Utilising the due diligence standard to interrogate Kenya’s accountability efforts with regard to survivors of sexual violence in the 2007-2008 post-election violence

in the 2007/2008 post-election violence. It finds that Kenya suffered from a number of structural and systemic shortcomings that resulted in its failure to meet its obligation to prevent, investigate, prosecute and compensate for such acts of sexual violence perpetrated by both state and non-state actors. Key among them are a lack of well-coordinated multi-sectoral approaches to address sexual violence; human capacity gaps in the provision of medico-legal services to survivors; and systemic failures in the investigation and prosecution of sexual violence cases. The article further highlights the hope for future accountability inherent in the recent ruling in Constitutional Petition 112 of 2013 which held the state accountable for all gaps and shortcomings in responding to sexual violence during the post-election violence. The article concludes by advocating community-based multi-sectoral approaches in prevention and response to sexual violence in the Kenyan context with an emphasis on improving both human and technical capacities for provision of medico-legal services to survivors.


Introduction
Since the restoration of multiparty politics in 1992, violence and displacements are a recurring feature of Kenyan elections. 1 This is largely attributable to the Kenyan political elite's ethnic mobilisation and sensitisation of voters based on both actual and perceived inequalities in the distribution of national resources. 2 They utilise these divisive ethnic strategies with the objective of securing political and economic resources. 3 Fjelde and Hoglund 4 observe that the dominance of ethnicity in political organisation in post-colonial Africa formed the basis for its prominence with the advent of multiparty politics in the 1990s. Similarly, Bates 5 notes that politics in Africa since independence has centred around control of patronage resources by ethnically-defined coalitions. Both observations hold true in the Kenyan context. The utilisation of ethnicity as a strategy for political mobilisation in Kenya is shaped by violence and repression, resulting in a strong sense of victimisation and injustice among ethnic minorities. 6 This is further aggravated by the 'winner take all' nature of the Kenyan electoral system which is an identified trigger for electoral violence. 7 Moreover, the centralisation of power in the presidency and weak political institutions further serve to normalise and institutionalise political violence. 8 Ultimately, ethnicity remains a key factor in political mobilisation and a critical source of patronage, thereby serving as a key influencer of voting patterns in contemporary Kenyan politics. 9 Consequently, ethnic mobilisation results in ethnic voting with violence as a means of deterring rival supporters and suppressing the undecided voters. 10 Since the formal installation of multi-party democracy in 1992, and despite the continued support for increased participation of women in Kenyan politics to date, Kenya's democracy has diminished neither ethnic nor sexual violence attached to politics with the two being a recurrent feature of every electoral cycle. The 2007-2008 postelection violence erupted when the presidential election results were disputed amid allegations of vote rigging. 11 It was the most violent and destructive cycle during which 900 cases of sexual violence were documented. 12 The most common perpetrators were reported to be both men and women affiliated with government or political groups. 13 The ethnicisation of state security responses to electoral violence 14  created a fertile ground for electioneering violence, in general, and sexual violence, in particular.
Sexual violence is defined as any sexual act or attempt to obtain a sexual act using coercion, threats of harm or physical force that results in physical or psychological harm. 16 It is a human rights violation and is addressed under a growing number of international agreements and treaties, including the 1981 Convention on the Elimination of All Forms of Discrimination against Women (CEDAW); the 1993 Declaration on the Elimination of Violence against Women (DEVAW); the 1994 Programme of Action (Cairo); the 1995 Beijing Declaration and Platform of Action; the International Covenant on Civil and Political Rights (1966)  Human rights principles require that sexual violence receives the same treatment, attention and resources as other serious violent crimes. 17 Although sexual violence affects both men and women, it disproportionately impacts women. DEVAW recognises that violence against women is a manifestation of historically unequal power relations between men and women. 18 Article 4(c) of DEVAW provides that states should 'exercise due diligence to prevent, investigate, and, in accordance with national legislation, punish acts of violence against women, whether those acts are perpetrated by the state or by private persons'. This due diligence standard can therefore be 16  interpreted to hold that the government must prevent and respond to sexual violence committed by either public or private actors.
Article 4(c) of DEVAW adopts the due diligence standard as a benchmark for assessing whether the state has met its obligation to prevent, investigate and punish acts of violence against women by either state or private perpetrators. 19 However, its application prior to Erturk's 20 elaboration in 2006 was state-centric and limited to responses to such acts as and when they occur. 21 The focus on response led to a neglect of the corresponding obligations to prevent the violence and compensate the victims. Also, the statecentredness led to further the neglect of the responsibility of nonstate actors. 22 It was because of these shortcomings that the UN Special Rapporteur on Violence against Women 23 elaborated on the four core components of the due diligence standard as a tool for the elimination of violence against women. This forms the international standard of the state's obligation regarding violence against women, which is one of due diligence to prevent, investigate, prosecute and compensate for such acts of violence. This obligation is binding on the state and cannot be delegated, even in instances where some functions are performed by another state or a non-state actor. 24 Based on the foregoing, the article subsequently uses the lens of the due diligence standard as elaborated on by the Special Rapporteur to interrogate Kenya's domestic accountability efforts regarding survivors of sexual violence in the 2007-2008 post-election violence. It utilises such interrogation to determine whether Kenya fulfilled its international human rights obligations with specific regard to prevention and response to election-related sexual violence in the post-election violence. Moreover, it seeks to analyse prevention and response mechanisms employed by duty bearers during the post-election violence with the aim of identifying priority gaps and proposing concrete measures that can be pursued to ensure effective mechanisms to election-related sexual violence.
The article consists of three parts. The first part elaborates on the four core components of the due diligence standard as elaborated on by the Special Rapporteur. The second part utilises this due diligence standard to critically analyse the prevention and response mechanisms deployed by duty bearers during the post-election violence. The third part concludes by highlighting priority gaps in the mechanisms deployed as identified in the critical analysis and proposing some concrete measures to prevent their recurrence in future elections.
2 Elaboration of the four core components of the due diligence standard as a tool for the elimination of violence against women

