of incipient modernity in the widest sense. Italy and Britain brought a (radimentary) state administration, new modes of governance and political action and the use of state force, as well as material changes, new technologies, motorized transport but also modem firearms, economie exploitation and colonization of land and other natural resources. Somalis often reacted negatively in a variety of ways but were also accommodating and tried to profit from new opportunities, especially in the coastal areas and the agricultural south. One important response, however, came to be couched predominantly in a religious idiom of résistance. In 1899 an anti-foreigner msurgence was formally declared by Sayyid Mohammed 'Abdulle Hassan (1864-1920), (abbreviated as MAH), a well-educated Islamic teacher (sheikh) and militant proto-nationalist from the Ogaden clan.41 Sayyid Mohammed was a member of the Salihiyya Sufi order, which originated in Saudi Arabia and was influenced by the puritanist Wahhabi movement. At a big northern Somali clan assembly in Burao that year, hè raised the flag of résistance against the 'Christian invaders' in Somali lands. At this point he had about 4,000 to 5,000 followers. His insurgency came to be known as the Dervish movement42 and had its strongest base among the D'ulbahante clan of the Darod clan family (see Figure 13.1). In the years to follow, his aim was to create an autonomous space, beyond the reach of the colonial powers, and to develop an alternative politico-religious order with the contours of a strict Islamic theocracy. To this effect, the message was spread with intimidation and brute force. In the border area of Italian and British Somalia, the movement built settlements and fortresses for its fighters and followers, A predatory 41 His mother was from the D'ulbahante (also part of the Darod clan family). Mohammed 'Abdulle Hassan was often called Mullah or Sayyid. The epithet 'Mad Mullah' (see Jardine, The Mad Mullah) did not originale with the British or the Italians as is often thought but is a translation of the Somali expression wadaad wal ('lunatic religious leader/priest') used in the late 1890s by Somalis opposed to him. See SheikAbdi, Divine Madness, 55. 42 The word darwish is Persian, onginally meanmg Islamic mendicant. It is generally used for members of Sufi orders. In Somalia, members of the Salihiyya Sufi order m particular were called 'dervish', and Mohammed 'Abdulle Hassan liked having his movement called thus. Dervishes, moryaan and freedom fighters 343 342 Abbmk economie system based on pastoralism, raiding livestock, looting caravans and confiscating trade profits from coastal populations émergée. Remarkably, the initial act of this revolt in August 1899 was not a massive attack on the British but on two religieus villages of the Ahmadiyya, a Sufi order that had not displayed total support for the Sayyi/fs call for a jihad.43 This is a telling event which foreshadowed the nature of the insurgency. Indeed, MAH's movement was directed as much against other Somali as against the British or Italians. Initially, MAH appeared to call for transcending segmentary clan divisions but soon he started playing upon them.44 In the years 1900-1905, the insurgency wreaked havoc in the north-central Somali areas, and was combated by the Italians and especially the British. After this initial phase of revolt, MAH reached a political compromise with the British and the Italians in 1905, known as the Illig Treaty, under which he was allowed an autonomous area in the Nugal Valley, and freedom of movement and religious practice. He then had an armed retinue of some 5,000 to 6,000 fighters, with their dépendants. For about four years things were quiet but in 1908 a new phase of violence began, again not targeting colonial troops but the coastal sultanate of Hobyo, the Majerteen clan in the northeast, the Hawiya area, and even the Rahanweyn in the southeast.45 Only in 1912 did MAH's troops attack the British in Berbera. After 1909 the insurgency again grew in strength. In 1910 the British had even decided to concentrate their forces in the coastal areas in order not to lose resources and men in fruitless forays into the interior, which was of no real interest to them. Due to their mobile warfare and the depth of their retreat lines on the Ogaden Plains, the Dervishes were hard to pin down and defeat. The repeated campaigns of the British and Italians could neither root out the movement nor prevent its dévastation of other Somali clan groups and their resources. Throughout the twenty-year insurgency, the majority of victims were other Somalis, not colonial troops and their supporters. Indeed, MAH's revolt, through its use of often callous and indiscriminate armed force against Somalis perceived to be 'against him' and his Ogaden clan, transformed existing patterns of clan coexistence and coopération, undermined the heer, led to internai disarray, and generated new antagonisms that proved to be a bürden in the future. Interestingly, the Dervish insurgence was denounced in a letter written in 1909 by Sheikh Mohammed bin Salih Rashid, the founder of the Salihiyya order of which MAH was a member. Although it might have been at the instigation of the Italians that some Somalis (among them 'Abdullah Shahari, a former close adviser to MAH) went to visit thé Sheikh in Saudi Arabia to get such a condemnation, the letter nonetheless had a significant impact on many of MAH's followers. It was not a 'somewhat mild denunciation', as Lewis terms it,46 but as is évident from thé text,47 a quite outspoken and threatening condemnation that even questioned MAH's Islamic identity. The second decade of the revolt continued with in-fighting and raiding of clan groups, religious leaders and régional chiefs that opposed him or tried to work out a compromise with thé colonial régimes. It did not bring the stated aims of MAH any closer. The year 1911-12 came to be known as the Haaraama 'Une, the 'Year of Eating Filth',48 reflecting the famine, chaos and communal hatred generated by the Dervish movement. In view of this, the British changed their policy in late 1912 and founded the Somali Camel Constabulary as an expeditionary force to end the Dervish movement. It would still take seven years to do so, partly because from 1914 to 1918 the British were hindered by their commitments elsewhere during the First World War. Towards 1920 the insurgence movement was weakened, and the social basis of MAH's authority and rule among clan groups in the north declined. He still had followers, and held out in his impressive fortified seulement at Taleh (see photo). The British wanted to put an end to his activities and after MAH declined negotiations, they undertook one last offensive in July that year, assisted by Somali troops of the Habr Tolja'elo and Habr Yunis clans and, ironically, of the D'ulbahante, who had once been MAH's strengest supporters. Several British planes from Aden were used to drop bombs on the Taleh fortress. While the military impact of this action (with 20-kg bombs) was limited, the psychological aspect was important, and MAH and his hard-core followers fled to Ogaden territory in Ethiopia. There hè was approached concerning new negotiations and a truce, but again hè refused, humiliating the peace délégation. In December 1920 hè died in Ethiopia either of malaria or of an infectious disease, leaving a Somali people more divided than ever. As noted above, the trans-clan Sufi brotherhoods (turuuq) in Somalia did not create sufficient crosscutting loyalties to prevent or neutralize clan-based rivalries, often generated on the basis of material interests (pasture, livestock, territory, trade profits and women). Throughout his insurgence, MAH remained at odds with the leaders of the Ahmadiyya and Qadiriyya, as well as with numerous clan groups across the country. Indeed his main base was always 46 43 See Lewis, A Modem History, 69. Prunier, 'Segmentante et Violence', 388-89. 