Political Violence in Kenya a Study of Causes, Responses, and a Framework for Discussing Preventive Action

Political violence in Kenya A study of causes, responses, and a framework for discussing preventive action

INTRODUCTION is therefore the pursuit of political objectives (and not the end result of gaining or not gaining political power) that Th is paper focuses on political violence in Kenya. Th e is at the core of the notion.4 Defi ning political violence rationale of the paper is to understand how such violence is typically contingent on a typology that separates the can best be prevented. To do this, we must fi rst under- politically motivated from economically and socially mo- stand the history of political violence in Kenya and what tivated forms of violence. In this sense, political violence factors in society contribute to a tendency of utilising takes place in the collective sphere where acts of violence violence as a means of obtaining political infl uence. are typically committed by a multitude of individuals Th e present study is informed by a number of from one group against individuals from another group, interviews with representatives of Kenyan civil society primarily because the targeted individual happens to organisations which are involved in human security.1 belong to this group. Political violence typically takes Moreover, it relies on an appraisal of studies and reports the form of murder, assaults, sexual abuse such as rape, available on political violence in Kenya. In its discus- forced pregnancy or sterilisation. Economic violence, sions, the paper draws on transitional justice discourses. on the other hand, is characterised by an individual (or Th e paper is structured as follows: Firstly, a defi nition a multitude of individuals) illegally pursuing fi nancial of political violence is provided for. Secondly, the paper enrichment by means of violence (or threats of violence), analyses how political violence has historically played and typically manifests as street crimes such as robbery, out in Kenya. Th irdly, the paper identifi es a variety drug related crimes or kidnapping. Social violence is of causes of political violence in Kenya. Fourthly, the said to pursue the empowerment of one individual over paper turns to an appraisal of which responses can be another, for example through domestic violence.5 deployed to prevent the recurrence (or persistence) of Many observers, however, recognise that political political violence. It does so by discussing initiatives that violence should not analytically be disintegrated from have already been implemented or which may be due to economic and social violence because all forms of implementation. Th e paper thus presents an account of violence interrelate.6 In Kenya, many of those consulted political violence in Kenya and the responses undertak- by the author understood political violence as violence en, while at the same time engaging more theoretically that takes place in relation to ‘political competition’, but with the question of how to prevent political violence. as a notion with strong economic and social underpin- nings that may be the determinant for the prevalence of CONCEPTUALISING political violence. POLITICAL VIOLENCE It is worth keeping in mind that the term political violence is used for a broad variety of situations, Political violence is a so-called ‘essentially contested ranging from terrorist attacks, armed revolution, violent concept’, and the premise that a universal understanding demonstrations or attacks by citizens aimed at less than ought to be utilised is not endorsed by all.2 Th is section the overthrow of their government to humanitarian attempts to clarify how the notion can be utilised to the intervention and intra-state wars.7 Political violence, Kenyan context. therefore, is not confi ned to non-state actors’ use of Where all forms of violence are said to relate to violent means to further a political agenda but can also power, political violence has been defi ned as ‘the com- relate to the state’s exercise of force, both against its own mission of violent acts motivated by a desire, conscious citizens and against other states and their citizens. It is or unconscious, to obtain or maintain political power’.3 It important to keep in mind that mass-scale violence is

Thomas Obel Hansen • ISS Paper 205 • November 2009 sometimes portrayed as essentially apolitical while in on building alliances with certain ethnic groups and reality being predominantly political. For example, some escalating tensions between these and other ethnic observers have characterised the Rwandan Genocide as a groups. Th erefore, besides its immediate connection with ‘crime of hate’, implying that it was brought about by ir- political violence, colonialism through ‘divide and rule’ rational ethnic hatred,8 when in fact the Genocide served policies has brought about or escalated inter-community (or was thought to serve) primarily a political agenda.9 At confl ict, the eff ects of which may still have importance the same time the Rwandan Genocide highlights how the for the prevalence of political violence. Likewise, the incentive to carry out violence that is essentially political construction of a colonial state is said to have sustained can be enhanced by ambitions to settle private scores eff ects on inequality, land ownership, and regional diff er- or obtain material benefi t, thereby exemplifying the ences. As discussed later on, these factors are relevant to diffi culties in operating with a clear distinction between take into account when examining political violence in political, economic and social violence.10 present day Kenya.12 Th e understanding that political violence relates to Also aft er the coming of independence in 1963, acts of violence that are carried out primarily as a means Kenya’s political history was marked by violent uprising of achieving political infl uence or power and usually and repression. Following the Kenya African National entails a ‘group-component’ is foundational for the Union (KANU) victory in the 1963 elections, Kenya present paper. It is, however, also accepted that political became a de facto one-party state, with its leader and violence is a relative concept that depends on the context President of Kenya, Jomo Kenyatta, for example banning to which it is applied. attempts of creating an opposition party associated with the Luo ethnic group. For some, this lead to the POLITICAL VIOLENCE IN KENYA: perception that Kenyatta, an ethnic Kikuyu himself, was 13 A BRIEF OUTLINE promoting Kikuyu interests over national interests. According to many observers, when Daniel Moi took Political violence has played out in diff erent manners offi ce in 1978 following the death of Kenyatta, he pursued throughout Kenya’s history. When in 1888 the British policies that benefi tted (parts of) his own ethnic group, East Africa Company (BEAC) obtained concessionary the Kalenjin, while excluding individuals from other rights to the Kenyan coast from the Sultan of Zanzibar, ethnic groups from gaining public offi ce or access to state Waiyaki Wa Henya, a Kikuyu chief was abducted and resources. To facilitate this, repression of the political killed by the British aft er having burned down the fort opposition was common practice. Repression included of a BEAC offi cial. Likewise, Kenyans’ opposition to the excessive use of force, torture, indefi nite detention, and building of the Uganda Railway, prompted the British to other measures. In 1982, following a coup attempt, the use violent means such as assassinations. Moi regime amended the Constitution and subsequently Kenya offi cially became a one-party state. Police and security forces dispersed demonstrations against this Political violence relates to acts move forcefully. Only aft er intense donor-pressure did Moi allow multi-party elections to be held in 1992.14 Th e of violence that are carried out election campaign, the election itself and its immediate aft ermath were characterised by threats, harassments primarily as a means of achieving and the occurrence of violent clashes between supporters for diff erent parties, claiming the lives of around 1,500 political infl uence or power Kenyans and displacing more than 300 000.15 Where Moi had warned that the return to multi-party politics would result in tribal clashes, some observers note that ‘far As a reaction to settlers’ dominance over economic from being the spontaneous result of a return to political resources and political exclusion, in 1921, Kenya’s fi rst pluralism, there is clear evidence that the government African political protest movement, the Young Kikuyu was involved in provoking this ethnic violence for politi- Association (later the Kenya African Union) was born. cal purposes and has taken no adequate steps to prevent Both the colonialists and indigenous Kenyans used it from spiralling out of control’.16 Moi maintained power violence in their pursuit of political objectives, most with the 1992 elections, and despite increased openness clearly evident from the 1952-1960 Mau Mau rebellion in the political system, commentators note that the Moi and its repression.11 As many of the Kenyans consulted regime continued to repress the political opposition.17 by the author pointed out, it is important to keep in Like the 1992 elections, the 1997 elections were mind that British colonialism in Kenya was founded associated with violence. Six months prior to the upon a strategy where eff ective rule of the colony relied elections, KANU party activists allegedly backed

2 Political violence in Kenya • ISS Paper 205 • November 2009 armed gangs who attacked ‘non-native ethnic groups’ Yet, the most recent post-election violence, because in the Coast Province, causing the death of more than of its relationship to the contested election results and 100 and leading to the displacement of more than 100 because of its speed and scale, seemingly came as a 000.18 Also aft er the elections, politically motivated surprise and shock for many Kenyans and the outside violence between ethnic groups took place. According world. Th e violence had erupted even before Kibaki was to Amnesty International, more than 120 Kenyans lost declared winner of the elections, but increased in scale their lives in the Rift Valley when KANU supporters aft er the announcement.24 Estimates of casualties vary, clashed with armed youths belonging to the Kikuyu but most observers cite the Waki Report’s number of ethnic group.19 Amnesty International notes how 1,133.25 Moreover, around half a million Kenyans were political violence predominantly occurred in those areas forced into exile by the violence,26 of which a signifi cant where the Kikuyu-dominated opposition party, the number remains in internal displacement camps.27 Democratic Party, had won over the Kalenjin-dominated It was in the Rift Valley – in particular around KANU.20 Human rights organisations indicate that the Eldoret – that violence fi rst erupted, seemingly taking violence in the Rift Valley, like the pre-election violence the form of ethnic-based clashes between Odinga’s in the Coast Province, was endorsed and supported by supporters, especially from the Kalenjin ethnic group, political leaders, and responses from security forces to and supporters of Kibaki, mostly from the Kikuyu halter the violence were non-existent or too reluctant or ethnic group. According to some observers, certain delayed to have any meaningful eff ect.21 In 2002, aft er local politicians incited the violence.28 Its carrying out Moi had held two terms and therefore was not allowed was oft en brutal and simply directed against individuals to run for president again, Uhuru Kenyatta (Jomo according to their ethnic affi liation.29 Partly in response Kenyatta’s son) was appointed as KANU’s candidate. to the violence in and around Eldoret, Kikuyus formed Dissatisfi ed with Moi’s choice, a number of KANU so-called ‘self-defence forces’. Th ese militias along with a members formed a faction, the Rainbow Coalition with criminal organisation formed in the 1980s, the Mungiki, Raila Odinga as its leader, which later formed a coalition carried out organised and large-scale violence in other with the Liberal Democratic Party (LDP). Th e 13-party parts of the country, including Naivasha, Nakuru, and coalition, the National Alliance Party of Kenya, joined Nairobi slums.30 Th e violence was in most cases followed fronts with the LDP under the name of the National by looting and sometimes perpetrators were supposedly Rainbow Coalition (NARC), headed by Mwai Kibaki rewarded with money for taking part in the violence.31 as its presidential candidate and with Raila Odinga as Th e level of the state agencies’ responsibility for the its prime minister candidate. Although the election occurrence of the election violence remains a disputed campaign (and its aft ermath) was characterised by a subject. Th e Waki Report notes that whereas the police signifi cant decrease in political violence compared to its in some areas of the country remained passive, either two predecessors, political rallies did on some occasions because they were unable or unwilling to stop the vio- lead to violence.22 lence, in other areas, police shootings seemingly resulted in many of the casualties.32 In some cases, such as in the slums of Kisumu, Human Rights Watch interviews with Kenya’s Daily Nation reported how police offi cers suggest that an unoffi cial ‘shoot-to-kill’ policy was applied, apparently resulting in unarmed assassins with political motives had civilians being shot dead if assumed to be rioters.33 Political violence in Kenya, although oft en at its killed 26 high-profi le Kenyans over extremes during election periods, also exists as an ‘everyday phenomenon’. Assassinations of political leaders, the course of only two months prominent businessmen, civil society leaders, and other fi gures that possess signifi cant infl uence on the allocation of resources or political developments in the country are As the Guardian’s Madeleine Bunting notes, observers far from exceptional. In early May 2009, Kenya’s Daily of the violence following the 2007 elections have oft en Nation reported how assassins with political motives had referred to the violence as in contrast to an otherwise killed 26 high-profi le Kenyans over the course of only politically stable and peaceful country. It is, however, two months.34 Most of those Kenyans consulted by the worth keeping in mind that most signifi cant political author view such form of political violence as related to the activities and changes in leadership throughout Kenya’s violence surrounding election processes because it is many history as an independent state have been followed, and of the same factors that allow for their occurrence. to some extent formed, by violence, usually framed along Whether violent crimes committed by criminal ethnic lines.23 groups, such as the Mungiki sect, should be categorised

