As part of the road map for criminal accountability for the post-election violence in Kenya, parties which were involved in the violence agreed categorically that the ICC’s intervention would be invoked if the agreed domestic judicial mechanisms failed. This agreement would appear to be evidence of a strong determination to break the “culture of impunity” which had become entrenched in Kenya as far as gross violations of human rights are concerned. However, when the domestic mechanisms actually failed and the ICC intervened, both legal and political “battles” ensued. This chapter addresses the legal issues relating to or arising from the Kenya situation before the ICC, covering four main areas. Firstly, it clarifies and examines the trigger mechanism employed; jurisdictional issues and the scope of the charges and ICC investigations. Secondly, it identifies and analyses contentious legal issues arising in relation to the definition of crimes against humanity and the principle of complementarity. Thirdly, it evaluates Kenya’s legal responses as well as political and diplomatic strategies and reactions to the ICC’s intervention. Fourthly, it examines the future of the ICC process in Kenya in light of Kenya’s 2013 presidential election results and other specific developments at domestic level.

6.1 Introductory Remarks

At the end of 2013, the second phase of proceedings, the trial stage, of one of the two cases derived from the Kenya situation¹ had just commenced at the ICC.² But the ICC process with respect to Kenya had commenced officially since 31 March 2010, the date on which the ICC Prosecutor was authorized by the Pre-Trial Chamber to commence an investigation.³ Following such an investigation, on 8 March 2011, the ICC’s Pre-Trial Chamber issued two summonses to appear against six suspects, all of whom Kenyan citizens. The first summons was issued against William Samoei Ruto (then suspended Higher Education Minister), Henry Kiprono Kosgey (then Chairman of the ODM party and former Minister for Industrialisation) and Joshua Arap Sang (a journalist who was then KASS FM radio executive).⁴ The second summons was issued against Francis Kirimi Muthaura (then Head of the Public Service, Secretary to the Cabinet and Chairman of the National Security Advisory Committee), Uhuru Muigai Kenyatta (then Deputy Prime Minister doubling as Finance Minister) and Major General Mohammed Hussein Ali (former police chief).⁵

The Pre-Trial Chamber conducted confirmation of charges proceedings from 1 to 8 September 2011 for the first case and from 21 September to 5 October 2011 for the second case. On 23 January 2012, the Chamber issued a consolidated decision confirming the charges against four of the six accused persons. Charges were confirmed against Ruto and Sang in the first case and against Kenyatta and Muthaura in the second case.⁶ These four accused persons were committed to the Trial Chamber for trial that was originally scheduled to commence in April 2013, but was later postponed to September 2013.⁷ However, before the trial commenced, the Prosecutor dropped all the charges against Muthaura, thereby leaving Kenyatta as the only accused person in the second case.⁸ In addition, while the trial of Ruto and Sang commenced in September 2013 as scheduled, that of Kenyatta was rescheduled to 7 October 2014, but was later postponed indefinitely in view of serious challenges encountered by the Prosecution.⁹

6.2 Issues Relating to Trigger of Jurisdiction

6.2.1 Proprio Motu Investigation

The ICC Statute provides for three modalities, commonly referred to as “trigger mechanisms”, through which the jurisdiction of the ICC can be activated. The first mechanism is where a State Party to the ICC Statute refers a situation in its own territory or in that of another State Party’s territory to the Court.¹⁰ The second mechanism is where a situation in a State Party or non-State Party to the Statute is referred to the ICC by the UN Security Council.¹¹ The third mechanism is where ICC’s jurisdiction is triggered by the Prosecutor proprio motu (i.e. acting in his or her own initiative).¹² Since the third trigger mechanism, namely proprio motu referral, was used in respect of Kenya, it is discussed more fully below.

A proprio motu investigation is regulated under Articles 15 and 53 of the ICC Statute, read together with Rules 46–50 of ICC’s Rules of Procedure and Evidence.¹³ It operates as follows: Different sources such as individuals, NGOs, human rights commissions, etc., which may have knowledge or belief that a crime within the jurisdiction of the ICC is being or has been committed, can send “information” (also known as a “communication”) to the ICC Prosecutor.¹⁴ The Prosecutor then assesses this information to decide whether or not it presents a reasonable basis to proceed with an investigation. If it does, the Prosecutor must request the Pre-Trial Chamber to authorize commencement of an investigation into that situation.¹⁵ If the Chamber is satisfied that in fact there is a “reasonable basis to proceed”¹⁶ with an investigation, it will grant Prosecutor’s request.¹⁷ Out of such general investigations the Prosecutor identifies individuals to be charged for the alleged crimes.

6.2.2 The Waki Commission on Trigger Mechanism

As shown earlier (see supra Sect. 3.4.4), the Waki Commission recommended that the ICC’s intervention should be invoked only if Kenya failed to create and put to operation the proposed special tribunal. Even though the inquiry by the Waki Commission was carried out pursuant to a proposal that originated from an externally brokered mediation process, the managers of that process did not deprive Kenya totally of its control over matters pertaining to criminal accountability, specifically those which would later lead to the trigger of the ICC jurisdiction. Kenya’s control in this regard can be seen in least two ways, both of which linking to its Parliament.

First, the inquiry into Kenya’s post-election violence was, for the most part, a Kenyan national process. The reason is that the Waki Commission, which carried out this particular inquiry, was established pursuant to a national piece of legislation, the Commissions of Enquiries Act.¹⁸ Second, the recommendation regarding why, how and when the ICC should be asked to intervene was endorsed by Parliament on 27 January 2009 when it adopted the Waki Report.¹⁹ Thus, these two ways in which the Kenyan Parliament was associated with the process indicate clearly that Kenya was not merely a “consumer” of an externally imposed idea as far as the road map for criminal accountability is concerned, but rather that Kenya became a “co-owner” of that idea and the ensuing processes.

It is due to the foregoing that when the ICC Prosecutor was asked to intervene in Kenya, some of the commentators viewed this invitation more as a self-referral rather than a call for a proprio motu investigation. Others, like Professor Kai Ambos, observed that it was “not clear” how exactly to classify the referral of the Kenya situation.²⁰ The fact that Kenyan Parliament had voted for the “The Hague Option” may have given the impression that Kenya as State Party to the ICC Statute had decided (through the Parliament) to “take” itself to the ICC voluntarily. In addition, the fact that hitherto there is no prescribed format to which a state referral must conform supports the thinking that a self-referral could take any form which the referring state deems fit, such as, for example, a parliamentary resolution.²¹ Thus, those who opined that Kenya had made a self-referral must have drawn such inference, inter alia, from the domestic agreements and parliamentary deliberations which, as shown above, indicated the Parliament wanted the government to embrace the “The Hague option”.

However, the Office of the Prosecutor (OTP) treated the receipt of the envelope that contained suspects’ names and the accompanying evidence compiled by the Waki Commission not as constituting a referral but rather as a mere communication calling for a proprio motu investigation. The Waki Commission had made it clear (supra Sect. 3.4.4) that if circumstances so dictated, its evidence would be transmitted to the ICC Prosecutor for him to “analyze its seriousness” and decide if he should commence investigations and prosecution. From the wording of this recommendation, it is clear that the Commission contemplated making a “communication” and rather than a (self) referral. In any case, on account of Kenya’s foreign policy and practice regarding international treaties, it is inconceivable that the Waki Commission would have been able to make a “self-referral” on behalf of the Kenyan government.²² Thus, the Prosecutor rightly treated the transmission of the envelope and the accompanying materials as a communication just like 30 other communications he had received by November 2009 in respect of Kenya.²³ However, as the following sections reveal, the Prosecutor would have been glad if indeed Kenya had made a self-referral.

6.2.3 Responses of ICC Prosecutor and Kenyan Government to Trigger Mechanism

The Kenyan government and the OTP had different preferences as regards the trigger mechanism. This was among the fundamental issues which seemed to affect the interests of each side. On the one hand, the Prosecutor thought that a proprio motu investigation was not the most convenient trigger mechanism for the OTP. He therefore wanted to avoid it at any cost. On the other hand, although the Kenyan government had failed to institute domestic proceedings, it would not easily let the ICC process ensue. But when it eventually became evident that the ICC’s intervention was unstoppable, the Kenyan government battled with the dilemma whether it should make a self-referral or let the agreed proprio motu route take its course.

6.2.3.1 Prosecutor’s Preference for Self-referral

There is no doubt that the then ICC Prosecutor Moreno-Ocampo was aware that a proprio motu investigation was the route recommended in the mediation process and that it had been endorsed by the Principals in the Kenyan coalition government.²⁴ However, he embarked on a “strategy” to avoid the proprio motu route and lobby for a self-referral. For example, in a meeting held on 3 July 2009 between the Prosecutor and the Kenyan government, Ocampo “manipulated” the Kenyan government into signing an undertaking to make a self-referral.²⁵ Here, a question arises as to why the Prosecutor wanted to avoid the proprio motu route in favour of a self-referral? Two main factors must have informed Prosecutor’s preference. These are the prosecutorial policy of his Office and the negative perception about the work of the ICC in Africa.

6.2.3.1.1 ICC’s Prosecutorial Policy

The prosecutorial policy adopted by the OTP from early days clearly encourages self-referrals more than the other forms of referral apparently on account of the advantages the former offers. A self-referral of a situation may guarantee more cooperation from the referring state as regards the facilitation of investigations, arrest and surrender of suspects (if necessary), witness protection, etc.²⁶

This policy echoes the controversy that surrounded the idea of proprio motu investigations at the Rome Conference for the adoption of the ICC Statute. According to the travaux préparatoires, while many states strongly supported state referrals, the idea of vesting the Prosecutor with proprio motu powers was very contentious. On the one hand, a group of so-called “like-minded states” favoured a powerful and independent Prosecutor with absolute prosecutorial discretion. On the other hand, another group of states argued that “excessive” powers would be prone to abuse by the Prosecutor. The second group feared that such an abuse would curtail a widespread acceptance of the Statute, or impair the willingness of States Parties to cooperate with the Court.²⁷ Given this contention, a compromise had to be reached. Consequently, although proprio motu powers were eventually included in Article 15 of the Statute, the Prosecutor can only invoke such powers after having been authorized by the Pre-Trial Chamber.²⁸

Therefore, in order to minimize or avoid criticism and to maximize cooperation, rightly, the OTP, through the prosecutorial policy, rightly made proprio motu investigations the least of its priorities.