45 The Warsangeh clan, however, declared support for MAH's cause in 1908. 44 Lewis, A Modern History, 75. Given in Jardine, The Mad Mullah, 184-85. Lewis, A Modern History, 77; S. Samatar, 'An Open Letter to Somalist Scholars', Hom ofAfrica, 13 (1990), 88-89. 47 48 Dervishes moryaan andfreedom fighters 345 344 Abbmk within the Ogaden clan Many opponents, whether religious men, clan eldeis or coastal sultans, were assassmated on his orders One notonous example is that of the prestigious Qadmyya sheikh Uways bin Muhammad al-Barawi, killed by a Dervish hit-team m April 190949 Sheikh Uways had been putting up résistance agamst the foreigners m his own way but he refused the violent, divisive path of the Sayyid, whom hè denounced in a series of poems In 1909 MAH had his own Qadi killed for dissent, which accordmg to Jardme50 caused the defection of about four hundred Dervish fighters M AH's mission also generated vehement polemics with other important Somah religious leaders, for example, Sheikh Abdallah Yusuf al-Qutbi51 This was expressed m poetic combat, often in stnngs of poems m response to each other A readmg of the poetry for which the Sayyid became famous52 shows it to be füll of références to the fightmg agamst foreign 'mfidels', but also to the bashing of those Somah individuals and clan or hneage groups that were opposed to him Farah Abduf and Kapteijns53 note that Somahs view his poems often as bnlhant, but frequently also as vulgär and divisive Presumably they mean his frequent références to slaughtermg and smashmg his enemies, and showing pnde in this Sexual insults of his opponents also abound Somah poetry has a genre of curse poems, the gabaar-awayti?* a véritable 'poetics of violence' marked by aggressive, stabbmg images and metaphors of uprooting and annihilation, and MAH's poems certamly fit into this category But m his case, these poetic images were also part of a politics of violence As many of his opponents found out, such passages were exhortations to be taken literally MAH's remarkable poetic gifts added much fuel to the force of his message Poetry was part of his struggle and was widely recited m the years of the insurgency His last recorded poem Dardaaran (wntten m the year of his 49 B G Martin, 'Muslim Politics and Résistance to Colonial Rule Shaykh Uways b Muhammad al-Barawi and the Qadinya Brotherhood m East Africa', Journal ofAfrican History, 10 (1969), 472 See also Lewis, A Modem History, 70, for mention of earlier assassmations of prominent Somah leaders 50 Jardme, The Mad Mullah, 186 51 Sheik-Abdi, Divine Madness, 9 52 See Samatar, Oral Poetry for an excellent and now classic study 53 Mursal Farah Afdub & L Kapteijns, "The Defeat of thé Infidels" A Poem by Sayyid Muhammad 'Abd Allah Hasan of Somalia', Sudamc Africa, 10 (1999), 27-47 See also Samatar, Oral Poetry, 202 54 Cf S Samatar, 'Sarbeeb The Art of Oblique Communication m Somali Culture', m J K Adjaye & A R Andrews (eds), Language Rhythm and Sound (Pittsburgh, 1997), 49 346 Abbink death) reveals what imaginative force and religious zeal he could display," although it is also a testimony to the utter failure of his project. Indeed, this failure lias made it clear that his poetry was the message; the political reality was always quite different. Although he started out äs a sheikh and a wadaad, MAH is not remembered äs such. Nor was he ever regarded as a saint by later générations of Somali. In the course of his violent campaigns, hè effectively lost any religious authority hè may have had, and his personal réputation declined markedly in later years.56 A wadaad is by définition a religious mediator between clans: MAH became the opposite. The end results of his Islamist insurgency avant la lettre^ were large-scale material and human destruction,58 deep divisions, new clan hatreds, and no end to the colonial présence. The transformation of armed résistance into politics: 1920-1960 In the years from 1 920 until the start of the Second World War, the impact of colonialism on Somali society increased notably. Both Italy and Britain became more committed to the colonial venture, modestly investing in road building, agriculture, well drilling and health facilities.59 But only Italy, under Mussolini, promoted a real colonization, intended to make the country a seulement colony. The effects of colonial policy on Somali land tenure, pasture and water rights, and thereby on transhumance patterns of clan groups, were problematic, especially in the south where Italy wanted to develop its agricultural projects. Land was taken from its traditional users, new boundaries were drawn, clan families and clans were divided and eut off from each other, and their modes of subsistence threatened.60 Indigenous patterns of leadership were devalued and 55 See the text in Sheik-Abdi, Divine Madness, 176-81. It was republished in a different translation by Mursal Farah Afdub, L. Kapteijns & Hasan Farah Warfa, '"Parting Words of Wisdom" (Dardaarari): A Poem by Sayyid Muhammad 'Abd Allah Hasan of Somalia', Sudanic Africa, 11 (2000), 25-34. 56 Cf. the accusations of contemporary poets cited by Samatar, Oral Poetry, 164-65. 57 The mode of thinking of MAH, as evident from his poems and letters (for example to the British authorities, see Lewis, A Modem History, 78-79) indeed shows remarkable similarity with present-day Islamists' discourse. 58 Jardine, The Mad Mullah, 315, (over)estimated the number of people killed directly or indirectly as a result of the Dervish war to be about 200,000. 59 Lewis, A Modern History, 104-5. 