01 Mar 2006: An Analysis and Comparison of Armed Groups in Somalia, Anthony Vinci ESSAY An analysis and comparison of armed groups in Somalia Anthony Vinci Doctoral student at the London School of Economics. Protracted state collapse in Somalia has led to a multiplication and diversification of armed groups. We can speak of at least five types of armed group: faction, warlord, business, court, and Islamic militias. These groups differ in important ways, yet often are simply classified as ‘militia’ or ‘warlord’. This essay seeks to add a measure of analytical rigour to the classification of armed groups and provides a comparison using a framework of purpose, motivations, logistics, and command, control and communication. It concludes with some observations about the importance of making these distinctions when formulating policy for this region. Introduction Protracted state collapse in Somalia has led to a multiplication and diversification of armed groups.1 We can speak of at least five types of armed group in Somalia: faction militias representing and protecting clans; warlord militias resembling personal armies; militias bought by businessmen to represent their business interests; the shari’a courts’ militias; and, in the case of al Ittihad, Islamist militias. These armed groups differ in important ways, in terms of their motives and their means of supporting themselves. The assumption, especially in the press, that armed groups in Somalia are all much the same is reflected in the use of the blanket terms ‘militia’ and ‘warlord’. This is misleading from an analytical point of view and may also lead to the prescription of ineffective or inappropriate responses to particular militias, which, in certain circumstances, may have dangerous consequences. Not all commentators are guilty of such oversimplification; some analysts do make distinctions between, for instance, faction militias and warlords.2 However, the distinctions usually have not been as rigorously examined as they could be. Context At the point of state collapse, the Somali situation differed in important respects from civil wars in which the state is taken over by one of the factions which sought to overthrow it. In Somalia, no faction was able to seize and retain the sovereignty of the state. Rather, the situation transformed into a series of skirmishes between militias. The eminent Somali scholar I M Lewis sums it up well: In 1991/92, reactively influenced by the example of the Somali Salvation Democratic Front (SSDF), Somali National Movement (SNM), United Somali Congress (USC) and Somali Patriotic Movement (SPM), the general tendency was for every major Somali clan to form its own militia movement. Thus clans were becoming effectively self-governing entities throughout the Somali region as they carved out spheres of influence in a process which, with the abundance of modern weapons, frequently entailed savage battles with a high toll of civilian casualties. The political geography of the Somali hinterland in 1992, consequently, closely resembled that reported by European explorers in the 19th century, spears replaced by Kalashnikovs and bazookas. These clan areas could only be entered or traversed by outsiders (people of other clans, foreigners), with the consent of the locals and, usually the payment of appropriate fees or ‘protection’.3 Indeed, it has been noted that since the collapse of the state, fighting has broken out between all clan groups.4 To understand how armed groups have evolved in the Somali context, one must appreciate the nature of the Somali clan system. Most Somalis identify themselves primarily in terms of their lineage or clan. These clans are segmented into six clan families that break down into various sub-clan units, all the way down to the level of the individual.5 Clans are led by elders who are responsible for negotiation and dispute mediation. An overlaid social unit is the diya-paying unit, composed of close relatives who contract to pay (or receive) blood-money, diya, if one of the members of the group kills an individual from, or raids the resources of, another group.6 The diya process is used as a method of conflict resolution, short of revenge killings, and requires clan elders to negotiate a payment in expiation of the crime.7 In Somalia, political entrepreneurs, in general, must mobilise segments of the clan system to support them. This mobilisation can be based on the proximity of the relationship, or may involve the use of less closely related groups, in alignment against a common opponent; this is referred to as gashaanbur in Somali pastoral society.8 Clans might also form alliances with related clans in order to counter the perceived power of a rival clan (which might have its own alliances). Inversely, factions may form for two reasons. Because segmentation possibilities are endless within the Somali kinship system, there is ample room for factionalisation