The traders were, however, prevented from taking more of the victims to hospital by the bandits who repeatedly returned to the village mounting six sporadic raids throughout the day. The death toll in this attack would have been higher had it not been for two administration policemen and three home guards who put up resistance before running out of ammunition. During the raid, the bandits also stole about 3,000 cattle, 5,000 sheep and 4,000 camels (Daily Nation, 13 July 2005:3; Daily Nation, 14 July 2005:5; Njeru, 2005:11).
In a revenge attack, ten people were killed at Bubisa trading centre, which is about 80 kilometres from Turbi. The ten, all members of a church group, were hacked and speared to death. They were pulled out of a church vehicle by a mob and killed as their Italian parish priest watched in horror. The priest was spared only because he was not a member of any of the warring communities. His pleas to the mob to spare the lives of the ten people went unheeded. The attackers accused the group of belonging to a community that killed their kinsmen at Turbi (DN, 2005b:5).
All told, 95 people – 23 of whom were children – were killed over two days in the massacre and revenge killings. About 6,200 people were displaced, out of whom 1,200 were children. Eleven primary schools were affected by the violence, and surviving pupils required security since most were too traumatised to return to school. Sixty-six children were orphaned by the massacre in Turbi alone. Many of the traumatised victims needed counselling. Given the region’s inadequate facilities in offering counselling services, civil society organisations at both grassroots and national levels, appealed to qualified counselling personnel nationwide to volunteer their services to those affected. About 12,000 animals were stolen out of which less than 50 per cent were recovered (Daily Nation, 19 July 2005:3; Njeru, 2005:11).
Marsabit is one of the arid and semi-arid districts of northern Kenya largely inhabited by pastoral communities. Violent conflicts among and between pastoral communities in the region have resulted in the loss of hundreds of lives, an indication that the state has failed in its responsibility of providing security and development in these regions. These are the ‘forgotten badlands’ of Kenya the country since the state’s presence is almost negligible. This particular violent conflict pitted the Borana and Gabra communities against each other. It also had cross-border dimensions into Ethiopia.
Violent conflict & human security in Kenya’s ‘Badlands’
Kenya's forgotten ‘badlands’ cover 80 per cent of the country's land mass, and account for 70 per cent of national livestock production, whose estimated value is US$1 billion (Kenya, 2004:vi). The areas contribute about 16 per cent of the country's Gross Domestic Product (GDP). However, violent conflicts involving pastoralists have become widespread and increasingly severe in northern Kenya. Dependent on livestock for their livelihoods, pastoralists are largely nomadic or transhumant. Access to pasture and water is essential for the survival of their livestock. These resources are, however, becoming increasingly scarce and over-utilised, yet they have to be shared amongst the pastoralist's communities. Conflicts associated with competition over scarce resources, stock theft, and large availability of illicit arms, are widespread in the region and becoming of increasing concern. Stock theft in the ‘badlands’, for example, costs the Kenyan economy over US$130 million annually (Pkalya et al., 2003:10; Goldsmith, 2005:9; Kerrow, 2005:9).
Violent pastoral conflicts in northern Kenya's ‘badlands’ have escalated not only in terms of the level of violence, but qualitatively: killing itself is an aim, deliberately targeting both women and children, often involving rape – traditionally this would have been taboo. It often has a commercial motive: there is a national and indeed regional livestock trading network. The participants are different today: hired fighters and warriors of specific ethnic groups as well as members of foreign liberation fronts. These are liberation movements based in neighbouring countries that are waging wars against their governments. As well as those on the raids, they include those who have invested by providing arms. Locally-based arm sellers supply weapons, sometimes on credit, to fighters or warriors. They acquire readily available light weapons and ammunitions from neighbouring countries, particularly Ethiopia, Somalia and Sudan that have experienced political unrest and internal wars. Other sources of weapons include those supplied to the locally-based Kenya Police Reservists, which find their way into the hands of these fighters or warriors. Some conflicts are more overtly political: prime movers may be politicians and warlords rather than businessmen or elders and prophets of specific ethnic groups. With regard to timing, traditional conflicts normally occurred after drought, during periods of serious impoverishment, following age-set initiations, and at the beginning of the rainy season. While commercial raids occur when livestock prices are high in large markets, in the case of political conflicts, timing is dependent on strategic considerations – for example cross-border raids and insurgencies (Kralti and Swift, 1999:16-18; Pkalya et al, 2003:55-56).