Prevention
Prevention addresses the systemic causative factors of sexual violence which include discrimination against women in all its forms, socio-cultural biases and a lack of adequate institutional responses. 25 Invariably, states seek to discharge this obligation through the adoption of specific legislation; the development of national action plans and multisectoral approaches to prevent violence; awarenessraising campaigns; and the training of specified professionals who attend to survivors. 26 Ultimately, prevention seeks to overcome gender stereotypes, biases and societal norms that facilitate violence against women. However, despite states initiating varied preventive programmes, Erturk 27 notes that scant evidence exists of state engagement in overall societal transformation to eliminate existing gender biases, or to support civil society initiatives on the same. 28

Protection
Protection addresses itself to keeping survivors safe, avoiding recurrences, and ensuring that they receive adequate and timely support services once the violence has occurred. Measures undertaken by states in discharging this obligation centre around the provision of services to survivors. These include telephone hotlines for reporting sexual violence, the provision of both medical and legal services to survivors, shelters and safe spaces, and financial aid to 25  survivors. 29 However, despite the adoption of such measures states continue to face major challenges in the enforcement of protective obligations. This is caused by, among other challenges, the lack of adequate enforcement of civil remedies and criminal sanctions and the absence or inadequate provision of medico-legal services and shelters. 30 States must therefore strive to ensure that survivors have access to justice as well as basic support services which include health care and counselling that (a) respond to their immediate needs; (b) help protect them against further harm; and (c) address the longerterm consequences of violence. 31 This can be achieved through the development and implementation of robust legal frameworks to create a safe environment that enables survivors to report violations and obtain effective protective measures. Such measures should not only be survivor centred but should also guarantee the rights of the accused perpetrators. 32

Punishment
Punishment as an obligation requires that states investigate and appropriately punish these criminal acts of violence against women. States have mainly perceived this as an obligation to adopt or modify laws alongside strengthening the capacities and powers of investigators, prosecutors and adjudicators. These efforts are thus focused on the police, the prosecution and the judiciary through enactment of specialised legislation and the formation of dedicated investigation and prosecution units, to ensure that such acts are met with appropriate punishments. 33 The effective and responsive investigation and punishment of offenders serve as a deterrent to potential offenders. An environment that allows offenders to get away with their crimes perpetuates individual recidivism and reinforces the normalisation of violence against women in society. There therefore is a correlation between the prevalence of violence against women and effective deterrence measures. Hence, thorough investigations and punishment of perpetrators are important measures of accountability and indicators of compliance with the due diligence standard. 34 However, the punishment should be in a manner consistent with the right to due process, also emphasising the dignity of survivors.

Reparation
The main objective of reparation is to address the harm or losses suffered by survivors and mitigate the effects of the violence to the extent possible, including compensation for physical and psychological injuries; the loss of employment; educational opportunities and other benefits; as well as any legal, medical and other costs incurred as a consequence of the violence. 35 States are also required to facilitate access to appropriate rehabilitation and support services. This aspect of due diligence is grossly underdeveloped and there is little information available regarding state obligations to provide adequate reparations to survivors of sexual violence. 36 into the Post-Election Violence (CIPEV) 40 documented incidences where mungiki used blunt objects such as broken glass to undertake forcible circumcision of Luo men, 41 and Kamau-Rutenberg 42 found that some of the survivors were as young as 11 and five years old. Male survivors also suffered sodomy, castrations and even mutilation of their penises. These sexual violations were meant to terrorise not only individuals but also targeted ethnic communities. 43 Therefore, it can be surmised that men and boys were targeted as an attack on the symbol of manhood for communities that do not practise circumcision to humiliate, traumatise and intimidate them both as individuals and as a community. 44 Such violations also fall within the purview of election-related sexual offences, which are the focus of this article.