60 Laitin & Samatar, Somalia: Nation in Search, 61-62. undermined. The insights of Italian scholars like Colucci61 into customary tenure and decision-making in Somali culture were not much heeded in colonial policy. The colonial regimes did not make any major investment in formal, modern éducation in Somalia and no new dass of urbanized educated Somalis emerged to formulate a nationalist movement. This only came during and after the Second World War. Resistance was fomenting but before the Second World War did not take on an organized, collective form. While a vibrant Somali local politics developed - as far as thé colonial régime allowed it - thé political scène showed a complex balancing act between clan-based politics and nationalist political organization. The violence and internai chaos seen in thé Dervish days did not return. There were small-scale rebellions but not wide-scale organized protests on a national basis. For instance, thé 1937 Camel Corps mutiny appears not to have been inspired by nationalist motives but primarily by direct material causes.62 The Second World War meant a radicalization of Somali politics and a growing unease with colonial governance. The Italian East African Empire collapsed in 1941 following the Allied military victory, and thé debate about thé future of its constituent parts began. Ethiopia regained its independence and part of thé Somali-inhabited Haud and thé Ogaden were ceded to it. In 1942, former British and Italian Somalia experienced a period of British Military Administration (BMA) until 1950. In thé report of the 1948 UN Four Power Commission, reporting on Somalia to help prépare its political future, thé opinions of a wide section of Somalis were presented.63 From this report one can deduce not only that thé économie interests and agricultural schemes of the Italians had created serious tensions concerning matters of land, labour,64 territory and 'clan borders', but also that British-Italian political compétition and Somali party politics had created great rifts. The effect was further opposition between clan families. For instance, people from the Somali Youth League (SYL), formed in 1943 and in favour of speedy independence, were not liked in the south. They were seen as a mainly Majerteen group and were opposed by the Hizbia Digil-Mirifle Somali party active in the riverine areas of 61 M. Colucci, Principi di Diritto Consuetudinario della Somalia Italiana Meridionale (Florence, 1924). 62 See Jama Mohamed, 'The 1937 Somaliland Camel Corps Mutiny: A Contrapuntal Reading', International Journal of African Historical Studies, 33 (2001), 615-34. The same was true of a revolt of the same corps in 1944. 63 See L.V. Cassanelli, 'Somali Perceptions of Colonial Rule: Testimonies to the Four Powers Commission', in R.W. Harms et al. (eds), Paths toward the Past: African Historical Essays in Honor ofjan Vansina (Atlanta G A, 1994), 143-55. 64 The Italian forced labour policy was especially hated by Somalis. Dervishes, moryaan 348 Abbink the south.65 This again illustrâtes the values attachée! by clan groups to their own traditions and to self-détermination in the game of clan politics and reciprocity. Colonial policy had aggravated rather than lessened such tendencies. In 1950 Italy - though defeated in thé Second World War and with a dubious colonial record - received a UN trusteeship to administer its former colony of Somalia for ten years. The British retained northern Somalia, also to prépare it for independence. Collective insurgency movements like thé Dervishes did not appear in either country. While thé public memory of the revolt of thé 'Mullah' (MÄH) was vivid, it was not a political example. But in the new party politics that emerged, factionalism and régional divisions among clans, clan families and caste groups were not overcome. Also, the Italian political model for Somalia was based on transposing a centralized state to a decentralized, rebellious country, an approach that had adverse effects on local politics and clan-group relations.66 In this state context, clan politics came to hâve a lasting appeal, despite thé objective challenge of colonialism to thé Somali subjects to unité, and despite thé growth of thé idea of Somali unity across thé various colonial frontiers. The colonial state arena, with its disregard for local leadership, autonomous structures and customary law, created a 'spoils System', whereby compétition for thé distribution of state resources on the basis of identifiable constituencies was the game. Accordingly, after the Second World War and the BMA period, new 'tribal' political parties emerged alongside the more pan-Somali SYL67 and the Hizbia Digil-Mirifle party: for example, the Marehan Union, the Progressive Majerteyn League, the Hamar Youth Club (in Mogadishu) and the Hawiye Youth League. They garnered to a spécifie clan-based constituency even at a time when an agenda of national unity might have been expected to be the best way to promote a viable independent Somalia. As Tripodi states: '...the new state organization became just an instrument in the hands of prédominant clans, who administered power according to their own interests'.68 In a way, the Somalis feil into a political trap set inadvertently by the Mandate authorities. 65 See Cassanelli, 'Somali Perceptions', 145-46, 148-49. See also his citation of the complaint of the Elai clan Community about all Somali political parties, ibid. 153-54. 66 Cf. P. Tripodi, 'Back to the Hörn: Italian Administration and Somalia's Troubled Independence', International journal ofAfriccm Historical Studies, 32 (1999), 79. 67 With mainly Darod and Hawiye membership. 68 Tripodi, 'Back to the Hom'. andfreedomfighters Independence and an attempt at state formation: The irredentist In 1960 thé two Somalias, ill prepared and divided by under-funded and topdown colonial raie, became independent. After five days of separate independent existence, they joined together as one state. A multi-party political System evolved, and for a while things looked promising in the new era of nationalist fervour. A careful politics of clan balancing at the national level started but soon factional manoeuvring and élite compétition, aimed at the sélective use and control of central state power, resurfaced leading to a creeping fragmentation of thé polity. The conditions for this lay partly in thé faulty political system not recognizing thé continued importance of clan identification at a social and cultural level but also in thé colonial land policies, that more than ever had tied clan rivalries to material interests around resources, generating conflicts no longer covered by the heer, the legal raie of clan elders and religious mediators (yvadaad). In 1969 General Mohammed Siyad Barre, chief of staff of the Somali army, took power in a coup, deposing a fragmented, faction-ridden government and an impotent parliament. He declared a state-socialist programme of development aimed at éducation, economie growth, literacy campaigns, an improvement in the rights of women, and neutralizing customary law and clan politics, even clan identities. Instead of constructively engaging them in a consociational or fédéral formula, he denied them and tried to forcefully repress them. As is known from the literature, this campaign of déniai took on hilarious forms, with people asking each other about their 