6.2.3.1.2 Negative Perception About the ICC in Africa

The second reason for the Prosecutor’s preference of a self-referral in relation to Kenya may have been the strained relationship between the ICC and Africa as a regional bloc. Negative perceptions had emerged in Africa about the ICC’s activities in the continent, especially after the issuing of warrants of arrest against President Omar Al-Bashir of Sudan. The ICC had already been accused by the African Union (AU) of being a biased institution, “hunting” or targeting only Africans, while perpetrators of similar crimes from Europe, America and allied countries are not targeted even where there is clear evidence against them.²⁹

A strong counter-argument is usually given in response to the foregoing perception. The argument is that in the instances where African states invite the ICC voluntarily, i.e. through self-referrals, to intervene into crimes committed in their territories, the claim that the ICC is targeting African states cannot be sustained.³⁰ The Prosecutor was aware that this specific argument would not apply to proprio motu investigations, hence his preference for a self referral from Kenya. On the one hand, a self-referral would have lent more political legitimacy to the ICC’s intervention, both at the domestic and international level, by creating the impression that it was Kenya itself which invited the ICC “voluntarily”. On the other hand, the Prosecutor seems to have feared that a proprio motu intervention would create more “antagonism” between the ICC on one side and the AU and Kenyan government on the other. The Prosecutor must also have feared that such antagonism would consequently impair Kenya’s voluntariness or readiness to cooperate with his Office. Given that Kenya was the first case in which proprio motu powers would be exercised, the Prosecutor had foreseen a mammoth challenge of having to justify why he only chose Kenya for proprio motu intervention and not any of the other similar situations which had been under preliminary investigations over a considerable period of time even prior to Kenya’s post-election violence.³¹

Thus, it was not a surprise that Ocampo tried to promote a self-referral from Kenya. In fact, this was not his first time to do so. In the first two situations before the ICC, namely those of Uganda and the DRC, the Prosecutor had been successful in “encouraging” the respective governments to make self-referrals, although he had also received a number of calls for the initiation of proprio motu investigations.³² He achieved his aim in the Uganda and the DRC situations despite the criticisms that his conduct elicited.³³

6.2.3.2 Why Did the Prosecutor Fail to Secure Self-Referral?

Despite his attempt, the Prosecutor could not secure a self-referral from Kenya for obvious reasons. It is noteworthy that in all the self-referrals made so far the primary motivation of the referring governments has been the crimes committed by rebel groups or government opponents.³⁴ However, the situation in Kenya was completely different: there were no rebels as such. Thus, although, as indicated above, the Kenyan government had agreed with the Prosecutor that it would make a self-referral if it failed to institute domestic proceedings, Kenya was not ready to honour this undertaking at the time it was expected to do so. The argument which emerged was that Kenya was unprepared to relinquish its sovereign right to investigate and try its own nationals.³⁵

It is true that no functional state would agree easily to waive its sovereignty over its own nationals through a simple agreement with another state or with a prosecutor of a foreign tribunal. The existence of several bilateral (impunity?) agreements between the USA and other states to ensure that no citizen of the USA will be arraigned before the ICC attests to this fact.³⁶ Notwithstanding the sovereignty argument, the Kenyan government’s reluctance to make a self-referral was influenced more by the prevailing circumstances in Kenya’s domestic politics at that particular time. The Kenyan government was, at that time, a coalition between two parties (see supra Sect. 3.3). Both parties to the coalition, the PNU and the ODM, had clearly been implicated in the crimes associated with the post-election violence. It would have been ironical to expect them to “invite” the ICC through a self-referral to investigate and prosecute the alleged crimes, given that up to the time such a self-referral was being sought, the leaders of the coalition government were not aware as to who would be arraigned before the ICC. This being the case, even on political grounds, a self-referral would have been counter-productive from Kenyan government’s point of view: It could have indeed been tantamount to inviting serious trouble to oneself.³⁷

6.3 Issues Relating to Parameters of ICC’s Investigation

Usually, the ICC Prosecutor’s investigation into any situation is confined to territorial, temporal and material parameters. The investigation authorized by the Pre-Trial Chamber was confined to the territory of Kenya, since the alleged crimes did not have cross-border character or effect.³⁸ There are no concerns or issues arising in this regard. However, with respect to the temporal and material parameters of the authorized investigation, a few specific legal issues that need more clarification and critical evaluation arise.

6.3.1 Temporal Scope of Investigation

6.3.1.1 Prosecutor’s Intention to Conduct Open-Ended Investigation

In the request for authorization of an investigation, the Prosecutor stated that the crimes which occurred in Kenya were “not limited to the time period between 27 December 2007 [and] 28 February 2008”.³⁹ However, in the remedy part of his request, the Prosecutor asked the Pre-Trial Chamber to authorize an investigation “in relation to the post-election violence of 2007–2008”.⁴⁰ In addition, the Prosecutor based his request on the findings of 11 “reliable” reports, most of which focused exclusively on the “alleged crimes committed during the post election violence”.⁴¹

The wording of the Prosecutor’s request, as shown above, entails two possible interpretations regarding the temporal scope of the investigation envisioned by the Prosecutor. Both interpretations lead to the conclusion that the Prosecutor envisioned an open-ended investigation that would stretch beyond the 2-month period of active violence.

6.3.1.1.1 First Interpretation of Prosecutor’s Request

One interpretation is that the Prosecutor intended not only to focus on the crimes committed during the post-election violence, but also on any criminal conduct incidental to the violence, namely atrocities which happened outside the 2-month time frame, but which nonetheless had a direct nexus with the 2007 general elections. Pursuant to this interpretation, Prosecutor’s proprio motu investigation would clearly have covered the planning, organization and incitement to commit violence which had started even before the election date. Crimes committed prior to the election date, such as the murder of 70 people and displacement of 2,000 others during the campaign period,⁴² would have been covered. Also, the investigation would have covered crimes committed after the official “ceasefire” agreement of 28 February 2008, but which crimes could have a retrospective nexus with the elections.

6.3.1.1.2 Second Interpretation of Prosecutor’s Request

The second interpretation of Prosecutor’s request begins from the premise that the main trigger of the ICC’s intervention in Kenya was the crimes associated with the 2007–2008 post-election violence. However, there could also have been other atrocities committed in Kenya which might not have had any clear nexus with the 2007 elections, but which could nevertheless be crimes falling under the jurisdiction of the ICC.

Certainly, the second interpretation is broader than the first. It would view the post-election violence only as a “door way” through which the ICC entered Kenya. It thus would have enabled the ICC to adopt a holistic approach to the fight against impunity in Kenya by also investigating other atrocities which had not been prosecuted domestically. As will be shown shortly (infra Sect. 6.3.3), the serious crimes alleged to have been committed in Mount Elgon district would have fallen under this scope.

6.3.1.2 Victims’ Views

A close look at the views of the victims reveals that they embraced both interpretations above. Most of the victims wanted a broadly defined temporal scope of investigations, stretching beyond the officially defined 2-month time frame for the post-election violence. Individual victims wanted redress for other past atrocities, irrespective of whether the particular atrocities had occurred in the context of the 2007 elections. For instance, victims variously wanted the investigations to span diverse periods, such as (i) 6 months before and after the elections; (ii) the campaigning period, during election and after election; (iii) from February 2007 “during the establishment and training of militia groups”; (iv) from 2006 to 2008, “as the violence was pre-planned” (v) from 2005 during the constitutional referendum “when the incitement started”; (vi) after February 2008, “because some crimes continue to be perpetrated to date”; (vii) since 1992 onwards; (viii) up to March and April 2008, etc.⁴³

6.3.1.3 Pre-Trial Chamber’s Decision

The Pre-Trial Chamber noted by a majority that it could not be established clearly from Prosecutor’s submission whether he wanted the investigation to be confined to the 2-month period of violence or to go beyond this time frame. However, considering the views of the victims, among other factors, the Chamber confined the temporal scope of the investigation to “the events that took place between 1 June 2005 (i.e., the date of the Statute’s entry into force for the Republic of Kenya) and 26 November 2009 (i.e., the date of the filing of the Prosecutor’s Request)”.⁴⁴ Furthermore, the Chamber was of the view that confining the investigation to the 2-month time frame would have been “inconsistent with (i) the purpose behind investigating an entire situation as opposed to subjectively selected crimes and (ii) the Prosecutor’s duty to establish the truth by extending the investigation to cover all facts and evidence pursuant to article 54(1) of the Statute”.⁴⁵

6.3.1.4 Evaluation

The decision to fix 26 November 2009 as the end date for the investigation was based on Pre-Trial Chamber’s understanding of Article 53(1)(a) of the ICC Statute. This Article requires that a proprio motu investigation can be authorized if the available information shows that a crime within the jurisdiction of the Court “has been or is being committed”. The Pre-Trial Chamber found that this particular provision prohibits authorization of an open-ended investigation. The Chamber stated:

It would be erroneous to leave open the temporal scope of this investigation to include events subsequent to the date of the Prosecutor’s Request. Article 53(1)(a) of the Statute, by referring to “a crime [which] has been or is being committed” makes clear that the authorization to investigate may only cover those crimes that have occurred up until the time of the filing of the Prosecutor’s Request.⁴⁶

6.3.1.4.1 Pre-Trial Chamber’s Interpretation Questioned

The Chamber’s interpretation of Article 53(1)(a) implied that whenever a proprio motu investigation is authorized, its temporal scope must always be sealed. It further appears that in setting an end date to the investigation, rather than leaving it completely open-ended, the Chamber wanted to achieve certainty as regards the scope of Prosecutor’s investigation. However, concerns have emerged about the practicability of this interpretation. What is particularly questioned is whether the contention that it is “erroneous” for the temporal scope of a proprio motu investigation “to include events subsequent to the date of the Prosecutor’s request” for authorization of investigation is the most correct and practicable interpretation of Article 53(1)(a).

Rastan opines that promotion of this interpretation will generally have negative implications on proprio motu investigations, contending that it will arguably affect all future Article 15 decisions. He further contends that such an interpretation will not be practicable in respect of all proprio motu investigations, more specifically those initiated in respect of ongoing crimes. According to him, this interpretation implies that whenever authorization for initiation of proprio motu investigations is requested with respect to a situation where the commission of crimes is still going on, several requests will have to be made serially or sequentially, i.e. each time a need arises to investigate events subsequent to the end date originally fixed, the Prosecutor must again submit a request for fresh authorization of a “new or extended investigation”. Rastan acknowledges that requesting for a new or extended investigation in this regard is possible under the Statute, but he rightly notes that it is “impracticable”, and that it entails “unnecessary restrictions” of the powers of the Prosecutor to investigate.⁴⁷

Interestingly, the Prosecutor did not bother to contend the Pre-Trial Chamber’s demarcation of an end date to the scope of the investigation authorized. This indifference seems to have stemmed from the fact that the demarcation of an end date per se would not have had any negative impact on Prosecution’s case in the two Kenyan cases. There is no express or implied indication that the Prosecutor had intended to extend his investigations even close to 26 November 2009 (the date of his request for authorization of investigation), as he did not include any incident beyond 28 February 2008 in the charges against the six suspects. As such, the Prosecutor may have chosen not to contend the demarcation of an end date not necessarily because he agreed to the Pre-Trial Chamber’s interpretation of Article 53(1)(a), but merely because such a demarcation did not have any limiting effect on the investigation of incidents or crimes he had resolved to investigate and charge. Thus, if the commission of the crimes had transcended the date of his request to commence an investigation, it is most likely that the Prosecutor would have contended the Chamber’s reasoning.

