Investigation and Case Selection
The International Criminal Tribunal for Rwanda (ICTR), a post-genocide court, was established on 8 November 1994 after the genocide was over. The chief prosecutor of the ICTY, based at The Hague, assumed that post for Rwanda as well. The seat of the ICTR was at Arusha, in northern Tanzania. The first ICTR Prosecutor, Judge Richard Goldstone, visited Rwanda for the first time in December 1994, but continued to reside at The Hague. These decisions had direct bearing on investigations in Rwanda.
The many delays and inefficiency that dogged the ICTR in its first few years have been well documented. Although many problems were of the sort endemic to any large bureaucratic organization, others stemmed directly from the United Nations’ lack of experience in the field of international criminal justice. The rules of the UN simply are not very well suited for the staffing and running of a tribunal expected to investigate and prosecute perpetrators responsible for atrocity crimes. When recruited, ICTR staff was posted in three continents – at New York in North America, The Hague in Europe, Arusha and Kigali in Africa – thus making coordination, consultation and team building problematic.
Delays in recruitment of staff members with experience in investigation of genocide cases resulted in the lapse of significant periods of time between events under investigation and when actual investigations began. This lapse created operational difficulties in the collection of evidence. Many witnesses were often interviewed more than once on the same subject matter, and by different investigators. The delays also meant that by the time the trial commenced, many witnesses were already engaged in the reconciliation process in Rwanda and were reluctant to testify in court. Some rape victims who had since started new families did not wish to testify anymore. Subsequently lacks of corroboration of witness statements were serious challenges and often affected the credibility of otherwise honest witnesses.
Again, it is important to remember that the prosecution was investigating the one genocide that occurred in 1994 albeit with many different incidents and crimes committed which were related to this genocide. Thus, it was expected that witnesses’ testimony describing one incident from different angles may differ; memory of witnesses who were either aged or very young at the time the crimes were committed may fail them, or witnesses may remember new facts not mentioned to past investigators in earlier interviews. These factors created additional challenges to investigators.
Under the ICTR Statute, the prosecutor is independent. However, the relationship between the tribunal’s judges and the Office of the Prosecutor (OTP) were complicated. Given the different systems from which the judges and prosecutors came, this was perhaps unavoidable. In the Anglo-American tradition the functions of a judge and prosecutor are closely separated. This is not so in civil-law systems where investigating judges play a leading role in criminal investigations and work closely with the prosecutor.
The ICTR practice is a hybrid between Anglo-American tradition and civil-law systems. The prosecutor may independently initiate investigations, and sought to protect that independence in his relationship with the judges. However, under the Rules of Procedure and Evidence (RPE), the prosecutor prepares an indictment containing a concise statement of the facts and the crime or crimes with which the accused is charged under the Statute and submits the indictment to a judge of a pre-trial chamber for review and confirmation. Thereafter, the prosecutor’s authority is subjected to the judge’s decision on whether to confirm or dismiss the indictment. This process provides a glimpse in the relationship between the prosecutor and the judges in managing investigations at the pre-trial stage.
Administratively, the chief prosecutor is appointed by the UN Security Council on the recommendation of the Secretary General. Within the Security Council, the permanent members with veto powers exercise the ultimate control on whether to endorse or reject a candidate recommended by the Secretary General. With respect to OTP staff, they are appointed by the Secretary General on the recommendation of the prosecutor. In practice, it is the Office of Legal Affairs (OLA) and the registrar, to who powers of the Secretary General to hire and fire staffs is delegated, who are responsible for the recruitment of OTP staff. Lastly, the OTP budget is approved by the General Assembly.
The above factors raise concerns on the extent of the prosecutor’s independence in recruitment of his staff and management of OTP. Aware of the role of the UN in the creation of the tribunal and its role in the management thereof, it is in that context that the chapter discusses the prosecutor’s policy on investigation and selection of cases. In section 2, early problems are identified to provide the context. In sections 3 and 4 policies on investigations and on selection of cases are examined. In section 5, the impact of resolutions 1503(2004) and 1534(2004), on prosecutorial discretion are discussed and concluding remarks are made in section 6.
- Background: a difficult beginning for the OTP
Investigations of serious crimes committed in Rwanda in 1994 started long before the Tribunal was established. When Judge Goldstone was appointed ICTR prosecutor, other actors, including the UN group of experts, UN Human Rights Commission, UN peacekeepers, local, regional and international organisation such as Africa Rights, Human Rights Watch, Amnesty International, as well as national and international journalists, national governments and independent researchers were already in Rwanda. These actors had interviewed potential witnesses, the same witnesses OTP investigators later sought to interview. Many witnesses interviewed by non OTP investigators were never informed of their rights or provided with adequate protection.
When the OTP recruitment process began, it was based on the UN policy which places great premium on geographical representation rather than competence. The policy, as noted by Bernard Muna, ICTR deputy prosecutor, was unsuitable. Muna observed that OTP recruitment policy was created in the image of the UN General Assembly rules and regulations. Staff members were recruited from around the world with strict adherence to geographic representation. While such geographic representation seemed like a good idea, Muna noted that this diversity brought a number of challenges in the creation of an investigating team, including those related to language, culture, and professional training and background. Additionally, the coming together of different legal systems created problems for the OTP as practitioners trained in different systems often have markedly different perspectives on issues such as relating to the statements by the accused.
Judge Goldstone was disappointed with the UN imposed recruitment policy because it resulted in unnecessary tussle between the UN secretariat in New York and the office of the ICTY/ICTR registrars at The Hague and Arusha. The UN Secretary General ‘delegated his hiring and firing authority with respect to the tribunal staff to the Registrar of the tribunal. This led to unfortunate problems and difficulties that were exacerbated by unimaginative and sometimes malicious officials who had been seconded to the tribunal from the other United Nations offices in order to advise the Registrar’. This bad policy required the prosecutor to spend many days at irritating and time-consuming meetings with UN officials in obtaining agreement on which staff members were suitable and levels at which they should be employed. The tedious process sometimes took so long that by the time competent and experienced lawyers were identified and offered positions at the OTP, they had lost interest in the job.
The competences of some senior lawyers recruited through the UN process were questionable. Prosecutor Carla Del Ponte observed that during her visits at Arusha, ‘whenever I discussed legal and factual issues with prosecuting counsels, I discovered the inexperience and incompetence of about a dozen of them’. When in court, the prosecutor ‘found some of the courtroom performances to be so lacking that I sensed a real possibility that serious killers were going to win acquittal’. Del Ponte also recalled ‘one white Westerner, an academic, who had found his way onto a prosecution team having insufficient practical knowledge of procedure or experience in the court room’. She continues, ‘another attorney was clearly not meant for prosecutorial work because he lacked the capacity to grasp the essence of cases on which he was working’. When Prosecutor Del Ponte sought to not renew contracts of non-performing lawyers, she discovered, to her surprise, that the procedure for terminating contracts within the UN systems was problematic because, in her words, ‘the United Nations bureaucracy imposes cumbersome procedures, including a complaint process that sometimes leads to investigations by officials from New York and entails wasting time I could not waste’. Overall, failure of the UN recruitment policy to give weight to qualifications, investigatory and prosecutorial experience of candidates created conditions that made it easy to recruit staff who were less qualified or had insufficient practical experience in international criminal justice. Had the prosecutor been given a supervisory role in the recruitment of his/her staff, perhaps these early challenges would have been avoided.