'ex-clan', it being forbidden to ask what their clan was.69 When hè came to power Siyad Barre did not instigate clan raie. His system of governance was initially led by the ideology of 'scientific socialism', and hè employed anyone who showed ability and personal loyalty, also those frorn marginal groups. For instance, his long-time defence minister was a member of one of the minority castes but his policy of alleged 'declanization' amounted to a déniai of people's primary identifications and of the structures of regional power and prestige. His Marxist-modernist policies were directed against heer, clan law, and local civil-society institutions. Over the years, Siyad Barre's orientation to his own clan (Marehan) and that of his mother (Ogaden) increased. A turning point was the debacle in 1977-78 in the Somalia Irredenta war against Ethiopia, aimed at taking the Ogaden. The 69 See also the joke cited by Ch. Geshekter & S.A. Warsama, 'An Introduction to Humour and Jokes in Somali Culture', in R.J. Hayward & I.M. Lewis (eds), Voice ana Power (London, 1996), 147. Dervishes, moryaan and freedom fighters 351 350 Abbink attempt to wrest this vast area up to the city of Jijiga from Ethiopia was a costly illusion that brought the regime to the brink of disaster. Instead of unifying the Somalis in a common cause, it led to internai récrimination, clan division and a décline in thé development commitments of the régime. Internai opposition movements with an agenda of libération emerged in various parts of the country. After 1988, the first year of the devastating war against the Somali National Movement in the Isaaq-dominated norm, Siyad Barre reverted to relying fully on the so-called 'MOD' alliance70 and began excluding people from other clans from power. In a predictable development, this led to an aggravation of clan rivalry,71 manifesting itself in pervasive distrust and enmity among people and perceived clan groupings, as well as in violent conflict. Subsequently, a speedy décline of the national state and the émergence of clan-based factional contestation for power went hand in hand. The various revolts in the country in the north, in the Majerteen area, in the Hawiye area and in the southern coastal cities - seemed to usher in an era of freedom and equality, but it eventually turned into the opposite. Armed revolt and insurgency became a way of life adopted by the clan-based militias and factions that saw easy ways to gain and profit. The new links of the Somali economy to export markets as well as external political forces (including international organizations and aid agencies) played an important rôle in enhancing such tendencies. It may not be fair to compare Mohammed 'Abdulle Hassan with Mohammed Siyad Barre but what can be said is that both their nationalist programmes failed because of their ruthless application of idéologies and violent practices that ignored and indeed actively suppressed the social structures and cultural values of Somali life. Foremost among mem were segmentary clan identification, customary law, the prestigious rôle of clan elders and clan councils, and the traditions of egalitarianism and local autonomy. Their policies evoked clan-based movements fighting for freedom as well as a culture of violence rooted in feuding, the settling of scores and looting. 70 71 See footnote 7. Cf. M. Jama, 'Trajectoire du Pouvoir', 423. Table 13.1 Approximate population distribution of the six clan families Digil Rahanweyn Dir Isaaq Hawiye Darod Ethnie and national minorities 4% 16% 7% 22% 23% 21% 4-6 % The last Somali government census was in 1987 so these figures are estimâtes only. The total population of Somalia is thought to be about 7-8 million. In Ethiopia there are about 3.2 million Somalis, and in Kenya the figure is estimated to be about 340,000. State disintegration - clan résurrection: The post-1991 period After 1991, the basis of collective political action was sought in the résurrection of clan identities, though shifting in scope according to the level of clan (tol) segmentation. Initially the element of freedom fighting was obvious. The southern Rahanweyn and Digil groups, choking after years of central state oppression, aimed for autonomy. The Isaaq-dominated north successfully fought free from southern, mainly Hawiye and Darod, domination but tensions remained with the D'ulbahante and Warsangeli clans (that had in the past largely supported the Dervish movement). In the south and northeast, factions based in a majority clan family were formed and became the units of insurgence and revolt, for example the opposing Abgal and Habr Gidir subclans within Hawiye, heading two wings of the USC movement. But hère the aim was no longer a national state or an Islamic-inspired restoration of order and justice but survival, prédation and the building of fiefdoms as large as possible. Almost all major clans saw the émergence of their own parties, militias or combatant groups. While these are manipulated by elite sections within the clans, they nevertheless were formed on the basis of kinship thinking and segmentation. In the past decade, several dozen such groups have emerged, even down to the smallest size of a lineage. For instance, the tiny group of Reer Barre in the north at one point also claimed to be a political party. The major ones still involved in political entrepreneurship and violence in Somalia today are listed in Table 13.2. The main clan-militia leaders/entrepreneurs active in southern Somalia especially in and around Mogadishu, are Ali Mahdi Mohammed (an Abga businessman), Hussein Mohammed Farah (a Habr Gidir), Osman Hassan Al Dervishes, morvaan andfreedom l l l 4 i If S § "i •IJ ij 'öS "o D 'S S •l g l rn ï—« s < < s i ll £53 3 73 ! l cc (S U l | K 3 U w D l l fighters 353 Dervishes, moryaan andfreedom fighters 355 354 Abbink Ato' (Habr Gidir), Muse Suudi Yahalow (Abgal), Muhammad Qanyare Afrah, Mohamed Hussein Addow, Bashir Raghe, and Husayn Haji Bod. They are predominantly from the Hawiye clan family and often head factions of the various organizations mentioned in Table 13.2. All have used moryaan, the heavily armed, roaming groups of young fighters living by intimidation and predatory activity, to further their own interests. Fierce resource compétition emerged between these various organizations and militias, be it for land, livestock, fragments of the state machinery, possession of former nationally-owned facilities like airports, contacts with NGOs and aid agencies, trade and export routes, or arms imports. There was also a marked but often ill-reported dimension of sexual politics involved, as the larger clan-faction groups claimed forceful access to women from weaker groups, especially in the southern agricultural areas and the big cities on the coast. This politics of humiliation aggravated clan divisions and communal anger. In addition, the traditional inequalities between Somali pastoral groups and urban elites, especially from the