6.3.1.4.2 Is Pre-Trial Chamber’s Reasoning Convincing?

The phrase a crime “is being committed” in Article 53(1)(a) refers to ongoing crimes. It is not entirely convincing to argue, as the Pre-Trial Chamber did, that generally a proprio motu investigation should be authorized in respect of crimes committed subsequent to the date of Prosecutor’s request for authorization thereof. Such an argument is particularly problematic because the phrase “a crime is being committed” not only suggests that there could be continuity in the commission of the crime or crimes in question, but also that there could be a link between the crimes already committed, those being committed at the date of Prosecutor’s request and those which might be committed subsequent to that date. The Pre-Trial Chamber’s reasoning overlooks such possibilities and will definitely not be practicable in all scenarios.

Experience from the other situations before the ICC, especially those that have resulted from state self-referrals and Security Council referrals, reveals that the Prosecutor’s investigations have never been sealed. It is noteworthy that even in cases where no crimes continued to be committed after referral of the situation, the practice, and apparently, the interest of the OTP has always been to keep the investigations open-ended.⁴⁸ This has been Prosecutor’s discretion over which the Pre-Trial Chamber has not exercised control, for it does not have any. The question is whether there is any justification for such a control with regard to proprio motu referrals.

The negotiators of the ICC Statute were, for various reasons, overly cautious about an “all powerful prosecutor” to the extent that States Parties to the ICC Statute did not want to grant excessive powers to the Prosecutor with regard to proprio motu investigations, and that in order to control such powers, they agreed to subject the exercise of proprio motu powers to the judicial control of the Pre-Trial Chamber (see supra Sect. 6.2.3.1.1). So, one needs to find out whether by extension those drafters also wanted the Chamber to exercise strict judicial control with regard to temporal scope of investigations commenced proprio motu. An affirmative answer to this question would then justify the demarcation of an end date as per the Pre-Trial Chamber’s reasoning.

Neither the available commentaries on Articles 15 and 53 nor the travaux préparatoires suggest anything to that effect.⁴⁹ Instead, literature shows that the manner in which the States Parties wanted to limit the temporal scope of investigations (jurisdiction ratione temporis), regardless of the type of trigger mechanism used, was strictly with regard to clearly demarcating the date in respect of which the investigation commences. This is clear under Articles 11(2) and 12(3) of the ICC Statute. These provisions clearly designate the date of entry into force of the Statute for a particular State, or the date on which such State accepts the jurisdiction of the ICC voluntarily, as the case may be, to be the dates prior to which, strictly speaking, the Prosecutor cannot investigate. And in any case, the Prosecutor cannot investigate any crimes prior to the date of entry into force of the Statute, which is 1 July 2002.⁵⁰

Thus, Pre-Trial Chamber II’s reasoning that the ICC Statute requires that the date of Prosecutor’s request must always be set as an end date for a proprio motu investigation is not entirely correct. However, for the reasons started above, as flawed as it may be, this decision will not have any effect on the two cases so far derived from the Kenyan situation. In addition, this flawed reasoning may not necessarily “affect all future” proprio motu situations as Rastan contends. Practice shows that rather than relying on previous decisions mechanically, the various Chambers of the ICC do consider each situation before them on a case-by-case basis. And in any case, the ICC is not bound by precedent.⁵¹

In fact, the reasoning of the Pre-Trial Chamber II in the Kenyan cases in this very aspect was departed from shortly later by the Pre-Trial Chamber III in yet another request for proprio motu investigation, the Cote d’Ivoire situation. In the latter case, the Prosecutor, apparently having been influenced by the “precedent” in the Kenya situation, requested the Chamber to authorize an investigation into the Cote d’Ivoire situation “up to the date of filing of the request”. However, distinguishing the decision of Pre-Trial Chamber II in the Kenya situation, the Pre-Trial Chamber III declined to fix an end date for the investigation as requested by the Prosecutor for practical reasons. The Chamber underscored the importance of considering the context of each situation, rightly noting that in the context of the Cote d’Ivoire situation, it was of paramount importance to ensure that:

Any grant of authorisation covers investigations into “continuing crimes”—those whose commission extends past the date of the application. Thus, crimes that may be committed after the date of the Prosecutor’s application will be covered by any authorisation, insofar as the contextual elements of the continuing crimes are the same as for those committed prior to [the date of the filing of Prosecutors Request]. They must, at least in a broad sense, involve the same actors and have been committed within the context of either the same attacks (crimes against humanity) or the same conflict (war crimes). Therefore if the authorisation is granted, it will include the investigation of any ongoing and continuing crimes that may be committed after [the date of the filing of Prosecutor’s request] as part of the ongoing situation.⁵²

Therefore, as it stands now, there are two positions emanating from the Pre-Trial Chamber with regard to the demarcation of an end date for proprio motu investigations. One position was set in the Kenya situation and the other position in the Cote d’Ivoire situation. The position set in the Cote d’Ivoire situation is more convincing and practicable. However, it remains to be seen which position will be taken by the Appeals Chamber if a need arises, especially by determining what “an ongoing situation” precisely entails with reference to a proprio motu investigation.

6.3.2 Subject-Matter Jurisdiction

Subject-matter jurisdiction determines which crimes are to be investigated or prosecuted. The victims of Kenya’s post-election violence argued that all types of violations of their rights should be investigated.⁵³ On his part, the Prosecutor submitted that the available information revealed that only crimes against humanity had been committed. However, he stated that this submission was “without prejudice to other possible crimes within the jurisdiction of the Court which may be identified during the course of an investigation”.⁵⁴ It appears that by referring to “other possible crimes”, the Prosecutor did not want to completely rule out war crimes and genocide, which other sources claimed that had been committed or at least attempted in Kenya.

Eventually, the Pre-Trial Chamber authorized the Prosecutor to investigate only crimes against humanity.⁵⁵ This, however, did not per se preclude the Prosecutor from subsequently asking the Chamber to broaden the scope of the investigation to include the other crimes in case there are facts pointing to their commission.⁵⁶

6.3.3 Locating Crimes in Mount Elgon Area in the Investigation

There was another wave of violence in Kenya’s Mount Elgon district that existed almost independently of the post-election violence.⁵⁷ In view of the temporal and material scope of the investigation authorized by the Pre-Trial Chamber, the crimes associated with this particular violence could be and should have been addressed by the Prosecutor. Despite the fact that most of the crimes associated with this specific violence are not wholly part of the defined time frame of the post-election violence, they nevertheless fall within the temporal scope of the investigation authorized by the Chamber. In addition, the nature, magnitude and gravity of the crimes committed during this violence warrant its special consideration.

6.3.3.1 Origins of Violence in Mount Elgon

The violence started in 2006, being a reaction to a three-phase resettlement programme implemented by the Kenyan government in Mount Elgon district between 1968 and 2005. The Sabot, a pastoralist ethnic community residing in the area, opposed the fact that their land was being allocated to the people of other communities resettled in the area. In 2002, the Sabot Land Defence Force (hereafter “Sabot militia”) was formed with a view to defending the Saboti land. From 2006, the militia, resisting phase III of the resettlement programme, launched attacks targeting those who, among other things, opposed its cause.⁵⁸ Heinous crimes were allegedly committed both by the Sabot militia and the Kenyan security forces which, in 2008, carried out an operation, operation Okoa Maisha (Save Lives), a “deadly” military campaign against the militia.⁵⁹

6.3.3.2 Gravity and Nature of Resulting Crimes

It is alleged that the Sabot militia killed more than 600 civilians, displaced between 66,000 and 200,000 others, and that it forcibly recruited about 650 child soldiers, mostly boys. It is further alleged that its members abducted, raped, mutilated and tortured many victims; destroyed property; and looted livestock.⁶⁰ Similarly, it is alleged that during the operation Okoa Maisha, the Kenyan security forces committed serious atrocities, as they arbitrarily arrested almost all men and boys as young as 10 years old, and “screened” them in a nearby military camp. The screening exercise, which Human Rights Watch describes as “systematic torture”, resulted in serious bodily injuries and many deaths.⁶¹ It has further been alleged that by July 2008, the Kenyan security forces had caused about 220 enforced disappearances, besides raping women and burning houses and property belonging to the suspected members of the Sabot militia.⁶²

Both Human Rights Watch and the Kenya National Commission on Human Rights have concluded that both the Sabot militia and Kenyan security forces violated international human rights law and humanitarian law, specifically the Geneva Conventions of 1949.⁶³ It has even been claimed that from 2006 to 2008 the Sabot militia was “in effective control” of the Mount Elgon district, because “there was no government in that area”.⁶⁴ Further documentation suggests that, just like the Kenyan security forces, the structure of the Sabot militia entailed a defined hierarchical chain of command. Accordingly, it is has been argued that the doctrine of command responsibility could also apply in dealing with the crimes of the militia.⁶⁵

After the end of the operation Okoa Maisha, about 700 members of the Sabot militia were charged domestically under the Kenyan Geneva Conventions Act with the crime of promoting “war-like activities”.⁶⁶ However, the crimes alleged to have been committed by the Kenyan security forces were not prosecuted.⁶⁷ Prospects of their being investigated and prosecuted domestically remain very slim.

The allegation that war crimes were committed is serious, but cannot be adequately scrutinized here to establish if all the legal requirements for such crimes could actually be established. However, as the Kenyan Truth, Justice and Reconciliation Commission also concluded, the nature of the crimes committed indicates that there could be a reasonable basis to believe that crimes against humanity were indeed committed.⁶⁸

6.3.3.3 Interim Conclusion

Most of the atrocities in Mount Elgon were not committed in connection with Kenya’s 2007 elections. However, there is no doubt that these crimes clearly fall within the jurisdiction of the ICC given the broader temporal and material scope of investigations as authorized by the Pre-Trial Chamber into Kenya. It is noteworthy that, when authorizing the investigation, the Chamber referred to the violence in Mount Elgon by passing, but the crimes committed in this area are conspicuously missing in the charges subsequently brought before the ICC. This deliberate “omission” seems to have been influenced by the Prosecutor’s express statement that the primary reason for his proprio motu intervention in Kenya was to deter future crimes associated with election violence. ⁶⁹ Hence, the atrocities committed in Mount Elgon remains one of the areas in which the prosecutorial discretion failed to meet the legitimate expectations of victims of serious crimes in Kenya.

6.4 Issues Relating to Substantive Criminal Law

There are two main issues of substantive law arising from the pre-trial phase of the Kenya situation that merit critical analysis. The first issue relates to the characterization of the facts in the charges. The second issue relates to the contention arising from the definition of crimes against humanity.

6.4.1 Prosecutorial Discretion Vis-a-Vis Scope of Charges

The indictment in the two Kenyan cases encompassed a total of five individual acts charged as crimes against humanity contrary to Article 7 of the ICC Statute.⁷⁰ The acts charged are only those with a direct nexus to the post-election violence, namely murder; deportation or forcible transfer of population; persecution; rape and other forms of sexual violence; and other inhumane acts.⁷¹ However, a contention emerged during the pre-trial proceedings as to how property crimes should have been characterized in the indictment. The main issue was whether, considering the circumstances in which they were committed, such crimes should have been characterized as “persecution”; or as “deportation or forcible transfer of population”; or as “other inhumane acts”.