Thomas Obel Hansen • ISS Paper 205 • November 2009 3 as organised economical or political when they do not election process, systematic documentation of its scope occur in connection with election processes depend and nature remains sparse. on the perspective taken. When the Mungiki sect was As should be clear from the above, both ‘ethnic vio- established in the 1980s, it had religious undertones and lence’ and organised economic crime can be intimately pursued a political agenda of ‘defending the culture and associated with political violence. Political violence traditions of the Kikuyu tribe’.35 At that point, the sect that unfolds along ethnic lines can serve as a ‘tool in is said to have aimed at protecting Kikuyu interests in the toolbox’ for political leaders who want to achieve the context of then president Moi’s affi liation with the infl uence.43 Like other instances of mass violence in the Kalenjin ethnic group. Th e Mungiki sect continues to be region, ethnic violence in Kenya connects to political associated with the Kikuyu ethnic group. However, the manipulation of identity. Likewise, organised economic agenda of the sect and its reasons for utilising violence crime is closely associated with political violence, in are now primarily economical. Th e Mungiki applies particular because agents of political violence are oft en ‘mafi a-like’ methods to control the Kenyan public involved in economic crime. transport sector and bribes local business owners.36 In order to establish a framework for understanding how political violence in Kenya may be countered it is useful fi rst to dwell by the causes of political violence. Like other instances of mass Th e following section aims at identifying a variety of underlying factors that may be determinant for a politi- violence in the region, ethnic cal culture that, at times, sanctions the use of violence to reach political objectives. violence in Kenya connects to POLITICAL VIOLENCE IN KENYA: political manipulation of identity IDENTIFYING THE CAUSES

Identifying causes of political violence is a problematic Yet, Mungiki’s actions remain associated with Kenyan task. Firstly, any attempt to ‘explain’ a phenomenon such politics in several ways. Firstly, Mungiki criminal as political violence can be subjected to the objection activities associate with politics not only in the sense that that attempting to establish causal connections to certain violence is oft en directed towards ‘political opponents’ of economic, social, or political realities is simply the wrong the sect,37 but also in the sense that both local vigilante way of approaching political violence. Viewing political groups and state security agencies have engaged in violence as an outcome of particular societal features a low-scale armed struggle with the sect, leading to easily gives way to the perception that the emergence of accusations of government-ordered extra-judicial kill- political violence is unavoidable.44 Nonetheless, the re- ings.38 Secondly, it was common perception among those versed position is also problematic. Mass violence is not Kenyans consulted by the author that Mungiki continues simply irrational and unexplainable, and it remains a fact to enjoy (moral and fi nancial) support from a number that political violence seems to occur persistently in some of parliamentarians. Th irdly – and closely related to the countries and not – or to a much smaller extent – in above – according to observers of political violence in others.45 In any case, some scepticism is justifi ed when- Kenya, intimidation and violence by the Mungiki sect ever studies pursue the ‘explanation’ or ‘identifi cation of and a number of other violent gangs have been employed causes’ of a phenomenon such as political violence. as political tools, both during election campaigns and In the following, an overview is presented of condi- beyond.39 Perhaps ironically, as one scholar observes, it tions that may be consenting to political violence in may well be that ‘the informal repression or quasi-legit- Kenya. imisation of sectarian violence for political ends by the state, has transformed a “moral ethnic” movement into a Strong executive powers, gaining 40 “political tribal” one’. Finally, the Mungiki sect exer- political offi ce as a ‘struggle for survival’, cises control as a ‘state within the state’ in certain parts and manipulation of ethnicity of Kenya and uses violence to maintain such control.41 In many ways, violent crime employed by the As implied in the above outline of political violence in Mungiki sect thus identifi es as political. While violent Kenya, support and resistance to political leaders of the crimes employed by organised gangs such as the Mungiki country have oft en followed ethnic lines. When forming sect and vigilante group retaliation receive intensive government, some political leaders have rewarded and cover in Kenyan newspapers,42 to the extent that violence ensured advantages to individuals from supportive committed by such groups does not relate directly to the ethnic groups, while marginalising or excluding

4 Political violence in Kenya • ISS Paper 205 • November 2009 individuals belonging to ethnic groups associated with these take over power, from taking jobs, land and entitle- political opponents. Th e Waki Report, for example, ments. All of this has led to acquisition of presidential notes how ‘Moi rewarded his supporters, particularly power being seen both by politicians and the public as a the Kalenjin, through appointments to political offi ces zero sum game, in which losing is seen as hugely costly and with jobs in the public service and the military’.46 and is not accepted’.54 Th e Waki Report also notes how during the 1980s and Yet, as many of the Kenyans consulted by the author 1990s ‘land grabbing and the allocation of public land pointed out, the perception that ethnic groups, as such, as political patronage were part of the gross corruption benefi t from ‘one of their own’ gaining presidency is of this period’.47 In this way land allocation was oft en highly miscalculated because only a very limited number turned into a reward to ‘politically correct individuals’.48 of individuals actually benefi t from these arrangements. Consequently, political power has been perceived as vital for obtaining access to public goods, and the distinc- Rule of law problems and the tion between individuals benefi tting or marginalised institutionalisation of political violence from such access is viewed in ethnic terms.49 Added to this, political leaders may have had a personal interest Strong executive powers are also said to circumvent in obtaining power because large-scale corruption has transparency and checks and balances, thereby decreas- become institutionalised.50 In a sense, gaining political ing the likelihood that the executive, or political sup- offi ce has been seen as ‘a struggle for survival’: if power is porters of the incumbent regime, are held accountable obtained, the perception is that access to sparse resources for political violence. Th e Waki Report notes that ‘checks is ensured, and if not; marginalisation and exclusion is and balances normally associated with democracies are reckoned to follow. very weak in Kenya and are deliberately so. Individuals in various parts of government whether in the civil service, the judiciary, and even in Parliament, under- The Waki Report notes that stand that, irrespective of the laws, the executive arm of government determines what happens’.55 Th e argument ‘checks and balances normally oft en goes that beyond dubious or extra-legal acquisition of public goods, the executive may have interest in associated with democracies maintaining structures that diminish the rule of law, thus facilitating government action (or inaction) that are very weak in Kenya may include political violence.56 Impunity in Kenya can be observed as playing out in three diff erent, but yet related, ways. Th is crisis of governance, where many leaders work Firstly, responsible agents of large-scale violence that not for the country as such but for themselves and correlate with a political agenda are seldom thoroughly their political supporters, is sometimes said to follow investigated, arrested, or prosecuted. Commenting on from a system where power has been centralised in the political violence surrounding the 1992 election, Africa hands of few since independence.51 Noting that Kenya’s Watch notes how ‘Kalenjin warriors’, who backed then constitution, despite continuous talks of a fundamental president Moi, were allowed to attack villagers from amendment, is still based on its colonial-era form where other ethnic groups with no or only little attempts of the the president is awarded with extensive powers, Human police to intervene. Not only did the police according to Rights Watch suggests that the risks of a ‘winner-takes- Africa Watch refuse to take statements from victims of all calculus’ increases.52 Th e Waki Report, in a similar the violence, in some cases they also stood by passively vein, notes: ‘power has been personalised around the while attacks were ongoing.57 ‘Kalenjin warriors’ who presidency and this has been increased by changes in the were actually prosecuted oft en received lenient sentences Constitution under each President since independence. or were acquitted because the executive interfered in the Laws are routinely passed to increase executive authority, work of the judiciary.58 Inaction from authorities again and those laws seen as being in the way are oft en changed prevailed when political violence erupted aft er the 1997 or even ignored’.53 Th is is said to result in the perception elections. Some have suggests that the police had received that ‘given the power of the president and the politi- orders from the political leadership not to intervene.59 cal class everything fl ows not from laws but from the Despite judicial inquiries into these outbreaks of political president’s power and personal decisions. Th is also has violence that named several persons as involved and led the public to believe a person from their own tribe recommended prosecutions, it has been suggested that must be in power, both to secure for them benefi ts and as neither the Moi-administration nor its successor, the a defensive strategy to keep other ethnic groups, should Kibaki-administration, had interest in prosecutions

Thomas Obel Hansen • ISS Paper 205 • November 2009 5 commencing.60 In a similar vein, violence committed of the commission established to investigate the prior to the 2007 elections did seldom result in perpetra- ethnic/ political violence in the 1990s (the Akiwumi tors being held accountable. Th e EU monitoring com- Commission) to further investigate certain political mission notes: ‘in most cases, abuses did not receive an leaders was halted by lack of commitment in government appropriate response from the police and the judiciary circles and by a legal sector structured in such a way and there was therefore impunity towards perpetrators’.61 that prosecutions of high-profi le political fi gures were When the most recent election violence erupted in unlikely to commence.70 December 2007, perpetrators had good reason to assume Th e rule of law has thus tended to be put aside in that also this time politically motivated violence would contexts of political competition.71 Th is is likely to have go unpunished.62 So far, they have not been mistaken: led to the presumption for next generation perpetrators the Waki Report concludes that out of more than 1,000 that committing acts of violence in a context that relates homicides related to the election violence, only 19 to political competition will remain not accounted for.72 were prosecuted.63 Representatives of Kenyan civil society consulted by the author all pointed to impunity and lack of commitment to the rule of law as imperative to take into account if Widespread poverty, unequal attempting to understand the prevalence of political violence in Kenya. distribution of resources, high Continuous political violence in Kenya may also have had a reinforcing eff ect in other ways. Political violence unemployment rates and land in the 1990s resulted in many of the persecuted Kikuyu’s leaving their homes in the Rift Valley and settling in disputes are pointed to as ‘root Nairobi. Some joined the Mungiki sect which – as pointed to above – has been an important agent of causes’ of political violence political violence.73 Such institutionalisation of political violence may have had the side-eff ect that violent behaviour has become normalised. According to some, Secondly, impunity has played out as a matter of state non-political violent crimes, especially in larger urban agencies’ excessive and sometimes extra-legal use of vio- areas, in part results from violent gangs being called lence, with other authorities ignoring or for other reasons upon by political leaders in the struggle for political failing to address responsible agents. For example, when infl uence.74 In this way, political violence is closely related the police responded to the violence following the 2007 to other forms of violence. elections, they did so in a partial manner where extensive use of force was deployed in areas dominated by Odinga- Socioeconomic causes: poverty, unequal 64 supporters. Th e Waki Report assumes that more than distribution of resources and land issues one third of the total casualties during the election violence results from police shootings.65 Th e report also Th e sidelining of the rule of law in contexts of competing concludes that in many instances the use of lethal force for political power, the impunity for both instigators and by the police targeted individuals who were seemingly on-the-ground perpetrators and a political tradition that posing no immediate threat. In Kisumu, for example, has, in many cases, implied that forming a government the Waki Report found that 30 out of 50 casualties of is deemed essential for obtaining access to resources police shootings had been shot from behind.66 Th e Waki cannot stand alone as explanations of political violence. Report also concludes that some police offi cers were Socioeconomic factors such as widespread poverty, involved in criminal acts such as sexual violence and unequal distribution of resources, high unemployment looting.67 Despite allegations that Kenyan police offi cers rates and land disputes have oft en been pointed to as have violated the criminal code, according to the Waki ‘root causes’ of political violence in Kenya.75 Report, the Kenyan police force has not initiated any Th ere seems to be a sense among some observers that comprehensive internal investigations into the behaviour agents of political violence in Kenya are fundamentally of police offi cers during the election violence,68 and, dissatisfi ed with social and economic conditions of apparently no police offi cers have yet been prosecuted for life. Kenya has an estimated two million unemployed extra-legal use of force related to the election violence.69 youth. Joining a gang may for some be seen as a ‘way of Th irdly, impunity has prevailed historically in Kenya life’ that can increase possibilities.76 When these gangs in the sense that political fi gures that have called for or engage in political violence, members are sometimes sponsored violence have seldom faced criminal account- promised payment for their activities, and also less ability. Th e Waki Report notes how recommendations organised perpetrators can benefi t from looting. Some