Conflict patterns in northern Kenya have become gradually more vicious and less controllable and are characterised by complex causality: stock theft; proliferation of illicit arms; inadequate policing and state security arrangements; the diminishing role of traditional governance systems; competition over control and access to natural resources such as pasture and water; land issues; political incitements; ethnocentrism; increasing levels of poverty; and idleness amongst the youth. Conflicts are aggravated by the failure on the state's part to understand traditional grazing boundaries. Further, the increasing failure of the state to provide adequate physical security has led to most pastoralist communities establishing armed militias to protect themselves. These communities lack confidence in the state's security apparatus given the corruption and indifference amongst security officers in the region. Administrators also lack proper understanding of the pastoralist communities (Ndegwa, 2001:2; Pkalya et al., 2003:10; Kerrow, 2005:9).
Violent conflicts have had very negative economic and social consequences and adversely effect human security: loss of human life and property; displacements of large sections of the affected communities; disruption of socio-economic activities and livelihoods; heightened tension and hatred among communities; environmental degradation and threat to water catchment areas; increased economic hardships; increased levels of starvation and severe malnutrition among displaced groups; and increased dependency syndrome on food relief. A total of 164,457 people have been displaced by conflicts in pastoralist areas in these regions; seventy per cent of them constitute women, and children below 14 years of age. Many women have also been widowed and children orphaned by the conflicts, increasing their susceptibility to poverty and human rights abuses. School enrolment rates are far below the national average while mortality rates are high. Further, water and sanitary services are inaccessible to the majority of pastoralists (CCR, 2002:23; Pkalya et al. 2003:10-11).
The pattern of violent conflicts in Marsabit district
To capture the nature, causes and impact of violent conflict in the district, it is important to understand its physical environment, and socio-economic activities. Much of the district is on a plain lying between 300 and 900 meters above sea level which slopes gently towards the southeast. The west and north plains are bordered by hills and mountain range; hill masses also protrude from the northern plains. The area around Lake Turkana is rifted and forms part of the Great Rift Valley system. The district is dominated by seasonal rivers.
Marsabit is located in the driest region of the country. It receives 200 mm to 1000 mm of rain annually for the lowest and highest elevations, respectively. The district is ecologically divided into four zones. Human settlement is concentrated around the humid and sub-humid mountain areas. Here, the main source of livelihood is agro-pastoral farming. Approximately 75 per cent of the district is classified as rangeland and the main form of land use is extensive grazing. Eighty per cent of the district's residents are pastoralists who derive their livelihood from livestock as well as livestock-based industries. About 10 per cent of the total population practice subsistence agriculture while 7 per cent is involved in commerce. The remaining are salaried employees (Pkalya et al. 2003:20-21; Haro et al. 2004:286).
The district's proximity to Ethiopia has increased contacts with Oromo groups, leading to conflicts over grazing resources. The infiltration of the Oromo Liberation Front (OLF) – a guerrilla outfit that has been waging ‘low-intensity’ war against the Ethiopian government into the regions where Marsabit borders with Ethiopia – has introduced another dimension of conflict commonly commonly referred to as insurgencies (Ibid. p. 54).
Pastoralist communities are in constant conflict especially during the dry spell when livestock tend to concentrate in the limited sections with pasture and water. But people no longer raid just to replenish their stock but to kill and maim while enriching themselves. Commercialisation of livestock raids has taken its toll in the district. The players in the conflicts in Marsabit include Rendille, Gabra and Turkana, who all live in the district, and Borana (from both Kenya and Ethiopia) communities as well as the Samburu who attack into the southern part of the district.