based on the needs of the clan. Factionalisation is also possible, however, based on the choice of particular political entrepreneurs who make their own rational choice. This seems to have been the case when Usman Ato split with Mohammed Aidid in 1994.9 As a result of factionalisation there has been a vertical multiplication of separate armed groups in Somalia since the early 1990s. Initially, clans joined together to fight the Somali government, which was roughly equated with the Darod clan and its allies, controlled by then President Mohamed Siad Barre. In 1991 Barre was ousted, and since then there has been a “devolution of warfare to lower and lower levels of clan lineages” as the higher level alliances factionalised.10 Somali armed groups Along with the multiplication in factions, there has been a lateral growth in the types of armed groups in Somalia since the civil war.11 With the collapse of the state, the clan or ‘faction’ militias became the providers of security and, often, the political representatives of the clans. These faction militias have continued to constitute the bulk of the armed groups in Somalia. Some of the militia leaders, however, were able to gain independence from the clan, thereby using the militias to represent their personal interests: they became warlords. Since the mid-1990s, businessmen have also raised their own armed militias to defend their business interests. These militias have sometimes been outsourced to the shari’a courts that have formed throughout Somalia. Finally, at least one set of Islamist militia has been set up by al Ittihad. The basic structure of each of these types is roughly similar and the nature of their strategy and tactics are also roughly similar. Militias are typically led by a ‘warleader’, who is the direct military commander of the militia. Below the warleader there is a smaller group of permanent, trained officers and soldiers, who are under his direct control. Further down is a larger group of paid or volunteer fighters. However, these types of armed groups all vary along set lines. These include the reason for their formation, their purpose, the way they motivate their fighters, and their logistical systems.  Formation and purpose Faction militias Faction militias are the most prominent type of Somali armed group. These militias can range in size from tens to hundreds of men and are formed along clan lines and within the clan structure. Daniel Compagnon notes that clan-based armed groups arose naturally in Somalia in the 1980s when armed groups such as the Somali Salvation Democratic Front (SSDF) were forming to combat Barre’s government. The clan-based militia organisation was easier as, “given the lack of resources (both money and weapons), the unit size requirement of guerrilla warfare, and the difficulty of giving military training to individualistic camel herders, it was more efficient to opt for a military structure based on kinship segmentation”.12 In faction militias, the warleader is the representative of the clan. He is assigned by the clan elders to lead an armed group, but although he commands the fighters, the latter remain loyal to the clan. Even though in stable periods warleaders are typically not very popular, during periods of potential conflict the clan will turn to them for help, for “the clan feel[s] that they are their savior … because they command the firepower”.13 Yet, his ability to please the clan is paramount and is based on his fulfilment of his duties. Failure to do so may result in the clan abandoning him. Regardless of the personal ambitions of any particular warleader, the faction militia represents the clan’s political aspirations and defends its territory and interests. With the loss of a state that plausibly could provide security for the people of Somalia, the clan became the relevant political community and the faction militia is what provides the clan with security. Even when it has additional goals, such as representation in high level political meetings, the first and foremost issue for the militia is security. Beyond this, the military power which the militia provides grants a method by which the clan can be represented in politics amongst other clans. Thus, the faction militia should be seen as an extension and representative of the clan. Warlord militias In some instances, a warleader may become independent of the clan and control a militia without being connected to the clan’s political, social and economic structure. Such men are known as warlords and we can refer to their armed groups as ‘warlord militias’. In most cases, militias exist on a continuum, with a varying mixture of clan and personal loyalty. However, in extreme cases, a warleader may come to monopolise the loyalty of fighters and use the militia for personal political ambitions. In other instances, the warlord may take the initiative in putting together a, usually small, militia made up of fellow