6 Political violence in Kenya • ISS Paper 205 • November 2009 commentators on the recent election violence argue that lasting problems of land ownership and distribution have poverty and opportunism had a signifi cant impact on created tensions between communities. In the Rift Valley, villagers’ and shantytown dwellers’ willingness to engage some Kikuyus have allegedly acquired land through in political violence.77 For many of the perpetrators, connections to the political elite.83 According to the political agendas may thus be less relevant than prospects Waki Report, Kenyans who belong to the Kalenjin ethnic of enrichment. group tend to view the most recent election violence as a Participation in political violence is said to be result of land injustices.84 furthered by a widespread sense of dissatisfaction in the Political violence in Kenya has tended to concentrate distribution of wealth. Distribution of wealth is extreme- in the Rift Valley, oft en with the Kikuyu ethnic group on ly unequal in Kenya. According to a 2004 report, Kenya the one side and the Kalenjin ethnic group on the other.85 is the 10th most unequal country in the world, and the Disputes over land ownership are historically related to richest 10 percent of the population controls 42 percent colonial and post-independent regimes’ unfair taking of the country’s wealth, while the poorest 10 percent own and allocation of land (and the colonialists’ forceful less than 1 percent. Access to resources varies highly resettlement of individuals from certain ethnic groups). from region to region.78 With these lenses, political vio- Such disputes remain an unsolved problem that political lence is a way of demonstrating dissatisfaction with the leaders can escalate by making reference to ethnicity. cruel conditions of life that stand in contrast to the elite’s By doing so, the problem acquires potential for inter- comfortable way of life which is perceived by the poor to community confl ict, as has indeed been a characteristic result from political connections and corruption.79 of political violence throughout Kenya’s history.86 In drawing on transitional justice discourses, the paper now turns to an appraisal of how political violence ‘land is a national cake, and in Kenya can be countered. Th e analysis should be seen as a contribution to ongoing discussions on what everybody should have actions are required to prevent the recurrence of political violence in Kenya. Th e analysis takes into account key at least a small piece’ developments in Kenya following the outburst of violence following the 2007 elections.

In particular, questions related to land distribution and RESPONDING TO POLITICAL VIOLENCE: ownership are central in explaining political violence in ACTION TAKEN AND A FRAMEWORK Kenya. Despite several attempts of reforms, land griev- FOR ENGAGEMENT ances dating back to the colonial era continue to constitute a major obstacle to peaceful cohabitation in Kenya. Transitional justice as a framework Th e problem can be summarised as follows: when Kenya for analysing preventive measures became independent in 1963, the most fertile areas that to political violence in Kenya had been occupied by white settlers were handed over to the new government instead of the people who had lived Th e idea that to avoid the recurrence of political violence there before. Besides selling pieces of this land on market and other forms of mass atrocity, societies must confront terms, Kenyan governments have continuously allocated these legacies through legal and quasi-legal measures these areas to shift ing supporters for patronage purposes. has only in recent decades obtained a central place in Moreover, because colonial laws were never fundamen- discourses on confl ict prevention, human rights, and tally changed, there are no provisions for collective land democratisation. Prior to the Latin American transitions rights, thus complicating ownership for communities in the 1980s, human rights activists tended to rely on that have traditionally been pastoralists.80 a methodology that emphasised the exposure of politi- At least two dimensions are relevant when consider- cal violence. In the early 1990s, activists and scholars ing land distribution and ownership as a cause of alike increasingly began to advocate for the need to political violence. First, many families can barely make a retrospectively confront perpetrators of state-sponsored living from the small piece of land they own. As pointed violence.87 It is along this activism that the so-called out by one of the Kenyans consulted by the author, ‘land ‘transitional justice’ scholarship has been formed. Th e is a national cake, and everybody should have at least scholarship, at this point dominated by international a small piece. But where some have plenty, most have lawyers, engaged with the question: ‘how should nascent little’.81 Shortage and unequal distribution of land can democracies address the human rights violations that lead to dissatisfaction with governance and violence may plagued their societies’ recent past?’88 A dichotomy be a means of expressing frustrations.82 Second, long- between punishment and amnesty provided the initial

Thomas Obel Hansen • ISS Paper 205 • November 2009 7 foundation for answering the question. Parts of the Although it is problematic to argue that a profound scholarship, however, soon started to advocate that political transformation has taken place in Kenya, ‘dealing with the past’ could entail other responses than transitional justice discourses off er useful tools for putting to trial perpetrators of state-sponsored violence, appreciating how Kenya can tackle political violence. or doing nothing. Th e paper now turns to an outline of what actions With the proliferation of truth commissions in the have been undertaken; is underway; or may prospectively 1990s, many academics started to suggest that trial and be embarked upon in Kenya to deal with legacies of punishment are insuffi cient in addressing the root causes political violence and prevent its recurrence. of confl ict. For these observers, truth commissions are seen as a response enabling enhanced roles of victims Trial and punishment and possibly paving the way to healing, reconciliation and peace, in a way that criminal justice cannot Most of the debates on how to address political violence facilitate.89 Others, however, suggest that compared to in Kenya have focused on prosecuting and punishing criminal trials, truth commissions constitute a ‘second- responsible agents for the violence surrounding the 2007 best option’.90 elections. Executive Director of Human Rights Watch, Besides criminal trials and truth commissions, repa- Kenneth Roth notes that for Kenya to overcome the lega- rations to victims of past violations; vetting processes cies of political violence, it has ‘only two choices: justice intended to rid state institutions from known human or impunity’.95 Roth argues that pursuing criminal rights perpetrators; and – but more marginally – legal justice for the most recent election violence is crucial for and institutional reform, are oft en considered within an ending a culture of impunity.96 analytical framework of transitional justice.91 Likewise, the parties to the Kenyan National Dialogue and Reconciliation (the parties) and a variety of Kenyan observers have noted the importance of using criminal For Kenya to overcome the justice to prevent political violence from recurring. Th e parties recognised that their fi nal goal is the achieve- legacies of political violence, ment of ‘sustainable peace, stability and justice in Kenya through the rule of law and respect for human rights’.97 it has only two choices: Th ey agreed that reaching this goal required the parties to conduct further discussions on how to ensure ‘the justice or impunity impartial, eff ective and expeditious investigation of gross and systematic violations of human rights and that those found guilty are brought to justice’.98 Th e need to Th e term ‘transitional justice’ was initially reserved prosecute perpetrators of the 2007 election violence was for justice that deals with gross human rights viola- further recognised in a public statement of the parties tions in times of fundamental (liberal) regime change. on February 14 2008: to solve the political crisis sur- Contemporary transitional justice discourses have rounding the election violence, the parties agreed that expanded to cover questions of how societies should reconciliation and healing was imperative, and reaching deal with past civil war and other forms of large-scale this end was said to require the ‘identifi cation and intra-state violence, also in cases where a fundamental prosecution of perpetrators of violence’.99 Likewise, the political transition is absent. Th e assumption is that Waki Commission set up by the parties highlighted the many of the challenges of doing justice which face role of impunity as a cause of the election violence and these societies are not that diff erent from cases of recommended the establishment of ‘special tribunals’ fundamental and liberalising political transition.92 From with specifi c jurisdiction over the election violence and that also follows that the fi eld now embraces dealing a judicial staff made up of both Kenyans and foreign- with large-scale violence committed by state actors as ers. Th e Waki Report requested the parties to reach an well as non-state actors. Th e use of transitional justice agreement on the establishment of such tribunals and discourses in connection with attempts to deal with put forward a bill in Parliament. Th e request was made atrocities in northern Uganda is illustrative of this under threat that failure to comply within a timeframe trend.93 Th e importance of dealing with past violations of 60 days aft er the Waki Report was made public, would is said to fl ow from a variety of normative claims, such result in a list of names with high-profi le Kenyans, which as reconciliation, victims’ healing, and the consolidation the Waki Commission suspected to be responsible for of the rule of law. Most of these objectives are related the violence, would be handed over to the prosecutor (or ought to be related) to preventing the recurrence of of the International Criminal Court (ICC).100 Yet, a bill large-scale rights-violations.94 on the establishment of special tribunals put forward in

8 Political violence in Kenya • ISS Paper 205 • November 2009 Parliament on February 12 2009 was voted down. Th e Central African Republic, at the time of writing, only one bill was criticised by some observers for failing to ensure case has reached its trial hearings (prosecutor vs. Th omas the proposed tribunals’ independence from the execu- Lubanga Dyilo). In other cases, however, arrest warrants tive and for being draft ed with insuffi cient input from have been issued and some cases are in their pre-trial Kenyan civil society.101 stage. Investigations against Lord’s Resistance Army in At the time of writing, debates continued on how, Uganda have resulted in the ICC issuing altogether fi ve where and whether eff orts to prosecute those responsible arrest warrants (but proceedings against one have been for the most recent election violence should commence, terminated due to his decease). Besides the case against but the Kenyan government appeared to be under Dyilo, two cases related to crimes committed in the DRC pressure from various sides to domestically deal with the have reached their pre-trial stage with the accused in question. Kofi Annan, the chief mediator of the Kenyan custody. One case against a citizen of the Central African National Dialogue and Reconciliation has handed over Republic is at its pre-trial stage. Finally, the UN Security the list of alleged high-profi le perpetrators to ICC pros- Council referral of crimes in Darfur to the ICC has led ecutor, Luis Moreno-Ocampo. Ocampo is attempting to the court to issue arrest warrants against four Sudanese push the Kenyans to take actions, but there is profound citizens, including President Bashir. One of the Sudanese disagreement among Kenyan decision-makers and it is accused has appeared voluntarily before the pre-trial far from certain that steps to prosecute domestically will chamber and the rest remain at large.105 As follows from be taken.102 this outline, it is extremely improbable that ICC investigations into the recent election violence in Kenya would lead to arrest warrants being issued against more than a Debates continue on how, couple of instigators. Should these Kenyans happen to be incumbent ministers, members of parliament, or high- where and whether eff orts to profi le civil servants, their actual handing over to Th e Hague to stand trial are far from guaranteed. Moreover, prosecute those responsible any possible conviction of Kenyans responsible for the election violence in Th e Hague is likely to occur only for the most recent election aft er several years. Besides these pragmatic concerns, international tribunals are oft en criticised for externalising violence should commence justice because those aff ected by mass violence, including victims, have little infl uence on (and perhaps understanding of) the proceedings; because trials take place far It is useful at this point to dwell by the implications of away from the communities aff ected, thus making them prosecuting perpetrators of the post-election violence inaccessible to the general public; and because those internationally or nationally. notions of justice that international tribunals enforce are Th e establishment of the ICC has oft en been praised not necessarily corresponding with notions of justice in by legal scholars. Cees Flinterman, for example, noted the communities aff ected by violence.106 on the existence of the ICC: ‘it is hoped [this] will end Finally, and perhaps most important for the question forever the culture of impunity, thereby deterring the of local vs. international prosecutions for the Kenyan commission of gross human rights violations in the election violence, one must realise that the key argument future’.103 Should criminal prosecutions related to the for pursuing criminal accountability for the violence Kenyan elections violence commence before the Hague surrounding the 2007 elections usually rests on the tribunal, the advantage, compared to domestic prosecu- assumption that it requires punishment of perpetra- tions, is usually perceived as deriving from the inde- tors to ‘eradicate impunity’.107 In the current debate pendence of the ICC from national authorities, thereby on political violence in Kenya, eradicating impunity is increasing the likelihood that high-profi le Kenyans will generally supposed to be a necessary step to prevent the be convicted if evidence confi rms their guilt.104 future occurrence of political violence, because only so On the other hand, pursuing accountability before will prospective perpetrators be deterred from engaging international tribunals is oft en said to have signifi cant in political violence and only so can the rule of law be fl aws. First, it is clear that prosecuting before the ICC is reinforced. Th e Waki Report notes: ‘the eradication likely to include only a very limited number of perpetra- of impunity will […] not only blow off the cover for tors and possible convictions are likely to occur only persons who break the law of the land but also deter aft er several years. It is worth keeping in mind that since others who may contemplate similar deeds in future’.108 its establishment in 2002, despite investigations into Logically, any deterrent eff ect must depend on how atrocities in the DRC, Northern Uganda, Darfur, and the prospective perpetrators perceive the chances of being