The economic and social impact of conflict in the district are many: reduction of farming activities, concentration of livestock in small areas where there is water and pasture, stock theft, looting and destruction of property, reduced business activities, environmental degradation, inaccessible health services, increased number of school drop outs, displacement of people, stalled development projects, and highway banditry. The impact has resulted in widespread poverty and adversely affected human security. Hunger is often severe and could last for as long as eight months, particularly between January and April and from September to December; food is available only between May and August. For the rest of the months, affected communities receive relief food from the state, civil society organisations and other aid agencies. The same is replicated during times of conflict (Ibid. pp.56-57).
Traditional conflict resolution measures have, however, been effected to contain the problems of violence and deaths caused by stock theft. The Modogashe Declaration, for example, was passed in 2001 by pastoral communities in Eastern and North Eastern provinces. Its objective is to address compensation for deaths during stock theft. It provides a framework for return of stolen livestock and even compensation for victims killed by bandits. The declaration states that for each man killed, 100 head of cattle is to be paid as compensation by the community where the bandits or invaders come from. In the case that a woman is killed, 50 cattle are to be paid while in the case of injuries 15 cattle are paid. In northern Kenya districts such as Isiolo and Wajir, the declaration is still in force and few cases of stock theft occur. It also worked in Marsabit until 2004 when the District Commissioner (DC) declared it had no legal backing. The Declaration is administered and enforced by respective district peace committees (ITDG, 2003: 12; DN, 2005a:3).
The Turbi massacre: State failure & human security
Reacting to the massacre, several civil society organisations, political and religious leaders blamed the state for failure to provide security. They emphasised that violent conflict in Marsabit and neighbouring districts was primarily a function of the state's failure to provide adequate development and security in the region since the country's independence in 1963. All post-independent governments, they said, had neglected the region, subjecting area residents to insecurity and human rights abuses. The Chairman and of the Law Society of Kenya (LSK) and the Secretary-General of National Council of Christian Churches in Kenya (NCCK) emphasised that the state had not done enough to ensure security in the region. The Chairman of the Kenya Human Rights Commission (KHRC) argued the fact,
That hundreds of criminals can terrorise a town for hours without the intervention of the country's security forces is a clear indication that the Government has little or no authority in the north-eastern region (Daily Nation, 14 July 2005:3).
It is quite obvious that the Government has to show more presence in the peripheral areas of this country … The politicians, who thrive on ethnic violence, must cease to do so, and the livestock industry of the people from these areas boosted (Daily Nation, 16 July 2005:4).
The violent conflict that occurred in Turbi is an indicator of the state's failure to provide security and development in the region and to establish effective early warning systems. This lack of institutional capacity or lack of political will to provide development and security, whether personal, community, human or political, in regions that make up most of the forgotten ‘badlands’ in the country has culminated in grave consequences that have led to the loss of innocent lives. The Turbi Massacre was planned and executed with military precision. Most of the casualties were women, children and elderly men (Njeru, 2005:11). The state did admit that it was partly to blame for the violence that has rocked Marsabit and Moyale districts. In the case of Turbi, the Eastern Provincial Commissioner (PC) admitted: ‘It is true … that we were slow in responding to the request to send in the GSU personnel and it is very painful’ (Daily Nation, 16 July 2005:4). The state's weakness in handling such crises was also pointed out by one commentator who stated that:
The Government, read police and military, can only respond after an attack … And quite often, the raiders outwit them … More accurately, the security forces allow themselves to be outwitted by the raiders … And once the heat dies off, leaders and Government officials will go back to their usual routine, until the next incident occurs. And that is where they are failing us (Murunga, 2005:8).
In fact, it took over 12 hours for the report of the Turbi Massacre to reach Marsabit Town. Security forces were dispatched to the affected area on the same day. By dusk they had killed ten of the bandits and recovered 5,000 sheep, 60 head of cattle and 10 camels. As the death toll rose, President Mwai Kibaki announced a security operation to track down the bandits and ensure there was peace in the region. Two thousand soldiers, troops from the paramilitary General Service Unit (GSU), and police officers were flown to Marsabit and the neighbouring districts to track down the heavily armed bandits. Military and police helicopters were also deployed to enhance aerial surveillance and assist the ground forces (Daily Nation, 14 July 2005: 2-4; Daily Nation, 15 July 2005:2; Kerrow, 2005:9). Local residents were however sceptical, emphasising that the security personnel were unlikely to arrest the bandits since most of them had already crossed over into Ethiopia. Residents and leaders accused the police and the army of taking hours to respond to the attack and of initially treating it too casually. The initial small group of security officers sent to Turbi from Marsabit Town following the massacre, residents pointed out, were no match for the large number of well-armed raiders (Daily Nation, 15 July 2005:2-3).