Another challenge was the funding of the tribunal. The ICTR has no independent source of funding apart from the UN. Approval of ICTR funding is through the United Nations General Assembly Advisory Committee on Administrative and Budgetary Questions, known by acronym ACABQ. With a tight grip on the recruitment process and funding, the prosecutor was always in a weak negotiating position vis-à-vis UN officials. The process of getting funds from the UN is tedious. The prosecutor and the registrar present the ICTR annual budget to ACABQ for approval. The OTP budget is only a portion of the ICTR overall budget which is jointly submitted with that of the chambers and the registry. It is the responsibility of the ICTR registrar and the chief prosecutor to justify budget proposals before the ACABQ. When not persuaded, the ACABQ, an advisory body of the UN, may reject the whole budget, part of it or seek further amendments or modifications to the budget. In that context, ACABQ exercises effective control over the ICTR budget. The ACABQ process is frightening. According to Goldstone, ‘Those knowledgeable in UN affairs advised me that it was a most unpleasant experience; that every item of our draft budget would be examined in detailed and that many of the members were difficult if not unpleasant’. Investigators have generally complained of a lack of funds. The financial year 2002/2003 was so bad for the OTP that there were no new recruitments and those who left service were not immediately replaced. Receiving funds from the UN is a frustrating exercise. 
With the power to appoint the prosecutor and his staff, and the authority to approve or reject the budget, one might question to what extent the prosecutor is still independent from the UN, particularly from the five permanent members of the Security Council? The prosecutor is appointed by the UN Security Council on nomination by the UN Secretary-General for a four-year period, being eligible for reappointment. The terms and conditions of service of the prosecutor are those of an under-Secretary-General. The staffs of the OTP are appointed by the Secretary-General on the recommendation of the prosecutor. Critics, including the international media, have referred to the close relationship between the UN and the prosecutor as evidence that the prosecutor is a tool of the Security Council and not at all independent. Successive prosecutors have fought against the stigma of being a tool of the Security Council. As observed by Goldstone, the international media had ‘[…] effectively written off the International Criminal Tribunal for the former Yugoslavia (ICTY) as the ‘fig leaf’ of the international community to hide its shame for inaction in the former Yugoslavia, particularly in Bosnia’. And, after documenting a catalogue of failings of the ad hoc tribunals, Ralph Zacklin, an assistant secretary general in OLA, argued that the tribunals were a ‘fig leaf’, created by the Security Council in response to the international community’s failure to intervene in armed conflicts in the former Yugoslavia and Rwanda. Carla Del Ponte noted: ‘The United Nations Security Council created the International Criminal Tribunal for Rwanda on November 8, 1994. This act was not just an effort to end the culture of impunity that had developed in Rwanda for generations before the genocide. It was a diplomatic mea culpa, an act of contrition by the world’s major powers to make amends for their gross failure to prevent or halt the massacre’. These concerns reflect a frustration in the relationship between the UN and the tribunals because of the apparent failure of the UN to comprehend how international criminal justice works, space that needs to be given to the prosecutor to freely function, and the public’s perception of the tribunals.
Could the argument that the ICTY/ICTR was a fig leaf, or that the Rwanda Tribunal was a diplomatic mea culpa, suggest that the prosecutor was not independent of the Security Council? Proponents of the Tribunal argue that although the ICTR was created by the Security Council and the chief prosecutor appointed by the same body, the prosecutor is an independent official and has publicly guarded his independence. To compromise his independence would render the prosecutor and his office irrelevant and inconsequential. It was therefore necessary for the prosecutor to respect the spirit and the law of the tribunal by recognizing that the objective of the tribunal is to render justice according to the law and not to appease great powers. However, various levels of contacts between the prosecutor on the one hand, and the UN, regional and national institutions, on the other hand, have reinforced perceptions that the prosecutor is not independent notwithstanding the prosecutor argument that he is. Examples of contacts between the prosecutor and these institutions are used to suggest that the prosecutor may not be as independent from the influences of these institutions as the OTP may claim.
Goldstone acknowledged that ‘Jan Pronk, the Dutch minister of development aid, played a major role in raising the start-up funds. The generosity of his government was quite out of the ordinary. The Netherlands’s provision of personnel, equipment, and funds for the Rwanda tribunal was second only to the United States’. There is, however, no evidence that Jan Pronk, or the Dutch government, used their position to influence decisions made by the prosecutor. Similarly, many states, including Pakistan, Italy, the United Kingdom, Belgium, Canada and the Nordic countries provided considerable financial, material and human support to the ICTR, but there is no evidence suggesting that they used their financial contributions to influence the prosecutor’s decisions.
The track record of the ICTR suggests that assistance offered by national governments was timely and appreciated by the OTP. There were states that assisted the OTP by providing a significant number of investigators. These officials were the first OTP investigators to go in the field and to interact with traumatised Rwanda survivors. The officials seconded to the OTP by their national governments performed commendable work although their presence in the hills of Rwanda also created curiosity from the local population. The states did not use their contribution to the ICTR as leverage to interfere with the prosecutor’s independence. On the contrary, the good intentions of those states, in some cases resulted in sending to the OTP overzealous investigators with limited or no relevant skills. Some of these officials, particularly a small group with no knowledge of criminal law, or respect for Rwanda culture and people, recorded statements that did not assist the prosecutor. Kinyarwanda is the national language of Rwanda. French and/or English are spoken by Rwandans who have had the benefit of education. Thus, language and cultural barrier exacerbated problems faced by investigators as they had to operate through Kinyarwanda interpreters. The working languages of the ICTR are French and English.
Administrative problems created another set of problems. The location of the seat of the ICTR at Arusha, and the sharing of a chief prosecutor with the ICTY until 2003, for example, created tension between the ICTR and the Rwandan government. Rwanda considered the ICTR a poor and subordinate institution compared to the ICTY. This view was more entrenched for the Chief Prosecutor resided at The Hague, and OTP personnel stationed at different stations, for example, the deputy prosecutor, investigators and legal advisors at Kigali, and prosecuting attorneys at Arusha. As there were still no electronic communications, coordination between staffs posted at different stations was problematic.
Jurisprudence on substantive and procedural law during the early period of the Tribunal was limited. It was jurisprudence of the Nuremberg and Tokyo International Military Tribunals created after WW II that was readily available. While the Nuremberg and Tokyo Tribunals had investigated and prosecuted international crimes, the principles and precedents of the two tribunals remained dormant for over fifty years. These precedents were not particularly helpful to the ICTR. The other precedent was that of the ICTY created in 1993. Until 2003, the ICTY and the ICTR shared the same chief prosecutor. Thus, from 1994 to 2003 the ICTY/ICTR policies on investigation and selection of cases were similar or at least formulated by the same chief prosecutor. Differences in policies of the two tribunals emerged after 2003 when Security Council adopted resolution 1503(2003) and later resolution 1534(2004). Security Council resolution 1503(2003) created an independent and separate chief prosecutor for the ICTR. It is pursuant to this resolution that the joint ICTY/ICTR prosecutor, Madam Carla Del Ponte was removed as ICTR prosecutor but retained at the ICTY in the same capacity. Security Council resolution 1534(2004) on the other hand outlined guidelines for the winding-up of the two tribunals, jointly referred to in the resolution as ‘Completion Strategy’.
The overview of challenges faced by the OTP during the early years of the tribunal provides context in which the policy on investigation was formulated and implemented. It also provides an understanding of the selection of cases, particularly why some perpetrators were prosecuted while others were not.
The prosecutor has authority to initiate investigations ex-officio or on the basis of information obtained from any source, including from governments, United Nations organs, intergovernmental and non-governmental organizations. The prosecutor also retains the right to independently assess information received or obtained from third parties and to decide whether there is sufficient basis to proceed. In exercise of his ‘prosecutorial discretion’, a prosecutor may choose to initiate investigations with respect to a given incident and select a case or cases for prosecution. The prosecutor may also choose to conduct guilty-plea negotiations and if successful, recommend lighter sentences. Or the prosecutor may choose to withdraw the case. These choices have different consequences for an accused.