Thomas Obel Hansen • ISS Paper 205 • November 2009 9 held accountable.109 If prosecutions in Th e Hague lead to was extra-ordinarily put aside, but instead reveals as a the conviction of one, two, or maybe three perpetrators phenomenon that has played out with impunity continu- of political violence, the chances of being ‘caught and ously since independence, putting to trial (a signifi cant convicted’ for future political violence – perhaps except proportion of) those allegedly responsible for more for those holding the highest offi ces – will be perceived recent acts of political violence may have a symbolic small since the signal sent is that the Kenyan state is eff ect on the rule of law. In this way, criminal account- unable or unwilling to deal with the question domesti- ability for recent acts of political violence may help bring cally and will not prosecute political violence; at worst about profound change in political culture. a few leaders are prosecuted internationally. Moreover, Th e question of whether and how criminal account- putting to trial perpetrators of political violence is ability as a preventive measure ought to be utilised as sometimes said to reinforce the rule of law symbolically a response to the latest outbreak of large-scale election by bringing about ‘collective appreciation of the rule violence in Kenya should be debated, but not from an of law’.110 Th e symbolic eff ect on the rule of law would ideological viewpoint where criminal justice is seen as be much greater if ‘Kenyans prosecute Kenyans’, and the solution to all problems. Instead, the question should in numbers far exceeding what the ICC would enable. be assessed from a viewpoint where possible benefi ts Nevertheless, reinforcing the rule of law by domestically of prosecuting alleged perpetrators are discussed in prosecuting perpetrators of the recent election violence conjunction with potential fl aws of the process’ set-up is dependent on the process’ legitimacy in the eyes of and with an assessment of other actions that can be taken the general public, and to the extent the set-up of special to ‘combat impunity’. tribunals fails to secure independence from the executive Should Kenya fail to prosecute domestically perpetra- (or others’ interference) and fails to bring to account also tors of the post-election violence in the near future, it members of the political elite, the process is likely to be is important to keep in mind that ICC action does not judged as fl awed and holds little potential for strengthen- preclude a domestic process at a later stage. ing the rule of law.111 Legal and institutional reform Eradicating impunity is generally Predominantly focusing on trial and punishment as the response that can end impunity and establish or supposed to be a necessary reinforce the rule of law, may lead to neglect for other structural changes that are profound for altering those step to prevent the future conditions that allow disrespect for the rule of law. It is obviously not a question of either/or: holding account- occurrence of political violence able perpetrators of the recent election violence can and should go hand in hand with a reform process that aims at transforming institutions into more accountable, Despite commentators’ enthusiasm for commencing transparent, effi cient and rights-oriented entities. But, processes of criminal justice, there are, however, good framing the transitional justice debate primarily as an reasons to question the assumption that trying perpetra- issue of ‘backward-looking’ punishment can lead to dis- tors of the most recent election violence will automati- regard for the importance of changing those structures cally lead to the prevention of future political violence that are generally believed to enable political violence. in Kenya. Th e key argument for trial and punishment A possible failure to see criminal justice unfold in the Kenyan context stipulates that only by punishing in Kenya risks bringing about the impression that perpetrators of past violations can we avoid future viola- transitional justice, as such, has failed. Whether or not tions. Th is utilitarian justifi cation for punishment fi nds prosecutions commence domestically, the debate ought support beyond the Kenyan context.112 However, some not to be disrupted from engaging further with questions commentators have pointed out that there is not neces- of how political violence can be prevented, for example sarily a correlation between failing to punish perpetra- through ‘deep’ legal and institutional change. tors of past violations (for example by granting them Most of the Kenyans consulted by the author suggest- amnesties) and the recurrence of large-scale violations, as ed that legal and institutional reforms are most central such. Post-Franco Spain and post-civil war Mozambique means in changing those conditions that make possible are oft en quoted as examples.113 political violence. Th e challenge is that certain structures On the other hand, it seems clear that in a country in Kenyan governance render diffi cult accountability for such as Kenya where political violence has been sparked political violence. Th e parties to the Kenyan National off not by one major single event where the rule of law Dialogue and Reconciliation recognised that addressing

10 Political violence in Kenya • ISS Paper 205 • November 2009 underlying causes of political violence (the so-called which established a committee of experts on constitution ‘long-term issues and solutions’ as entailed in ‘agenda review to prepare a draft constitution and present it item 4’) is essential and depends on the undertaking of to parliament.118 constitutional, legal and institutional reform.114 Since Controversy has surrounded the committee’s com- then, the parties have confi rmed the importance and mencement. On 23 May 2009, for example, Kenya’s Daily their commitment to profound legal and institutional Nation reported that the committee was complaining changes. Th e parties’ statement of 14 February 2008 con- that it had not yet seen disbursement of funds to make cludes that a broad reform agenda is necessary to address the committee fully operational.119 Moreover, some the root causes of the political crisis that had enabled commentators have questioned whether parliament, political violence to erupt following the 2007 elections. based on its current composition, can agree on putting to According to the parties, this reform agenda includes referendum a constitution that complies with the ‘deep’ for example ‘comprehensive constitutional reforms’; reforms fi rst envisioned by the parties.120 ‘comprehensive electoral reform’; ‘parliamentary reform’; Constitutional reform is deemed important by a ‘police reform’ and ‘legal and judicial reforms’.115 large majority of Kenyans. According to a January 2009 survey, 90 per cent strongly agree or agree that ‘the coalition government must give Kenya a new constitution’.121 A possible failure to see criminal Reforming the Kenyan police is by many observers deemed among the most crucial aspects of the reform justice unfold in Kenya risks process. Such reforms are said to hold potential for signifi cantly strengthening respect for the rule of law.122 bringing about the impression In May 2009, a task force was established to facilitate this reform but there seemed to be lack of clarity on exactly that transitional justice has failed what the task force should address and how ‘deep’ its proposals for police reform should be.123 A key challenge facing the prospects for legal and Refl ecting this acknowledgement, on 4 March 2008, institutional change is that, unlike measures that pursue the parties issued a statement that acknowledges the accountability for perpetrators of past political violence, importance of a constitutional review process based such reforms relate less directly to ‘confronting the past’ on inclusiveness, and recalled the need for its urgent and their implementation is typically a long-term and undertaking by stating it should be completed within technical process. Th is may make a reform process seem 12 months.116 Th e July 2008 Implementation Matrix less urgent than criminal trials. At the same time, while laid down the overall framework according to which putting to trial perpetrators of gross human rights viola- state institutions should be reformed and put forward tions is supported by requirements in international law, deadlines for their undertaking. Besides reconfi rming requirements to undertake legal and institutional reform the commitment to concluding the constitutional reform have a much looser legal foundation, if any.124 What is process within 12 months, it was noted that reforms more, some stakeholders have little interest in ‘deep’ of the judiciary should strengthen its independence structural change taking place. Strong executive powers, and commitment to human rights. Moreover, a review for example, can be seen as enabling a system where those process related to police reform was to be fi nalised who gain power can benefi t excessively from holding within six months; the constitutional review to establish offi ce, therefore potentially leaving incumbent cabinet an independent police commission within 12 months; members and top-level civil servants with little interest in and recruitment and training of police offi cers to bring strengthening the independence of the judiciary and in the police-to-population ratio to UN standards by 2012. other ways reforming the set-up of state institutions. Finally, a number of legislative reforms targeting the Kenyans consulted by the author generally held that civil service were envisioned to take place following the many politicians are part of an elite group that benefi t coming into force of a new constitution, and the Kenyan from state structures as they are, and will therefore Parliament was to be subjected to reforms that aim at pursue a status quo – or at best a limited and superfi cial increasing its oversight with the executive. Th is review reform agenda. Compared to criminal justice it is not process was stipulated to take place within six months.117 only more diffi cult for civil society to push for legal and Because most profound reforms are preconditioned institutional reform because the legal language of state on a new constitution, many of these commitments to obligations is diffi cult to evoke, but stakeholders in the institutional change are, however, yet to be implemented. reform process may also perceive profound reform as To facilitate the constitutional reform, parliament passed off ering few short-term benefi ts and prefer to deal with the Constitution of Kenya Review (amendment) Act 2008 the past in a lighter and less structural way.