Prevention interventions prior to the 2007-2008 postelection violence
CIPEV Chairperson, Justice Waki, notes that 'sexual violence is silent and preying because it is underreported, under-investigated and insufficiently addressed'. 45 Testimony adduced during the hearings revealed that women were the worst hit and constituted the most sexual violence victims. This was mainly attributed to the existing inequalities between men and women in Kenya. 46 Moreover, it is reported that women are traditionally perceived as lesser human beings which promotes sexual violence and insensitive police response to the crime when reported. 47 This creates fear in survivors to report violations in the belief that they could be turned away or even killed if they did so.
CIPEV found that 82 per cent of the victims did not formally report to the police, the reasons for non-reporting ranging from being attacked by the police; a fear of being attacked again; thinking that nothing would be done; not being able to identify the rapist; not knowing where to report the incident; being asked not to report the matter; and not knowing how to do so. 48 Moreover, a national population-based assessment of the 2007-2008 post-election violence revealed, among other things, that 68 per cent of Kenyans were unaware of the Sexual Offences Act; survivors lacked access to the formal justice system in many areas that substitute it with traditional court systems that do not recognise sexual violence as a crime; and that police had limited training in the documentation of sexual violence. 49 Despite all these existing shortcomings there was no evidence of official state engagement in interventions aimed at the prevention of sexual violence prior to the 2007 elections, as discussed below.

Lack of internal coordination and multi-sectoral approaches to prevent sexual violence
CIPEV faulted the government for their failure to anticipate sexual violence as a possible threat during the elections, and their delayed response once it began. 50 Furthermore, officers in the public administration and security agencies admitted being overwhelmed and their lack of coordination resulted in a breakdown of law and order, hence perpetuating the spread of violence. 51 Sexual violence is often higher during conflict because the corresponding breakdown in law and order creates both opportunities and incentives for perpetrators to engage in such criminal acts. 52 Thomas et al 53 observe that the suspension of the rule of law during times of transition is a potential cause of sexual and gender-based violence, and this proved to be the case during the post-election violence.
The Kenyan government failed to set up a national coordinating mechanism to promote peacebuilding and conflict management prior to the elections. Such a mechanism would have enabled it to coordinate conflict prevention efforts by both state and nonstate actors and put in place early warning systems that could have prevented or contained the violence. Its key objectives would have been to identify possible flashpoints for violence, to anticipate and mitigate potential threats, and to enhance the government's capacity to deal with moments of vulnerability through targeted responses and community-based approaches.
Moreover, the state's own organisational inconsistencies and internal politics prevented it from effectively mitigating and responding to incidences of sexual violence, as well as violence in general once they started to occur. An effective coordination mechanism would have enabled the state to identify security, social, political and cultural factors that could undermine the electoral process as well as potential causes of sexual violence. The identification of the risk factors would have enabled the state to come up with security counter-measures involving cooperation between different government law enforcement agencies. This would have enabled the government to reach out in good time to the security agencies for better inter-agency collaboration and coordination; to implement an enhanced security programme of training security agencies on electoral security; and to ensure that security issues affecting women's participation in the electoral process are monitored and addressed.
In addition, a major challenge to Kenyan government ministries, departments and agencies is a silo mentality which leads even those pursuing similar objectives, for instance, law enforcement agencies, 54 to work separately rather than together. This greatly affects their capacity to implement crime prevention measures by creating overlaps and a duplication of efforts. Moreover, most government bodies do not allocate adequate resources towards sexual violence prevention and response programmes. 55 This is attributable to the fact that the results of such programmes are not tangible and hence the benefits are not obvious to the electorate, thus making them unpopular with policy makers.

Challenges in implementing a specialised sexual violence law
CIPEV reported that, whereas Kenya had ratified several international and regional instruments prohibiting violence against women, it had not incorporated these into its laws or acted on those parts of the law that reflected them. 56 Moreover, the law comprehensively addressing sexual violence was only enacted the year prior to the elections, in 2006. Therefore, despite its existence it was established that many government officers handling sexual violence matters were ill-informed on its provisions and therefore were not able to provide either quality or standardised services to survivors of sexual violence. 57 Additionally, it was only in March 2007 that the Attorney-General appointed and launched the National Task Force on the Implementation of the Sexual Offences Act (TFSOA) to prepare and recommend a national policy framework and guidelines for the implementation and administration of the Act. 58 TFSOA was also mandated to propose effective measures to secure acceptable schemes, programmes and other mechanisms for the protection, treatment and care of sexual violence victims; as well as the treatment, supervision and rehabilitation of sexual offenders.
Therefore, although a specialised sexual violence law was in place, the lack of a national policy framework and guidelines for its implementation and administration greatly hindered its effectiveness in addressing cases of sexual violence. This further contributed to the ineffective investigation, prosecution and adjudication of such cases, resulting in an injustice to survivors. Such injustice took many forms, including insufficient courts, police and prosecution officers, cases not being reported, and cases taking too long to be resolved. 59