The ICTR prosecutor adopted a hybrid of Anglo-American tradition and inquisitorial civil law model. The exercise of prosecutorial discretion has been criticized by some accused who view it as an abuse of power on the ground that any accused may be manipulated by the prosecutor to testify against co-conspirators in exchange of favours, such as lighter sentence. The prosecutor, on the other hand, views guilty-plea negotiations as positive contributions to judicial economy. The process is cost-effective and witnesses do not suffer secondary trauma since they are not requested to testify during guilty plea hearing and sentencing.
Based on my experience at the OTP, it was apparent that even with the prosecutor’s broad powers, the conduct of investigation would still be problematic for many reasons. The ICTR, a post-genocide creation, was established after the crimes were already committed and many of the perpetrators considered most responsible had fled Rwanda. The scene of crimes, that is, the entire territory of Rwanda, was not preserved, and many non-OTP investigators had interfered with primary evidence. These were the first groups of ‘investigators’ who arrived in Rwanda during the armed conflict and eventually interviewed many witnesses during the genocide itself. These investigators included officials from various UN agencies, human rights organizations, journalists and representative of various foreign governments. While they gathered contemporaneous statements that were useful, unfortunately most were not trained investigators. Best practice in the investigation of serious crimes was not followed. Potential witnesses, for example, were interviewed more than once and often by different persons from different organizations without coordination. In many cases, questions relating to relevant elements of crimes were never put to a witness or simply glossed over. Some statements gathered by third-party investigators were passed on to the OTP. The statements were reviewed and evaluated by investigators who then submitted their assessment to the prosecutor. The prosecutor made decisions on which of the incidents and suspects were to be further investigated. Investigators to whom cases were assigned were divided in teams. Each investigating team was made aware that the objective of investigation was to gather evidence that is sufficient to satisfy legal requirement for proving a case beyond reasonable doubt.
During the early stages of investigation, the prosecutor’s priority was to identify suspects who bear the greatest level of responsibility for genocide and other serious crimes. The targeted suspects were senior politicians, government officials and military leaders. As investigations proceeded, it soon became apparent that there was insufficient or no direct evidence connecting senior officials who ordered or instigated commission of the crimes with the criminal acts committed by their subordinates, that is, the low-level perpetrator. Lack of nexus between orders issued, or planning made by some of the senior leaders and criminal acts of actual perpetrators meant that unless methodology of planning the genocide was meticulously examined, it would be problematic to establish responsibility of superiors. It would even be more difficult to prove liability of civilian superiors as opposed to military commanders since civilians may not necessarily be in the chain of command and, as civilians; they may not directly exercise de facto or de jure authority over subordinates, the direct perpetrators.
To address these problems, the prosecutor developed a policy which assisted investigators in establishing links between direct perpetrators and senior military and political leaders who planned or issued orders for the commission of the alleged crimes by rigorously investigating activities of notorious killers notwithstanding that they were low-level perpetrators. Information obtained from low-level perpetrators, in many cases corroborated existing evidence, or linked senior political and military leaders to serious crimes alleged in indictments, or provided the prosecutor with opportunity to amend indictments and to make additional charges. Additionally, to empower investigators, they were encouraged to read widely on the history of Rwanda, its people and culture. Publications by African Rights and Human Rights Watch were extremely helpful.
Due to the complexity of international crimes, particularly of specific elements of these crimes, integration of various fields of expertise – for example, individuals with higher level of education, professional qualifications, and prosecutorial experience in an investigation team – was necessary. It was also useful for experts from relevant disciplines to support investigators. Thus, the establishment of multi-disciplinary groups of investigators supported by local expertise with ability and competence to interpret information in its authentic social context, for example, culture, politics, economy and linguistics. Additional support was received from experts commissioned by the prosecutor to write expert reports or to testify as expert witnesses. Dr Alison De Forges is perhaps the best known OTP expert on Rwanda.
In the course of collecting the evidence, different types of evidence were gathered. Witnesses’ statements, mainly from survivors, were the primary source. Investigators were instructed to treat witnesses humanely, with respect and dignity. Documentary evidence was useful for two reasons. First, it had probative value and enhanced judicial economy. Contrary to witness testimony, a document cannot legally change its content through passage of time, and it is not vulnerable to threats, blackmail, faulty recollection or other human weaknesses. Second, documentary evidence was reliable because, unlike witnesses who on intrusive cross-examination may present different versions of testimony, documentary evidence, once admitted, its contents remained unchanged. Documentary evidence also assists in limiting the exposure of vulnerable witnesses, particularly witnesses who are also victims of rape and sexual violence.
Financial records, though difficult to gather, when available, enables the prosecution to establish net-works and command structure related to the organization and commission of crimes. Personal records of transactions or of companies owned by perpetrators or their associates are also important in establishing the nexus between the actual perpetrators and senior officials who issued orders, instigated or aided and abetted the commission of the crimes. However, a challenge to the prosecutor was admissibility of documents obtained from suspects or witnesses in the course of investigations. In some instances the prosecutor had to rely on copies of the documents rather than the original versions. It is the responsibility of the prosecutor to authenticate the relevant documents as well as to establish unbroken movement of documents from suspects or from witnesses who handed it to investigator(s). The investigator must then explain circumstances under which he/she kept custody of the document prior to handing it over to the evidence section of the OTP for safe custody. At trial, prosecuting counsel who retrieved the document from the evidence section must then explain to the satisfaction of the Chamber the chain of custody of the documents before it is admitted as evidence. The movement of documents from the point of discovery, to the time it is submitted as evidence in court demands that investigators carefully document source(s) of documents, including providing evidence of detailed records of chain of custody, namely, how the evidence was moved from one point to the next, and eventually submitted in evidence for the court’s evaluation and determination.
A much more complex situation is interviewing and recording statements from an ‘insider witness’. It is helpful to recall that investigation into genocide, crimes against humanity and war crimes in Rwanda was complicated partly because perpetrators and their co-conspirators did not meet in public places, kept official minutes of the meetings or discussed such delicate matters with third parties when planning to commit atrocity crimes. On the other hand, some co-conspirators who were insiders in the decision-making process, or very close to the inner circles, were sometimes, under certain conditions, prepared to cooperate with the prosecutor. Those who chose to cooperate with the prosecutor as insider witnesses often had personal knowledge about broader issues, for example, methodologies used in planning or the persons who attended those preparatory meetings leading to the commission of genocide. An insider witness is expected to testify against their former colleagues in return for a guilty plea and, subject to the Chamber acceptance of a negotiated sentence between the prosecution and the defence, impose a lighter sentence.
To that extent, insider witnesses are usually accomplices to crimes with which the accused are charged. Due to this accomplice relationship, use of insider witnesses is controversial and their statements are treated with caution. Some insider witnesses retract their earlier confessions and subsequently refuse to continue to cooperate, or to testify against their fellow co-conspirators; others tend to lie systematically concerning their own crimes while generally truthful concerning crimes committed by other co-perpetrators; while some insider witnesses tend to seek friendship, embellish acts attributed to co-conspirators and in return make requests for assistance to be extended to his family in exchange for his cooperation. It is not for investigators to make decisions on whether a suspect is an insider witness. That responsibility is the prosecutor’s.