freelance militiamen. The warlord militia has no other purpose than to reflect the personal, often political, goals of the warlord himself. Although he is often concerned with raising funds, generally, this is used only as a method of retaining his fighters; the warlord is concerned with gaining power; clan loyalties and other issues of public good are completely discarded, or used solely to further personal goals. Yet, the warlord militia is structured much like the faction militia. The warlord is at the pinnacle of the hierarchy and below him are fighters. Because there is no clan to represent and form the basis of the militia, however, rather than these troops being irregulars who are members of the clan, they are privately hired freelance militiamen. Business militias Initially, Somali businessmen funded factions and warlords, but as the conflict progressed, they found this counterproductive. For example, the two main factions controlling Mogadishu in the 1990s typically received backing from businessmen. However, these administrations never established the security and predictability businessmen needed and they began to lose money on their ‘investment’. More generally, private security was not always adequate in an economy based on services and cross-regional and cross-border trade. Clearly, greater order and security were needed to promote commerce. For these reasons, many businessmen ceased their support for warlords in the late 1990s.14 Thereafter, businessmen directly employed freelance militiamen, along the same militia structure outlined above, but with the businessman himself as the warleader. The businessman might also outsource this role to another experienced person. The militia was otherwise organised around the standard structure and members might be drawn in whole or part from the businessman’s clan or be privately hired. In some cases the businessmen formed coalitions in order to accomplish militia formation and management. The raison d’être of the business militia is the defence and representation of its patron’s business interests. This may include defending property or defeating a potential rival, but the logic behind its actions will always go back to the businessman’s interests above those of clan or any other factor. Court militias Often businessmen have outsourced their militias to the shari’a courts that have emerged to fill the legal vacuum left in the wake of state collapse.15 These courts were established in Mogadishu and elsewhere to provide some order through a shari’a law system. Typically they and their militias are clan-based and their rulings hold within a clan’s territory. They are financed by ‘taxes’ from businessmen, in exchange for providing the latter with armed guards and for maintaining social stability and providing (public) protection services.16 These services range from relations with groups such as the Rahanwein Resistance Army (RRA) to clearing the road from Mogadishu to Merka.17 In general, the court militia has a similar structure to the business militia. The courts are focused on defending the interests of their rulings and of their backers. Unlike the faction or warlord militia, they do not generally seem to have their own strategic political goals. Certainly the businessmen who fund these courts exercise some influence, but it is difficult to assess this accurately. Finally, the Islamist goals of such groups tend to be weak or non-existent, except where they have been infiltrated by the more fundamentalist al Ittihad organisation.18 Islamist militias Related to the court militias are Islamist militias, in particular those of the fundamentalist organisation al Ittihad, which has controlled militias at various times since 1991. Al Ittihad has controlled areas including Merka and Kismaayo and the town of Luuq in the Gedo region.19 Although its militias are organised like other types, al Ittihad bases its support on religion rather than clan or other factors. Though the group initially functioned like faction militias in looking to control territory, more recently it has switched its strategy to focus on controlling courts.20 The true intentions of al Ittihad are not completely clear. The organisation, which is comparable to the Muslim Brotherhood of Egypt, is concerned with the creation of an Islamic state. When al Ittihad managed the administration of towns, it did carry out this promise and instituted Islamic law and practices.21 However, the implications of these goals are debated: they range from an innocent wish to uphold Islamic values to accusations of creating a base upon which transnational Islamic militants could attack the Ethiopian government to being a front for al Qaeda.  