Gaps in training of police officers to address cases of sexual violence
Going into the 2007 elections the Kenyan police were those mandated with the maintenance of law and order, the preservation of peace, the protection of life and property, the prevention and detection of crime, and the apprehension of offenders. 60  known) was still enthralled by its colonial past which emphasised the protection of the Crown 61 above its subjects. Murunga notes that in the post-independence period the police retained their historical position as functionaries of the authoritarian state. 62 In the 2007 elections they were subsequently deployed specifically for purposes of regime maintenance as a key cog in the ruling regime's Party of National Unity (PNU) vote manipulation strategy. 63 Consequently, they were ill prepared to maintain law and order in the ensuing mayhem and actually ended up being key perpetrators of atrocities against Kenyans. CIPEV heard testimony that in some instances police were the perpetrators of sexual violence 64 and were also identified as the sector most responsible for post-election violence deaths. 65 Overall CIPEV found that the police totally failed to 'engage in professional, timely and quality planning and preparation for the elections event'. 66 Moreover, the police were not adequately trained on human rights and over time had proved to be the greatest violators of the rights of the citizens they were supposed to protect. More particularly, during the Kenyatta and Moi regimes they had formed the habit of abusing their wide discretionary powers to maintain law and order to control and brutalise citizens as a mode of regime maintenance. 67 Similarly, in the months leading up to the 2007 elections the police had been cited for extra-judicial killings of mungiki 68 members with the Kenya National Commission on Human Rights (KNCHR) having investigated and established police participation or complicity in the killings. 69 Correspondingly, CIPEV heard testimony that of the 24 rapes reported during the post-election violence seven were committed by police by way of gang rapes with officers being variously described as being drawn from the General Service Unit (GSU); the Kenyan police and the administration police. 70 Ultimately, the police were irredeemably deficient in their understanding and internalisation of international human rights standards and consequently were ill-prepared to identify and redress human rights violations, including sexual violence, during the post-election violence.