Before interviewing an insider witness, whether through a formal or informal approach, it is necessary that an investigator must have full command of the facts of the case and a clear knowledge of the insider witness’s political, military and social background. In situations where the insider witness has a better knowledge of the subject than the investigator, he will control the interview and will only provide the investigator the information that he chooses to and not necessarily what the investigator needs to build his case. In Rwanda, a country where extended family structure is the norm, most witnesses are either, relatives, friends, in-laws or acquaintances of the accused, and each may have a reason to protect the accused and therefore lie or offer half-truths to investigators.
Investigators must avoid asking leading questions loaded with preconceptions on the role of the insider witness or other perpetrators under investigation. Questions that provide a ‘yes’ or ‘no’ answer only tend to obstruct more objective findings. When interviewing witnesses, including insider witnesses, an investigator must not (a) lose control during the course of the interview; (b) ask leading questions if he does not know the answer; (c) change the language of interview from, for example, English to French while the interview is in progress; (d) must note the body language, or shifting behaviour of the witness and (e) work in pairs and never alone. Best practice requires that an investigator (a) conducts prior extensive study and analysis of background, motivation and context; and (b) not allow the witness to control the interview through his substantial knowledge or superior status. OTP investigators interviewed many accused who included a prime minister, several cabinet ministers, senior political leaders and military generals, diplomats, university professors and other professionals. Some of the most responsible accused had higher educational qualifications and substantive experience in public life than investigators hence the importance of thorough preparatory work.
As a multi-disciplinary team, the investigators are able to cover different aspects of the planning of crimes and background of suspects including insider witnesses. Interpreters are part of the multi-disciplinary team. They are Rwandans or persons fluent in speaking, writing and reading Kinyarwanda. The selected interpreters are screened to determine their relationship with the accused and other key witnesses, including determining their reliability and credibility. Since the genocide in Rwanda was ethnically based, it was necessary to discreetly determine the ethnicity of a local interpreter. The interview of a witness, particularly an insider witness, is not a cross-examination but is grounded on building trust without becoming too friendly to the witness. The objective is limited to obtaining information that may incriminate or exculpate an accused but not to challenge the witness or to judge his actions. In other words, an investigator seeks to collect as much accurate information as the circumstances permit without judging the witness. However, if other evidence from prior interviews of other witnesses, documentary evidence or any credible source are available, and such information may contradict what the witness states, the investigator must confront the witness with such information and seek explanations of the context and the role of the witness vis-à-vis that of the accused. The purpose of the interview is not to determine guilt or innocence of the witness, but to gather as much relevant and credible information as is possible. An investigator must gather both incriminating and exculpatory evidence.
During the interview, it is necessary to keep a record of the statement, whether audio or video, to assure utmost accuracy of the process and to be able to confront the witness during the course of the interview or at subsequent meetings. An investigator must be alert and caution the witness if he is about to, or indeed incriminates himself. Upon completion of the interview, the investigator must be de-briefed, preferably by a senior investigator or a prosecuting attorney for critical evaluation and assessment of the quality of the statement and any discrepancy or contradictions that may exist or emerge. Information provided by insider witnesses assists in determining the pattern of crime. Often, the testimonies of insider witnesses are supported by reports, notes or diary that they produced at the relevant time. These contemporaneous reports tend to have high credibility and veracity as they were prepared at a time when the insider witness did not anticipate any possibility of it being used at a trial.
As regards relevance and credibility of statements recorded from witnesses, when interviewing potential witnesses, investigators had to always bear in mind that each of the counts in an indictment must disclose distinct elements of a contextual nature. For example, in the case of crimes against humanity, evidence must seek to prove the existence of a ‘widespread or systematic attack against a civilian population’; proof of material element, such as ‘torture’ or assault, and proof of the mental element, that is, ‘intent’ to commit the crime as charged. These elements of crimes against humanity, for example, are well grounded in international criminal law and do not change from case to case, although their precise meaning, depending on the factual and contextual evidence as presented by the prosecution, is subject to interpretation by the Chamber. For these reasons, findings of guilt at trial will only follow where the prosecutor has proved each of the elements of an alleged crime beyond a reasonable doubt.
It logically follows that to assist prosecuting attorneys in the presentation of their cases before the chambers; investigators must have not only reasonable knowledge of substantive law, but specifically good knowledge of the elements of the crime which in turn allows them to ask relevant questions. Spending time listening to horror stories from witnesses without knowledge of the essential elements of the crimes does not enhance the quality of the statement gathered. Further, poorly conducted interviews tend to lead to multiple interviews of potential witnesses by different investigators and prosecuting attorneys as they constantly return to the same witnesses to clarify past statements. As more statements seeking clarification are recorded, more risks of contradictions in earlier statements made by the same witnesses to different investigators will emerge. The result is that such statements tend to lose their credibility and do not assist prosecuting counsels in presenting credible cases.
Besides knowledge of substantive elements of crimes, investigators must be conversant with various modes of participation as provided in Article 6 of the Statute; an article that sets forth forms of liability. Article 6(1), for example, imposes individual criminal responsibility on a person who planned, instigated, ordered, committed or otherwise aided and abetted in the planning, preparation or execution of a crime of genocide, crimes against humanity and war crimes. Article 6(3), on the other hand, imposes individual criminal responsibility on a superior for acts or omissions of his subordinates if he knew or had reason to know that his subordinate was about to commit such acts or had done so and the superior failed to take the necessary and reasonable measures to prevent such acts or to punish the perpetrators thereof. It is pertinent that an investigator understands basic legal principles for establishing individual criminal responsibility. Article 6 on modes of responsibility for example, is characterized by a distinct set of legal requirements, each of which must be proved by the prosecutor through the introduction of sufficient evidence before a Chamber finds an accused guilty.
Overall, it is the duty of the prosecutor to prove, beyond reasonable doubt, the substantive elements of crimes as stipulated in the Statute. Investigating serious international crimes is always problematic because of the complexity of substantial and procedural law as well as human trauma experienced by victims who are often prosecution witnesses. Investigating genocide in Rwanda was exceptionally difficult because the OTP began investigations long after the crimes were committed.
- Selection of cases
From the outset, the prosecutor faced unprecedented problems in the selection of cases because crimes committed in Rwanda were too massive, widespread and systematic to be subjected to full investigations and prosecution on every incident and crime. The United Nations Commission of Experts, Human Rights Watch, Amnesty International and Africa Rights for example, have all documented atrocities committed during the Rwanda crisis. The numbers of victims was mind boggling. There were many perpetrators who deserved to be investigated. There were many witnesses who were ready and willing to testify. Ultimately it was humanely impossible to investigate every incident and interview all witnesses. To prosecute those most responsible, the prosecutor adopted a policy of selecting cases that involved participation of senior political and military leaders. Also included were individual perpetrators, regardless of their ranks provided their actions involved incidents and offences that were considered serious. In selecting incidents and cases, the prosecutor identified representative samples that covered the geographical spread of Rwanda. This approach required the prosecutor to limit and select evidence that was considered, for pragmatic reasons, (such as limited financial and human resources, security of investigators, secondary victimization of victims and witnesses; and witness fatigue), to exclude other incidences and crimes that were equally serious and deserving of further investigations even when evidence was available or could be obtained.
