Violence & social order beyond the state: Somalia & Angola

Abstract

This paper examines the activities of non-state actors in war in Somalia and Angola. Arguing that prolonged wars are characterised by the emergence of social orders of violence beyond the state, our analytical focus is on how actors establish and sustain these orders. A core influence is the insight from research on war economies that war is not equal to the breakdown of societal order, but represents an alternative form of social order. We therefore examine the economic activities of insurgents in regard to their embeddedness in social and political spheres. The central question in this paper is how economic, political and symbolic aspects interact and determine as well as transform social orders of violence. With the examples of Somalia and Angola, two rather distinct cases of non-state orders of violence are examined. It is argued that these orders represent forms of authority with fundamental structural aspects in common. We suggest that these orders can be systematised on a continuum between two poles of institutionalisation of authority beyond the state: a warlord system and a quasi-state system of violence.

Notes

1. We here reflect upon classical social theory (Weber, Elias, Bourdieu) as well as political economy (Marx, Polanyi). Despite their differences, these authors have stressed economic activities as a form of social interaction, an argument recently emphasised within economic sociology; cf. e.g. Swedberg & Granovetter,1992; Martinelli & Smelser, 1990:1-50.

2. The greed perspective is explored in a research project of the World Bank, addressing whether Greed or Grievance is the primary motivation for the outbreak of conflict, thereby reducing the complexity of the causes of war to a mechanical cause-and-effect relation tested on its support for greed or grievance as a cause of war; Collier, 1999; Collier & Hoeffler, 2000.

3. The first protagonists of this view were Keen (1998, 2000) in the English-speaking and Georg Elwert (1997, 1999) in the German-speaking context.

4. By re-writing and departing from Clausewitz’ idea, Keen argues that war may then become ‘a continuation of economics by other means’; Keen. 1997:69.

5. Reno and Duffield have both published detailed and well-researched empirical studies on societies at war and actors of violence, on cases such as Liberia, Sierra Leone, Nigeria (Reno; 1995, 1998, 2002) and Sudan (Duffield; 2001). These studies have contributed considerably to the understanding of the complexity of orders of violence.

6. See e.g. Chingono, 1996 and Mwanasali, 2000. See also Mustapha's fundamental critique of Bayart's notion of the ruled, the people, in an African context as representing a passive mass of victims, thus emphasising ‘the agency of leaders and strongmen, whilst ignoring wider and more important popular processes that limit the options of the former’; Mustapha, 2002.

7. Although a quasi-state system differs from a formal state in regard to the lack of international recognition, armed groups are able to establish diplomatic relations with official state representatives; Cf. Clapham, 1996:223.

8. This comprehension of the term warlord deviates from its traditional meaning in the context of Chinese militarism, according to which a warlord exercises effective control in a fairly demarcated territory by military means (cf. Adam, 1992:21).

9. Depending on how you count, Somalia is divided into 4 or 6 Clan families. The traditional nomadic Clans are the Dir (the Isaaq are originally a sub-clan of the Dir, but some scholars count them as an own Clan family), Darood, Hawiye, while the clans of Digil and Mirifle (summarised as Rahanweyn) are considered to be sedentary. However it is important to note here, that the distinction between nomadic and sedentary clans does not fully fit with the Somali reality since most of the ‘sedentary’ clans adopted an agro-pastoral system, i.e. a combination of mobile livestock-keeping and farming.

10. The social transformation of the Somali society was initiated and the direction of this transformation determined during colonialism. Due to limited space we focus on the late post-colonial system before the state collapsed. For a detailed analysis see Bakonyi, 2001:50-80.

11. As a result of the Somali war against Ethiopia (1977/78), Somalia lost its support from the Soviet Union, its main ally and donor. Although Western donors soon replaced the Soviet Union as main beneficiary, the support was henceforth linked to structural adjustment programmes. As a consequence, approximately 20% of the state employees were suspended (Brons, 2001:190, footnote 28).

12. With the extension of the patrimonial system as well as the nationalisation of the land, the state apparatus became the political as well as the economic centre of Somalia. The business contracts of wholesalers and landowners were allocated by the state. The position of these entrepreneurs became directly linked to the state.