Comparison These five types of armed groups in Somalia have all evolved under similar circumstances. Moreover, they have all had to meet similar requirements in order to perpetuate their organisations. First, they have had to motivate individuals to fight for them. They have had to obtain weapons and equipment with which to fight. Finally, they have had to institute command, control, and communication structures with which to direct their forces. The following section will compare the types of armed groups in Somalia along the lines of this tripartite framework.22 Motivation In general, there are two methods of motivating soldiers in Somalia. Either the militia leader may rely on clan loyalty (or, in the case of al Ittihad, religious loyalty) or on direct economic incentive. Loyalty is tied into the nature of the clan system and is dependent, at least tacitly, on the consent of the clan. Kinship is used as a “ready-made ideology”23 and provides a strong incentive for individuals to fight voluntarily. It has also become possible for militia leaders directly to enlist fighters following the increase in the numbers of freelance militiamen in Somalia, and this has allowed for the growth of non-clan-based armed groups, including warlord, business, and court militias. Since the late 1980s, many young, unemployed and disenfranchised Somalis, or mooryaan, have come to urban areas, especially Mogadishu, in order to find jobs, adventure, or to fulfil their ambitions. These men are distinct from other parts of Somali society. Usually they have come from nomadic backgrounds or are otherwise separated from their clan. Many are addicted to drugs, beyond the ubiquitous khat. They sometimes even speak their own distinct dialect that is “virtually unintelligible to outsiders”.24 Since these men usually come from non-local clans, they are able to circumvent many of the rules of interclan rivalry.25 Because they are ‘foreign’, in that they are not from the local clans, they are free from traditional commitment issues associated with clan life, such as xeer or diya payments for transgressions.26 This means that they consider looting and other generally unacceptable activities permissible.27 Initially, these mooryaan were concerned with finding a place from which to prosper within Barre’s patronage networks.28 This practice continued after the collapse of the Somali state, as their loyalty could be sold into the patronage networks of warlords and other militia leaders.29 In sum, this meant that there was a reservoir of freelance gunmen into which aspiring militia leaders could tap. Although militias, and faction militias in particular, continued to use clan-based fighters, the mooryaan fundamentally changed the balance of power in Somalia by allowing militias to form outside the clan’s control. Faction militia The faction militias and their warleaders use clan-based loyalty to motivate voluntary recruits. The members of faction militia are typically recruited directly from a single clan.30 The council of clan elders will send representatives to each family’s house looking for young men to serve. There is no history of forced conscription in Somalia and the men are all voluntary recruits. They will likely already be armed or have been mobilised before as part of a standby militia. Similar to conventional reserve systems, they are called up to fight and then can demobilise back to civilian life. Thus, faction militias are closest to what we think of as true militias in that they are the temporary armed extension of a larger civilian political community. Warlord militia The independence of a warlord depends on his ability to support himself without the clan’s backing. Since the involvement of clan elders is necessary to motivate young men to fight voluntarily, they have to be ‘kept in the loop’. What truly shifts the balance towards allowing a warlord to break away from the clan is the use of freelance gunmen. The mooryaan allow the warlord to monopolise fighters’ loyalties because these can be bought on a pay-for-service basis. In Somalia, where the average young man lives at, or below, the poverty line (usually calculated at US$1 per day), the motivation to be paid for combat services is strong. Immediate economic incentives thereby provide a warlord with the tools necessary to obtain his independence. However, the trade-off is that the warlord cannot rely on the loyalty of these troops and must keep them from selling their services to a rival. In the absence of economic incentives, such fighters may move their loyalty to another patron, set up an economic enterprise of their own, or even rebel against the patron. As one aid worker put it, “all is well if the militias are getting paid … if not, they run wild, setting up their own checkpoints etc”.31 Consequently, the reliance on economic incentives deepens a reliance on sustainable resource exploitation. Business militias As in the case of warlords, business militias use immediate economic incentive to motivate their fighters. Even though they are usually affiliated with one sub-clan, they rely even less on clan loyalty because business, by its nature, crosses clan