Protection measures taken by the state during the 2007-2008 post-election violence
The World Health Organisation (WHO) guidelines for medico-legal care for victims of sexual violence provide that the appropriate management of sexual violence requires a standardised clinical evaluation alongside an effective interface with law enforcement for the handling of forensic evidence. 71 Moreover, the effects of sexual violence cut across aspects of both public health and human rights as applied to the survivors with public health concerns being guided by legal procedures to be observed in the collection of forensic evidence from survivors to be used in the prosecution of the perpetrators. 72 Guedes et al 73 argue that gender-based violence should be viewed as both a public health and human rights problem, hence any response should entail government collaboration with the health services (public or private). 74 Such response would ensure sensitive and appropriate care for survivors respectful of their human rights and promoting their human dignity. Guedes et al 75 further note that such a human rights-based approach is essential since it gives health workers a clear set of principles for understanding the wider context of gender-based violence and the need to protect the survivors' dignity and rights.
All Kenyans are entitled to the highest attainable standard of health, which includes the right to health care services as provided for in article 43(1)(a) of the Constitution. This is also provided for in article 12 of ICESCR which Kenya has ratified and therefore is bound by it. The UN Committee on Economic, Social and Cultural Rights (ESCR Committee) which monitors compliance with ICESCR adopted a General Comment on the Right to Health in 2000. It defined four elements of the right to health, namely, availability, accessibility, acceptability and quality (AAAQ). 76 The Kenyan government must therefore ensure that it provides health services that meet these four key elements of the right to health. CIPEV 77 indicated that many victims of sexual violence could not get to hospitals to receive treatment owing to the lack of security and means of transport because of the breakdown in law and order. The lawlessness emboldened people to set up illegal roadblocks and engage in criminal acts such as sexual violence. Moreover, those who were able to get to hospital did not do so within the prescribed 72hour window for post-rape care services. This is the critical window period for survivors to receive post-exposure prophylaxis (PEP) to prevent HIV infection as well as for the collection and storage of medical evidence necessary for investigation and prosecution. As a result, some contracted HIV whereas those who were already infected could not obtain their daily doses of anti-retroviral drugs (ARVs) to manage the disease. Also, some victims fell pregnant because of the violations 78 while other harmful effects suffered included physical injury; psychological trauma; desertion by their spouses; unwanted pregnancies; and a loss of trust they previously had in state security agencies. 79 Furthermore, owing to the lawlessness many medical staff were unable to go to work or access their work stations. There were also not enough trained personnel to assist the survivors of sexual violence and drugs often were in short supply. 80 The problem was further compounded by a lack of a coordinated government policy on the provision of medico-legal services to survivors of sexual violence. CIPEV notes that the state was undecided on whether to keep systematic figures on casualties and incidences of sexual violence or to ignore it altogether. 81 Furthermore, the state did not communicate instructions to both hospitals and citizens that medical fees had been waived. As a result, some survivors did not seek medical attention because they thought they would have to pay when they could not afford to do so. A survey by the Centre for Rights Education and Awareness (CREAW) 83 found that in the 2007-2008 post-election violence female rape survivors were only able to access free medical treatment from private hospitals whereas they were charged a fee in public hospitals. Moreover, the number of women and girls seeking such treatment in public hospitals diminished in the two months subsequent to the outbreak of the post-election violence. 84 Health care providers interviewed by CREAW attributed this decrease to challenges encountered by women and girl survivors in accessing services while fleeing the violence and in the circumstances of internal displacement owing to the violence. 85 It was also observed that 25 per cent of health centres were temporarily closed as a result of staff shortages during the height of the post-election violence. The staff shortage was a direct result of the post-election violence since they too were caught up in flight and internal displacement because of the violence. From the foregoing it is evident that there was an inadequacy in the state response in terms of provision of medical treatment to survivors through the government hospitals because of cost and staff constraints.
Interestingly, Ossome 86 observes that even where women's organisations stepped in to ensure survivors access to free medical treatment, this did little to change the survivors' own negative perceptions of these organisations. They viewed them as exploitative and self-serving, which was a result of the high expectations formed by the survivors of long-term support and economic sustenance from the donor funds channelled through the women's organisations. The survivors had hoped that the funds would benefit them on an individual basis beyond the generalised access to free treatment. However, this was not possible since the 'rapid-response and urgent action' interventions undertaken in the immediate aftermath of the post-election violence focused on material provisioning and were time-bound, as well as being heavily influenced and enabled by the donor-funding objectives. They were restricted to emergency funding to provide for the needs of highly-affected populations, particularly the internally-displaced persons (IDPs). Neither state nor non-state actors had imagined the extent or form the violence and dispossession would assume, hence all were caught unprepared. Moreover, all forms of such material assistance ceased with the cessation of hostilities after the 2008 peace accord. 87 Consequently, these interventions left a larger vacuum of unfulfilled expectations and demands by the survivors, many of whom lamented the betrayal and abandonment by both the state and women's organisations that had stepped in to assist them. 88 These barriers to survivors' access to medical treatment and care during the 2007-2008 post-election violence is a pointer to the fact that the state did not meet the first two elements of the right to health, namely, availability and accessibility. Regarding the element of availability, this means that the state was unable to guarantee functioning public health and health care facilities, goods and services, as well as programmes in sufficient quantities. As for the element of accessibility, this means that the health facilities, goods and services were not accessible to everyone within the state. The lack of sufficient and proper equipment to cater for survivors means that the local health facilities available did not meet the fourth element of the right to health, namely, quality, which requires that the health facilities, goods and services must be scientifically and medically appropriate and of good quality.
Moreover, it should be noted that accessibility has four overlapping dimensions, namely, non-discrimination, physical accessibility, economical accessibility (affordability) and information accessibility, 89 none of which was met in the context of the 2007-2008 post-election violence. The failure to meet these first two elements automatically means that the state could not meet the subsequent elements of acceptability and quality. Therefore, it can be concluded that the state failed to meet all four elements (AAAQ) of the right to health in the context of the 2007-2008 post-election violence. Both state and non-state actors working on sexual violence should have cooperated to map out medico-legal service providers across the country. They ought to have thereafter assessed their institutional capacity to adequately support them both technically and financially to provide medico-legal services to survivors of sexual violence at the local level.
Specific attention ought to have been given to low-income communities living in informal settlements who constituted the main source of both perpetrators and survivors. 90 Communities needed to be empowered so that their youth could be engaged in gainful economic activity to avoid them joining the criminal gangs exploited by politicians to perpetrate acts of violence during the elections. On the other hand, women and girls, being the most vulnerable to sexual violence, also needed to be empowered so that they could use their economic might to improve their lives and those of their children. This would have helped them reduce their vulnerability since they would have been independent and able to seek justice and medical treatment without the need for intermediaries to foot the costs. An empowered community is also an informed community and community members therefore need to be empowered to be more knowledgeable on how to prevent and respond to cases of sexual violence when they occur. Armed with such knowledge they could have formulated community-based prevention and response mechanisms to ensure that members were protected from sexual violence and assisted to access medical treatment and justice when they became victims of sexual violence.

Punishment
The due diligence standard also requires governments to investigate sexual violence when it does occur, to punish and to hold to account perpetrators for their actions, and to adhere to the principles of accountability and non-impunity. 91 Thorough investigations and arrest, prosecution and punishment of perpetrators of sexual violence are important measures of accountability and indicators of compliance with the due diligence standard. 92 CIPEV established that the perpetrators of sexual violence during the 2007-2008 post-election violence included state security agents; members of organised gangs; neighbours; relatives; friends; and individuals working in the camps for internally-displaced persons. 93 The Commission further found that perpetrators committed the sexual violence crimes for different reasons, such as to pressure people to abandon their homes; as retaliation against those perceived to have voted for the wrong candidate; and to dominate, humiliate and degrade those belonging to the 'wrong' tribe. 94 Survivors' recollections of insults directed at them and the language of their attackers led many to believe that they had been deliberately targeted due to their party affiliations, political loyalties and ethnicity. 95 However, in other areas criminals simply took advantage 91 ACLU Women's Rights Project et al (n 17). 92 As above. 93 Government of Kenya (n 12) 252. 94 As above. 95 Ossome (n 15) 7. of the lawlessness and power vacuum to engage in sexual violence as a purely criminal act. 96 Furthermore, in the IDP camps victims were preyed upon by individuals from neighbouring communities, security personnel and humanitarian workers. 97