big clans like Hawiye, Majerteen and Marehan on the one hand and the Waable (caste minorities) and Bantu groups on the other, were reproduced, if not reinforced.72 Together with the agricultural Digil-Rahanweyn, these vulnérable groups were a major target in the civil war and the warlord fighting, with most of their resources being taken or claimed. In this respect, nothing had changed since the pre-colonial era except for the severity of resource exploitation and destruction. Several authors such as Prunier and Schlee73 have tried to see some 'order in the chaos', i.e. the regularities underlying the violence of armed groups, shifts in their alliances, economie activities etc. There are indeed notable regularities and patterns but the only predictable factor is the orientation of Somalis towards clan identity and their structuring of political action in sirailar terms. The Somali central state of 1960-1991 has gone forever. The various independent regional administrations in Puntland,74 in Somaliland, and since 2002 in the South West State in the Rahanweyn area75 can be seen äs having emerged from résistance against any central government dominated by clans other than the majority ones in their own région. As no région is homogeneous, 72 Cf. Mohammed-Abdi, 'Les Bouleversements Induits'. Schlee, 'Régularités'; Prunier, 'Segmentante et Violence'. 74 See M. Doornbos, 'When is a State a State?: Exploring Puntland', in P. Konings, W. van Binsbergen & G. Hesseling (eds), Trajectoires de Libération en Afrique Contemporaine. Hommage à Robert Buijtenhuijs (Paris, 2001), 125-39. 75 See the déclaration on www.somalilandnet.com/warya/2002/somalianews/10532. shtml 73 there are constant tensions within these new units for political and économie control, often fuelled by claims of kin loyalty, segmentation and the idiom of rivalling clan sub-units. As thé work of Compagnon and Marchai76 has shown, in most cases thé use of violence was instrumentalized by clan-faction groups without a clear programme, except an economy of violent loot and intimidation. The traditional Somali waranle ideal - that of warriors legitimately defending thé interests of their clan and territorial group - had transformed into the entrepreneurial violence of moryaan, thé armed youths active both independently and in thé service of clan élites or so-called warlords, thé entrepreneurs of violence. Most major Somali traders at some point began supporting or organizing armed clan-faction groups to advance their interests and monopolize resources like import-export lines.77 Towards the end of the 1990s, this culture of living by looting was not only limited to thé militias and moryaan but had also spread to the général population, especially in thé urban areas. Revolt and résistance thus dissipated into a generalized culture of violence based on shifting alliances and without any wider socio-political aims. Any distinction between moryaan and politically inspired fighters became blurred. The situation in many areas of thé south is still marked by this kind of everyday violence by looting rebels without a cause except for intimidation and extortion. In this situation, thé underlying continuities of clan and lineage distinctions and of locally based interests in Somali society are notable: clanbased groups form the units of survival and reproduction.78 They keep providing people, also in the cities, with a basic identity used for collective identification and instrumental gain. 76 D. Compagnon, 'Somali Armed Units: The Interplay of Political Entrepreneurship and Clan-Based Factions', in C. Clapham (ed.), African Guerrillas (Oxford & Bloomington, 1998), 73-90; R. Marchai, 'Formes de la Violence et de son Contrôk dans un Espace Urbain en Guerre: Les Mooryaan de Mogadishu', Cahiers d'Etudes Africaines, 33 (1993), 295-320. 77 Compagnon, 'Somali Armed Units', 85. 18 Several authors hâve stressed thé rôle of thé ideology of expansive reproduction il Somali culture. Cf. I.M. Lewis, Understanding Somalia (London, 1993), 13. Tb Somali population growth figure is high at 4.43%, with a fertility rate of 7.01 childre per woman. See www.wifak.uni-wuerzburg.de/fact98/so.htm Dervishes, moryaan andfreedomflghters 356 357 Abbink Resistance and revolt as a way of life? After the fall of the Siyad Barre government in 1991, the famous statue in Mogadishu of Mohammed 'Abdulle Hassan, the declared proto-nationalist hero of Somali nationhood, was '...defaced and later toppled, to be sold for scrap'.79 This was no doubt a poignant symbolic event, underlining in the eyes of many the total démise of Somalia by violent interaal rivalry and factional fighting. It was simply the outcome of the same processes of territorial and clan-based division that had plagued Somalia in the time of Sayyid Mohammed 'Abdulle Hassan's insurgence. Kin groups based on tol remain the ideational and material survival mechanisms in times of state collapse and economie décline. Such groups, be they clans, sub-clans or mag-paying (lineage) groups, provide an element of social and psychological identity even - or perhaps especially to uprooted people estranged from their socio-cultural orders. But these groups are not based on ancient hatreds. Clans and clan-thinking appear not to be atavistic éléments of the past reasserting themselves. They are, and always were, 'identity vessels' in a stateless environment that give agency to people. In this sense, the fall of the MAH statue in the capital symbolizes another turn in the cycle of internai disarray and rébellion that has characterized Somali society for the past 150 years - a cycle whereby only the names of protagonists, the alliances, and the number and extent of violent conflagrations differ while éléments of continuity are notable. Briefly recalling the various periods of revolt in the past Century, we see similarities. In both episodes described in this chapter powerful political movements arose, the First ihstigated by a charismatic but destructive and divisive religieus leader (yvadaad) turned war leader (Mohammed 'Abdulle Hassan), and the second by a warrior (waranlë) gone awry (General M. Siyad Barre), whose rule led to the growth of libération movements, civil war and the rise of the moryaan génération in a. sort of stateless limbo. After 1991, these various movements quickly transformed into, or engendered, clan-based factions whose leaders appealed to clan identity and/or interests that placed them against each other, mocking both traditional and Islamic ideas of peace or médiation. In any of these periods, it is therefore difficult to label all these movements of revolt as 'freedom fights'. Sayyid Mohammed 'Abdulle Hassan may have had an agenda of Islamic revivalism against a foreign colonial présence but he created a regime of pervasive repression against his countrymen, leading to chaotic violence and unprecedented clan reckonings that were self-defeating vis-à-vis the colonial powers. Siyad Barre, in the Marxist-socialist spirit of the times (1960s-1970s), Moryaan in Mogadishu Source: Hollandse Hoogte, Amsterdam 79 Sheik-Abdi, Divine Madness, 212. (A photograph of the statue appears on p. 210.) 