By adopting a policy of selection of cases as opposed to prosecuting all perpetrators, the prosecutor was aware that some senior perpetrators may never be tried. Some perpetrators were probably already dead, or if alive, could never be found; some may not be tried due to lack of resources; and some due to criteria that were subsequently developed by the prosecutor to identify, and prosecute those most responsible before the closure of the tribunal. Absent a policy on selection of cases, it would have probably taken more than fifty years to prosecute every single perpetrator. It was logical to leave the bulk of the cases for investigations and prosecutions to Rwandan national courts, because unlike the ICTR, national courts do not close However, in the interests of justice and transparency, an open selection process was necessary to permit victims and perpetrators to satisfy themselves of the fairness of the process. A public and open process would demonstrate to the public, for instance, that no political pressure or bias was exerted on the prosecutor to influence his decision to select or reject other equally major perpetrators for prosecution. Arguably, the manner of case selection can substantially affect the way in which the criminal justice process is perceived by victims and other persons affected by the atrocities. It is for this reason that formal criteria are an essential tool for a more rational and coherent selection of cases for trials even if victims and other affected persons may not necessarily agree with the criteria adopted. Significantly however, the process can assist the prosecutor in mapping and ranking cases so that those most suitable for trial are prosecuted first. To meet these objectives, the prosecutor considered whether criteria for the selection of cases, if adopted, should remain an internal OTP document or should be made public. During several meetings with his senior staff, Prosecutor Jallow considered what the criteria should consist of and how flexible, if at all, the criteria should be?
In the course of his meetings with OTP senior staff the prosecutor revised the old Gamma list which had more than one thousand perpetrators identified for prosecutions. The Gamma list included senior leaders who gave orders, failed to prevent the commission of the crimes, aided or abetted the commission of those crimes, or were part of a group with a common criminal purpose. There were also mid-level perpetrators who coordinated works of actual perpetrators by re-issuing instructions given to them by their superiors to direct perpetrators. Some of the low-level perpetrators, most of whom were sadistic and vicious were also on the Gamma list. At an OTP management meeting at Amboseli, the Gamma list was reviewed and revised downwards, considerably reducing the numbers of perpetrators identified for prosecution.
Apprehending those most responsible was difficult because after the genocide, the majority of senior perpetrators had fled to states where the political and military leadership protected them. By default, it became easier for the prosecutor to arrest and prosecutes mid and low-level perpetrators because many of them had no safe and secure places to hide. Thus, Jean-Paul Akayesu, a mid-level perpetrator, was the first to be prosecuted and convicted for, inter alia, genocide. As a result, based on the Akayesu precedent, many mid-level perpetrators were selected for prosecution. For those in hiding, the prosecutor created a Tracking Team, under the command of a senior investigator to search for, and in cooperation with national government officials, arrest them.
The prosecutor had another unexpected challenge: there was too much evidence of genocide, crimes against humanity and war crimes. After careful consideration, the prosecutor opted to prioritize prosecution of genocide crimes. The rationale was that genocide was the most profound – the crime of all crimes – compared to crimes against humanity and war crimes. The deaths of an estimated eight hundred thousand civilians or more were mainly caused by the genocide and it is primarily this fact that persuaded the prosecutor to focus on the prosecution of genocide perpetrators.
Prior to appointment of Hassan Jallow as ICTR prosecutor in 2003, Carla Del Ponte, then ICTR/ICTY prosecutor, made serious attempts to arrest and prosecute senior RPF officials who were identified as being responsible for crimes against humanity and war crimes based on the evidence gathered by OTP special investigations unit. According to Del Ponte, her efforts to prosecute alleged RPF perpetrators were frustrated by the Rwanda government and her allies, the United States and the United Kingdom, amongst others. As a cover, the US and the UK pushed through adoption of Resolution 1503(2003). The resolution created separate offices for ICTY and ICTR prosecutors. Carla Del Ponte was then removed as the ICTR prosecutor but retained as ICTY prosecutor with the result that she was taken off RPF cases.
Hassan Jallow, Del Ponte’s successor, opted not to prosecute RPF at the ICTR but to transmit their cases to Rwanda for trial by the Rwanda judiciary. This decision was criticized and considered by some critics as encouraging impunity by not prosecuting RPF alleged perpetrators at the ICTR. Human Rights Watch (HRW) urged Prosecutor Jallow to investigate and prosecute all sides to the armed conflict. Absent any prosecution of the RPF, HRW argued that the ICTR process would in future be viewed as ‘victor’s justice’. Prosecutor Jallow, writing to Kenneth Roth, executive director of Human Rights Watch, defended his decision. Prosecutor Jallow reasoned that since it is not practically possible to prosecute every person who committed serious crimes including genocide, or every Rwandan who committed genocide, crimes against humanity and war crimes in the neighbouring states, the focus should be on those who bear the highest responsibility. This focus group involved senior political and military leaders. However, there were also many mid and low-level perpetrators who were alleged to have committed serious crimes. This category of perpetrators was also selected for prosecution. The prosecutor sought to ensure that selected suspects represented every level of the hierarchy in the political parties and the military establishment. The objective was to seek to guarantee, as far as possible, that perpetrators who participated at every level in the commission of the atrocities were covered and the totality of different types of involvement documented.
Rationale for including or excluding a perpetrator from the prosecutor’s list of selected cases included consideration of (a) interests of victims (b) the Rwanda community at large; (c) whether the evidence was sufficient to present a reasonable prospects for conviction; (d) whether upon objective evaluation of the credibility and mental stability of a witness, such a witness was reliable; and, (e) whether the evidence was admissible, substantial, and reliable to the extent that it could establish a prima facie case at the indictment confirmation hearing. This policy, although it covered many perpetrators, it also excluded some of the one thousand suspects in the original Gamma list and several thousand perpetrators who deserved prosecution by the ICTR, Rwanda government and other national jurisdictions but were not included in the original Gamma list. The gap was filled by the use of Rule 11bis, a process which permitted the prosecutor to transfer indicted accused, by order of Referral Chamber, to national jurisdictions, including Rwanda. An additional process used by the prosecution was transmission of cases of accused persons who had not been indicted by the ICTR, or suspects who were resident in foreign countries but also not indicted, to national jurisdictions for prosecution. The majority of such cases, however, were transmitted to Rwanda.
The policy on the selection of cases resulted in identification of leaders of political parties; ministers and prefects for prosecution; senior FAR officers, leaders and organisers of Civil Defence and the Interahamwe. A second category was the media. The prosecutor chose to prosecute owners, publishers, directors and senior employees of the media with focus on RTLM radio and Kangura newspaper. A third category was church officials. Many civilians, both Tutsis and moderate Hutus, were killed in churches and schools under control and management of the church. A fourth category involved perpetrators of rape. Individuals identified as having committed rape, directly or as a superior, were selected for prosecution regardless of whether they were mid or low-level, or senior officials in the military, political, media or church leadership. This strategy was, however, adopted after the Akayesu judgement which ruled, inter alia, that rape and sexual violence may be used as means in the commission of genocide, crimes against humanity and war crimes.
Due to the widespread and systematic crimes committed throughout Rwanda, the prosecutor ensured that there was geographical representation of all areas of Rwanda to avoid the perception of bias, favouritism or discrimination which could suggest that the prosecutor chose certain perpetrators from a given region and ignored equally serious perpetrators from other regions of Rwanda. The prosecutor therefore took a conscious decision to identify perpetrators across the geographic spread of Rwanda ensuring that perpetrators selected for prosecution covered the entire territory and to the extent possible, represented the different categories of perpetrators in the Rwanda society. A disadvantage of this policy meant that some targets in one location in Rwanda may have to be excluded from the list of perpetrators selected for prosecution to accommodate other perpetrators from other administrative regions of Rwanda to meet the criteria of geographical representation. Overall, it is submitted that the policy of selection of cases met the objective of the ICTR Statute to prosecute persons most responsible for serious violations of international humanitarian law.