Shortcomings in the investigation of sexual offences committed during the post-election violence
Witnesses at the CIPEV hearings singled out members of the security agencies as perpetrators. They reported that security officers would enter their houses under the pretext of looking for weapons used by the young men barricading the roads or for members of mungiki and then sexually violate them when they could not find what they were looking for. 98 Aside from such acts of commission, the security officers were also cited for acts of omission such as refusing to record cases of sexual violence when they were reported to them. Additionally, it was reported that some of the security agents were ethnically biased and hostile towards survivors who did not belong to their own ethnic groups or political parties. 99 CIPEV hence concluded that the involvement of the security agents in perpetrating sexual violence and fear of self-incrimination are what led the police to omit sexual violence data on the reports they presented to the Commission. Such omission amounts to injustice to the victims since it is the police that were mandated to detect and investigate crimes before subsequently arresting the perpetrators so that they may be prosecuted and, ultimately, punished in a court of law. 100 The failure to act on reports of sexual violence presented to them meant that such offences could not be investigated and prosecuted with the definitive objective of punishing the offenders, thereby denying justice to the victims.
Nevertheless, the then Commissioner of Police, Major General Ali, was emphatic that only the police had the authoritative figures. 101 Moreover, CIPEV expressed disappointment that Major General Ali testified that he could not assess whether sexual violence incidents were fit to be included in the reports and was not aware of anyone who had been arrested and charged for the commission of such crimes. 102 Consequently, despite implications of state security officers as perpetrators there was no pressure on the police to either investigate or act against officers who were perpetrators of sexual violence in the 2007-2008 post-election violence. An investigation by a task force formed at the instigation of the then first lady, Lucy Kibaki, saw the police investigate 66 complaints of rape by security agents. However, in all the cases the police recommended closure for lack of evidence and when these files were submitted to the Director of Public Prosecutions (DPP) they were returned to the police for further investigation and this marked the end of the matter. The task force never submitted a final report. 103 CIPEV also found numerous shortcomings on the part of the police that hindered reporting and the investigation of sexual violence. Among these shortcomings was the fact that no complaints of rape were recorded by the police during the period and no internal investigations into police conduct was ever undertaken. 104 This blatant failure to investigate occurred despite a robust and facilitative legal framework. At the time the police held the sole mandate for the prevention and detection of crime, as well as the apprehension of offenders and the enforcement of all laws and regulations as provided for in section 14 of the Police Act. 105 The failure to investigate was an abdication of this mandate. Similarly, a specialised sexual violence law was in place which, albeit new, had already come into force and could be used in the aftermath of the post-election violence to pursue justice for the victims of sexual violence. This is the Sexual Offences Act 106 which provides for sexual offences, their definition, prevention, and the protection of all persons from such offences, and prescribes punishment for perpetrators. However, as discussed earlier, its major challenge was the lack of a national policy framework and guidelines for its implementation and administration. Aura notes that the enactment of the Act was not matched with adequate training and dissemination to law enforcement officers and relevant criminal justice actors as well as the public. 107 As a result, most were not even aware of its existence. justice sector. Article 48 of the Constitution specifically mandates the state to ensure access to justice for all persons, yet apart from one murder investigation many months after the post-election violence, and despite the many failures of the police to investigate, there were no internal investigations into the conduct of culpable police officers. 108 The International Centre for Transitional Justice (ICTJ) 109 also observed that despite the vast documentation on sexual and gender-based violence during the 2007-2008 post-election violence, no one has ever been convicted.

Resultant challenges in prosecution and securing convictions in court
Ambani 110 notes that failures by the state investigative agencies subsequently led to very few arrests, prosecutions and convictions of perpetrators of sexual violence in the 2007-2008 post-election violence. Moreover, Asaala 111 observes that the post-election violence also highlighted prosecutorial weaknesses and judicial shortfalls since of the few cases the police prosecuted, most of them ended up in acquittals with only six convictions. She attributes this to the following factors: poor investigations, police corruption and incompetence, a lack of local ownership, and a lack of political goodwill. 112 Similarly, CIPEV found the overall challenges in securing accountability for post-election violence to be a lack of political will, a lack of credible, independent and timely investigations, political interference in investigations, and other failings within the criminal justice system. 113 For his part the Director of Public Prosecutions, Keriako Tobiko, faulted the files forwarded to him for not being prosecutable owing to a lack of evidence. 114 This was occasioned by the lack of proficient investigations by the police, if any; victims reporting many days after violations hence hindering the collection of forensic evidence; and the failure of victims to identify perpetrators as well as forgetting the exact dates when they were violated. 115 Tobiko further acknowledged that the prosecution of sexual offences was severely hampered by poor investigations owing to a lack of equipment to collect and preserve evidence. He also stated that prosecutors lacked appropriate training and skills to prosecute such cases. 116 Consequently, from 2007 to the present, despite several attempts to review cases of sexual violence during the post-election violence, the Director of Public Prosecutions is yet to achieve the successful conviction of perpetrators. 117 Moreover, there has been no discernible effort to secure accountability for those affected during that time through the criminal justice process.