358 Dervishes, moryaan andfreedom Abbink came to power on a ticket of anti-clanism, equality and social development, but his rule ended in utter tyranny with the state being seen as a mafiose association.80 The period of clan-based factions after 1991 yielded a precarious libération of certain hard-hit régions (Somaliland, Puntland and the DigilRahanweyn South) by their own forces while the rest of the country saw a descent into mayhem or violent predatory rule. There is both continuity and change in the nature and political relevance of clan segmentation and in the use of violence in Somalia. The big différence with the traditional, pre-state situation now is that most of the violence since 1991 has been fully instrumentalized, released from its social context of legitimate self-defence and of compétition for basic interests of clan/lineage groups in the pastoral or urban trade economy, regulated by law, heer or mediating elders. As the persistent présence of clan militias and warlords as well as the many groups of moryaan (a new term and phenomenon in Somali history) makes clear, predatory violence and the culture of looting could not easily be 'retrieved' within the socio-cultural frameworks of Somali society, all the more since there was no real ideological rationale to be discerned. While many violent militia groups active today were once part of political movements - or still claim to be — their wider political aims have become lost in nebulous clan-based conflict, opportunism and violently defended business interests. Violent action became an entrepreneurial activity - a rational means of accumulating resources, though at the expense of other Somalis (and foreign aid agencies). The destabilizing effects of this soon became clear but such activities could not easily be stopped. Continuities with pre-modern conditions are evident in that clan-based politics proved to be quite adaptable to the modern setting of a stateless Somalia, perhaps even more so since the two meta-narratives of Islamic revolt (Mohammed 'Abdulleh Hassan's movement) and of state socialism (in the Siyad Barre era) failed so dramatically After January 1991, a process of conciliation and rehabilitation was slow in coming, except in Somaliland. Hère independence was declared under the SNM and a national project of rebuilding a state administration was initiated. Puntland, under the Somali Salvation Democratie Front (the SSDF, founded in 1978 by Majerteen), and Southwest State (in the Digil-Rahanwein area) followed in early 2002, but were weaker institutionally. During the late 1990s, social rebuilding and grassroots initiatives for civic and economie recovery were largely initiated by people tired of the daily violence, especially clan elders, customary law experts, women's groups, Islamic (Sufi) orders and some ' Prunier, 'Segmentarité et Violence', 396. flghters 359 other local Islamic groups and courts.81 But southern Somalia, including Mogadishu, has lagged behind. Paradoxically, this is also the area where the claim to restore national order and unity through a 'transitional national government' (TNG) for all Somalia was put forward following the 2000 Arta Conference, a meeting of reconciliation of Somalis called by the President of Djibouti with UN assistance and held in the town of Arta in Djibouti. The above claim seems weak in view of the ongoing conflict in that area, divided among six clan-militia leaders, and the limited representativeness of the conference. Interestingly enough, the TNG is backed by the local Islamic movement in Mogadishu82 that has adopted methods of force similar to those of the warlords, with armed militias driving around in 'technicals' (pick-up trucks mounted with machine guns) to intimidate opponents. The TNG, led in 2002 by Abdiqassem Salât Hassan (a Hawiye Habr-Gidir), is also regularly accused, mostly by non-Hawiye opponents, of trying to win support in other régions, such as Puntland, by subversion. In many parts of Somalia today, revolt or résistance against any form of political or other authority, especially if it is perceived as clan-based, seems to have been institutionalized as a way of life: a whole génération knows nothing else. Conclusion: Cultural structure and political conjuncture The past decade of statelessness and regional conflict after President Siyad Barre's downfall shows similarities with the colonial period during and after the Dervish insurgency. At that time, internecine fighting, generated by external influences and the challenges of modernity, also produced chaos, thousands of casualties, economie disruption and deep clan divisions instrumentalized by war leaders. While the present political conjunciure in Somali history shows new forms of violence and conflict, continuities in culrural and social structure are remarkable, going back to the frameworks not only of clan organization and kin-group identification but also to traditional disputes and material interests related to land and other resources. But a century later the cycle has gone further, with Somalia's constituent clan-based 81 See S. Arnold, 'Islam and Somali Clan Wars', 25 May 1996, on www.netnomad.com/ lam-clanwar.htm islam-clanwar.htm 82 °' There were crédible reports that Arab countries bankrolled the TNG and its associated Islamic groups. See Reuters news message of 26 June 2001, 'Chief Says Saudi Arabia is Destabilising Somalia'. Also A. Le Sage, 'Somalia: Sovereign Disguise for E Mogadishu Mafia', Review ofAfrican Political Economy, 29 (2002), 134, 37. 360 Dervishes, moryaan andfreedom Abbink units, especially in the urban areas and the South, showing little inclination to come together, even if these vanous movements have achieved some sort of political order in their own areas. The perception is that a Hawiye has no understanding of Majerteen problems, that the Darod should keep away from the Dir, and an Isaaq has nothing to teil the Rahanweyn. A pervasive and distrustful localism has settled in Somalia. In a typical phenomenon of globalization, Somalis in the diaspora are enhancing this discourse of localism and clan-thinking, as is evident from the highly polemic and sectarian nature of Somali Internet website contributions. Indeed, these exchanges on the numerous sites on Somalia would merit a separate study of divisive and often aimless 'diaspora politics'.83 Somalia in recent years has shown that in permanently unstable political conditions new models of socio-political organization and conflict management need to be sought, based on the centrifugal, loosely 'federative' character of Somali society. Under some conditions a centralized légal-bureaucratie state is one option, but not necessarily the best. This, in a way, 'post-modem' situation of contested authority and decentred state