lines and is not overtly political. Private security, however, is dangerous because, at times, the security force might turn on their employer and extort from him. Court militias Like warlord and business militias, court militias rely on freelance fighters, whose loyalty is secured through economic incentives. In the same way this is necessary, as without the clan to provide a source of loyalty, the fighters would have little incentive to continue fighting. Islamist militias Islamist militias can rely on religious belief for loyalty. This loyalty can cross over clan lines. For example, in its administration of Luuq, al Ittihad also invited its members who were not part of the locally dominate Marehaan clan to participate in governance.32 However, this Islamic loyalty was more based in the upper echelons of the al Ittihad leadership. The fighters themselves were “composed mainly of young gunmen [mooryaan, known locally as jiri], whose devotion to tenets of fundamentalist Islam was negligible and who fought in the name of al Ittihad only because al Ittihad paid them”.33 Logistics The sources of militias’ weapons and other goods are varied and depend on the region and particular history of a militia. Nevertheless, since the collapse of the state, a common method for supplying weapons and other equipment has been looting. Other frequent approaches to supply needs have been to take part in various economic entrepreneurial activities or to be funded by businessmen or external actors. Somalia’s strategic importance during the Cold War made it one of the best-armed states in sub-Saharan Africa, and through Barre’s policy of divide and rule, various clans were armed. After Barre’s fall the ‘public’ goods of the state, formally controlled by individual clans, were extensively looted by the factions, especially the large state armoury. Businessmen have also been apt to back militias, especially the faction militias of their own clans. These are businessmen who do not have their own militia as discussed above. In general, militias are financed by businessmen for two reasons. In the short term, there are financial rewards from looting or establishing control over certain properties. In the long term, financing a militia is an investment gamble on potential political power. Militias have also created a range of enterprises that are valued because of their ability to command force. Common practices include setting up checkpoints and demanding payment for safe passage, or occupying ports or airports and demanding ‘taxation’ for their use. In some instances, land occupation may become institutionalised, where militias set up fiefdoms and trade taxation for protection from other militias.34 In some cases they even ‘normalised’ relations, as militia groups intermarry with local women. International aid can be used by militias in both tactical and strategic ways to make profit.35 Tactically, by pretending to be non-combatants, militias can coerce, extort, or manipulate humanitarian aid agencies into giving them aid. Strategically, population movements can be used to dictate the placement of aid. An ‘aid economy’, such as providing housing or security for the international agencies, can be created, which is then tapped by the militias for their own funding. In particular, it is common practice for UN agencies and NGOs to hire private guards. In some cases, the militia effectively forces the agency to rent their protection services in order to safeguard the agency.36 A related issue concerns militias being hired by multinational corporations. This is a different situation, however, because the private corporations, unlike NGOs and the UN, are themselves in competition. An example is the competition between Dole and Somalfruit in southern Somalia.37 The companies hired local militias and used them to threaten farmers into selling them bananas.38 The large Somali diaspora also provides funding to the militias within Somalia using the worldwide networks of the numerous Somali remittance companies. Support may also come from external Somali political actors; for instance, the Somali National Front (SNF) was funded by former Barre government figures living in Kenya.39 In some cases, the militias acted as proxies of the foreign policies of external states, especially Egypt, Libya and Ethiopia. This led to Egyptian and Libyan funding of Aidid and Ali Mahdi in Mogadishu.40 Ethiopia, for its part, has funded the Puntland administration, Hadawdle (USC/SPM) militia, the RRA, and the SNF.41 In fact, the International Crisis Group has noted that formally independent and powerful warlords Hussein Aidid and General Morgan have been completely dependent on Ethiopia.42 Faction militias Faction militias have a diverse set of means of support that arise from their clan connections. The militias are extensions of clans, many of which have farming or nomadic roots and can provide food and other basic survival