The state's failure in the case of violent conflicts in the ‘badlands’ is partly due to the fact that it is accorded a low degree of popular legitimacy by area residents. In the case of Turbi, the state was irresponsive and insensitive in heeding the advice of area leaders and residents. The utter disregard of community leaders’ views by the state's security apparatus was responsible for the failure to prevent violence. Area leaders and residents warned the state that tension between the warring communities was building in the district since January 2005 due to sporadic killings that were arising more than retaliatory attacks. These sporadic conflicts should have sent early warning signals to the government to put in place conflict prevention measures and provide adequate security in the area. Leaders and residents emphasised that this would have averted the violence at Turbi (DN, 2005c:3; Kerrow, 2005:9). Leaders from the area had been anticipating an attack for some time and that is why most of the young and middle aged men were not at the scene when the bandits struck and killed innocent children and women. Sources say the men were in the field where they had anticipated the battle would be fought. One MP said he was informed of the imminent attack and went ahead to call a press conference on 22 June 2005 where he accused some Marsabit politicians of using the OLF to kill his people and destroy their property. He pleaded with the local security committee to send in the GSU to deal with the problem of mercenaries from the OLF but the provincial administration disregarded his pleas. Nineteen days after the press conference, the bandits attacked Turbi, demonstrating a weakness in both the state's intelligence and crime prevention systems. The Parliamentary Committee on Security announced that it would visit Marsabit on a fact-finding mission over the wave of violence in the area. Its Chairman said that it would investigate weaknesses in the intelligence and crime prevention systems before reporting back to Parliament (DN, 2005e:3; Njeru, 2005:11). Many area leaders and residents, therefore, not only question the legitimacy of the state in providing personal and community security, but also and more importantly, its role in conflict prevention.
Though the state acknowledged that it was partly responsible for the violence that occurred, it defended itself against accusations that the Turbi massacre was an outcome of its gross negligence in the region. The state shifted the blame on political ethnocentrism, emphasising that it had held a series of meetings with local leaders and their communities which were aimed at easing ethnic tensions, prior to the massacre. The Government of Kenya spokesperson said that a series of meetings to defuse tensions among clans in the region had been taking place since January 2005 where the clans were urged to stop engaging in violence and assist the state instead by restoring order in the region. The state emphasised that the hostilities were fuelled by political leaders in the area. The Assistant Minister in charge of Internal Security said the state was aware of the hostility particularly among the politicians. He noted: ‘We had done our part to bring them together but it does not appear to help’ (DN, 2005b:4). In a ministerial statement on the Turbi Massacre and related killings, the Minister in charge of Internal Security pointed out that the atrocious acts were a culmination of ethnic and political hostilities and tensions that had been building up in the district in the previous months. He emphasised that the state's security apparatus had made a series of attempts to reconcile and encourage harmonious co-existence among the hostile communities in the district, and that it was concerned that some leaders had been making irresponsible statements bordering on incitement to violence. The irresponsible statements, he said, had served to heighten tensions in the district. The state appealed to MPs and other local leaders to desist from making inflammatory, reckless, and divisive utterances. The Eastern PC also said politicians from both Borana and Gabra communities were to blame since they were not committed to peace initiatives (DN, 2005b:5; DN, 2005d:4).