13. Here and in the following the first name refers to the clan family, while the second is the sub-clan.

14. The SSDF was established by a handful of Daarod/Majerteen army members, who fled to Ethiopia after a failed coup in 1978. To begin with, the clan Hawiye was also integrated both in the SNM and the SSDF. Internal differences and competition for power led to disintegration, and in 1989 the Hawiye formed their own clan-based militia, the United Somali Congress (USC).

15. Many leading cadres of the SSDF, including the founder Abdullahi Yussuf, were detained in Ethiopia at the beginning of the 1980s. This led to the de facto collapse of the SSDF. The detainees were not released until the downfall of Mengistu in 1991, but they subsequently reorganised their militias. In October 2004, as a result of the ongoing peace negotiations amongst the Somali warlords in Nairobi, Abdullahi Yussuf was elected the new president of Somalia.

16. In the late 1980s, almost all male members of the Marehan in Mogdishu absolved military training and received a weapon (Bakonyi's own interviews). Subsequent to the defeat in the Ogaden war against Ethiopia, members of the Ogaden-clan in Ethiopia fled to Somalia. The Ogaden resettled in northwestern Somalia, received weapons and were mobilised against the SNM.

17. According to recent estimates, Somali remittance companies transfer between US$750m and US$1bn annually (UNDP & European Commission, Somali Unit, 2004:9). The three biggest remittance companies, the Dahabshiil, Al-Barakat or Hormuud and Amal, transfer approximately US$500m a year; this sector is a booming business in Somalia. The biggest remittance company today is Dahabshil, located in Hargeisa. After the bombing of Hargeisa and the mass migration in 1988, Dahabshil managed to organise the money transfer of the northern Somali refugees. During the war, its activities were extended to the southern regions (own interviews with Dahabshil management in Hargeisa, August 2003). Prior to 2001, the company Al-Barakat was more important than Dahabshil. But after the September 11th terrorist attacks, the USA accused Al-Barakat of funding terrorism, and froze the international assets of the bank. Meanwhile many branches of Al-Barakat have reopened under the disguise of a new name. Another important remittance company is Amal, but there are also many more companies engaged in this global Somali enterprise.

18. Ali Mahdi is a member of the Hawiye/Habr-Gedir/Abgal, while Aideed is a Hawiye/Habr-Gedir/Sa’ad.

19. For example, Osman Ato, a businessman and the former financer of General Aideed, shall have made a fortune through the organised export of looted goods from former state-run enterprises. Similarly, Compagnon (1998) describes how a businessman systematically exported looted office facilities to the Gulf states. In Kismayo, wholesalers bought and exported looted goods, including parts of the port and airport infrastructure, the fish factory and even whole ships (own interviews in Kismayo, July 2003).

20. During the Barre regime, the cultivation of Qat was forbidden thereby halting the profitable Qat production in Northern Somalia (today Somaliland). At the same time, the smuggling of Qat from Ethiopia to Somalia offered an alternative source of income, and this profitable trade has expanded during the war, albeit in an altered form.

21. The Juba Valley Alliance, for example, extracts their income almost exclusively from taxes collected at the harbour and airport in Kismayo. The export of charcoal has developed into a lucrative business. The warlord Qanyareh Afrah in Mogadishu finances his militia through the taxation of imported goods, as well as from fees paid by traders and international organisations using his landing strip, Danilieh, in Mogadishu. Mohamed Dhere, whose militias control parts of the Middle Shabelle region, cooperates closely with a businessman in Mogadishu, who controls the Eisleh landing strip in Mogadishu, frequently used by humanitarian flights, and the nearby sea port, where charcoal is the main traded good (own interviews, July and August 2003).

22. A detailed study of the reorganisation of the livestock trade, with particular emphasis on the border trade between Kenya and Somalia is Little, 2003.

23. A famous example is Osman Ato, the former financier of General Aideed who, since 1995, is in control of his own USC-faction.

24. There is a vast literature on the causes of the Angolan conflict; see in particular Heimer (1979), Clarence-Smith (1980), Guimarães (1998), Heywood (2000) and Luansi (2001).

25. The majority of the people in Angola were neither under the authority of UNITA nor the state, but were continuously exposed to attacks from both sides.

26. This presentation of the social order of UNITA is based primarily on the following sources: Minter, 1990/1994; Heywood, 1989/2000; Bridgland, 1986. The three special reports of the UN Monitoring Commission also provided important insights about the UNITA system; see Stuvøy, 2002.