goods. The clans also turn to their, often extensive, overseas diasporas for support. Like other militias, faction militias may develop various economic enterprises on their own. These include looting, the diversion of humanitarian aid, support from businessmen, and occupation of land. However, these enterprises may represent the beginning of the slippery slide into independence for a warleader as they provide a source of funding which is potentially separate from the clan. Warlord militias Warlords are concerned with gaining and maintaining their independence, and this is only possible if they have the economic resources to fund their militias. They participate, therefore, in most of the economic activities listed above with the goal of monopolising the economic activities or relationships at the expense of the clan. This may mean having a private port to tax or being the sole connection for external supporters. A good example is Mohamed Quanyare Afrah, based in the Deynle neighbourhood of Mogadishu, who receives funds from fishing interests and a private airstrip, among other ventures, and uses these to maintain his militia.43 By controlling finances, warlords buy the loyalty of the men they need as well as the equipment to arm them. Large cash reserves are necessary for warlords to be effective, as a firefight can cost $4,000 an hour in ammunition alone.44 Thus, we should not necessarily see the warlord’s penchant for economic exploitation as an end, but as a means to securing power through the control of a militia. Business militias Businessmen, on the other hand, are fully focused on economic exploitation. As detailed by Mark Bradbury and Peter Little, among others, the economic sector in Somalia is doing relatively well, in spite of state collapse.45 Moneymaking enterprises include ‘grey’ and ‘black’ market activities such as checkpoints and weapons trading, as well as ‘whiter’ activities which run the gambit from trade in cattle, to money transfers, to providing utilities. Broadly speaking, in the early 1990s, businessmen focused on grey-area trade in looted goods, weapons and food aid, but following the UN intervention in 1993/94, they made huge sums of money through real-estate rentals, money exchange and transport, which allowed them to shift their focus to other businesses such as hospitality and telecommunications.46 As with warlords, these enterprises provide the cash to buy militia loyalty, weapons and other necessary equipment. Court militias Court militias have been backed by businessmen and, to some degree, by the clans themselves. This is unlike warlord and business militias, which have come to find their own wealth-generating activities. In exchange for their management of the militias, the courts are expected to create and maintain as much predictability as possible. This arrangement clearly limits the independence of the court militias. Islamist militias Al Ittihad also engaged in entrepreneurship in order to raise funds. For instance, they controlled the ports in Kismaayo and Merka. There has also been some attempt by al Ittihad to incorporate businessmen into their organisation to find funding.47 What differentiates them from other types of militias is that they also are funded by international, non-Somali Islamist sources. This is possible because of their cross-cutting ideology. It is not clear to what extent al Ittihad had terrorist or other subaltern funding, but they certainly have been funded by Islamic NGOs in various guises. Command, control, and communication Interestingly, the different militia types have essentially the same command, control, and communication structure. In particular, the militias have a relatively decentralised power structure. This is partially because of the nature of the clan system, in which there is little application of discipline in the Western military sense. Militia fighters are volunteers, and severe discipline, especially executions, would fall under the traditional system of diya, in which grievances must be recompensed for harm or loss and is therefore not frequently practised. This is also true of non-clan-based militias as a result of the inherently undisciplined nature of the mooryaan. The relative lack of discipline and direct control is combined with the presence of cellular telephones, making it possible to coordinate troops across long distances. Consequently militias are able to maintain a relatively undisciplined and decentralised command and control structure but keep units in communication. Furthermore, even though the militias have different motivational and economic attributes, their strategies are essentially similar. Though the goals of Somali warfare may differ, from controlling economic resources to gaining the status necessary to take part in peace processes, the strategies towards obtaining these goals are convergent. Military power is used to control territory or populations, force