6.4.1.1 Characterization of Crimes Relating to Property

The charge, by either design or oversight, treated the property crimes of looting and destruction of property differently in the two cases.

In the case of Ruto, Kosgey and Sang, the Prosecution argued that these acts constituted the crime against humanity of “deportation or forcible transfer of population” contrary to Article 7(1)(d) of the ICC Statute. Accordingly, the Prosecutor alleged that these acts were committed by the “Network Perpetrators” (see infra Sect. 6.4.2.2.1.1) in order to “permanently expel PNU supporters from the Rift Valley”.⁷² However, in the second case, Muthaura, Kenyatta and Ali, the Prosecution argued that the looting and destruction of property allegedly committed by “the Mungiki Perpetrators” (see infra Sect. 6.4.2.2.1.2) constituted the crime of “other inhumane acts” under Article 7(1)(k) of the ICC Statute. Accordingly, the Prosecutor submitted that these acts and several others affected the “physical and mental health” of the victims.⁷³

6.4.1.1.1 Victims’ Arguments

A total of 327 victims indicated that they had suffered theft, looting or destruction of their property during the post-election violence. They raised a concern that the charges brought by the Prosecutor “did not indicate clearly and expressly” whether they would actually cover such property crimes. In particular, the common victims’ representative argued that in both cases, the Prosecution had characterized the facts in relation to property crimes erroneously. In her view, under the ICC Statute, property crimes committed in Kenya were neither “other inhumane acts” nor “deportation or forcible transfer of population” as characterized in the charges. Rather, the she submitted that these acts constituted a crime against humanity of “persecution” under Article 7(1)(h) of the ICC Statute. Thus, the victims asked the Pre-Trial Chamber to advise the Prosecutor to consider amending the charges accordingly, since the error would adversely affect their individual interests at the end of the trial.⁷⁴

6.4.1.1.2 Reasoning and Decision of Pre-Trial Chamber

In the case of Ruto, Kosgey and Sang, the Pre-Trial Chamber approved of the Prosecutor’s approach. It ruled that the acts of looting, theft and destruction of property were correctly characterized in the charging document as constituting “deportation or forcible transfer of population” under Article 7(1)(d). The crime of “deportation or forcible transfer of population” is defined as “forcible displacement of the persons concerned by expulsion or other coercive acts from the area in which they are lawfully present, without grounds permitted under international law”.⁷⁵ Pursuant to this definition, the Chamber endorsed the Prosecutor’s argument that the acts of burning and looting of property belonging to PNU supporters in Rift Valley could amount to “coercive acts”, because they appeared to have been committed with the intention to force such supporters to vacate the area. Accordingly, the charges were confirmed as framed.⁷⁶

However, in Muthaura, Kenyatta and Ali, the Pre-Trial Chamber found that there was a mistake in the manner in which the Prosecutor characterized the facts relating to property crimes. The Chamber rejected Prosecutor’s characterization of facts which had suggested that the looting and destruction of property belonging to the ODM supporters constituted “other inhumane acts”. In particular, the evidence produced by the Prosecutor was insufficient to persuade the Chamber that such conduct met the definitional requirements of “inhumane acts” under Article 7(1)(k) of the Statute. The Chamber noted that the crime of “other inhumane acts”, being a “catch-all” category of crimes against humanity, can only cover conduct which is not covered by the preceding paras a to j of Article 7(1). But for such conduct to qualify as “inhumane acts” under the provision, the acts must, inter alia, be capable of “causing great suffering or serious injury to body or mental or physical health”. Applying this standard, the Chamber rejected the Prosecution’s characterization, stating that while the Prosecution had succeeded in showing that such property crimes had actually occurred, it had failed to establish the mandatory link between these acts and “serious injury to body or mental health”.⁷⁷

Finally, the Chamber also rejected the victims’ views that the proceedings should be adjourned so that the Prosecutor could be requested to amend the charges accordingly. The Chamber noted that in characterizing the facts, however wrongly, the Prosecutor had exercised his prosecutorial discretion. Ultimately, the Chamber concluded that to request for an amendment “would mean to go beyond the factual ambit of the charges, and would therefore be tantamount to requesting the Prosecutor to consider adding a new charge”.⁷⁸

6.4.1.1.3 Evaluation

Certainly, the Prosecutor exercised his prosecutorial discretion in framing the charges the way he did with respect to property crimes committed during the post-election violence. However, the fact that property crimes were confirmed in the first case and rejected in the second case will potentially entail different implications for the victims in the two cases.

At the trial stage of a case before the ICC, the Trial Chamber can only adjudicate on charges that have been confirmed by the Pre-Trial Chamber at the end of the pre-trial stage. This is the reason why, for example, Article 61(9) of the ICC Statute provides that if, after confirmation of charges decision and before trial begins, the Prosecutor feels that there is a need to amend the confirmed charges by, for example, adding more charges or substituting more serious charges, a separate confirmation of charges hearing must be conducted in respect of the additional or substituted charges. It is only when these new charges are also confirmed that they can be adjudicated upon by the Trial Chamber, the reason being that during the trial stage, the Trial Chamber cannot “exceed the facts and circumstances described in the charges and amendment to the charges”.⁷⁹

In addition, the Regulations of the Court only allow the Trial Chamber to “change the legal characterisation of facts to accord with the crimes under Articles 6, 7 or 8”, but strictly speaking, this must “not exceed the facts and circumstances described in the charges and any amendments to the charges”.⁸⁰ In fact, the Appeals Chamber has stated clearly that such re-characterization “must not exceed the factual circumstances that were identified in the confirmation decision as supporting each of the legal elements of the crimes charged”.⁸¹

The implication of the foregoing paragraph is that in Muthaura, Kenyatta and Ali, the facts relating to acts of destruction and looting of property, which the Pre-Trial Chamber declined to confirm, can no longer be part of the charges to be adjudicated upon during trial. As a result the victims, who expressed their wish to seek remedy against such acts will no longer be able to claim individual reparation for these acts in case of a conviction. On the other hand, the victims of similar crimes in Ruto, Kosgey and Sang will have the property crimes adjudicated upon during trial, because these crimes will remain part of the charges. Thus, the victims in the second case will have the opportunity to claim and be granted reparations for the personal loss they incurred in this regard in line with the ICC reparation principles which have been adopted recently in the Lubanga case. According to these principles, “economic harm”, such as “loss of, or damage to, property”, can be compensated, provided the harm is sufficiently quantifiable.⁸²

The above-mentioned mistake regarding characterization of facts in the Muthaura, Kenyatta and Ali case could have been rectified, and thereby mitigate the “injustice” which could occur to the victims of the acts in question. Rather than outrightly declining to confirm the wrongly characterized facts, the Chamber should have, pursuant to Article 61(7) of the Statute, and as proposed by the victims’ representative, considered requesting the Prosecutor to re-characterize the facts correctly. Article 61(7) provides that during the confirmation of charges hearing, the Pre-Trial Chamber has three mandates. It can (i) confirm the charges wholly or partly; or (ii) decline to confirm all or part of the charges; or (iii) propose or advise an amendment of a charge which would not have otherwise been confirmed so that it can be confirmed. Particularly, on the third mandate, Article 61(7)(c)(ii) further provides that the Chamber “shall adjourn the hearing and request the Prosecutor to consider amending a charge because the evidence submitted appears to establish a different crime within the jurisdiction of the Court”. It appears that such amendment may also rectify a wrong characterization or labelling of facts. Moreover, the use of the word “shall” may, in this case, be taken to imply that an adjournment is mandatory. Accordingly, whenever the Pre-Trial Chamber notices that certain facts have been wrongly characterized in the charges and that on the basis of such a mistake the facts are unlikely to be confirmed unless they are re-characterized, then the Chamber should not simply decline to confirm such facts before advising the Prosecutor to consider amending the charge.

It is submitted that while the Pre-Trial Chamber’s reasoning is partly correct in that the Prosecutor had the discretion to frame the charges as he deemed fit, two aspects of its decision are questionable regarding what should have been the correct application of Article 61(7)(c)(ii) of the ICC Statute. Firstly, taken plainly, the wording of Article 61(7)(c)(ii), which allows for re-characterization of facts, retains the prosecutorial discretion as it provides that the Pre-Trial Chamber will only “request” the Prosecutor “to consider” amending a charge. This means that even if he had been requested to do so, the Prosecutor would not have been obligated to act as per the request of the Chamber.⁸³ Here, one must draw a distinction from the controversy that emerged from the Lubanga case in which the Pre-Trial Chamber “usurped” the prosecutorial powers and purported to re-classify the charges suo motu (in its own initiative) instead of requesting the Prosecutor to do so. This tendency was reversed by the Appeals Chamber.⁸⁴ Secondly, the count entailing property crimes was already part of the indictment, except that the facts had been wrongly characterized as constituting “other inhumane acts” instead of “deportation or forcible transfer of population” and “persecution”, which were also contained in the indictment.⁸⁵ Thus, contrary to what the Pre-Trial Chamber suggested, had an amendment to the charge been requested and effected, the result would not have been tantamount to adding a “new charge”.

Therefore, the failure to provide an opportunity for the amendment of the charges with respect to property crimes in Muthaura, Kenyatta and Ali case was an oversight on the part of the Pre-Trial Chamber. This is despite the fact that it still remains unclear if the Prosecutor would in fact have considered re-characterizing the facts had he been requested to do so. The Prosecutor himself did not take any initiative to have the facts re-characterized so that they could be adjudicated upon during trial.⁸⁶ This shows clearly that the Prosecutor became indifferent about the legitimate concerns of the victims. This indifference gives rise to the question whether, in view of the ICC’s unique procedure allowing victims’ representation, the OTP should always view issues of victims’ welfare as an entire responsibility of the victims’ representative, or whether such issues should also be viewed as an integral part (a concern) of the Prosecution’s case.

6.4.2 Whether Acts Committed During Post-Election Violence Amounted to Crimes Against Humanity

6.4.2.1 Introductory Note

Although the definitions of crimes against humanity under customary international law and under the ICC Statute have many similar elements, the two are not entirely the same. In view of this fact, even before the ICC intervened in Kenya, the Kenya National Commission on Human Rights had already contended that the crimes committed during the post-election violence might not qualify as crimes against humanity under the ICC Statute, although they obviously did qualify as such under international customary law.⁸⁷ This contention per se touches on a crucial question of substantive law which must be examined. The reason is that in view of Article 21 of the ICC Statute, the ICC can try only crimes whose elements meet the definitional criteria set out in its Statute. Indeed there are many overlaps between the ICC Statute and international customary criminal law that make the two not mutually exclusive. However, as Cassese notes, in some aspects, the ICC Statute is either broader or narrower than the customary law.⁸⁸ For that reason, in terms of substantive law, the ICC enforces international customary law only to the extent the latter is reflected in the ICC Statute.