- Security Council Resolutions 1503(2003) and 1534(2004) and prosecutor’s discretion on selection of cases
After almost ten years of the ICTR and ICTY experiment with a joint chief prosecutor, on 28 August 2003 the Security Council, by Resolution 1503 amended Article 15 of the Statute, created a separate and an independent prosecutor for the ICTR, and laid down guidelines for closing the tribunal. The resolution ‘Calls on the ICTY and the ICTR to take all possible measures to complete investigations by the end of 2004, to complete all trial activities at first instance by the end of 2008, and to complete all work in 2010 (the Completion Strategies)’. Six months later, on 26 March 2004 the Security Council adopted Resolution 1534 and directed both the ICTY and ICTR prosecutors to, inter alia, (a) ‘review the case load of the ICTY and ICTR respectively in particular with a view to determining which cases should be proceeded with and which transferred to competent national jurisdictions, as well as the measures which will need to be taken to meet the Completion Strategy referred to in resolution 1503(2003) […]’, and (b) ‘in reviewing and confirming any new indictments, to ensure that any such indictments concentrate on the most senior leaders suspected of being most responsible for crimes within the jurisdiction of the relevant Tribunal as set out in resolution 1503(2003)’.
Critics argue that Resolution 1534 (2004) was adopted by the Security Council because Carla Del Ponte, the ICTY prosecutor, had made her view known that she would issue thirty new indictments, contrary to the wishes of the Security Council. To block the prosecutor, the Council, relying on the UN Secretary General’s report on the establishment of the SCSL, recommended the adoption of the phrase ‘those who bear the greatest responsibility’. The phrase was to be understood as a limitation on the number of accused identified for prosecution by reference to their command authority and the gravity and scale of the crime. This meant that the ICTY shall prosecute fewer perpetrators by focusing on ‘those who bear the greatest responsibility’. A second report relied on by the Council was communications between the president of the Security Council and the UN Secretary General. It suggested that the phrase ‘those who bear the greatest responsibility’ must refer only to those who played a leadership role in the conflict. Resolutions 1503(2003) and 1534(2004) narrow the scope of Article 1 of the ICTR Statute by use of the phrase ‘[…] to prosecute persons responsible for serious violations […]’ with the result that it limits a prosecutor’s discretion on whether to charge criminal offences and what those offences should be. Cote persuasively argues that in international justice, the discretionary powers of the prosecutor lie at the very heart of the system. It is the prosecutor who decides whether or not to conduct investigations, which individuals should be charged, and with what offences.
The prosecutor has the freedom to decide whether to commence investigations, to determine whether investigations reveal commission of any crime as defined under the law, to determine which person committed the crimes, what crimes and which perpetrator to charge. Even after a trial has started, a prosecutor still has the discretion to decide what evidence to lead, what witnesses to call, what questions to ask, what interlocutory applications to make, what objections to raise and whether or not to drop some charges in return for a guilty plea; and sometimes to withdraw charges altogether. Under common law or Roman-Dutch jurisdictions, the prosecutor has discretion whether to prosecute and with what offence. The prosecutor also has the power to withdraw or discontinue any prosecution, and in doing so, the prosecutor does not have to give any reason. Under civil or continental law systems however, the exercise of prosecutorial discretion is more restrictive. The prosecutor is, for example, duty bound to prosecute provided there is sufficient evidence to justify prosecution. He or she has no discretion in the matter because the duty to prosecute is based on the principle of legality and therefore failure to prosecute is a criminal offence. To that extend, the Security Council’s resolutions seeking to limit powers of the ICTY/ICTR prosecutor is, prima facie, not illegal or unlawful because the Rules of Procedure and Evidence (RPE) already impose a number of limitations on the prosecutor.
Under the RPE, while the prosecutor may amend an indictment without an order of a judge any time before the confirmation of the indictment, after confirmation, the prosecutor may only amend with the leave of the judge who confirmed the indictment or, during trial, the Trial Chamber. Similarly, the prosecutor has the power to withdraw an indictment at any time before its confirmation, but after confirmation, it can be withdrawn only with leave of the confirming judge or, at trial, the Trial Chamber. Further, the prosecutor’s discretion is limited in respect of his authority to determine whether the evidence proves the existence of a prima facie case because a Trial Chamber retains the right to review or reverse the prosecutor’s findings. It is mandatory that a judge reviews and confirms an indictment before an accused person can be tried on it. The reviewing judge, or at trial, a Trial Chamber may decline to confirm an indictment on the ground that it does not disclose a prima facie case, dismiss the indictment and discharge the accused. Overall, requirement that an indictment is submitted by the prosecutor and confirmed by a judge helps to check the prosecutor’s exercise of his discretionary powers and guard against possible abuse. Besides limitation of the prosecutor’s authority already provided in the Statute and RPE, Resolutions 1503(2003) and 1534(2004) introduced additional limitations to the scope of prosecutorial discretion. For example, by directing the prosecutor to focus only on the highest level of political and military leadership most responsible for crimes committed, if strictly adhered to, prosecution of lower-level perpetrators responsible for mass murders would fall outside the scope of the resolutions.
Pursuant to Resolution 1534(2004), the ICTY judges, through the ICTY bureau, a body elected by the same judges, revised Rule 28 of RPE by adopting a new Rule 28A of the RPE. The new rule gave to the judges authority to review all instruments submitted by the prosecutor to determine whether ‘the indictment, prima facie, concentrates on one or more of the most senior leaders suspected of being the most responsible for crimes committed within the jurisdiction of the Tribunal’, failing which it will not be confirmed. However, the prosecutor objected to Rule 28A on the ground that it was ultra vires the ICTY Statute. The prosecutor’s objection did not stop the judges to proceed with implementation of Rule 28A. Although the purpose of Rule 28A was laudable and in line with the Security Council Resolution 1534(2004), the amendment raised legal issues as to whether Rule 28A is indeed an amendment of the ICTY Statute itself; and if so, whether the judges, based on Resolution 1534(2004) were competent to adopt rules that limit the exercise of prosecutorial discretion. Legal opinion is divided on the extent to which Rule 28A may have limited the ICTY prosecutor’s discretionary authority. There is, however, a general consensus that there was some limitation to the prosecutor’s authority.
When the ICTR prosecutor issued a policy statement on case selection, he explained that he would focus on those ‘in position of leadership’ consistent with Resolution 1534(2004). In the statement, the prosecutor listed criteria he meant to apply to determine who are ‘the most senior leaders’ and those who may be classified as mid or low-level perpetrators but bore ‘the greatest responsibility for genocide’. Thus, unlike restrictions imposed on the ICTY prosecutor by Rule 28A, the ICTR judges did not adopt similar rules. They permitted the prosecutor to continue to use his discretionary authority in the selection of cases. However, as Cote argued, the narrowing of the mandate of the ICTY and the ICTR was a reminder to every one of the ad hoc nature of these tribunals. The impact of Resolution 1534(2004) on the ICTY/ICTR Statutes originally intended as ephemeral instruments now had to take the new resolution into account when selecting cases for prosecutions, failing which they would be taken to task by the judges first of all and then by the Security Council.
Absent similar provision in the ICTR Statute, it remained the responsibility of the prosecutor to identify suspected perpetrators for trial. In theory the prosecutor could include for trial perpetrators who fall outside the class of senior leaders depending on whether crimes committed by ‘minor’ perpetrators have significant impact for victims. The ICTR prosecutor retained the power to indict low-level perpetrators who did not necessarily hold senior positions or were part of a formal chain of command but exercised power or influence beyond their official position. Unlike his counterpart at the ICTY whose powers were limited by Rule 28A, the ICTR prosecutor continued to exercise prosecutorial discretion.