other actors into making concessions, or obtain further resources. Indeed, the militias have so converged in their methods of warfare that we may speak of a Somali way of war. The Somali way of war The strategy and tactics employed by all militias constitute an evolution in the combination of traditional clan raiding practices, conventional army tactics, and new warfare methods developed after the collapse of the state. The militarisation of Somali nomadic life has left Somalis with a heritage of raiding practices easily deployable in the semi-anarchy of post-state Somalia. The ample funding of Barre’s army by both superpowers during the Cold War left the country with a large military class, including militia leaders such as General Morgan and Mahumud Aidid, who have lent their expertise to the militias. Combined with these influences, there is the development of new tactics based on the unique environment of Somalia, in particular the use of ‘technicals’ (pick-up trucks mounted with heavy machine-guns or grenade launchers) and the use of khat by fighters. This has resulted in a special brand of Somali tactics and a unique Somali way of war. Ambushes are the most common form of attack, since they demand that the militia be brought together for only a short period and promise looting after the attack. Larger, coordinated attack or defence operations are much more difficult. In general, it is a very fluid warfighting method. The UN illustrates it nicely: … armed clashes tended to take the form of wild, chaotic exchanges of fire, featuring front-lines which could shift fifty or one hundred kilometers in a day as lines of defense disintegrated and regrouped. Supply lines were ad hoc to nonexistent, relying mostly on looting; occasionally a [technical] was captured or destroyed simply because it ran out of fuel.48 This evolved out of the more conventional warfare which took place early in the war, in which “clan militias held neighborhoods tenaciously and shelled one another’s positions for weeks on end”.49 More recently, with the growing expense of buying, rather than looting weapons and ammunition, and more limited political goals, fighting has mostly become smaller in scale and more concerned with the defence of checkpoints and other economic resources. Conclusions This essay is intended to counter the monolithic view of Somali armed groups and to analyse the types of armed groups that have arisen in the ongoing conflict. Admittedly, these are idealised categories and merely represent trends in the different groups’ means of mobilisation. There is also a significant amount of overlap and reversion. Nevertheless, we can draw some conclusions about the security implications of these groups. There has not been space here to compare and contrast all the types of armed groups, but it is instructive to draw some insights from contrasting the warlord militias with the others. The warlord militia differs from others in terms of its purpose for fighting, the motivation of its fighters, and its logistics system. The warlord controls his fighters for his own personal, political purposes, drawing the forces together in a patronage system. The faction militia, on the other hand, can turn to clan loyalty – a more lasting means of motivating fighters. In consequence, the warlord has to be concerned primarily with economic exploitation, for this is essential if he is to retain power over his militia, since he cannot rely on loyalty. This contrasts with the businessman and his militia, for whom economic ends in themselves are sufficient. The warlord militia also differs from the faction militia, which can turn to external diaspora supporters with less fear of a loss of authority, since the faction militia is only an extension of the clan. It is important to make these distinctions, because they help explain the warlord’s interaction with domestic and international actors. For instance, it may be possible to bribe the warlord, since he needs economic resources to maintain power, though there will be limits to the effectiveness of this tool in that he will not dissolve his militia because its maintenance remains essential to his own personal power. Appeals to the warlord’s sense of clan loyalty will be less effective than in the case of a faction militia. In general, we must keep in mind that there are specific responses to warlords which may be valuable in specific cases when they are being dealt with. This more nuanced way of considering armed groups in Somalia is particularly valuable when considering security. Acknowledgement The author would like to thank Mark Bradbury for valuable discussions concerning Somalia as well as the many helpful sources from his research trip who must remain anonymous. He would especially like to thank Dominique Jacquin-Berdal for her mentorship – she will be remembered. Notes The term ‘armed group’ is used here to refer to organised, non-state