One area in which the ICC Statute differs slightly from international customary law is in relation to the definition of crimes against humanity. In view of such difference, a disagreement emerged as to whether crimes against humanity under the ICC Statute occurred during the post-election violence in Kenya, or whether what happened could only qualify as crimes against humanity under international customary law. Legally speaking, therefore, the ICC would not have jurisdiction over the criminal acts committed in Kenya if such acts did not meet the criteria for crimes against humanity under the ICC Statute.

The said disagreement gained prominence when the judges who sat in the pre-trial proceedings of the Kenya situation expressed divergent opinions on the matter. The disagreement permeated the entire pre-trial phase, clearly manifesting itself in three important Pre-Trial Chamber’s decisions. It emerged for the first time in the decision on the authorization of an investigation. It then recurred in the decisions relating to both issuance of summonses to appear and the confirmation of charges. The main bone of contention was on the contextual requirements of crimes against humanity as provided for in the ICC Statute.

Before outlining and evaluating the disagreement at length, the following section will first outline the definition of crimes against humanity with a view to giving a clear perspective to the subsequent discussion.

6.4.2.2 Outline of the Definition of Crimes Against Humanity

The definition of crimes against humanity in the Statute constitutes the material elements of the crime, namely any of the 11 categories of individual acts (actus reus) listed under Article 7(1)a–k ⁸⁹ of the Statute “when committed as part of a widespread or systematic attack directed against any civilian population, with knowledge of the attack”.⁹⁰ Furthermore, “an attack directed against a civilian population” entails “a course of conduct involving the multiple commission of acts referred to in [Article 7(1)] against any civilian population, pursuant to or in furtherance of a State or Organizational policy to commit such attack”.⁹¹

Put more precisely, the above definition contains four main cumulative criteria, also known as the “contextual elements”, under which the material elements (individual acts) constituting the crime must occur if the individual acts are to qualify as “crimes against humanity” under the Statute. These elements are that: (i) the acts must be part of a widespread or systematic attack⁹²; (ii) the attack must be directed against any civilian population⁹³; (iii) the perpetrator must have the knowledge of the attack⁹⁴; and (iv) the attack must be pursuant to or in furtherance of a policy of a State or an organization.

6.4.2.3 “State or Organizational Policy”: A Source of Disagreement

There was no any disagreement among the judges of the Pre-Trial Chamber as regards the existence of the first three contextual elements in the definition of crimes against humanity identified above in relation to the criminal acts committed during the post-election violence in Kenya. However, the question whether the fourth element, namely a “State or organizational policy”, was present, created a sharp division not only among the judges, but also between the parties to the cases as well as among scholars. On the one hand, a two-judge majority of the Pre-Trial Chamber, constituted by Judges Ekaterina Trendafilova and Cuno Tarfusser, maintained that the acts could qualify as crimes against humanity under the Statute, for they satisfied all the contextual elements. On the other hand, Judge Hans-Peter Kaul maintained that the acts would not qualify as such. He expressed a “fundamental disagreement” with the majority, arguing that the acts were not committed pursuant to or in furtherance of a policy of State or an organization.⁹⁵ Hence, while the majority affirmed that the ICC had jurisdiction ratione materiae over the Kenya situation in general and the two cases in particular, the minority’s view was that the ICC lacked such jurisdiction under the ICC Statute.⁹⁶

6.4.2.3.1 Prosecutor’s Submission

In each of the two cases, the Prosecution maintained that the crimes had been committed pursuant to an “organizational policy”. There was no any explicit allegation that there was a State policy per se to commit crimes, although the Prosecution tried to link individual officials in the government and State agencies, such as the police, to the organizational policy of private entities such as the Mungiki.⁹⁷ Building its case on organizational policy, the Prosecution argued as follows:

6.4.2.3.1.1 The Network as an “Organization”

In the Ruto, Kosgey and Sang case, the Prosecution submitted that the alleged crimes were committed by “the Network”, an “organization” which was allegedly in support of the Orange Democratic Movement (ODM) party. This “organization” was allegedly linked to the Kalenjins, having been created already in December 2006.⁹⁸ Furthermore, it was submitted that the Network had five identifiable components, namely political, media, financial, tribal and military components, which acted interdependently during the post-election violence to implement a policy. The objective of the policy under implementation was allegedly twofold: (i) “to punish and expel from the Rift Valley” the civilians from the ethnic groups perceived to be supporters of the Party of National Unity (PNU); and (ii) “to gain power and create a uniform ODM voting bloc”.⁹⁹

As to the structure of the Network, the Prosecution alleged that William Ruto and Henry Kosgey were not only the organization’s top political leaders, but also acted as its main sponsors and mobilizers. Joshua Arap Sang allegedly headed the media component of the Network, and, as such, he was responsible for spreading propaganda, hate speech and incitement through the Kalenjin radio station KASS FM.¹⁰⁰ It was further alleged that the military component of the Network had a defined chain of command, William Ruto being its ultimate supreme commander. Reporting directly to the supreme commander were allegedly three commanders (generals) in charge of one of the three different “military zones” created in Rift Valley. Each zonal commander was allegedly in charge of subordinates, mainly tribal and local leaders. It was these tribal and local leaders who allegedly mobilized the direct perpetrators (mainly youths), facilitated training, provided weapons and guidance on how the attacks should be implemented.¹⁰¹

6.4.2.3.1.2 The Mungiki as an “Organization”

In the Muthaura, Kenyatta and Ali case, the Prosecution argued that the crimes were committed under the auspices of the Mungiki, an organization that was acting in support of the PNU. It was submitted that this organization implemented an “organizational policy”, namely “to keep the PNU in power through every means necessary, including by orchestrating a police failure to prevent the commission of crimes”. The policy was allegedly implemented through a common plan to attack ODM supporters “by (i) penalizing them through retaliatory attacks; and (ii) deliberately failing to take action to prevent or stop the retaliatory attacks”.¹⁰²

In terms of its organizational structure, the Prosecutor alleged that the Mungiki was hierarchically organized, with a known national leadership over which Mr. Kenyatta “had control … due to his wealth and privileged background”.¹⁰³ It was further alleged that the Mungiki had local and regional branches whose leadership had executive and judicial powers, and that it also had a political wing called the “Kenya National Youth Alliance” and a quasi-military wing known as the “Mungiki Defence Council”.¹⁰⁴

It was the prosecution’s case that the logistical, material and moral support from the six suspects in each respective case, facilitated by the hierarchical structures of the two entities, the Mungiki and the Network “organizations” acquired the capacity to organize and implement attacks. The prosecution concluded that these organizations actually utilized that capacity and launched attacks on civilian populations pursuant to and in furtherance of their respective policies.

6.4.2.3.2 Defence Submission: There Were No “Organizations”

The defence teams persistently refuted the submissions by the prosecution on the policy element in the definition of crimes against humanity. Their main arguments revolved around one point: the Prosecutor had failed to establish “even a reasonable basis to believe” that there was the existence of an “organization” which was capable of adopting an “organizational policy”. Consequently, the defence argued that there were no crimes against humanity committed in Kenya and, as a result, the ICC lacked jurisdiction over the Kenya situation.¹⁰⁵

In their submission the defence teams banked on the arguments raised in the dissenting opinions of Judge Hans-Peter Kaul. Notably, when Judge Kaul expressed his dissenting opinion for the first time in response to Prosecutor’s request for authorization of an investigation, the defence teams had not been constituted, neither had any suspect been identified yet. This being the case, it is doubtful whether the meaning of “organization” as used in the Statute would have become so controversial had Judge Kaul not stirred the debate in the first place. Subsequent to the first dissenting opinion of Judge Kaul, the defence teams cleverly picked his arguments with a view to capitalizing on the fundamental divide already noticed in the Pre-Trial Chamber.

For that reason, and in order to avoid repetition, the core of the defence’s argument regarding “organizational policy” should further be inferred from Judge Kaul’s opinion outlined below (infra Sect. 6.4.2.3.4).

6.4.2.3.3 Pre-Trial Chamber’s Majority Opinion

The majority (two judges) of the Pre-Trial Chamber agreed with the Prosecutor’s submissions regarding organizational policy. The two judges stated expressly that they were in favour of a broad interpretation of Article 7(2)(a) of the ICC Statute, such that both the Mungiki and the Network would indeed qualify as “organizations” under that provision.¹⁰⁶ The majority enumerated six factors which they insisted “may assist” in determining whether a certain group or entity qualifies as an “organization” capable of authoring a policy. These factors are whether the entity:

(i) is under a responsible command, or has an established hierarchy; (ii) possesses, in fact, the means to carry out a widespread or systematic attack against a civilian population; (iii) exercises control over part of the territory; (iv) has criminal activities against the civilian population as a primary purpose; (v) articulates explicitly or implicitly an intention to attack a civilian population; and (vi) is part of a larger group, which fulfils some or all of the above-mentioned criteria.¹⁰⁷

Furthermore, the majority insisted that the factors enumerated above do not constitute a rigid legal definition, nor must they always be fulfilled exhaustively. Rather, an independent determination and assessment about the nature of the entity has to be made on a case–by–case basis. Moreover, they noted that the decisive criterion of whether a group qualifies as an organization should not be its “formal nature” or even “the level of its organization”. The criterion, according to them, should be the group’s “capability to perform acts which infringe on basic human values”. Consequently, the majority concluded that an organization envisioned by Article 7(2)(a) must not necessarily be linked to a state, nor must it be a state-like entity.¹⁰⁸

The majority added that Article 7(2)(a) of the ICC Statute envisions (i) policies of states which may be adopted at the highest level or by regional or even local organs (of the State); or (ii) policies of any non-state actor capable of adopting and implementing a policy to commit widespread or systematic attacks against a civilian population.¹⁰⁹ They noted that the policy envisaged under Article 7(2)(a) must not necessarily be formal or written: It suffices if the policy can, as the case was in Kenya, be deduced or inferred from an attack that exhibits the characteristics of being “planned, directed or organized, as opposed to spontaneous or (consisting of) isolated acts”.¹¹⁰ Thus, the majority were of the view that in each of the two Kenyan cases a policy to attack a civilian population could be established.¹¹¹

6.4.2.3.4 Dissenting Opinion of Judge Kaul

In his dissenting opinion, Judge Hans-Peter Kaul expounded his “fundamental disagreement” with the majority with regard to how the word “organization” in the definition of crimes against humanity should have been construed. While the majority favoured a “broad” or non-restrictive interpretation, Judge Kaul argued that the word should have been interpreted more narrowly or restrictively. Accordingly, Article 7(2)(a) of the Statute should have been viewed as envisaging not just any organization but only State–like organizations, of which neither the Mungiki nor the alleged Network would qualify.