Thomas Obel Hansen • ISS Paper 205 • November 2009 11 Yet, because of the centrality of the institutional re-populated’. It is argued that ‘transformation must set-up of state institutions in allowing political violence, occur at an ethical level where the level of conduct is engagement with preventing political violence in Kenya being considered’, and constitution-making should ought to place legal and institutional reform at the be seen only as a fi rst stage in creating some of the very core. Doing so requires that we scrutinise how the conditions in which […] negotiation and re-negotiation reform process unfolds. It also requires that political of how we wish to govern ourselves is carried out’.126 leaders remain under pressure to put into eff ect their Besides the importance of a participatory process that commitments to profound reform. Because ‘deep’ aims at substantial societal change, perhaps the most reform is not an overnight undertaking but a long and important lesson for the Kenyan reform process lies in complicated process, it also requires some amount of asking the question whether a technical approach to patience. Sticking to deadlines can be less important legal and institutional change will actually transform the than pursuing a compromise that is broadly accepted and way institutions function and relate to one and another, at the same time engages substantially with the issues and to Kenyan citizens. Constitutional and legal change at stake. In any case, the active involvement of Kenyan sometimes looks good on paper, but yet creates little civil society is imperative for the reform process to gain on-ground change. According to some of the Kenyans legitimacy and for eff ectively scrutinising that political consulted by the author this is already a problem. One leaders use its context not as a means of gaining personal civil society representative noted that ‘Kenya already has or group advantages or as a smoke-screen for maintain- good laws; what we need is their implementation and ing compromised structures and institutions under the the government’s respect for the rule of law’.127 From one cover of ‘reform’. Th e reform process must ensure that perspective this is exactly the problem that a constitu- state institutions are profoundly transformed; an under- tional review can solve, for example by strengthening the taking that will eventually benefi t Kenyans, as such. It independence of the judiciary and facilitating that state is, as always, important that democracy realises through institutions work in a transparent way. engagement and works in a transparent way – and not However, this is not enough. As also recognised as an exclusive top-down and oblique process. It is also by the parties to the Kenyan National Dialogue and important, however, that the compromises of democracy Reconciliation, it is vital for strengthening the rule of do not end up functioning as an excuse for diminishing law that state offi cials are suffi ciently trained and that the agenda of ‘deep’ reform. recruitment is based on qualifi cations.128 For example, it makes little sense to reform the police by establishing legal provisions for increased accountability and internal It is vital for strengthening oversight mechanisms if police offi cers are not trained to implement these changes. the rule of law that state Some transitional justice scholars have suggested that to reform state institutions and prevent the recurrence offi cials are suffi ciently of human rights violations, it is necessary to rid these institutions of known human rights perpetrators. Th e trained and that recruitment argument goes that only if high-level offi cials who have accepted and perhaps endorsed an institutional culture is based on qualifi cations where it is common practice that its agents violate the law and well-known on-the-ground perpetrators are vetted can these institutions undergo the needed change and For these reforms to eff ectively change structures again be perceived as legitimate in the eyes of citizens.129 that enable political violence, transitional justice Heated debates about vetting leaders of certain state discourses can provide valuable input. It is increasingly institutions in Kenya have already unfolded. Th e UN acknowledged that a legalistic top-down approach to Special Rapporteur on extrajudicial, summary and transitional justice is insuffi cient and entails a number arbitrary executions, Philip Alston, recommends that of risks.125 While constitution-making is obviously both the Police Commissioner and the Attorney General about law and inevitably will involve some degree of resign. Th e Kenyan government responded that the ‘top-down’ decision-making, scholars such as Kirsten Rapporteur is exceeding his mandate by making such McConnachie and John Morrison have argued that ‘if recommendations,130 but has later transferred the Police constitution-making is to maximise its transformative Commissioner to another post.131 For state institutions potential in the reconstitution of societies, it must seek that have a reputation for failing to respond eff ectively a more dynamic task than merely a re-shuffl ing of elites and in a rights-oriented manner to political violence, as the traditional institutions of formal government are or are themselves a part of the problem, the starting

12 Political violence in Kenya • ISS Paper 205 • November 2009 point must be that agents who are proved responsible for violence is not only a question of doing justice for the violating the criminal code are punished according to the victims, but is also seen as serving purposes of reconcili- law, and that those who are not will maintain their post ation and peace.137 Reparations, even if mainly justifi ed unless they for other reasons disqualify. Vetting should as a means of restoring victims’ dignity, can therefore not be applied as a measure of punishment but as a tool potentially also play a role in preventing the recurrence that may strengthen state institutions’ respect for the rule of political violence. of law, integrity, and legitimacy in the eyes of the general Questions related to victims’ redress have received public. Because vetting state offi cials for the individual in little attention in debates about addressing political question in many ways resemble a court conviction, any violence in Kenya. Th e Commission of Inquiry on process that aims at removing rough elements known or Post-Election Violence was established to ‘prevent any assumed to have been involved in serious human rights repetition of similar deeds and, in general, to eradicate violations must provide individuals with guarantees impunity and promote national reconciliation in Kenya’ against arbitrary or collective dismissal. A possible and was mandated to ‘recommend measures of a legal, vetting process in Kenya should therefore not be based political or administrative nature, as appropriate, on removing entities simply because they are alleged including measures with regard to bringing to justice to have been involved in political violence or because those persons responsible for criminal acts’.138 Yet, its they are seen as constituting an obstacle to reform,132 recommendations focus primarily on how to bring these but ought to be based on due process guarantees, in perpetrators to justice and on how a variety of legal and particular the presumption of innocence.133 In any case, it institutional reforms should be undertaken. However, should be up to Kenyan stakeholders to initiate a possible the detailed Human Rights Watch report on the 2007 vetting process, and foreign involvement must be careful election violence entails recommendations for the not to advocate for importing off -the-shelf models. international community to provide support for initia- Nonetheless, vetting state institutions can be imperative tives aimed at ‘compensation for historical and current for bringing about legitimacy for compromised and human rights violations’, and for the Kenyan government disregarded institutions. to ensure the establishment of ‘a process of compensation for those who have lost their homes and property’ in connection with the election violence.139 Remedying victims of mass Th e parties to the Kenyan National Dialogue and Reconciliation in their February 2008 agreement stated violence is not only a question that ‘discussions will be conducted to identify and agree on the modalities of implementation’ of measures aiming of doing justice, but also at ‘ensuring that the assistance to the aff ected communities and individuals is delivered more eff ectively’.140 serving the purposes of Eff orts to redress victims of political violence in Kenya have mainly concentrated on facilitating the return reconciliation and peace of IDPs following the most recent election violence. A Humanitarian Fund for Mitigation of Eff ects and Resettlement of Victims of Post 2007 Election Violence Victim redress and confronting has been set up. Th e fund aims at reconstructing houses socioeconomic causes of political violence and replacing household eff ects for those people who suff ered material damages during the violence.141 Transitional justice discourses oft en emphasise the Compensating individual victims of political violence, need to redress victims in the aft ermath of gross human including victims of sexual violence have not been rights violations. Questions related to reparation for prioritised to the same extent.142 victims have historically been central to the scholar- Diffi cult questions of reparation oft en arise in the ship.134 International standards recognise victims’ aft ermath of large-scale political violence: should the rights to reparation and stipulate how reparation can focus be on compensating individual persons who have take a variety of forms including, but not limited to, been victims of rights-violations or should a collective fi nancial compensation for harm suff ered, restitution approach be taken where victims are defi ned in group and diff erent forms of rehabilitation.135 Recent accounts terms and obtain increased access to certain resources, tend to extend victims’ rights and oft en formulate such as health care and education; what rights-violations transitional justice processes such as criminal trials as a should result in compensation (only violations of civil right of victims to see perpetrators brought to account.136 and political rights or also violations of economic, For some commentators, remedying victims of mass social and cultural rights); how far back should we go in

Thomas Obel Hansen • ISS Paper 205 • November 2009 13 defi ning who is a victim; who should pay for compensa- It has oft en been a problem that recommendations tion schemes; and many more diffi cult questions arise.143 of truth commissions are not suffi ciently implemented. From a preventive perspective, it is important that In South Africa, for example, commentators regret that Kenya addresses questions of how victims of political many of the Truth and Reconciliation Commission’s violence can be redressed. Th e question must be ap- recommendations have not been attended to.150 proached not only in terms of the most recent election Th e Kenyan commission has potential to contribute violence but also as a wider problem where groups that to the reform agenda discussed above as well as to other have historically been exposed to political violence or forms of preventive actions. For this to happen, it is marginalised receive reparations. Th e development of important that the commission’s work is followed closely such policies requires much more than the adoption of a and debated in public. It is also important that its recom- backward-looking approach where victims of the recent mendations are discussed openly and leaders take their election violence are compensated. It requires also a implementation seriously. future-oriented approach where the key objective is to Beyond the Truth, Justice and Reconciliation ensure that individuals and groups who have historically Commission, a number of important initiatives have been subjected to injustices become less marginalised, been discussed which may further an agenda of attend- for example by increasing access to education, health ing to socioeconomic causes of confl ict. Agenda Item 4, care, and land. Reconciliation between communities as decided upon by the parties to the Kenyan National can be important for preventing future political violence Dialogue and Reconciliation, recognises the need for because it potentially leaves both individuals and groups ‘tackling poverty and inequity, as well as combating with a sense that they can overcome tensions of the past regional development imbalances’; the need for address- and move forward.144 ing ‘unemployment, particularly among the youth’; and the need for undertaking land reforms’.151 According to the parties’ Implementation Matrix, fundamental From a preventive perspective, it land reforms are to be facilitated by the Constitutional review process and a number of other mechanisms to be is important that Kenya addresses implemented to deal legally and administratively with problems related to land allocation and ownership.152 Th e victims of political violence same document envisions that action be taken to address poverty, inequality, regional imbalances and unemployment. Among many other activities, the Implementation Th e parties’ agreement to set up a ‘Truth, Justice and Matrix stipulates that action must be taken to ‘ensure Reconciliation Commission’ may prove important for equity and balance are attained in development across these objectives.145 Th e commission is mandated to all regions in job creation, poverty reduction, improved inquire into historical injustices, including political income distribution and gender equity’; ‘increase violence, community displacement and grand corrup- availability of aff ordable and accessible credit, savings tion, and it is envisioned that it can ‘promote peace, programmes and appropriate technologies to create an justice, national unity, healing, and reconciliation among enabling environment for poor communities to take part the people of Kenya.146 Th e commission is intended to in wealth creation’; and ‘generate an average of 740 000 provide a platform for victims to be heard and restore new jobs each year from 2008-2012’.153 their dignity; make recommendations on how to While transitional justice as an analytical framework redress victims of these injustices; and more generally has historically tended to neglect dealing with the past by make recommendations for the prevention of political addressing socioeconomic causes of confl ict, the impor- violence and other violations.147 Th e commission is also tance of such means is increasingly being realised in the mandated to grant amnesties to perpetrators who make scholarship.154 Th e fact that the Kenyan framework for full disclosure of human rights violations and economic preventing the recurrence of political violence outlines crimes committed.148 a detailed agenda and set of action needed to reverse Th e process that has led to the establishment of the socioeconomic factors that are recognised as contribut- commission has been criticised for failing to consult ing to confl ict in many ways stand out in comparison suffi ciently with civil society, and the law itself has been to attempts in other countries of dealing with past described as fl awed, for example because it allows for injustices. amnesties, because it fails to secure independence from Accepting that prevention depends on more than other state institutions, because it lacks provisions for eradicating a culture of impunity and reforming state implementation the commission’s recommendations.149 institutions by paving the way for confronting problems of poverty, unemployment and land distribution is an