Consequences of the systemic failures in investigation and prosecution
Given the foregoing, it is evident that Kenya did not fulfil the punishment component of the due diligence standard. Investigatory and prosecutorial agencies should be aware of laws, regulations, avenues, measures and considerations that have been put in place to overcome the above-mentioned challenges, to deliver justice to victims of such violations. For the internalisation of human rights standards, particularly in the police, continuous engagement is required. This should have been done alongside continuously sensitising communities so that they know about sexual violence in general and what to expect from the investigative, prosecutorial and judicial agencies in terms of resolving such cases. This multi-sectoral approach involving engagement and capacitation of duty bearers, on the one hand, and engagement and sensitisation of communities, on the other, contributes towards effectively combating sexual violence both in and outside the context of elections. CIPEV recommended the training of the police in handling cases of sexual violence but noted that this would only be of use if accountability mechanisms within the security agencies are also established and acted upon. Additionally, the state must always strive to punish the perpetrators through prosecution and conviction in accordance with existing laws 119 to achieve justice for the survivors. Weru 120 found that calls for peace by government in the aftermath of the 2007-2008 postelection violence were viewed with mistrust as it was perceived to be either unwilling or incapable of responding to the violence since government efforts did not go far enough in punishing the perpetrators of violence, hence leaving an unresolved issue in the violence affected areas. This had the negative effect of concealing tensions the cumulative nature of which built up over time resulting in a prolonged state of potential and unpredictable political violence that has the potential to explode at subsequent elections. The overall question therefore remains as to the commitment of relevant agencies and officers to genuinely investigate and prosecute acts of sexual violence.

Reparation
The Truth, Justice and Reconciliation Commission (TJRC) was established in 2008 to investigate historical injustices and gross violations of human rights since independence . 121 It was a non-judicial body and did not have the power to prosecute but could recommend prosecutions, reparations for victims, institutional changes and amnesty for perpetrators who confessed, were willing to testify and had not committed gross human rights violations. Even though it could not pronounce itself on criminal accountability for perpetrators of sexual violence in the 2007-2008 post-election violence, the TJRC documented crucial findings, confirming the findings of CIPEV that numerous cases of sexual violence occurred during the 2007-2008 post-election violence. 122 Importantly, the TJRC corroborated the CIPEV finding that state security agencies, specifically the police and the army, were the main perpetrators of sexual violence alongside other human rights violations, such as massacres, enforced disappearances, and torture and ill-treatment. 123 Moreover, the TJRC made comprehensive recommendations regarding the establishment of a responsive reparations framework, and the provision of appropriate reparations to victims of gross human rights violations; including victims of sexual violence. 124 Additionally, the TJRC recommended further investigations into the conduct of state security agents, politicians and other individuals who were responsible for gross human rights violations in order to ensure accountability for these crimes. The TJRC also recommended the establishment of the Office of the Special Rapporteur on Sexual Violence. 125 This office was meant to play a key monitoring and evaluation role in the implementation of legislative, policy, administrative, policy, educational and other measures towards addressing sexual violence in Kenya.
The TJRC recommendations were never implemented by the state and the report served only as an archive on historical injustices in Kenya. Reparation, not only for victims of sexual violence in the 2007-2008 post-election violence but also for victims of gross human rights violations since independence is an issue that to date remains unaddressed. 126 Kenya therefore did not fulfil this component of the due diligence standard and this is attributable largely to the lack of political goodwill stemming from the aftermath of the 2007-2008 post-election violence.