Overall, while the ICTR prosecutor took into account relevance of the phrases ‘in positions of leadership’ and those who bore ‘the greatest responsibility for genocide’, as stipulated in Resolution 1534(2004), he also considered severity of the crime or its massive scale alleged to have been committed in the case selection process. Thus, perpetrators ‘in positions of leadership’ and those who bore ‘the greatest responsibility for genocide’ denotes both a leadership or authority position held by the accused at the time the offence was committed, and a sense of the gravity, seriousness or massive scale of the crime. A low-level official who engages in systematic criminal acts such as widespread rape and sexual violence could be indicted regardless of his official position in the chain of command.
The prosecutor was of the view that strict adherence to a policy of only investigating and prosecuting those in senior leadership positions could have the effect of fostering a sense of impunity among lower-ranking perpetrators and set a bad precedent by encouraging them to continue committing abuses. By contrast, pursuing those officials further down the chain of command, of civilians outside the official chain of command, could have a positive impact for victims. At a practical level, building cases against those in senior leadership positions was extremely difficult due to challenges associated with establishing links between crimes committed by the actual perpetrators and the decisions taken by senior officials, most of the time, miles away from the scene of crime. Aggressively pursuing lower-level perpetrators tends to provide the necessary links and lead to senior perpetrators. Thus, it may be necessary to pursue some lower level perpetrators down the chain of command.
Ideally, the process of identifying senior perpetrators, or ‘those most responsible’ based on the ICTR experience, starts when investigators consciously and deliberately commence tracing victims or eye-witnesses to the crimes and encourage them to describe in considerable detail circumstances under which the crimes were committed, where, how and by whom. Thus, victims are the primary source of evidence relied on in criminal prosecutions. Without testimony of victims describing their inhumane treatment at the hands of the perpetrators, the trial itself would become abstract and detached from the real facts as it occurred. In practice however, only a small fraction of the number of victims gave testimony to investigators as demonstrated by the genocide of the Tutsi where more than 800 000 victims could not be interviewed because they were dead. These constrains while challenging, can be overcome by evidence of survivors, documentary evidence, particularly contemporaneous materials provided by human rights organisations, journalists, UN agencies, particularly as and when confidentiality agreements permit these organizations and individuals to do so. Overall, Resolutions 1503(2003) and 1534(2004) did not have a negative impact on the ICTR prosecutor’s discretionary powers to initiate investigations and to select cases.
Investigation and selection of cases are two important components for successful prosecution. If not properly handled, many senior perpetrators are more likely to escape prosecution. In the conduct of investigation and selection of cases, the ICTR faced serious problems at the beginning of its work despite the good intentions of the United Nations, national governments and human rights organizations, journalists including individuals who freely handed over information to the prosecutor and provided support. Gradually, as experienced lawyers joined the OTP, the quality of the work of the OTP improved. This category of qualified and experience lawyers continued to bear disproportionate burden in the investigations and conduct of cases compared to the less-performing members of staff who were recruited under the liberal UN recruitment rules. Besides their routine investigations and court work, the more experienced lawyers continued to provide necessary in-house training to other staff thus improving the quality of the OTP work and performance. Successive prosecutors continued to use their expertise and that of other outside experts by regularly organizing in-house training on substantive and procedural law. The result is a rich body of jurisprudence developed by the ICTR for the good of humanity. The ICTR prosecutors’ tenacity created conditions for the success of the tribunal after a very difficult start.
*Chief of Appeals and Legal Advisory Division, Office of the Prosecutor, International Criminal Tribunal for Rwanda. The views expressed and personal and do not reflect that of the Prosecutor, the OTP, the ICT or the United Nations.
 ICTR Statute annexed to UN Doc. S/RES/955 (1994).
 Goldstone, R.J, For Humanity – Reflections of a War Crimes Investigator (Yale University Press 2000), 113.
 Zacklin, R ‘The Failings of Ad Hoc International Tribunals’ (2004) 2 Journal of International Criminal Justice 541-545.
 Goldstone 2000, supra note 2, 81.
 Article 15 ICTR Statute.
 Goldstone 2000, supra note 2, 89.
 Article 17 ICTR Statute.
 Article 17(4) ICTR Statute.
 Article 18 ICTR Statute.
 Article 15(4) ICTR Statute.
 Article 15(5) ICTR Statute.
 Article 30 ICTR Statute.
 UN Doc. S/RES/ 1503 (2003).
 UN Doc. S/RES/1534 (2004).
 My experience as chief of appeals and legal advisory division (ALAD) at the OTP ICTR, among other things, in the conduct of investigations and case selection, provide examples relied on in writing this chapter.
 Muna, B, ‘The Early Challenges of Conducting Investigations and Prosecutions before International Criminal Tribunal for Rwanda’ Colloquium of Prosecutors of International Criminal Tribunals – The Challenges of International Criminal Justice (A Publication of ICTR in collaboration with Brandeis University, USA 2005), 15.
 Goldstone 2000, supra note 2, 86.
 Ibid, 85.
 Del Ponte, C in collaboration with Sudetic, C, Madame Prosecutor – Confrontations with Humanity’s Worst Criminals and the Culture of Impunity (Other Press 2009), 134.
 Ibid, 134.
 Ibid, 134.
 Ibid, 135.
 Article 30 ICTR Statute.
 Goldstone 2000, supra note 2, 104-105.
 Goldstone narrates one of the many examples of his frustration with the UN in New York as follows: ‘I remember to my surprise when I was told after arriving in New York that the funding for my visit there and to The Hague had not yet been approved. I could not be reimbursed in New York, as previously informed; I would have to await payment in The Hague. At the request of the Office of Legal Affairs I had paid the air fare out of my own pocket.’ Goldstone 2000, supra note 2, 77.
 Article 15(4) ICTR Statute.
 Article 15(5) ICTR Statute.
 Goldstone 2000, supra note 2, 77.
 Zacklin 2004, supra note 3, 541.
 Del Ponte 2009, supra note 21, 69.
 Goldstone 2000, supra note 2, 114.
 Ibid, 80.
 Article 31 ICTR Statute.
 UN Doc. S/RES/1503 (2003).
 UN Doc. S/1534 (2004).
 Article 17(1) ICTR Statute.
 Jallow H.B, ‘Prosecutorial Discretion and International Criminal Justice’ (2005) 3 Journal of International Criminal Justice 145-161.
 See supra note 15.
 Del Ponte, C, ‘Prosecuting the Individuals Bearing the Highest Responsibility’ (2004) 2 Journal of International Criminal Justice 516-519.
 The Prosecutor v. Vincent Rutaganira ICTR-95-1C-T (Judgement, 14 March 2005). Rutaganira was a conseiller communal of Mubuga sector; The Prosecutor v. Joseph Serugendo ICTR-2005-84-I (Judgement, 12 June 2006). Serugendo was a member of the governing board of Radio Television Libre des Mille Collines (RTLM) and advisor to RTLM radio station. He was also a member of the Interahamwe.
 See, for example, two books that were very useful to the OTP: Rwanda: Death, Despair and Defiance (Revised edition, Africa Rights 1995); Genocide in Rwanda: Leave None to Tell The Story (Human Rights Watch 1999; International Federation of Human Rights, Paris 1999).
 The Prosecutor v. Michael Bagaragaza ICTR-05-86-S (Judgement, 17 November 2009). Bagaragaza agreed to plead guilty for committing crime of complicity in genocide (para. 24-26); agreed to cooperate with the prosecutor as an insider witness (paras. 39-40) and in return he was given a light sentence of 8 years imprisonment and sent to Sweden, a country with excellent prison conditions. He has since completed his sentence and released from prison.
 The Prosecutor v. Jean Kambanda ICTR-97-23-S (Judgement, 4 September 1998). After his guilty plea, Kambanda was sentenced to life imprisonment. He then appealed against both conviction and sentence. The appeal was dismissed and thereafter he declined to cooperate with the Prosecutor and to testify against his former ministers and government officials including military officers. See Jean Kambanda v. The Prosecutor ICTR-97-23-A (Judgement, 19 October 2000).