14 Political violence in Kenya • ISS Paper 205 • November 2009 important fi rst step in addressing holistically the root political leaders must take responsibility, even if in the causes of political violence in Kenya. However, infl uential short-term it might seem easier to move on without stakeholders in Kenya may have personal interest in some addressing what is needed: ‘deep’, structural changes of the reforms not taking place. For example it has been that can strengthen the rule of law, and at the same time noted that draft land reform legislation is being lobbied address socioeconomic issues such as poverty, inequality, against by large landowners and certain members of and land distribution. Civil society must be allowed, cabinet.155 For the reforms to foster substantial change and willing, to participate and engage substantially in it is obviously not enough that they are included in a the further shaping and implementation of the process. broader process that aims to prevent political violence If not, risks are not only that the undertakings will lack and confl ict on paper. Both politicians and civil society legitimacy but also that it becomes oblique, superfi cial, must continuously push for their actual implementation and unsuccessful in reaching the objective of preventing and ensure that other pending questions of transitional political violence. Although the present debate about justice do not lead to the neglect of these important preventing political violence in Kenya tends to take its structural changes. Th e commitment announced starting point in addressing the recent election violence, by the parties to the Kenyan National Dialogue and it is clear that by undertaking the reforms envisaged, Reconciliation to a reform agenda that includes altering many of the causes of political violence could be con- socioeconomic causes of confl ict is vital but should be fronted eff ectively. translated into concrete and far-reaching initiatives and action. Again, structural changes in society are not an overnight-undertaking. Patience and long-term commit- NOTES ment is required from all involved stakeholders. 1 Th e interviews were carried out confi dentially and the identity of those organisations consulted will remain on fi le with the author only. A window of opportunity 2 F N Aolín, Political violence and gender during times of transition, Columbia Journal of Gender and Law 15(3), 2006, has been created where 829–849. 3 C Moser and F Clark, Victims, perpetrators, or actors? legacies of political violence Gendered armed confl ict and political violence. New York: St Martin Press, 2001, 36. in Kenya can be countered 4 D A Hibbs Jr, Mass political violence: a cross-national causal analysis, New York: John Wiley and Sons, 1973. 5 Moser and Clark, Victims, perpetrators, or actors? 36. On its CONCLUSION ‘group’ or ‘mass’ character, see generally Hibbs Jr, Mass political violence. With the adoption of a broad reform agenda by 6 C D L Rios, Understanding political violence, http://www. the parties to the Kenyan National Dialogue and umbc.edu/llc/llcreview/2004/Understanding_Political_ Reconciliation following the most recent election vio- Violence.pdf accessed 12 May 2009, 30–32. lence, a window of opportunity has been created where 7 C A J Coady, Morality and political violence, Cambridge: legacies of political violence in Kenya can be countered. Cambridge University Press, 2007, 3. Th is commitment to confront past injustices and bring 8 J E Alvarez, Crimes of state/crimes of hate: lessons from about substantial change is imperative. Aft er all, without Rwanda, Th e Yale Journal of International Law 24,1999, leaders’ acknowledgement that certain features in 365–483. society such as impunity, the set-up of state institutions, 9 M Mamdani, When victims become killers, Kampala: Fountain and socioeconomic factors have allowed for political Publishers. violence, profound change is unlikely to occur. Yet, the 10 L Waldorf, Mass justice for mass atrocity: rethinking local process of dealing with political violence in Kenya seems justice as transitional justice, Temple Law Review 79(1), 2006, threatened by a lack of sincere commitment among vital 1–88, 30–32. stakeholders. 11 B Berman and J Lonsdale, Unhappy valley: confl ict in Kenya Despite the existence of a civil society in Kenya that and Africa, London: James Curry Publishers, 1992. reminds the political leadership of its pledges for reform 12 J Overton, Th e colonial state and spatial diff erentiation: Kenya, and scrutinises action taken – and the international 1895–1920, Journal of Historical Geography 13(3),1987, 267–282. community following closely the developments – there 13 Africa Watch, Divide and rule: state-sponsored ethnic violence are risks that important parts of the reform process will in Kenya, New York: Human Rights Watch, 1993, http://www. end up neglected. To avoid these risks materialising, hrw.org/legacy/reports/1993/kenya1193.pdf, accessed 15 May,

Thomas Obel Hansen • ISS Paper 205 • November 2009 15 2009, 6–8. See also Commission of Inquiry into Post-Election 36 E Totolo, Kenya: the Mungiki mess; Waki report, for example Violence, Report of the Commission of Inquiry into Post- at 27. Election Violence ‘Waki Report’, Nairobi, 2008, 24. 37 See generally D M Anderson, Vigilantes, violence and the 14 Africa Watch, Divide and rule, 8-10. See also Waki Report, politics of public order in Kenya, African Aff airs, 101, 2001. 24–26. 531–555. 15 Human Rights Watch, Playing with fi re: weapons prolifera- 38 UN Human Rights Council, Report of the Special Rapporteur tion, political violence, and human rights in Kenya, New on extrajudicial, summary or arbitrary executions, Mr. Philip York: Human Rights Watch, 2002, www.hrw.org/legacy/ Alston advance unedited version, UN Doc. A/HRC/11/2/Add.6, reports/2002/kenya/, accessed 15 May 2009, 20–21. May 26, 2009, for example para. 8 and 10. See the Kenyan 16 Africa Watch, Divide and rule, 1. See also Kenyan National Government’s response to the accusations (denying the Assembly, Parliamentary Select Committee to Investigate existence of ‘death squats’ and political support for extra- Ethnic Clashes in Western and Other Parts of Kenya, Report judicial killings) in Government of Kenya, Response to the of the Parliamentary Select Committee to Investigate Ethnic report of the Special Rapporteur on extrajudicial, arbitrary or Clashes in Western and Other Parts of Kenya, Nairobi, 1992. summary executions, professor Philip Alston, on his mission to Kenya from 16-25 February, 2009, 22 May 2009, http://www. 17 Africa Watch, Divide and rule, 12–16. protectionline.org/spip.php?article8295, accessed 10 June 2009. 18 Human Rights Watch, Playing with fi re, 24–59. 39 D Anderson, How violence infected Kenya’s democracy, Th e 19 Amnesty International, Kenya: political violence spirals, Independent, 30 January 2008, http://www.independent. London: Amnesty International, 1998, www.amnesty.org/en/ co.uk/opinion/commentators/david-anderson-how-violence- library/info/AFR32/019/1998, accessed 1 May 2009, 1. infected-kenyas-democracy-775619.html, accessed 21 May 20 Amnesty International, Kenya: political violence spirals, 4. 2009. See also Waki report, for example at 27–28. 21 Amnesty International, Kenya: political violence spirals, 7–13. 40 P M Kagwanja, Facing Mount Kenya or facing Mekka? Th e Mungiki, ethnic violence and the politics of the Moi succession 22 Human Rights Watch, Ballots to bullets, 6. in Kenya, 1987–2002, African Aff airs 102, 2003, 25–49, quote 23 M Bunting, Th e violence in Kenya may be awful, but it is not from abstract at 25. senseless ’savagery’, Th e Guardian, 14 January 2008, http:// 41 For example by collecting taxes and providing security for www.guardian.co.uk/commentisfree/2008/jan/14/kenya. those who pay. See for example Mungiki: the growing crisis, world, accessed 3 May 2009. Daily Nation, 15 June 2009, 1 and 6. 24 Human Rights Watch, Ballots to bullets, 23, 35. 42 See for example the Standard’s cover of Mungiki attacks 25 Waki report, 383. on villagers in late April in: Slaughter of the innocent, Th e 26 Human Rights Watch, Ballots to bullets, 2. Standard, 22 April 2009, 1, 4–5; and Daily Nation’s cover of the same event in Killings: Mungiki massacre, Daily Nation, 22 27 As of April 2009, the number of displaced households in the April 2009, 1–2. Rift Valley was estimated at more than 14,000. See Kenyan Ministry of Special Programmes, Status of satellite camps 43 Human Rights Watch, Playing with fi re, 20. in Rift Valley as at 1st April, 2009, http://www.sprogrammes. 44 Bunting comes close to perceiving the post-election violence as go.ke/index.php?option=com_content&task=view&id=150&It ‘economically rational’, and thus as a ‘natural result’ of certain emid=1, accessed 10 June 2009. socio-economic features of the Kenyan society and a political 28 Human Rights Watch, Ballots to bullets, 35–39. culture that endorses violence. See M Bunting, Th e violence in Kenya may be awful. 29 Kenya Human Rights Commission, Violating the vote: a report of the 2007 general elections, Nairobi, 15 September 2008, 45 For a discussion of these issues, see for example M Mamdani, www.khrc.or.ke/documents/violating_the_vote.pdf, accessed When victims become killers, preface. 10 June 2009, for example at 8. 46 Waki Report, 25. 30 Human Rights Watch, Ballots to bullets, 43–48. 47 Waki Report, 31. 31 Human Rights Watch, Ballots to bullets, 48. 48 Ibid citing Republic of Kenya, Report of the Commission of 32 See generally Waki Report, 417–420. Inquiry into the Illegal/ Irregular Allocation of Pubic Land: Th e Ndungu Report, Nairobi: Government Printer, June 2004, 33 Human Rights Watch, Ballots to bullets, 24–31. See also Waki 14. Report, 383. 49 Human Rights Watch, Ballots to bullets, 12–15. 34 D Wabala and F Mukinda, Trail of death as gunmen run wild, Daily Nation, 8 May 2009, 1 and 4. 50 Human Rights Watch, Ballots to bullets, 11–12. 35 E Totolo, Kenya: the Mungiki mess, ISN Security Watch, 17 51 Ibid. November 2008, http://www.isn.ethz.ch/isn/Current-Aff airs/ 52 Human Rights Watch, Ballots to bullets, 15–17. Security-Watch/Detail/?ots591=4888CAA0-B3DB-1461-98B9- E20E7B9C13D4&lng=en&id=93861, accessed 20 May 2009. On 53 Waki Report, 28. Mungiki’s association with ‘indigenous beliefs’ and the sects 54 Waki Report, 28. political agenda in earlier years, see G N Wamue, Revisiting 55 Ibid. our indigenous shrines through Mungiki, African Aff airs, 100, 2001, 453–467. 56 Waki Report, for example at 26.