military actors. The work of Mark Bradbury, Ken Menkhaus and the War Torn Societies Project provides examples of such distinctions. See M Bradbury, Living with statelessness: The Somali road to development, Conflict, Security and Development 3(1), 2003; M Bradbury, interviewed by S Feldman and B Slattery, Living without a government in Somalia: An interview with Mark Bradbury, Journal of International Affairs 57(1), 2003; K Menkhaus, Traditional conflict management in contemporary Somalia, in I W Zartman (ed), Traditional cures for modern conflicts: African conflict ’medicine’, Lynne Rienner, London, 2000; K Menkhaus, Somalia: Combating terrorism in a failed state, International Crisis Group (ICG), Nairobi, 2002; K Menkhaus, Political Islam in Somalia, Middle East Policy 9(1), 2002; K Menkhaus, State collapse in Somalia: Second thoughts, Review of African Political Economy 97, 2002; K Menkhaus, Warlords and landlords: Non-state actors and humanitarian norms in Somalia, Liu Institute for Global Issues, University of British Columbia, 2003; K Menkhaus, Somalia: State collapse and the threat of terrorism, Adelphi Paper 364, Oxford University Press, Oxford, 2004; War Torn Societies Project, Towards social integration of the militias and armed youth (draft), WSP Somali Programme in Puntland, 2000; War Torn Societies Project, Somalia: Path to recovery building a sustainable peace, WSP Somali Programme, 2004. I M Lewis, Blood and bone: The call of kinship in Somali society, The Red Sea Press, Lawrenceville, NJ, 1994, p 231. A M Issa-Salwe, The collapse of the Somali state, Haan Associates, London, 1994. These clans are the Darod, Isaaq, Dir, Hawiye, Rahanwein and Digil. I M Lewis, A modern history of Somalia: Nation and state in the Horn of Africa, Westview Press, Boulder, Colo, 1988, p 11. K Menkhaus, Traditional conflict management in contemporary Somalia, op cit. Clans can also interact in peaceful ways through xeer, a set of customary laws. The laws are precedent-based and passed down through oral transmission. D Compagnon, Somali armed movement: The interplay of political entrepreneurship and clan-based factions, in C Clapham (ed), African guerrillas, James Currey, Oxford, 1998. Ibid. Menkhaus, Warlords and landlords, op cit, p 20. In particular, there has been significant intra-clan fighting amongst the sub-clans of the Hawiye and Darod clans. The following typology is based loosely on War Torn Societies Project, Somalia: Path to recovery, op cit. Compagnon, op cit, p 77. Interview, UN Source, Nairobi, 30 March 2005. International Crisis Group, Somalia: Countering terrorism in a failed state, ICG, Nairobi/Brussels, 2002, p 3. Menkhaus, Warlords and landlords, op cit, p 30. United Nations Development Program (UNDP), Somalia human development report 2001, UNDP, Nairobi, 2001. Ibid. Menkhaus, Political Islam in Somalia, op cit. Ibid. Ibid. Ibid. For a more detailed description of this framework, see A Vinci, The ‘Problems of mobilization’ and the analysis of armed groups, Parameters 36(1), 2006. Compagnon, op cit, p 83. War Torn Societies Project, Somalia: Path to recovery, op cit, p 29. Not all freelance militiamen might be considered mooryaan but mooryaan make up the bulk. W Reno, Somalia and survival in the shadow of the global economy, Working Paper 100, Queen Elizabeth House, Oxford, 2003, p 15. Another notable effect of mooryaanism is that the Somali traditional respect for the safety of non-combatants in conflict, known as birimageydo (’to be spared from the iron’) has broken down (WSP 2004). Reno, op cit, p 15. R Marchel, Forms of violence and ways to control it in an urban war zone: The mooryaan in Mogadishu, in H M Adam and R Ford (eds), Mending rips in the sky: Options for Somali communities in the 21st century, The Red Sea Press, Lawrenceville, NJ, 1997 for more on mooryaan. Though it is possible that some are former professional fighters or freelance fighters from other clans. War Torn Societies Project, Somalia: Path to recovery, op cit. Interview with aid worker, Nairobi, 1 April 2005. Menkhaus, Political Islam in Somalia, op cit. Ibid, p 113. United National Development Office for Somalia (UNDOS), Study on governance in Gedo region, United Nations Development Programme (UNDP), Nairobi, 1998, p 38. M LeRichie, Unintended alliance: The co- option of humanitarian aid in conflicts, Parameters 34(1), 2004, pp 104–120. Human Rights Watch, Somalia faces the future, HRW, London, April 1995. Ibid. Apparently both were from the same Sa’ad subclan (Habar Gidir clan). UNDOS, op cit. Ibid. Ibid. International Crisis Group, op cit, p 4. Menkhaus, Political Islam in Somalia, op cit. Cost estimate from ibid. Bradbury, Living with statelessness, op cit; P D Little, Somalia: Economy without state, The International African Institute, James Currey, Oxford, 2003. International Crisis Group, op cit, p 3. Menkhaus, Political Islam in Somalia, op cit. UNDOS, op cit, p 75. Ibid.