Ethnopolitics, particularly political ethnocentrism, therefore, also contributes to state failure in the region. The state's inability to create or enhance societal cohesion, and consensus among leaders and residents of various ethnic communities, has also led to its failure in fulfilling its basic functions. Ethnic and political intolerance, and hostilities among political leaders from the region, indicate that political ethnocentrism was also a major contributing factor in the Turbi Massacre. MPs from the region traded accusations over the perennial violence between the Borana and Gabra communities, claiming that the killings were political. The MP for the area was accused by fellow MPs from neighbouring communities of failing to condemn previous attacks by his own Gabra people against theirs, particularly the Borana. Responding to the accusations, the MP also blamed fellow MPs from the region claiming that they were actively involved in the violent conflict (DN, 2005b:2-3). He pointed out: ‘The Government is aware of activities that have been taking place in the district and it knows who has been working with foreign militias’ (Ibid. p. 4). The MP further said it was important that the politicians be investigated so that the truth is established. The Parliamentary Committee on Security also acknowledged the hostile political differences among the area MPs when it held meetings with the Marsabit District Security Committee and all MPs from the district. This was prior to the attack. Civil society leaders and organisations urged the state to take stern action against local leaders who were fuelling ethnic and political animosity (DN, 2005b:3; DN, 2005e:3). Having placed blame on political ethnocentrism, the state began investigating the political dimensions of the conflict and, in the process, the three MPs from the district were summoned by police to record statements over the massacre. The MPs voluntarily recorded statements but accused the state of political persecution and the police of planning to use them as scapegoats over the killings (DN, 2005e:1; DN, 2005g:3). While it is acknowledged that pre-existing and on-going political feuds contributed towards the Turbi Massacre and related killings, the state should avoid faulting political incitement for its own shortcomings if it is at all serious about providing security to residents in the region (Goldsmith, 2005:9). The legitimacy of the state also becomes questioned when it is unable to create societal cohesion and consensus among warring pastoral communities. It is also the case when it is unable to put in place effective conflict prevention mechanisms.
The state's vulnerability to external political forces is also a contributing factor to its failure in the region. In the case of Turbi and surrounding areas, an external force that was a contributing factor to the violent conflicts and subsequent loss of lives was the OLF. Area residents, grassroots organisations, local leaders and social commentators attribute the cause of the Turbi Massacre to external forces, particularly the OLF, and the general elections held in Ethiopia in June 2005. According to the residents, all the raiders came from neighbouring Moyale District and included suspected elements of the OLF. The fundamental objective of the Oromo liberation struggle, which is led by OLF, is to exercise the Oromo's people's inalienable right to national self-determination. The Oromo (Ethiopian), Borana and Gabra are part of the wider Oromia community that straddles across the Kenya-Ethiopia border. The claims of the residents were supported by the local leaders who pointed out that the bandits came from a neighbouring country. In its ministerial statement, the state also acknowledged that some of the bandits are believed to have crossed over to Ethiopia (DN, 2005b:2-5; Goldsmith, 2005:9; Munene, 2005:5; OLFa, 2005:11).
The activities of OLF, backed by the Oromo-speaking communities – the largest ‘nationality’ in federal Ethiopia are, arguably, a source of conflict in the region. OLF military activities have negatively impacted on the security situation in northern Kenya. OLF sees itself as pursuing the right of self-determination and has proved to be a thorn in the flesh for the Ethiopian government (Munene, 2005:5). However, as one conflict and peace-building consultant with the Centre for Minority Rights pointed out:
In Ethiopia, the Oromia community feel marginalised and have always wanted a Borana state. The trouble is that the Gabra refused to join in the initiative while the Borana agreed and this has created enmity that spills across the border and involves their brothers on the Kenyan side (Ibid. p.5).
OLF, however, denied any role in the massacre. It emphasised that exacerbating ethnic conflict between the Borana and Gabra communities of Oromo, living on both sides of the Kenya-Ethiopia border, was a technique of counter-insurgency used by the EPRDF government to discredit OLF. It argued that EPRDF attributed its losses in the June 2005 general election, especially in the South, to the activity of OLF, thus devised this campaign to cross the border and instigate this particular conflict between the two groups in Kenya. OLF stressed it has neither strategic nor temporary interest in provoking Kenyans and their government, nor any motive to commit such atrocities on their kin and kith across the border, even though some news media in Kenya had, allegedly, reported that it massacred innocent Kenyans (OLF, 2005b:1-2).