Constitutional Petition 112 of 2013
The case was filed by eight post-election sexual violence survivors, six women and two men, who primarily sought to hold the government accountable for its failure to prevent or mitigate the post-election violence, investigate and prosecute the offenders, and provide redress to survivors. It was their contention, among other things, that the failure to prepare adequate and lawful police responses to the anticipated post-election violence contributed to sexual and genderbased violence and that the failure to prosecute the perpetrators or provide effective remedies to victims violated their fundamental rights. These fundamental rights were guaranteed under the former Constitution and included the right to life, the right to security of the person, the right to protection of the law and the right to remedy and rehabilitation. 127 The case was filed in February 2013 and concluded in December 2020, seven years after it had been filed and 13 years after the postelection violence during which the sexual violence was suffered. This in itself was a case of justice delayed. Fortunately, the Court ultimately ruled in favour of the petitioners and issued, among others, a declaratory order to the effect that the failure to conduct independent and effective investigations and prosecutions of sexual and gender-based violence-related crimes during the post-election violence was a violation of the positive obligation on the Kenyan state to investigate and prosecute violations of the right to life, the prohibition of torture, inhuman and degrading treatment, and security of the person. 128 The Court also issued a declaratory order that the petitioners' right to life, the prohibition of torture, inhuman and degrading treatment, the right to security of the person, the right to protection of the law, the right to equality and freedom from discrimination and the right to a remedy were violated during the 2007-2008 post-election violence, as a result of the failure of the government to protect these rights. 129 This case is ground-breaking since it establishes a precedent whereby the state was found liable for a violation of its international and national human rights obligations towards its citizens with specific regard to the investigation and prosecution of sexual and gender-based violence-related crimes. It therefore holds the government to account for all the gaps and shortcomings in responding to incidences of sexual violence that occurred during the 2007-2008 post-election violence. Moreover, in terms of reparations the Court awarded four of the petitioners general damages of Kshs 4 million each for the violation of their constitutional rights. In future, other survivors can use this case to hold the government liable when it fails to investigate and prosecute sexual violence in any context, even outside the election cycle.

Conclusion
The due diligence standard as elaborated by the UN Special Rapporteur on Violence against Women is useful in holding states accountable when they fail to fulfil their obligations to protect victims and punish perpetrators of sexual violence. Utilising the due diligence standard to interrogate state accountability for sexual violence has the potential of ensuring the full implementation of generalised obligations of prevention and compensation, alongside the effective realisation of existing objectives to protect and punish. 130 When evaluated against the due diligence standard as adopted for sexual violence cases, it is evident that Kenya did not meet its obligations to the 2007-2008 sexual violence victims and had shortcomings in fulfilling all four components of prevention, protection, punishment and reparations, as discussed above. To redress these shortcomings both state and non-state actors can consider some of the recommendations discussed hereafter.

Develop, implement, and monitor joint prevention mechanisms
Both state and non-state actors working on sexual violence at both the national and county level should collaborate in mapping prevention and response actors at both levels of government. They should then assess their institutional capacity to adequately support them both technically and financially to develop, implement and monitor prevention and response mechanisms ahead of any election cycle. These efforts can utilise the line ministry responsible for gender affairs as a secretariat since government ministries tend to have officers working in all levels of the administrative units and their officers can hence be effective and permanently present coordinators of sexual and gender-based violence cluster working groups in these units. Direct financial support can be given to non-state actors whereas both national and county governments can be supported to prioritise budgeting and the establishment of prevention and response mechanisms that can serve Kenyans even during the nonelection period.

Utilise existing community-based structures in developing effective protection mechanisms
Both the perpetrator and the survivor come from within the same community, especially in the context of election-related sexual violence. Protection mechanisms should therefore be geared towards ensuring that survivors are supported at the community level to report violations and access medical treatment. It would hence be beneficial to involve the communities in the planning, implementation, monitoring and evaluation of all sexual violence protection mechanisms. Involvement should go beyond having community representatives on planning committees dominated by government and civil society officials.
Counties can use a model such as the public participation framework currently utilised for planning, implementation, monitoring and evaluation of county development programmes. This is specifically provided for in article 10(2)(a) and the Fourth Schedule Part 2(14) of the Constitution and is stipulated as a function of the county government.
This would ensure that participation in these forums is purely voluntary with no benefit whatsoever accruing to members because of their engagement, hence eliminating rent seekers and coming up with a truly and robustly community-based protection mechanism specifically tailored to the needs of the community.

Training, internalisation and implementation of international human rights standards by law enforcement officers
The capacity building of police, prosecutors and magistrates using a multi-sectoral approach (joint trainings) would be best placed in terms of having them train on, internalise and implement international human rights standards. This can be done through support to existing training institutions in this sector such as the Judicial Training Institute (JTI). Partners can offer both technical and financial support to the JTI to develop a curriculum for such a joint training specific to gender-based violence and undertake to train on the same to the various actors in the criminal justice system.
For the internalisation of human rights standards, particularly in the police, continuous engagement is required. This can be done alongside continuously sensitising communities so that they know about sexual violence in general and what to expect from the investigative, prosecutorial and judicial agencies in terms of resolving such cases. This multi-sectoral approach involving engagement and capacitation of duty bearers, on the one hand, and engagement and sensitisation of the communities, on the other, contributes towards effectively combating sexual violence both in and outside the context of elections. The UN Special Rapporteur on Violence against Women argues that whereas international human rights law provides the guiding principles for state action, there should be alternative discussions and innovative strategies to eliminate sexual violence against women. 131 Therefore, both state and non-state actors must utilise a multiplicity of approaches in their interventions to prevent, protect, prosecute and compensate survivors of sexual violence.