The dissenting Judge agreed with the majority opinion that an “organization” is fundamentally different from a “State”, and that the former can include non-state entities. He also agreed with them that a policy can be simply inferred and need not be formalized.¹¹² However, he disagreed with the majority as regards the attributes of the organization or entity that can author such a policy. He argued that the mere juxtaposition of the notions of “State” and “organization” under Article 7(2)(a) suggests that the “organization” contemplated by that provision must have one specific attribute: it must “partake of some characteristics of a State”. As such, it must be an entity which “may act like a State or has quasi-State abilities”.¹¹³ Such characteristics or abilities, according to Judge Kaul, could include, but are not limited to, the following:

(a) a collectivity of persons; (b) which was established and acts for a common purpose; (c) over a prolonged period of time; (d) which is under a responsible command or adopted a certain degree of hierarchical structure, including, as a minimum level, some kind of a policy; (e) with capacity to impose the policy on its members and to sanction them; and (f) which has the capacity and means…to attack any civilian population on a large scale.¹¹⁴

The dissenting judge also listed the characteristics that in his view make certain groups not to qualify as “organizations” within the meaning of Article 7(2)(a) of the Statute. In other words, these entities or groups are not “State-like”. These are:

groups of organized crime, a mob, groups of (armed) civilians or criminal gangs…[such as those] formed on an ad hoc basis, randomly, spontaneously, for a passing occasion, with fluctuating membership, and without a structure and level to set up a policy [do not qualify], even if they engage in numerous serious and organized crimes.¹¹⁵

Applying the above understanding, Judge Kaul concluded that the features of the Mungiki and the Network, as alleged by the Prosecutor, simply placed the two groups in the category of “organized armed criminal gangs”, and not organizations within the meaning of Article 7(2)(a). He thus flawed the majority decision for holding that the two groups qualified as “organizations”. He stated that the majority set a bad precedent which, if not reversed, will have far-reaching implications, such as rendering the ICC an incredible institution that usurps the jurisdiction of the domestic courts over “ordinary serious crimes”. To Judge Kaul, the criminal acts that were committed in Kenya during the post-election violence were undoubtedly serious but strictly “ordinary” criminality under the Kenyan domestic laws and cannot be elevated to the status of crimes under international law, particularly “crimes against humanity” under the ICC Statute.¹¹⁶

Judge Kaul expressed the view that a restrictive interpretation of the word “organization” is important when applying Article 7(2)(a), in order to avoid a “banalisation” or “trivialization” of the crimes that the ICC is meant to handle.¹¹⁷ Finally, the judge concluded that the ICC was not the right forum to investigate and prosecute the crimes in Kenya, for it lacked jurisdiction.

6.4.2.4 Evaluation

6.4.2.4.1 Policy Element in Crimes Against Humanity

Four major approaches to the policy element in crimes against humanity can be identified so far. These approaches are that crimes against humanity: (1) do not require a policy at all; (2) require a state policy for their commission; (3) require a policy of either a state or a state-like organization; (4) require a policy of a state or any organization (not necessarily state-like), provided such organization has the capacity to carry out widespread attacks against a civilian population.¹¹⁸ Whereas the first two approaches feature more prominently in the jurisprudence of the ad hoc tribunals, the latter two approaches dominate the debate on the ICC Statute as the foregoing discussion has already shown.

However, there is no dispute that the explicit requirement of a policy as part of the legal requirements for crimes against humanity came with the ICC Statute.¹¹⁹ A policy was not, for example, an express requirement in the definitions in the IMT Charter or in the statutes of the ad hoc Tribunals or even those of the hybrid courts.¹²⁰ This notwithstanding, the question whether a policy was required for crimes against humanity still emerged in the jurisprudence of the ad hoc Tribunals, always stirring a heated debate. Initially, some scholars, as well as the early judicial pronouncements of the Tribunals, maintained that a policy was an implicit requirement under the customary law definition of crimes against humanity.¹²¹ Others, however, contested this position, arguing that proof of a policy or plan had no relevance in the customary law definition of crimes against humanity.¹²² After several inconsistencies in the pronouncements of the Tribunals, especially the ICTY, it came to be agreed that the existence of a policy was merely a demonstration of the systematic character of an attack directed against a civilian population. As to its legal status, the ICTY’s Appeals Chamber settled the controversy by ruling that for the customary law definition of crimes against humanity, “the existence of a policy or plan may be evidentially relevant, but it is not a legal element of the crime”.¹²³

Therefore, the extent to which the jurisprudence of the ad hoc Tribunals on policy element is relevant to the application of the ICC Statute must be taken cautiously. Indeed the ICC Statute made a radical shift from the position under international customary law applied by the Tribunals on this issue. It follows that the jurisprudence on “State or organizational policy” as a legal requirement for crimes against humanity is in its infancy, and evolving. Its shape will be defined largely by the ICC. In this regard, there are several issues which are yet to be completely settled at the ICC.

One such issue is whether a “State or organizational policy” in the ICC Statute is an independent contextual element for crimes against humanity, or whether it is simply an indicator of a systematic attack.¹²⁴ Regarding this issue, the Pre-Trial Chamber III stated in the Bemba case that “the existence of a State or organisational policy is an element from which the systematic nature of an attack may be inferred”.¹²⁵ However, when the Prosecutor adopted this line of reasoning in the request for authorization of an investigation into Kenya,¹²⁶ the Pre-Trial Chamber II, in a unanimous decision, adopted a slightly different position: It decided that an organizational policy is a separate contextual element independent of a systematic attack.¹²⁷ Interestingly, two of the Pre-Trial judges in the Kenya situation, Trendafilova and Kaul, were present in the bench for the Bemba case.

Another issue which is yet to be completely settled is the one that divided the Pre-Trial Chamber judges in the Kenyan cases, i.e. the meaning of “organization” in the definition of crimes against humanity in the ICC Statute. In the Bemba case, Judge Kaul agreed with the other two judges that an “organization” under Article 7(2)(a) of the ICC Statute can be “any organization with the capability to commit a widespread or systematic attack against a civilian population”.¹²⁸ Yet when the Prosecutor adopted this line of reasoning before the Pre-Trial Chamber in the Kenya situation, only the majority, excluding Judge Kaul, upheld the reasoning of the Chamber in Bemba. Kaul no longer aligned himself with that reasoning, arguing that the meaning of “organization” that was adopted in the Bemba decision was apparently too broad. He thus wrote his dissenting opinion in favour of “State-like organizations”. Justifying a change of the position he had previously endorsed in Bemba, Judge Kaul said that the Bemba case concerned “military-like organized groups in the context of an armed conflict” while the Kenyan scenario was of a different context.¹²⁹

6.4.2.4.2 Key Factors Towards Interpretation of “Organization”

6.4.2.4.2.1 Article 21 of the ICC Statute

Article 21 of the ICC Statute contains a hierarchical order of law applicable before the ICC, which is a self-contained legal regime.¹³⁰ It provides that the ICC “shall apply” (i) the ICC Statute itself, Elements of Crimes, and its Rules of Procedure and Evidence; (ii) applicable treaties and the principles and rules of international law, including those of armed conflict; (iii) principles derived from national laws of legal systems that would normally try the crimes in the ICC Statute, provided they are consistent with the laws listed in i and ii above; and (iv) principles and rules created in previous decisions of the ICC (precedents). As a common denominator, it the Statute requires that any law applied according to the order above must be, inter alia, consistent with internationally recognized human rights.¹³¹

6.4.2.4.2.2 Other Factors to Consider

Being a treaty, the ICC Statute must be interpreted according to the Vienna Convention on the Law of Treaties.¹³² The Convention requires, inter alia, that when interpreting a treaty, its terms must, first and foremost, be given their “ordinary meanings”, unless there are justifiable reasons not to do so.¹³³ In this regard, it is noteworthy that although the word “organization” is not defined in the Statute or in the Elements of Crimes,¹³⁴ resort could be taken to dictionaries. The Black’s Law Dictionary defines “organization” as “a body of persons (such as a union or corporation)”.¹³⁵ Non-legal dictionaries define it more broadly. For example, according to the Shorter Oxford English Dictionary, contextually, the verb “organize”, from which the noun “organization” is derived, means to “form into a whole with mutually connected and dependant parts; give a definite or orderly structure to [something]; frame and put into working order (an institution, enterprise, etc.); arrange (something involving united action); become a systematic whole; become coordinated; attain orderly structure or working order”. Accordingly, the noun “organization” is defined as “an organized body, system or society”.¹³⁶

6.4.2.4.2.3 Observation

The word “organization” as used in Article 7(2)(a) is not a term of art. It is thus important to give it its ordinary meaning, but which is nevertheless consistent with the main intent and purpose of the States Parties to the ICC Statute. As Werle and Burghardt rightly observe, neither the ordinary meaning of the word “organization” nor the grammatical context of the phrase “State or organizational policy” in which the word has been used supports the conclusion that the word envisioned a state-like entity. They further rightly note that the disjunctive “or” juxtaposes “state” and “organization” as equals. Thus, it is grammatically incorrect to infer from the formulation “State or organizational policy” that the “organization” must share the definitional characteristics of a “state” such that the former is necessarily a “state-like” organization.¹³⁷

An analogy could also be drawn from the chapeau of the Article 7(1) of the ICC Statute. This provision defines a crime against humanity as a “widespread or systematic” attack on a civilian population. Here, too, the disjunctive “or” has always been understood to indicate that the attack can either be widespread or systematic, and more importantly, a systematic attack need not share the characteristics of a widespread attack. Hence, the two attributes of the attack and could exist independently of each other.¹³⁸

This leads to the conclusion that even on grammatical grounds, the minority’s interpretation of the word “organization” cannot stand.

6.4.2.4.3 Teleological Approach of the Dissenting Opinion

At another level, in his dissenting opinion, Judge Kaul Judge attempted to give a “teleological” justification of his restrictive interpretation of the word “organization”. His main arguments revolved around four points, but which were followed by contradicting conclusions.

6.4.2.4.3.1 Main Arguments

Firstly, Judge Kaul noted that the earliest definition of crimes against humanity in the IMT Charter was adopted to respond to the crimes committed by the Nazis during the Second World War.¹³⁹ As he rightly noted, from the context in which the Nazi crimes happened, the acts were indisputably mass atrocities committed by a “sovereign state against the civilian population … according to a State plan or policy, involving large segments of the State”.¹⁴⁰ Secondly, he argued that subsequent to the Nazi crimes experience, crimes against humanity committed in other parts of the world again proved that crimes of such nature and magnitude were possible only “by virtue of an existing State policy”. Thirdly, he asserted that by expressly including “State policy” as part of the definition of crimes against humanity, the ICC Statute clearly “embraces” all the historic considerations referred to above.¹⁴¹ Fourthly, and more importantly, he admitted the fact that, despite the historical role of the State in the commission of crimes against humanity, the definition of the crime under Article 7(2)(a) of the ICC Statute focuses beyond that role mainly for one reason: “to accommodate new scenarios of threats” which, although might not have links to a state, “may equally shake the very foundations of the international community and deeply shock the conscience of humanity”.¹⁴²

The notion of “new threats”, as referred to by Judge Kaul, was given consideration for the first time by the International Law Commission (ILC) in its 1996 Draft Code of Offenses against the Peace and Security of Mankind.¹⁴³ The ILC proposed an express inclusion of “State or organizational” policy in the definition of crimes against humanity as a response to “new developments after Nuremberg”.¹⁴⁴ In this regard, therefore, the ICL recognized the dynamic nature of crimes against humanity (i.e. its form and manner of commission) from when they were defined for the first time in history.