power is füll of challenges for both the notion of a viable social order in Somali society in particular and for political theory in général.84 There is no direct prospect of resurrecting a central state order among the southern Somalis, and authority structures that emerged elsewhere - while promising in Somaliland and fledgling in Puntland or South West State - remain fragile and conflict-prone. The Transitional National Government of Somalia in Mogadishu (since 2000) remains a shaky and localized structure with only a façade of national authority. It is beset by rebellious and/or criminal groups, is incapable of maintaining public order or regulating illegal economie activities, and is contested even within its own ranks.85 Power and the control of resources in Somalia - including land, livestock, trade profits, women,86 and qat - are very fluid but remain based on the wielding of armed force. This is reinforced by the near-universal spread and possession of automatic weapons across the country, leading to a 'democratized' and instrumental use of violence.87 These resources are a continuous bone of contention between local groups and clan formations vying fighters 361 for survival and prédominance in a world ill-contained by traditional frameworks of national law and custom. There is an emerging consensus among Somalia observers that imposing a unitary state based only on abstract individual rights and top-down government is less fruitful than a policy of récognition of the existence of these collective identifies based on clan thinking and buttressed by traditional Systems and customary laws related to land use and territory, especially in the interior of the country. As caste and ethnie minorities as well as people of the coastal cities have different identities and economie concerns, a forced unitary regime will only perpetuate division and centrifugal conflict. In addition, the economie situation in southern Somalia is now one resembling the 'ultimate free market' with no regulatory institutions, gréât entrepreneurial freedom, and deals based on trust (clan members) or on extortion and intimidation (non-clan members). It is, however, a system where the bills are not paid: environmental pollution and exhaustion continue unabated. The visibly devastating effects of this might eventually press the various clan and territorial groups to work towards reaching an agreement. Ultimately, a durable socio-political order in Somalia will have to take into account traditional institutions like the recognized clan elders, religious mediators, and the heer.™ There is significant transformation in Somali society, but also notable continuity in 'tradition'. A historical and cultural approach to the politics of Somalia shows that such institutions would contribute by addressing the concerns of most Somalis with underlying issues like tol segmentation, the rhetoric and practice of (clan) group equality, and their désire of mutual respect and having rights and duties met. More intense inter-clan communication and negotiation in an institutional manner would be equally necessary. Institutionalizing regulär meetings of local stakeholders in the various régions also seems necessary. In recent years, new possibilities have been opened up by NGOs or women's groups mediating and organizing peace activities as well as small-scale economie ventures.89 Somalia has thus to be re-imagined anew, not only as a political idea or state but also as a cultural and social order in line with the values of clanship and customary law. That a reinvention of traditional Somali notions of values, religious identification, clan belonging and civilizational identity should hereb> 83 For some samples, see: www.somaliawatch.org; www.mudug.org; allpuntland.s5.com; somalitalk.members.easyspace.com and www.xamarcadde.com 84 Cf. M. van Notten, 'From Nation-State to Stateless Nation: The Somali Expérience', (forthcoming). 85 See Le Sage, 'Somalia: Sovereign Disguise'. 86 Cf. Lewis 2001. 87 See a recent report on the continuing flow of illegal arms to the Horn by the UN IRIN news service of 15 July 2002 on www.irinnews.org 88 S. Samatar, 'Somalia: The Politics of Poetry' emphasized thé rôle of poetry (as ir Mohammed 'Abdulleh Hassan's time) but this genre has virtually disappeared fron Somali public life in thé last 20 years. 89 For examples, see Asha H. Elmi et al 'Women's Rôles in Peacemaking m Somal Society', in D.L. Hodgson, Rethinking Pastorahsm m Africa Gender, Culture and tht Myth ofthe Patriarchal Pastoralist (Oxford & Kampala & Nairobi, 2000), 121-41. 362 stand central - and not imported Western, Arab or Islamist ideas - seems reasonable but is by no means sure in view of constant external pressure by the UN and powers like the US, the EU, Saudi Arabia and some Arab Gulf states. Southern Somalia is still in a 'cycle of rébellion' that is becoming increasingly vicious. The entrepreneurs of violence and prédation dominate the scène, and any overarching authority structure or normative framework is absent. What was achieved in the rest of the country should, however, not be underestimated. There has been a remarkable movement towards restoring a social order based on local courts, heer, clan elder médiation and in some cases a résurgence of religious (Islamic) mediators and women's NGOs. But even these positive developments have been largely based on clan-group identities. In view of the Somali game of clan prestige and the façade of status and nominal equality, as previously emphasized by Hussein Adam,90 an 'integrative structure of constructive conflict' should be sought if the county intends to move forward.91 Even then, some conflicts in Somalia may not be solved except by ignoring them. Disagreements should be given room to be expressed, though without the prospect of immédiate resolution. The main thing would be to build structures that could contain the escalation of violence. It is also an illusion to think that a non-violent consensus structure can be built in a country with persistent natural resource scarcity and decentred, kin-based politics with deep-rooted traditions of autonomy and collective honour. This is evident from the simmering conflicts within Somaliland and Puntland and along their border, cutting across certain clan territories.92 It is interesting that the Puntland administration, dominated by the Majerteen, is trying to win over parts of the D'ulbahante and Warsangeli sub-clans in Somaliland, appealing to clan unity under a higher level of segmentation: all three form the Harti clan that is opposed to the Isaaq dominant in Somaliland. For the construction of a more comprehensive political order, récognition of cultural continuities in Somali society would remain necessary, despite their transformed, violent articulation in the post-state era. In this respect, a down-up 'building block' approach to conflict resolution and the création of a wider 90 See Adam, 'Müitarism and Warlordism'. As all observers of contemporary Somalia feel bound to offer their advice, I follow suit. For a recent example, see also M. Doornbos, 'Somalia: Alternative Scénarios for Political Reconstruction', Afneem Affairs, 101 (2002), 93-107. 