In an apparent contradiction of the claims that the bandits came from Ethiopia, the Government spokesperson emphasised that the attacks were the product of historical banditry and revenge missions among communities in the region, and ruled out an external hand in the killings at Turbi. He noted that, ‘The Government would like to point out that this was a local issue. It is time we dealt with our problems instead of blaming others’. The spokesperson, nonetheless, noted that diplomatic contacts with Ethiopia had been put in place in case the bandits had slipped into the neighbouring country (Daily Nation, 14 July 2005:2).
The problem of boundary dispute and scarcity of resources, is arguably, yet another source that led to the violent conflict at Turbi. This is partly due to the state's failure to provide sound environmental and natural resource management policies. This is, however, debatable given the nature of the killings. The question of boundary dispute is intricately intertwined with that of competition for scarce resources, particularly pasture and water, as the warring communities dispute the boundaries that identify the location of water points. At two separate meetings held at Sololo on 16 July 2005 by the Eastern provincial security committee, and chaired by the PC, the issue of the boundary between Moyale and Marsabit cropped up. The meetings were called to find answers to the genesis of the clashes. A councillor from Moyale requested the provincial administration to clarify on which side of the border Turbi village lay. The civic leader claimed that the water point at Turbi, the primary school and other social amenities had been developed by his constituents but they were pushed out by the previous Kenya African National Union (KANU) government through the influence of a local politician from a neighbouring community (Njeru, 2005:11). The civic leader emphasised that:
We are still bitter for having been thrown out of our area. It is common knowledge that it is us who developed the area and drilled the water hole … We want you, Bwana (Sir) PC, to tell us whether Turbi is in Moyale or in Marsabit (Ibid. p.11).
The state's immediate response in providing and ensuring human security needs was minimal. It only provided, alongside civil society organisations and international aid agencies, relief food to the displaced. It did not assist in burying the dead nor even to set up an emergency fund to assist the bereaved, displaced, and those whose houses were destroyed by the attackers. Had the state responded immediately and provided adequate security in the area, less people would have died in retaliatory or revenge killings (DN, 2005c:3; DN, 2005g:3). The conflicts negatively impacted on the economic well-being, food security, health, personal, community and political security of the area residents in general, thereby endangering their human security.
Conclusion
The Turbi Massacre and related killings are largely the result of the state's failure to fulfil its basic functions and responsibilities in the ‘badlands’ of the country. The state does not have the capacity or political will to provide meaningful security and development in the region. Ethno-politics and external political forces also contribute to this failure.
The nature of this particular violent pastoral conflict indicates that it was more political than traditional or commercial. A large number of people were deliberately killed – many of whom were women and children. Many of the attackers were from a specific ethnic group and included members of OLF. Given that it was planned and executed with military precision and that the state had been forewarned of the attacks indicates that the timing of the conflict was strategic. The fact that the violent conflict occurred one month after the general elections in Ethiopia is a further indication that its timing was strategic.
The state must enhance its capacity or political will to provide development and security. To reduce its vulnerability to ethno-politics, the state must create and enhance societal cohesion and consensus among the warring communities especially among their leaders. This can be partly achieved by putting in place effective conflict prevention, management and resolution mechanisms particularly at the grassroots level. This will create the necessary conditions for the political leaders to engage in mutual cooperation rather than conflict in matters related to development and security. But community participation in such security arrangements with the state is essential.
The state must also put in place necessary developmental and security arrangements that will reduce its vulnerability to external political forces. Given the expansive and hostile terrain of the ‘badlands’, the state should set up and provide adequate communications and transport infrastructure that will enable it to rapidly respond to cross-border insurgencies or raids. Infrastructural development must be accompanied by an increase in the provision of broader development initiatives.
The state must ensure that it guarantees food security of the regions’ residents through the implementation of appropriate agro-pastoral farming and range management techniques that involve community participation. This will reduce dependency on food aid. The state must also provide environmental security through the implementation of appropriate or sound natural resource management techniques that also involve community participation. This is likely to reduce the problem of competition over scarce resources. By guaranteeing personal, community, food, health and environmental security, the state will be in effect guaranteeing human security. The general populace will, therefore, be in a position to accord it a higher degree of popular legitimacy. This in turn strengthens the state's capacity to fulfil its basic functions of providing development and security in the ‘badlands’ of the country.