6.4.2.4.3.2 Weaknesses and Contradictions

Despite acknowledging the important facts above, especially the emergence of “new threats”, it is difficult to comprehend why the dissenting Judge came to the assertion that only the “historic origins are decisive in understanding the specific nature and fundamental rationale of this category of international crimes”.¹⁴⁵ This assertion further led him to a flawed conclusion that in interpreting the policy element in the ICC Statute, absolute reliance should be placed on the “historic experience”, which, he argued, is the only “logical lesson” that influenced the drafters when adopting the definition of crimes against humanity in the ICC Statute.¹⁴⁶

Of course, Judge Kaul is not alone: his views on “state-like organizations” resonate with those of prominent scholars.¹⁴⁷ However, such views sharply contradict those of other prominent scholars.¹⁴⁸ But these divergent views aside, Judge Kaul’s approach contains intrinsic contradictions which make it flawed. Kaul’s reasoning is internally inconsistent, for it partly disregards the “new threats” that the judge expressly acknowledges to have kept emerging even after the experience of the Nazi criminality. His approach exhibits too much obsession with the definitions of crimes against humanity prior to the inception of the ICC Statute to the point of overlooking the need for a dynamic interpretation and application of the ICC Statute.¹⁴⁹ This (dynamic interpretation) should have been an important consideration, given the fact that the crime itself and the techniques of committing it are equally dynamic.

It is submitted that one could still rely on the historical evolution of crimes against humanity to counter-argue the historical–phenomenological justifications on which Judge Kaul based his dissenting opinion. The following section does just that.

6.4.2.4.4 To What Extent Is the Historic Experience Relevant to the ICC?

6.4.2.4.4.1 Evolution of the Definition of Crimes Against Humanity

Of all the definitions of the core crimes under international law, that of crimes against humanity has been most dynamic, having undergone a clear chronological metamorphosis since its first formulation. Such evolution is evident in terms of both the material and contextual elements of the crime.¹⁵⁰ For example, as regards the original definition in the IMT Charter, contextually, crimes against humanity would only result if the material elements were committed “in execution or in connection with” crimes against peace and war crimes.¹⁵¹ The immediate subsequent definition in the Control Council Law No. 10 dispensed with the requirement for a nexus with the other crimes, thereby effectively treating crimes against humanity as an independent crime.¹⁵²

Then, while the ICTY Statute expressly required that crimes against humanity be committed in the context of an “armed conflict, either of internal or international character”, the Statutes of the ICTR and the Special Court for Sierra Leone dispensed with this requirement. Instead, the latter two Statutes required that the material elements of the crime be “committed as part of widespread or systematic attack”, and the ICTR Statute further required that the attack be made “on national, political, ethnic, racial or religious grounds”.¹⁵³ Thus, from the IMT Charter to the ad hoc Tribunals and hybrid courts, there has always been a new element in the definition of crimes against humanity, and this has been not a mere accident.

For example, the definition of crimes against humanity in the IMT Charter responded to the atrocities committed by the Germans alongside the aggression war they waged. For that reason, it required a nexus with the war or the acts of aggression. The definition in the ICTY Statute was crafted to respond to the armed conflict which had taken place in the former Yugoslavia. For that reason, it specifically required a nexus with an armed conflict. The definition in the ICTR Statute was crafted to respond broadly to the 1994 Rwandan genocide in which close to one million people had been murdered. For that reason, it focused mainly on the widespread and systematic nature of the atrocities and dispensed with the need for a nexus with an armed conflict which the ICTY Statute had required. The definitions in the ICTY and ICTR Statutes differed in this fundamental way despite the fact that both Statutes were adopted at about the same time (1993 and 1994, respectively) and by the same body, the UN Security Council.

One clear thing from the foregoing paragraph is that all the definitions of crimes against humanity prior to the adoption of the ICC Statute were retrospective in nature; they were backward looking.¹⁵⁴ They responded to crimes which had already been committed, namely crimes whose specific circumstances and contours were clearly known even before their definitions were adopted. Thus, in all these cases, it was the known characteristics of the already-committed acts that shaped or determined the content and scope of the definitions of the crimes against humanity. But as the following two sections demonstrate, the ICC Statute entails a completely different paradigm.

6.4.2.4.4.2 The Prospective Nature of the ICC Statute

Unlike the IMT Charter and the Statutes of the ad hoc Tribunals, the ICC Statute is prospective; it is forward looking.¹⁵⁵ This is not to argue that the historic experiences that underpin the foundations of crimes against humanity have no relevance today as far as the ICC is concerned. Rather, it is to agree with Sadat that over–reliance on the Nuremberg precedent today is counter-productive; it amounts to retrogression to almost seven decades ago. Similarly, agreeing with Judge Kaul’s argument that “organization” means “military-like organized groups” will in effect “reverse nearly two decades of progressive development by effectively re-linking the commission of [crimes against humanity] to a finding of armed conflict (which requires organized fighting forces)”.¹⁵⁶

The fight against impunity for “serious crimes of concern to the international community” can be more efficient if the ICC Statute is applied progressively, i.e. in a manner that responds to or takes into consideration the new or contemporary experiences. Such experiences include the emergence of organized entities, permanent and ad hoc in nature, which are neither states nor state-like, but which have got the capacity to commit very serious crimes.¹⁵⁷ Indeed, as Sadat rightly puts it, regardless of “the canonical status of the Nuremberg precedent in international criminal law”, and even if it may seem “heretical” to object it, Judge Kaul’s argument “does not [adequately] respond to the terrible suffering of today’s victims of [crimes against humanity], nor does it accurately describe the modern law”.¹⁵⁸

Moreover, unlike the ICTR and ICTY, which were created as ad hoc institutions, the ICC was established with a view to being be a “permanent court”.¹⁵⁹ As such, no “completion strategy” was contemplated for the ICC. Logically, therefore, it is expected that this institution will be able to render international criminal justice effectively and adequately for an indefinite future. Given that the procedure for the amendment or review of the ICC Statute is long and cumbersome,¹⁶⁰ it is illogical if the ICC Judges were to start introducing untenable restrictions in the application or interpretation of the Statute even where the Statute itself does not warrant such restrictions.¹⁶¹

6.4.2.4.4.3 Interim Conclusion

The Pre-Trial Chamber’s majority view on the interpretation of the word “organization” under Article 7(2)(a) is welcome, for it largely responds to the reality of crimes under international law today. This was another area and opportunity in which the Chamber adopted a pragmatic view in relation to a contentious legal issue involving the interpretation of the ICC Statute.¹⁶² In the circumstances of (or similar to) the crimes committed in Kenya, placing more reliance on the capacity of the entity to organize and implement attacks rather than on the organizational structure of the entity is a correct approach; it is not a mere judicial activism as critics might think or argue. Indeed to the international community, whose interest is to ensure that serious crimes of international concern do not go unpunished, and more importantly, to the victims of the post-election violence in Kenya, justice does not depend on the strict organizational structure of the entities which may have orchestrated the commission of the crimes.

6.5 Issues Relating to Complementarity

6.5.1 Meaning of Complementarity

Legally speaking, the ICC and national courts have concurrent jurisdiction over the core crimes in the ICC Statute. Such a jurisdictional relationship required further regulation in order to avoid competition for cases between the two legal regimes. The principle of complementarity¹⁶³ was agreed upon to serve this purpose, and also to safeguard States’ sovereignty and increase their willingness to accept the jurisdiction of the ICC.¹⁶⁴ Both the Preamble and Article 1 of the ICC Statute, in a similar wording, stipulate that the ICC “shall be complementary to national criminal jurisdictions”. It is clear from this formulation that precedence is given to national courts over the ICC as far as the prosecution of the core crimes is concerned. The national courts are the primary jurisdictions while the ICC remains a secondary jurisdiction—a court of last resort. Therefore, the complementarity model clearly differs from the primacy model that governs the relationship between the ad hoc Tribunals and the national courts.¹⁶⁵

The content of the principle of complementarity is embodied in the provisions of Article 17 of the ICC Statute which deals with “issues of admissibility”. The Article enumerates circumstances in which a case becomes inadmissible before the ICC. Accordingly, the ICC can admit a case on complementarity grounds if the state with primary jurisdiction is inactive, unwilling or unable to prosecute in good faith.¹⁶⁶ Before analysing complementarity in relation to the Kenya situation, the relevant parts of Article 17 are quoted verbatim below for ease of reference.

1.

Having regard to para 10 of the Preamble and Article 1 [of the ICC Statute] the Court shall determine that a case is inadmissible where:

(a)

The case is being investigated or prosecuted by a State which has jurisdiction over it, unless the State is unwilling or unable to genuinely carry out the investigation or prosecution;

(b)

The case has been investigated by a State which has jurisdiction over it and the State has decided not to prosecute the person concerned, unless the decision resulted from unwillingness or inability of the State genuinely to prosecute;

(c)

The person concerned has already been tried for conduct which is the subject of complaint, and a trial by the Court is not permitted under Article 20, para 3.¹⁶⁷…

2.

In order to determine unwillingness in a particular case, the Court shall consider…whether one or more of the following exist, as applicable:

(a)

The proceedings were or are being undertaken or the national decision was made for the purposes of shielding the person concerned from criminal responsibility;

(b)

There has been an unjustified delay in the proceedings which in the circumstances is inconsistent with an intent to bring the person concerned to justice;

(c)

The proceedings were not or are not being conducted independently or impartially, and they were or are being conducted in a manner, which, in the circumstances, is inconsistent with an intent to bring the person concerned to justice…

3.

In order to determine inability in a particular case, the Court shall consider whether, due to a total or substantial collapse or unavailability of its national Judicial system, the State is unable to obtain the accused or the necessary evidence and testimony or otherwise unable to carry out its proceedings.

As the following discussion will show, the ICC’s exercise of jurisdiction over the Kenyan cases was not, legally speaking, due to Kenya’s unwillingness or inability to investigate or prosecute, but was rather due to Kenya’s “inaction”.

6.5.2 Evaluation of “Unwillingness” and “Inability” in Relation to Kenya

The post-election violence in Kenya per se did not have any direct effect on the functioning of the country’s judicial system. After the violence, Kenya’s ability to arrest, investigate and prosecute those involved in the violence remained the same as it was prior to the violence.¹⁶⁸ The allegations and perceptions that Kenya’s judiciary continued lacking independence or being corrupt¹⁶⁹ do not change this reality.

Although the judicial system remained functional, Kenya, as already shown, did not show any political will to investigate or prosecute those who bear the greatest responsibility for the crimes. Yet, as also shown earlier, some of the suspects had been identified in various commission reports, and also the available domestic legal framework could sufficiently be used to prosecute them. Although it has been argued that this state of affairs per se could be taken as constituting “unwillingness” or “inability” to prosecute on the part of the Kenyan government,¹⁷⁰ such an argument cannot meet the threshold of unwillingness or inability if tested against the provisions of Article 17 reproduced above, for reasons explained below.