92 See http://news.bbc.co.uk/hi/english/world/africa/newsid_576000/576721.stm (on the 1999 tensions) and the IRIN news item of 15 July 2002 ('Somalia: Troop Build-Up Créâtes Tension in Sool') on www.irinnews.org 91 Dervishes, moryaan andfreedom fighters 363 Abbmk social order - as advocated by I.M. Lewis, Ahmed Yusuf Farah93 and others in thé early 1990s but rejected by thé UN - still offers the most reasonable way ahead. It implies avoiding a top-down, externally funded model of politics, and rediscovering traditional units of decision-making and légal discussion, some régulation of economie life, and creating working administrative structures at the local level of clans and clan-family régions. This is already happening. In Somaliland, Puntland, South West State and several coastal cities, some precarious progress was made in the last few years, despite the internai segmentary tensions. These units - in themselves also based on a föderative association of sub-clans and urban elites and agreement to respect their interests, including at some point the heers - might at a future date décide to form a state confédération. Unfortunately it seems that the 'Arta process', leading to the TNG, has subverted this long-term development because it posited the need for a state at a point when most Somalis were not in agreement about its desirability and shape. The unitary state of the thirty-year independence era had left an image of internai division and disaster that worked against any résurrection of such a structure. Finally, there is also a wider regional context of the Somali conundrum as it concerns neighbouring countries in the Horn. The prime example is Ethiopia, again deeply involved in Somali affairs,94 as it was in the time of Mohammed 'Abdulle Hassan when it encroached in the Ogaden and staked claims with Italy and Britain.95 Ethiopia was sucked into the post-1991 Somali problems precisely because of the country's internai divisions. For example, Hawiye faction leaders invited or allowed Eritrean troops and insurgents of the separatist Oromo Liberation Front (OLF) to operate against Ethiopia from their territory during the Ethio-Eritrean war of 1998-2000. Also, al-Ittihad groups in the Luuq area made attacks into Ethiopia in the late 1990s. Ethiopia came to oppose the transitional govemment in Mogadishu and to support the 93 See Ahmed Yusuf Farah & I.M. Lewis, Somalia- The Roots of Reconciliation. PeaceMakmg Endeavours of Contemporary Lineage Elders A Survey of Grassroots Peace Conferences in 'Somaliland' (London, 1993). 94 In February 2002, the Ethiopian ambassador to the UN, Dr Abdul-Mejid Hussein who is of Somali origin himself, spoke about the instability in Somalia and was quoted by Reuters as saying: 'If the Somalis don't solve their problems, then we will do it for them...We won't wait for ever.' He was strongly criticized for this by Djibouti and by the Transitional National Government of Somalia in Mogadishu (see http://www.irinnews.org/report.asp?ReportID=23245). There is, however, no doubt that hè meant what hè said and tensions still remain high. 95 See Lewis, A Modern History, 57-58, 69, 71, 72. Ethiopian interest in Somaliinhabited territories must be seen in the light of the Italian policy of systematic infiltration into Ethiopia. Cf. Lewis, ibid. 89-90. 364 Dervishes, moryacm andfreedom Abbmk Somaliland government, as well as the forces of the Rahanweyn Resistance Army (RRA) and the Somali Reconciliation and Restoration Council (SRRC) based in the Digil-Rahanweyn area. Since 11 September 2001 the Somali case has assumed added relevance in the aftermath of the global terrorist challenge from the Al Qa'eda network, which allegedly has had training camps and a network of activists in the country. According to American and Ethiopian intelligence sources, small Somali Islamist groups (for example Al-Ittihad al-Islamiyyà) were connected to this network and in 1993 participated in thé torture and killing of American servicemen during thé failed 'Opération Restore Hope'.96 Hence, what happens in Somalia concerns more people than only thé Somalis and their immédiate neighbours, although that is probably the least of their worries. The wider conflict potential of a fragmented, segmentary society is much greater than its constituent parts can imagine. This is part of the tragedy of what are often called 'collapsed' state orders (cf. Sierra Leone, Congo, Afghanistan, parts of Pakistan). Such societies are nevertheless part of globalizing networks with economie, religieus and political interests. Somalis, at least those clan factions interested in an Islamist or anti-Western agenda, may corne to join new international forms of résistance against Western and other domination.97 Thus, another cycle of conflict and violence would be initiated, creating new formations of clan power and identity but boding ill for thé emerging social order and économie rehabilitation within Somali society itself. The above account may illustrate that gaining an understanding of the politics of contemporary Somalia requires a rethinking of définitions of rébellion and résistance and their socio-cultural basis, as does any imagining of a 'functional' Somali state in the classic sensé. The twentieth-century Somali state certainly 96 When the Al-Ittihad group became active in southern Ethiopia in the late 1990s this provoked a major retaliatory campaign by the Ethiopian army m the Luuq area, leading to the collapse of the group's military wing and the élimination of several of its core leaders by Ethiopian or Ethiopian-supported troops. Since then, the Al~Ittihad has followed a largely domestic Somall agenda, although it mamtains active links with an international Al Ittihad network existing in the Gulf States, Sudan, Pakistan and other countnes. It is highly unlikely that Al Qa 'eda remnants will regroup in the unpredictable environment of Somalia. (After Pakistan and Muslim aeas in the Caucasus, they would perhaps, in Africa, be most likely to choose northern Nigeria.) 97 For a conservative US view, see J. Philips, 'Somalia and Al-Qaeda: Implications for the War on Terrorism', The Héritage Foundation Backgrounder, 1526 (2002), 1-12. www.foreignpolicy-infocus.org/comrnentary/2002/0201 somterror_body.html provides a more liberal viewpoint - see K. Menkhaus, 'Somalia at the Crossroads of American Policy'. View publication stats î ^ fighters 365 was made through cycles of rébellion, but 'who resisted who and what' at crucial moments was unclear. The définition of the state was always in question. While résistance seems to have been an inhérent part and thus constitutive of the state in many parts of Africa, in Somalia it was as much against the state - any state - itself.