16 Political violence in Kenya • ISS Paper 205 • November 2009 57 Africa Watch, Divide and rule, 61–67. 74 Author’s interview with representatives of civil society organisation, Nairobi, 21 May 2009. Notes on fi le with author. 58 Africa Watch, Divide and rule, 67–71. 75 See for example A Suich, Beyond tribalism: Kenya’s violence 59 Amnesty International, Kenya: political violence spirals, 10–13. is not just about ethnicity. Age and poverty are factors too, 60 Th e Akiwumi Report was completed in 1999, but only made Newsweek, 3 January 2008, http://www.newsweek.com/ public in 2002 before Kibaki’s taking offi ce. Human Rights id/84257, accessed 20 May 2009. Watch notes how ‘the report’s fi ndings were completely 76 Waki Report, 33–35. ignored by the incoming administration’. See Human Rights Watch, Ballots to bullets, 18–19. 77 Author’s interview with representatives of civil society organisation, Nairobi, 21 May 2009. Notes on fi le with author. 61 EU Election Observer Mission: Kenya, General Elections, Preliminary statement, 27 December 2007, http://issuu.com/ 78 See generally Society for International Development, Pulling tfk 2/docs/kenya_2007_fi nal_preliminary_statement/11, apart: facts and fi gures on inequality in Kenya, Nairobi: accessed 22 May 2009. Society for International Development, 2004, http://www. scribd.com/doc/2223295/Pulling-Apart-Facts-and-Figures-on- 62 Human Rights Watch, Ballots to bullets, 17-20. See also Waki Inequality-in-Kenya, accessed 14 May 2009. Report, 26–27. 79 Author’s interview with representatives of civil society 63 Waki Report, 420. organisation, Nairobi, 19 May 2009, notes on fi le with author. 64 Human Rights Watch, Ballots to bullets, 17-20; Waki Report, 80 Human Rights Watch, Ballots to bullets, 17–20; Waki Report, 59–60. 12-15. See also Republic of Kenya, Report of the Commission 65 Th e Waki Report estimates that 405 of the 1133 victims were of Inquiry into the Illegal/ Irregular Allocation of Pubic Land; killed by gunshots and while the commission ‘acknowledges Th e Ndungu Report, Nairobi: Government Printer, June 2004; the possibility that some of those killed and wounded by Kenya Land Alliance, Th e national land policy in Kenya: gunshots may have been the victims of people other than addressing historical injustices, Issues paper No.2/2004, http:// the Police’ it concluded that ‘no evidence to this eff ect was www.caledonia.org.uk/land/documents/kla_issues_paper. received’. See Waki Report, 385 pdf,accessed May 12, 2009. 66 Waki Report, 387. 81 Author’s interview with representatives of civil society 67 Waki report, 396-398, 421. organisation, Nairobi, 21 May 2009, notes on fi le with author. 68 Waki report, 404. 82 Author’s interview with representatives of civil society organisation, Nairobi, 21 May 2009, notes on fi le with author. 69 Report of the Special Rapporteur on extrajudicial, summary or arbitrary executions, Mr. Philip Alston (advance unedited 83 See for example Human Rights Watch, Ballots to bullets, 12–15. version), para. 68–73. According to the Kenyan Government, 84 Waki Report, 32. however, since 2000, 81 police offi cers have been prosecuted for 85 See for example Africa Watch, Divide and rule, 22–24. murder or manslaughter. It is unclear how many – if any – of these are held to relate to the election violence. See Government 86 See for example A Suich, Beyond tribalism: Kenya’s violence is of Kenya, Response to the report of the Special Rapporteur on not just about ethnicity. extrajudicial, arbitrary or summary executions, professor Philip 87 See for example A Neier, Rethinking truth, justice, and guilt Alston, on his mission to Kenya from 16–25 February, 2009, 1. aft er Bosnia and Rwanda, in C Hesse and R Post (eds), Human 70 Waki report, 446–460, concluding at 456 that in ‘our view is rights in political transitions: Gettysburg to Bosnia, New York: that the lackadaisical manner in which the government dealt Zone Books, 1999, 39–52. with the Akiwumi Report only goes to illustrate that the state 88 Feher, Michael, Terms of reconciliation”, in Hesse, Carla and was not particularly interested in resolving once and for all the Post, Robert, Human Rights in Political Transitions: Gettysburg issue of ethnic violence’. to Bosnia, New York: Zone Books, 1999, 325-338. 71 As Kenya Human Rights Commission states: impunity has 89 For a comprehensive overview of truth commissions and for been ‘a recurrent problem during election periods, when Kenya a discussion of their merits and challenges, see generally P seems to suspend laws or provide “collective impunity” for B Hayner, Unspeakable truths: facing the challenge of truth breaches of the same’. See Kenya Human Rights Commission, commissions, New York: Routledge, 2002. Violating the vote: a report of the 2007 general elections, Nairobi, 15 September 2008, www.khrc.or.ke/documents/ 90 C S Nino, Radical Evil on Trial, New Haven and London: Yale violating_the_vote.pdf, accessed 10 June 2009, 11. University Press, 1996, 146. 72 Such correlation between impunity and political violence is 91 For a discussion of these ‘transitional justice mechanisms’, a general assumption in scholarly writings on transitional see for example M Minow, Between vengeance and forgiveness: justice. See for example J E Mendez, In defense of transitional facing history aft er genocide and mass violence, Boston: Beacon justice, in J A McAdams (ed), Transitional justice and the rule Press, 1998; or R G Teitel, Transitional justice, Oxford: Oxford of law in new democracies, Notre Dame: University of Notre University Press, 2000. Dame Press, 1997, 1–26. 92 For the argument that transitional justice discourses can 73 Author’s interview with representatives of civil society inform both responses to past repression in periods of transi- organisation, Nairobi, 28 May 2009. Notes on fi le with author. tion from authoritarianism to democracy and responses to See also Waki Report, 27–28. past widespread violence in so-called ‘confl icted democracies’,

Thomas Obel Hansen • ISS Paper 205 • November 2009 17 see generally F N Aolaín and C Campell, Th e paradox of to international accountability: the search for justice in a world transition in confl icted democracies, Human Rights Quarterly of states, Tokyo: United Nations University Press, 2004, 266. 27 (2005), 172–213. 104 Th is for example was the general opinion among those 93 See for example J O Latigo, Northern Uganda: tradition-based Kenyans consulted by the author. practices in the Acholi region”, in L Huyse and M Salter (eds), 105 See the International Criminal Court, Situations and cases, Traditional justice and reconciliation aft er violent confl ict: http://www.icc-cpi.int/Menus/ICC/Situations+and+Cases/, learning from African experiences, Stockholm: International accessed on 10 June 10 2009. Institute for Democracy and Electoral Assistance, 2008, 85–122. 106 For a profound critique of international tribunals and international criminal justice, see generally Drumbl, Atrocity, 94 On the normative claims of transitional justice see for example punishment, and international law. Clark’s discussion in P Clark, Establishing a conceptual framework: six key transitional justice themes, in P Clark and Z D 107 Waki report, 472. Kaufman (eds), Aft er genocide: transitional justice, post-confl ict 108 Waki report, 444. reconstruction and reconciliation in Rwanda and beyond, London: Hurst, 2008, 191-205. On the rule of law assumption 109 M J Aukerman, Extraordinary evil, ordinary crime: a frame- and liberalisation, see more specifi cally Teitel, Transitional work for understanding transitional justice”, Harvard Human justice. On the preventive rationale, see for example UN Rights Journal 15 (2002), 39–97. Secretary-General, Th e rule of law and transitional justice 110 Nino, Radical Evil on Trial, 146–147. in confl ict and post-confl ict societies: report of the Secretary- 111 At least this was common perception among those Kenyans General, UN Doc. S/2004/616, 2004, for example para. 4. consulted by the author. 95 K Roth, Justice or impunity: what will Kenya choose, 3 April 112 See for example D F Orentlicher, Settling accounts: the duty to 2009, http://www.hrw.org/en/news/2009/04/03/justice-or- prosecute human rights violations of a prior regime, Th e Yale impunity-what-will-kenya-choose, accessed 12 June 2009. Law Journal 100 (1991), 2537–2615. 96 Ibid. 113 H Cobban, Amnesty aft er atrocity: healing nations aft er 97 Parties to the Kenyan National Dialogue and Reconciliation genocide and war crimes, Boulder: Paradigm Publishers, 2007. on the resolution of the political crisis and its root causes; the For a comprehensive empirical study of amnesties in political Government of Kenya/Party of National Unity and the Orange transitions, see L Mallinder, Amnesty, human rights and politi- Democratic Movement, Public statement, Nairobi, 1 February cal transitions: bridging the peace and justice divide, Oxford: 2008, http://www.dialoguekenya.org/docs/Public%20 Hart Publishing, 2008. Statement%20from%20Dialogue%20Feb%201%20FINAL.pdf, 114 Parties to the Kenyan National Dialogue and Reconciliation accessed 29 May 2009, preamble. on the resolution of the political crisis and its root causes; the 98 Parties to the Kenyan National Dialogue and Reconciliation Government of Kenya/Party of National Unity and the Orange on the resolution of the political crisis and its root causes; Democratic Movement, Kenyan National Dialogue and the Government of Kenya/Party of National Unity and the Reconciliation: annotated agenda and timetable, agenda 4. Orange Democratic Movement, Kenyan National Dialogue 115 Parties to the Kenyan National Dialogue and Reconciliation and Reconciliation: annotated agenda and timetable, Nairobi, on the resolution of the political crisis and its root causes; the 1 February 2008, http://www.dialoguekenya.org/docs/ Government of Kenya/Party of National Unity and the Orange Signed_Annotated_Agenda_Feb1st.pdf, accessed 1 February Democratic Movement, Agenda item 3: how to solve the 2009, agenda 2. political crisis, para. 3. 99 Parties to the Kenyan National Dialogue and Reconciliation 116 Parties to the Kenyan National Dialogue and Reconciliation on the resolution of the political crisis and its root causes; on the resolution of the political crisis and its root causes; the the Government of Kenya/Party of National Unity and the Government of Kenya/Party of National Unity and the Orange Orange Democratic Movement, Agenda item 3: how to solve Democratic Movement, Longer-terms issues and their solu- the political crisis, Nairobi, 14 February 2008, http://www. tions: constitutional review , Nairobi, 4 March 2008, http:// dialoguekenya.org/docs/14_Feb_08_TsavoAgreement.pdf, www.dialoguekenya.org/docs/Signed_Annotated_Agenda_ accessed 10 June, 2009,para. 3 Feb1st.pdf, accessed 1 February 2009, agenda 4. 100 Waki report, 472-475. 117 On all of these components of institutional reform, see Parties 101 Human Rights Watch, Kenya: swift ly enact special to the Kenyan National Dialogue and Reconciliation on the tribunal: International Criminal Courts should be a last resolution of the political crisis and its root causes; Statement resort for justice, 25 March 2009, http://www.hrw.org/en/ of principles on long-term issues and solutions, Nairobi, 30 news/2009/03/24/kenya-swift ly-enact-special-tribunal, July 2008, http://www.dialoguekenya.org/docs/S_of_P_with_ accessed 12 May 2009. Matrix.pdf, accessed 5 June 2009. 102 E Th omasson, Ocampo: I will work with local courts, Sunday 118 Constitution of Kenya Review Act, 2008 Act (Act no. 9 of Nation, 31 October 2009. See also P Opiyo and B Gikandi, 2008), article 23. Raila: we want local tribunal, Th e Standard, 15 June, 1 and 5. 119 O Mathengep, No funds yet for review team, Daily 103 C Flinterman, Th e International Criminal Court: obstacle or Nation, 23 May 2009, http://www.nation.co.ke/News/poli- contribution to an eff ective system of human rights protection, tics/-/1064/602162/-/xy2bq6z/-/index.html, accessed 15 June in R Th akur and P Malcontent (eds), From sovereign impunity 2009.