According to Nouwen, in determining the admissibility of a case before the ICC, examination of state’s unwillingness or inability to prosecute is, strictly speaking, a secondary question or step. The primary step, as Article 17 above presupposes, is to establish an existence of tangible “proceedings”, namely concluded or ongoing investigations or prosecutions in respect of the case. It is only if such proceedings do not pass the genuineness test—e.g. if they were or are intended for shielding the perpetrator(s), or if they lack independence or impartiality—can the state concerned be said to be unwilling or unable to investigate or prosecute.¹⁷¹

As already shown, and will be explained below, Kenya did not even initiate proceedings. This is to say that even the first step that would have paved way for the legal assessment of unwillingness or inability was not reached at the domestic level. The mere non-existence of proceedings in this regard, therefore, renders irrelevant a determination of unwillingness or inability for purposes complementarity with regard to the Kenya situation. However, since it was Kenya’s failure to initiate proceedings (inaction) which was the basis of admissibility of the situation and the cases before the ICC, it (inaction) merits further evaluation.

6.5.3 Inaction as a Component of Complementarity

6.5.3.1 Meaning of Inaction

The notion of inaction (also “inactivity”) is not mentioned explicitly under Article 17 of the ICC Statute. However, it is now settled that inaction is implicit in this provision, being a third ground (in addition to unwillingness and inability) which makes a situation or a case admissible before the ICC for complementarity purposes. The fact that “inaction” is part and parcel of Article 17 was raised for the first time in 2003 by a group of prominent scholars (experts) in a paper commissioned to them by the ICC.¹⁷² Their argument was endorsed by the ICC’s OTP¹⁷³ and later by both the Pre-Trial Chamber¹⁷⁴ and the Appeals Chamber.¹⁷⁵ Accordingly, “inaction” as a ground for admissibility of situations and cases before the ICC comes into play in circumstances where unwillingness or inability cannot practically or legally apply, namely where there are no national proceedings at all encompassing the same person and the same conduct forming the subject of the case before the ICC.¹⁷⁶

Inaction occurs in two scenarios. The first scenario is where the state with jurisdiction is doing nothing as regards investigations or prosecutions, even though, for example, prima facie, such a state appears able to do so. The second scenario where inaction comes into play is in a case of a self-referral, i.e. where the state of commission relinquishes or waives its primacy of jurisdiction voluntarily in favour of the prosecution before the ICC believing that, due to specific circumstances, justice will be delivered or rendered more effectively by the ICC. As a practice, the second scenario is now embedded in the ICC’s legal regime¹⁷⁷ despite facing a strong criticism when it was first endorsed.¹⁷⁸ In cases of inaction, the ICC intervenes in order to fill the impunity gap that could have resulted if it (the ICC) did not do so.¹⁷⁹

6.5.3.2 Analysis of Inaction in Relation to Kenya

Complementarity assessment in investigations triggered proprio motu, like the case of the Kenya situation, takes place at two levels, i.e. the situation and case levels. Each level entails different thresholds and contexts in which the assessment is done.¹⁸⁰ For example, complementarity requires a higher degree of specificity when assessed at a case level than when it is assessed at a situation level. This being the case, Kenya had an opportunity to claim its primacy of jurisdiction at each level with different implications. Although Kenya did not bother to invoke complementarity at the situation level, the Pre-Trial Chamber did a suo motu assessment. Kenya tried to invoke complementarity only at the case level, albeit unsuccessfully.

6.5.3.2.1 Inaction at Situation Level: The Notion of “Potential Cases” and Deferral of Investigations Under Article 18

Article 18 of the ICC Statute covers complementarity at the level of a situation. It provides that even when the Prosecutor has a reasonable basis to commence a proprio motu investigation into a situation, he or she cannot embark on it immediately. The Prosecutor shall first notify all the States which could have jurisdiction over the alleged crimes, including the state of commission.¹⁸¹ Then, a State so notified is given up to 1 month to assert its primacy of jurisdiction by: (i) informing the Court if “it is investigating or has investigated” the crimes relating to “the information provided in the notification”; and (ii) requesting the Prosecutor to defer to that state’s investigation of those persons. In fact, this is not a “request” per se; it is a demand to which the Prosecutor must comply, unless he or she justifies before the Pre-Trial Chamber as to why he or she should be authorized to continue with investigations despite the demand. One such justification could be that the deferral was asked for in bad faith.¹⁸²

Therefore, Article 18 makes it possible for a state with any jurisdictional link with a situation to prevent the Prosecutor from initiating an investigation, “because even initiation of an investigation might interfere with the exercise of national jurisdiction”.¹⁸³

6.5.3.2.1.1 Kenya’s (Lost) Opportunities Under Article 18

Had Kenya invoked Article 18, it would have got several advantages. At this particular stage, complementarity applies vaguely or generally, since no specific suspects have been identified yet. The basis of its assessment is “possible or potential cases” as opposed to “specific cases”. So all Kenya had to do at this stage was to show genuinely that it was generally investigating the “group of persons” and “incidents” which were likely to form Prosecutor’s future “potential cases”.¹⁸⁴ There would not have been a strict requirement to show a high degree of specificity as regards the identity of the persons or the crimes or conduct being investigated. For example, it would have been sufficient for Kenya to prove that it was investigating “events”, such as “the massacre in a certain village or a campaign in a particular geographic area during a particular time period” with regard to the post-election violence.¹⁸⁵

Even if Kenya had not yet started “actual investigations” at this level, it would still have been possible to argue that it “intended” to do so because the Prosecutor’s notification that he wanted to intervene, together with any additional information furnished by the OTP, had been the motivation to trigger such domestic investigations. Such arguments would have been sufficient for a deferral under Article 18 in view of the general nature of applicability of the complementarity principle at this (situation) stage.¹⁸⁶ In this regard, Kenya could have presented its plan for domestic investigations and prosecutions. As it will be shown shortly, Kenya attempted to present such a plan at a later stage, but this was already too late.

Lastly, had Kenya requested and obtained a deferral under Article 18, it would have got a period of up to 6 months before the Prosecutor could review the deferral again, provided that it continued to show constant genuineness in its investigations.¹⁸⁷ However temporary this “grace period” might seem, it would have enabled Kenya to commence or continue with investigations while continuing to put its house in order with regard to prosecutions.¹⁸⁸ Thus, there would have been time, for example, to re-engage Parliament with regard to the creation of the proposed local tribunal or the designing of any other domestic forum, such as a special division of the judiciary, where the cases linked to the post-election violence could be tried.

6.5.3.2.1.2 Reasons for Kenya’s Failure to Invoke Article 18

Of all the situations he handled, the former ICC’s Prosecutor Moreno-Ocampo is said to have acted, arguably, most transparently with regard to Kenya.¹⁸⁹ For example, the Prosecutor did not hasten the ICC’s intervention in Kenya even when matters so dictated. Even prior to the official notification of his intention to commence investigations into Kenya, the Prosecutor had insisted that “if the Kenyan authorities [carried] out genuine judicial proceedings against those most responsible, the OTP [would] not have ground to intervene”.¹⁹⁰ This suggests that even before the ICC’s intervention, the Kenyan authorities were already aware, or at least had a clue, of the “potential cases” that the Prosecutor had in mind. In any case, it must have been clear that the main source of the Prosecutor’s potential action would be, among others, the publicly available reports of the Waki Commission and that of the Kenya National Commission on Human Rights, and that the Prosecutor’s main list of suspects would, first and foremost, be derived from the twenty names in the famous Waki envelope.

A question that arises at this juncture is why then did Kenya forgo the opportunity under Article 18, while, as it will be shown shortly, it did not like the ICC’s intervention? It is interesting that 2 years after losing this opportunity, the Kenyan government argued in retrospect that it would “not [have been] possible” to invoke Article 18 “before the adoption of the new Constitution and the legislative and other reforms”.¹⁹¹ However, this argument is not convincing since, as already stated, even before the claimed reforms, Kenya’s judicial system had by all domestic standards remained functional. Hence one fails to see why investigations, let alone prosecutions, should have been completely “impossible”. One could even go further to ask: What if the adoption of these particular reforms (mainly constitutional) failed to materialize? Would it then have remained forever “impossible” for Kenya to investigate and prosecute?

Considering the prevailing Kenyan domestic political situation at that time, there are several reasons that contributed to Kenya’s failure to ask the Prosecutor to defer to its investigations at the situation stage. One such reason is that Kenya underestimated the opportunity available under Article 18, given that there had not been any previous precedent where this provision had been invoked. Another reason is that the ICC was still perceived by the Kenyan political elite as a remote threat, especially because until then the names of the specific suspects to be charged by the ICC had not been revealed by the Prosecutor. Lastly, the two main political parties in the coalition government, PNU and ODM, might have failed to agree on a common position as regards the immediate response or reaction to the Prosecutor’s notification of his intention to open an investigation. This seems to be the case is in view of the fact that even after the Prosecutor had intervened officially, the two sides of the coalition government opposed each other openly with regard to Kenya’s request to the Security Council asking for a deferral of investigation under Article 16 of the ICC Statute.¹⁹²

6.5.3.2.1.3 Assessment by Pre-Trial Chamber

Assessing admissibility suo motu, the Pre-Trial Chamber asked the Prosecutor to submit additional information on, inter alia, “admissibility within the context of the situation in the Republic of Kenya”, specifically in relation to: (i) the incidents that [were] likely to be the focus of an investigation; (ii) the groups of persons involved that [were] likely to be the target of an investigation for the purpose of identifying the potential cases under consideration; and (iii) domestic investigations, if any, with respect to those potential cases as constituted by the previous two elements.¹⁹³

On the basis of this information, the majority of the Pre-Trial Chamber ruled that neither Kenya nor third states were “active” with regard to investigating elements that were likely to shape the Prosecutor’s “potential cases” to be derived out of the situation. The Chamber further ruled that although Kenya submitted a report supposedly showing that some domestic prosecutions had taken place, these prosecutions were only in respect of “minor cases”. More importantly, it found, as a bigger flaw, that even these minor cases themselves did not cover the potential cases against “senior business and political leaders” related to the two political parties involved in the violence, the PNU and the ODM.¹⁹⁴ Consequently, the Chamber concluded that there was “inaction” at situation level which made the Kenya situation admissible. But after the situation had been declared admissible, the question of “inaction” emerged again at case level.

Unlike a situation, a “case” refers to “specific incidents during which one or more crimes within the jurisdiction of the Court seem to have been committed by one or more identified suspects”.¹⁹⁵ Thus, a case arises out of an admitted situation. It comes into being when an arrest warrant or a summons to appear has been issued against a specific suspect. Complementarity (admissibility) has to be determined again at this stage by specifically assessing both the individual and conduct charged.¹⁹⁶ Kenya invoked complementarity at this stage by filing an admissibility challenge pursuant to Articles 19 and 17 of the ICC Statute.

< div class='tao-gold-member'>

Only gold members can continue reading. Log In or Register to continue

Fastest Law Search Engine

The Kenya Situation Before the ICC

6. The Kenya Situation Before the ICC