27. The control system comprised three units: first, Brigada de Informaçao Geral (BIG) in charge of internal stability through observation of suspicious behaviour; second, the military policy Dragons, observing the behaviour of the soldiers; and third, a personal security guard for Savimbi (cf. Monitoring Mechanism, 2001: paragraphs 26-27; Bridgland, 1986:107, 110).

28. The richest diamond mines in Angola are located in the northeast, but there are mines also in the central Highlands and in the south. Diamonds are found both in mines and in rivers (cf. Dietrich, 2000c).

29. Garimpeiros worked as diamond diggers in Angola since the late 1970s. In the early 1990s, UNITA began to control the border crossing of garimpeiros, and their numbers increased from 10,000 to 100,000 between 1991-1996 (Dietrich, 2000b:174; Monitoring Mechanism, 2000: paragraph 157).

30. UNITA was, for example, able to buy oil from the state oil company Sonangol due to their informal, personal contacts with the company.

31. Although warlord systems differ from bandits and other criminals, warlords can develop from such criminal structures: ‘the transition from outlaw pure and proper to warlord suggests the development and maintenance of a social order, clearly understood, among the outlaws, as well as the explicit recognition of a purpose in rebellion’ (Chan, 1999:165).

32. A similar logic was observable in Afghanistan, where by 1992 the parallel collapse of the pro-communist regime of Nadschibullah and the opposition insurgent movement Mujaheddin lead to a power takeover by warlords who drew their legitimacy mainly from the local group Qaum. This era of warlord authority in Afghanistan ended in 1994 when the Taliban established control.

33. According to Trutz von Trotha (1994, 1995), a traditional system of violent self-help is contrasted by the monopolisation of violence in the modern state. The order of violent self-help is characteristic of less differentiated societies. Here, autonomous political groups and communities such as households, kinships, cities etc., claim their right to collective defence. In these societies, the use of force is an acknowledged means of conflict resolution, and in cases of murder or blood vendetta, it is even an obligation to use violence to settle conflicts. Under such conditions, a certain stability and security is established through a ‘balance of fear’ (Trotha, 1994:38).

34. If the supply of resources to a warlord system is problematic, the system will most likely transform into a criminal gang. An exception to this rule is the focus on robberies of aid deliveries. The continuation of international aid supply structures is a relatively secure resource base.

35. Such a ‘tax’ on the security of certain actors is not necessarily based upon use of physical force. These payments may represent a strategic relationship between the actors of violence and business representatives. In his study of the Italian mafia, Gambetta (1993) recognised that the provision of security through the mafia was linked both to monetary payment and to a promise of loyalty, but that the previous threat of use of force considerably contributed to the establishment of the security relationship. Nonetheless, relations between the mafia and its clients were not fully determined either by violence or coercion but depend also on the establishment of reciprocal advantages for both sides.

36. With the disappearance of the state, the terms legal and illegal are of no real worth in the local context.

37. Such a development was observed in Lebanon, where the various insurgent movements reestablished localised state structures, and thus a number of different quasi-state structures of authority emerged on the basis of distinct warlord systems. The organisational basis of the militias was the structural foundation upon which the more or less tranquil transformation from a society at war to a post-war society was made possible. The various local and regional quasi-state structures were successfully integrated into and subordinated to the centralised state system.

38. Similar to UNITA, the Eritrean Peoples Liberation Front (EPLF) established a legal system and provided health services to the civilian population in Eritrea. The Maoist rebels in Nepal established a banking system and a court system (Humphreys, 2002:12). In Columbia, the rebel group, Armed Revolutionary Forces of Colombia (FARC), provided several state services, and contributed to the improvement of the provision of basic needs in rural areas (Labrousse, 1999:318).

39. Another example is Lebanon, where almost all functions of the formal state were taken over and performed by insurgent groups, such as police and intelligence services, but also the organisation of the transportation system (Naylor, 1993:16f.). The establishment of quasi-state systems of authority was particularly successful in the areas controlled by the drusian hizb at-taqaddumi al-ishtiraki in the mountainous area of Schuf and the Christian al-quwat al-lubnaniya in Eastern Beirut, on the Coast between Beirut and El-Batroun.

40. Kopp (1999) has examined how economic embargos contribute to further informalisation, criminalisation and political conflict within states, and draws a link between embargos and the rapid expansion of illegal trading.