18 Political violence in Kenya • ISS Paper 205 • November 2009 120 J Kwayera, Reforms far from being realised, Th e Standard, 1 131 C Rwenji, Former police boss takes over at the GPO, Daily June 2009, http://www.eastandard.net/archives/InsidePage. Nation, 14 September 2009, http://www.nation.co.ke/News/- php?id=1144015629&cid=289&, accessed 15 June 2009. /1056/657852/-/umuqqh/-/index.html, accessed 15 September 2009. 121 Th e Kenya National Dialogue and Reconciliation Monitoring Project, National baseline survey, January 2009, 27. 132 Report of the Special Rapporteur on extrajudicial, summary or arbitrary executions, Mr. Philip Alston (advance unedited 122 See for example M Machuka, US urges Kenya to reform police; judiciary, Th e Standard, 15 June 2009, http://www.eastandard. version), para. 19. net/InsidePage.php?id=1144016732&cid=418&, accessed 15 133 See also F Andreu-Guzmán, Due process and vetting, in A June 2009. Th e same view was put forward by many of the Mayer-Rieckh and P de Grieff (eds), Justice as prevention: Kenyans consulted by the author. vetting public employees in transitional societies, New York: 123 Editorial, Speak in one voice on Kenya police reforms, Social Science Research Council, 2007, 448–481. Daily Nation, 13 May 2009, http://www.nation.co.ke/oped/ 134 See for example Minow, Between vengeance and forgiveness, Editorial/-/440804/597934/-/q0km7lz/-/index.html, accessed 91–117. June 15 2009. 135 UN General Assembly, Basic principles and guidelines on the 124 A duty to prosecute perpetrators of international crimes and right to a remedy and reparation for victims of gross violations of other gross human rights violations is generally said to fl ow international human rights and serious violations of international both from treaty law such as the International Covenant on humanitarian law, UN Doc. A/RES/60/147, 2006, principle 15–23. Civil and Political Rights and, for some crimes, from custom- 136 M C Bassiouni, International recognition of victims’ rights, ary international law. See Orentlicher, Settling accounts: the Human Rights Law Review 6 (2006), 203–279; R Aldana- duty to prosecute human rights violations of a prior regime. Pindell, An emerging universality of justiciable victims’ rights Exactly how and when such duty requires states to punish in the criminal process to curtail impunity for state-sponsored perpetrators of political violence that amount to gross human crimes, Human Rights Quarterly 26 (2004), 605–686. rights violations is still a matter of dispute. For a recent account of the question see L Mallinder, Can amnesties and in- 137 Teitel, Transitional justice, 127 ternational justice be reconciled?, Th e International Journal of 138 Parties to the Kenyan National Dialogue and Reconciliation Transitional Justice 1 (2007), 208–230. Legal and institutional on the resolution of the political crisis and its root causes; the reform, on the other hand, is seldom debated as an obligation Government of Kenya/Party of National Unity and the Orange of international law. Democratic Movement, Agreement establishing a commission 125 See for example K McEvoy, Letting go of legalism: developing of inquiry on post-election violence, Nairobi, 4 March 2008, 1. a ‘thicker’ version of transitional justice, in K McEvoy and L 139 Human Rights Watch, Ballots to bullets, 10. McGregor (eds), Transitional justice from below: grassroots activism and the struggle for change, Human rights law in 140 Parties to the Kenyan National Dialogue and Reconciliation perspective (vol. 14): transitional justice from below: grassroots on the resolution of the political crisis and its root causes; the activism and the struggle for change, Oxford and Portland: Government of Kenya/Party of National Unity and the Orange Hart Publishing, 2008, 15–45. Democratic Movement, Kenyan National Dialogue and Reconciliation: annotated agenda and timetable, agenda 2. 126 K McConnachie and J Morrison, Constitution-making, transition and the reconstitution of society, in K McEvoy and 141 Kenyan Government, Kibaki: Launch of public fundraising L McGregor (eds), Transitional justice from below: grassroots towards the resettlement of internally displaced persons, 12 activism and the struggle for change, Human rights law in May 2008, http://www.polity.org.za/article/kenya-kibaki- perspective (vol. 14): transitional justice from below: grassroots launch-of-public-fundraising-towards-the-resettlement-of- activism and the struggle for change, Oxford and Portland: internally-displaced-persons-12052008-2008-05-12, accessed Hart Publishing, 2008, 97. 16 June 2009. 127 Author’s interview with representatives of civil society 142 CARE, Kenya: Women and girls victims of post-election organisation, Nairobi, 21 May 2009 (notes on fi le with author). violence still seeking justice one year later, Nairobi, 23 February 2009, http://www.reliefweb.int/rw/rwb.nsf/db900sid/ 128 Parties to the Kenyan National Dialogue and Reconciliation MUMA-7PP59J?OpenDocument, accessed 16 June 2009. on the resolution of the political crisis and its root causes; Statement of principles on long-term issues and solutions. 143 See generally J Elster, Closing the Books: Transitional Justice in Historical Perspective, Cambridge: Cambridge University 129 A Mayer-Rieckh, On preventing abuse: vetting and other Press, 2004. transitional reforms, in A Mayer-Rieckh and P de Grieff (eds), Justice as prevention: vetting public employees in transitional 144 V C Charles, Reconciliation, in V C Charles and E Doxtader societies, New York: Social Science Research Council, 2007, (eds), Pieces of the puzzle: keywords on reconciliation and 482–520. transitional justice, Cape Town: Institute for Justice and Reconciliation, 2004, 3-9; and M R Amstutz, Restorative 130 Report of the Special Rapporteur on extrajudicial, summary justice, political forgiveness, and the possibility of political or arbitrary executions, Mr. Philip Alston (advance unedited reconciliation”, in D Philpott (ed), Th e politics of past evil: version), para. 85 (a) and 93. On the government of Kenya’s religion, reconciliation, and the dilemmas of transitional justice, response, see for example B Namunane and A Shiundup, Notre Dame: University of Notre Dame Press, 2006, 151–188. Death squads: Kenya plots to oust Alston, Daily Nation, 7 June 2009, http://www.nation.co.ke/News/-/1056/607960/-/item/1/-/ 145 Parties to the Kenyan National Dialogue and Reconciliation k1hajv/-/index.html, accessed 16 June 2009. on the resolution of the political crisis and its root causes; the

Thomas Obel Hansen • ISS Paper 205 • November 2009 19 Government of Kenya/Party of National Unity and the Orange the Government of Kenya/Party of National Unity and the Democratic Movement, Truth, Justice and Reconciliation, Orange Democratic Movement, Kenyan National Dialogue Nairobi, 4 March, 2008, http://www.dialoguekenya.org/docs/ and Reconciliation: annotated agenda and timetable, KenyanNationalDialogue_Truth&Justice.pdf, accessed 29 May agenda 4. 2009. 152 Parties to the Kenyan National Dialogue and Reconciliation 146 Th e Truth, Justice and Reconciliation Bill, 2008, http://www. on the resolution of the political crisis and its root causes; kenyalaw.org/Downloads/Bills/2008/Th e_Truth_Justice_and_ Statement of principles on long-term issues and solutions, Reconciliation_Commission_Bill_2008.pdf, accessed 9 May 4–5. 2009, article 5 (a). 153 Parties to the Kenyan National Dialogue and Reconciliation 147 Th e Truth, Justice and Reconciliation Bill, 2008, article 5 (l), on the resolution of the political crisis and its root causes; (p), and (r) (respectively). Statement of principles on long-term issues and solutions, 5–6. 148 Th e Truth, Justice and Reconciliation Bill, 2008, article 5 (m). 154 For example Z Miller, Eff ects of invisibility: in search of the 149 N Wainaina, Th e truth, justice and reconciliation commission: economic in transitional justice, International Journal of a fl awed law, Pambazuka News, 29 January 2009, http://www. Transitional Justice 2 (2008), 266–291. pambazuka.org/en/category/comment/53657, accessed 16 June 2009. 155 Th e Kenya National Dialogue and Reconciliation Monitoring Project, Agenda item 4: long-standing issues and solutions, 150 See generally the essays entailed in C Villa-Vicencio and F 19; M Mutiga, Kenya: how powerful forces have blocked du Toit, Truth and reconciliation in South Africa: 10 years on, land reforms, Th e Zimbabwean, 24 May 2009, http://www. Cape Town: Institute for Justice and Reconciliation, 2006. thezimbabwean.co.uk/2009052421356/africa/kenya-how- 151 Parties to the Kenyan National Dialogue and Reconciliation powerful-forces-have-blocked-land-reforms.html, accessed 16 on the resolution of the political crisis and its root causes; June 2009.

20 Political violence in Kenya • ISS Paper 205 • November 2009 SUBSCRIPTIONS TO ISS PUBLICATIONS

If you would like to subscribe to ISS publications, please complete the form below and return it to the ISS with a cheque, or a postal/ money order for the correct amount, made payable to the Institute for Security Studies (marked not transferable). Please note that credit card payments are also welcome. You can also deposit your payment into the following bank account, quoting the reference: PUBSPAY. ISS bank details: ABSA, Brooklyn Court, Branch Code: 632005, Account number: 405 749 8921 Please mail, fax or email this form to: ISS Publication Subscriptions, PO Box 1787, Brooklyn Square, 0075, Pretoria, South Africa. ISS contact details: (Tel) +27 12 346 9500, (Fax) +27 12 460 0998, Email: [email protected] Website: www.issafrica.org

➤ PERSONAL DETAILS Title: ...... Surname: ...... Initials: ...... Organisation: ...... Position: ...... Postal Address: ...... Postal Code: ...... Country: ...... Tel: ...... Fax: ...... Email: ...... Method of Payment: Visa ■ Master Card ■ Other ■ Specify: ...... Card Number: ...... Expiry Date: ...... / ...... Cardholder Name: ...... Signature: ......

➤ PUBLICATIONS South Africa African Countries* International

African Security Review R 200,00 US$ 40,00 US$ 55,00 (4 issues per year) ISS Monographs R 370,00 US$ 75,00 US$ 95,00 (Approx. 15 per year) ISS Papers R 150,00 US$ 30,00 US$ 40,00 (Approx. 12 per year) SA Crime Quarterly R 115,00 US$ 25,00 US$ 35,00 (4 issues per year) Comprehensive subscription R 800,00 US$ 160,00 US$ 210,00 (African Security Review, Monographs, Papers and SA Crime Quarterly)

➤ SUBSCRIPTIONS Indicate Cost African Security Review only Knowledge ISS Monographs only empowers Africa ISS Papers only

SA Crime Quarterly only La savoir Comprehensive subscription émancipe l’Afrique

TOTAL

* Angola; Botswana; Burundi; Congo-Brazzaville; Democratic Republic of the Congo; Gabon, Kenya, Lesotho, Madagascar; Malawi, Mauritius; Mozambique; Namibia; Reunion; Rwanda; Seychelles; Swaziland; Tanzania; Uganda; Zambia; Zimbabwe (formerly African Postal Union countries).

Please visit the magazine website for more information www.the-african.org

Our readers: ‘Informed, connected and infl uential’

Our features: ‘Original, thought-provoking and independent’

For more information or to subscribe contact the editor Liesl Louw-Vaudran at [email protected] or write to the ISS at PO Box 1787, Brooklyn Square, 0075, South Africa As a leading African human security research institution, the Institute for Security Studies (ISS) works towards a stable and peaceful Africa ABOUT THE PAPER characterised by sustainable development, human rights, the rule of law, democracy, collaborative Following the 2007 elections, Kenya suff ered political security and gender mainstreaming. The ISS realises violence. More than 1 000 Kenyans lost their lives and many this vision by: were displaced. While oft en portrayed as standing in contrast to an otherwise politically stable country, the fact remains ■ Undertaking applied research, training and that large-scale political violence has occurred on several capacity building other occasions. Th is paper looks into the legacies of political ■ Working collaboratively with others violence in Kenya, and points to some of its main causes. By ■ Facilitating and supporting policy formulation drawing on transitional justice discourses, the paper discusses the action taken by the parties to the Kenya National Dialogue ■ Monitoring trends and policy implementation and Reconciliation and other stakeholders. In analysing these ■ Collecting, interpreting and disseminating measures, it is argued that for Kenya to prevent the recurrence information of political violence, priority must be given to profoundly ■ Networking on national, regional and reforming institutions and ensuring that accountability international levels measures are set up.

Copyright in the volume as a whole is vested in the Institute Th omas Obel Hansen is a PhD researcher with Aarhus for Security Studies, and no part may be reproduced University Law School. His research focuses on the fi eld of in whole or in part without the express permission, in transitional justice, and is based on fi eldwork in Rwanda. He writing, of both the authors and the publishers. has conducted research stays with African civil society organi- The opinions expressed do not necessarily refl ect those of the Institute, its trustees, members of the Council or donors. Authors sations, including a stay with ISS’ Nairobi offi ce in February- contribute to ISS publications in their personal capacity. June 2009. He has published and lectured on human rights, transitional justice, and international criminal law. Published by the Institute for Security Studies, P O Box 1787, Brooklyn Square 0075 Pretoria, South Africa FUNDER Tel: (27-12) 346 9500 Fax: (27-12) 460 0998 [email protected] Th is publication was made possible through funding provided www.issafrica.org by the Governments of Denmark, the Netherlands, Norway and Sweden. Design, layout and printing Marketing Support Services

ISS Paper No 205 Price R15,00

Political violence in Kenya • ISS Paper 205 • November 2009