 The Prosecutor v. Omar Serushago, ICTR-98-39-S (Sentencing Judgement, 5 February 1999). Serushago pleaded guilty and was sentenced to 15 years. He appealed and the appeal was dismissed. See Omar Serushago v. The Prosecutor ICTR-98-39-A (Reasons for Judgement, 6 April 2000). While he continued to cooperate, though reluctantly, Serushago consistently downplayed his role in the genocide; see also Georges Ruggiu who pleaded guilty but consistently downplayed his role while embellishing those of his co-perpetrators. The Prosecutor v. Georges Ruggiu ICTR-97-32-I (Judgement and Sentence, 1 June 2000).
 United Nations, Commission of Experts Established Pursuant to Security Council Resolution 935 (1994) on Rwanda, Final Report, Geneva (25 November 1994).
 Prior to 2005, the term ‘Rwanda genocide’ was often used rather than the ‘Genocide of the Tutsi’. See Prosecutor v. Karemera, Mathieu Ngirumpatse, Joseph Nzirorera ICTR-98-44-AR 73(C) (Decision on Prosecutor’s Interlocutory Appeal of Decision on Judicial Notice, 16 June 2006). The Chamber held that mass killings that took place in Rwanda constituted ‘Genocide of the Tutsi’, paras. 25-38. See also Prosecutor v. Semanza ICTR-97-20-A (Judgement, 20 May 2005), para. 194. In other words, after 2005, it was legally recognized that in Rwanda, genocide was committed against the Tutsi, thus debunking the theory of ‘double genocide’ advanced by most Accused at the United Nations Detention Facility (UNDF) at Arusha.
 Arbour, L, ‘Progress and Challenges in International Criminal Justice’ (1997) 2 Fordham International Law Journal 531.
 Internal OTP document (prosecutor’s daily briefing of December 2004 and January 2005) (on file with the author).
 Jean-Paul Akayesu was bourgmestre of Taba commune from April 1993 until June 1994. In the political and administrative hierarchy in Rwanda during 1994, he was a low-level administrator. See The Prosecutor v. Jean- Paul Akayesu ICTR-96-4-T (Judgement, 2 September 1998); and the appeal judgement see Jean- Paul Akayesu v. The Prosecutor ICTR-96-4-A (Judgement, 1 June 2001).
 Del Ponte 2009, supra note 21, 224-241.
 Reyntjens F, ‘Prosecutorial Policies in the ICTR: Ensuring Impunity for the Victors’, Institute of Development Policy and Management (unpublished copy with the author).
 See Roth, K Executive Director of Human Rights Watch, ‘Letter to the Prosecutor of the International Criminal Tribunal for Rwanda Regarding the Prosecution of RPF Crimes’ dated 26 May 2009, available at: <http://www.hrw.org/node/83536> accessed on 25 March 2015.
 Jallow, H.B, Chief Prosecutor, ICTR’s response to Kenneth Roth’s letter of 26 May 2006 Ref: OTP/2009/P/084, dated 22 June 2009 (copy with the author). Kenneth Roth’s responses to the prosecutor’s reply ‘Letter to ICTR Chief Prosecutor Hassan Jallow in Response to His Letter on the Prosecution of RPF Crimes’ dated 14 August 2009, available at: <http://www.hrw.org/en/news/2009/08/14/letter-ictr-chief-prosecutor-hassan-jallow-respo> accessed on 25 March 2015.
 As required by Article 1 of the ICTR Statute, even lower-level perpetrators of genocide could be excluded from prosecution because the prosecutor’s focus are ‘those most responsible’ for the genocide.
 Rule 11bis(A) of the Rules provides: ‘If an Indictment has been confirmed, whether or not the Accused is in custody of the Tribunal, the President may designate a Trial Chamber which shall determine whether the case should be referred to the authorities of the State: (i) In whose territory the crime was committed; or (ii) in which the accused was arrested; or (iii) Having jurisdiction and being willing and adequately prepared to accept such a case, so that those authorities should forthwith refer the case to appropriate court for trial within that state.’ On prosecutor’s challenges in implementing Rule 11bis, Obote-Odora, A, ‘Transfer of cases from the International Criminal Tribunal for Rwanda to Domestic Jurisdictions’ (2012) 5 African Journal of Legal Studies 147-180.
 Senior political leaders selected for prosecution include Prime Minister Jean Kambanda, supra note 52; cabinet ministers and leaders of various political parties in the interim government that included Joseph Nzirorera, secretary general of MRND party; Edouard Karemera, president of MRND party; cases against cabinet ministers included The Prosecutor v. Emmanuel Ndindabahizi ICTR-2001-71-I (Judgement, 15 July 2004) and on appeal see: Emmanuel Ndindabahizi v. The Prosecutor ICTR-01-71-A (Judgement, 16 January 2007); The Prosecutor v. Eliezer Niyitegeka, ICTR-96-14-T (Judgement, 16 May 2003) and on appeal see: The Prosecutor v. Eliezer Niyitegeka ICTR-96-14-A (Judgement, 9 July 2004); The Prosecutor v. Andre Rwamakuba ICTR-98-44C-T (Judgement, 20 September 2006).
 Senior military leaders selected for prosecution include Colonel Theoneste Bagosora, Gen. Gratien Kabiligi, Lt. Col. Anatole Nsengiyumva, Lt. Col. Francois-Xavier Nzuwonemeye, Maj. Innocent Sagahutu, see: The Prosecutor v. Theoneste Bagosora, Gratien Kabiligi, Aloys Ntabakuze and Anatole Nsengiyumva ICTR-98-41-T (Judgement, 18 December 2008); Lt. Col. Tharcisse Renzaho, see: The Prosecutor v. Tharcisse Renzaho ICTR-97-31-T (Judgement, 14 July 2009); Gen. Augustine Bizimungu, Gen. Augustin Ndindiliyimana, see: The Prosecutor v. Augustin Ndindiliyimana, Augustine Bizimungu, Francis-Xavier Nzuwonemeye and Innocent Sagahutu ICTR-00-56-T (Judgement, 17 May 2011) and Interahamwe leader George Rutuganda, see: The Prosecutor v. Georges Anderson Nderubumwe Rutuganda ICTR-96-3-T (Judgement, 6 December 1999); The Prosecutor v. Georges Anderson Nderubumwe Rutuganda ICTR-96-3-A (Judgement, 26 May 2003).
 Radio and newspaper (media) owners including journalists selected for prosecution include: Ferdinand Nahimana, Jean Bosco Barayagwiza, Hassan Ngeze, see: The Prosecutor v. Ferdinand Nahimana, Jean Bosco Barayagwiza, Hassan Ngeze ICTR-99-52-T (Judgement, 3 December 2003); and Ferdinand Nahimana, Jean Bosco Barayagwiza, Hassan Ngeze v. The Prosecutor ICTR-99-52-A (Judgement, 28 November 2007) and Georges Ruggiu, see: Ruggiu ICTR-97-32-I.
 The church leaders prosecuted at Arusha include Bishop Samuel Musabyimana; see: The Prosecutor v. Samuel Musabyimana ICTR-01-62 (died before trial); Pastor Elizaphan Ntakirutimana, see: The Prosecutor v. Elizaphan Ntakirutimana & Gerard Ntakirutimana ICTR-96-10-T & ICTR-96-17-T (Judgement, 21 February 2003) and on appeal, see: Elizaphan Ntakirutimana & Gerard Ntakirutimana v. The Prosecutor ICTR-96-10-A & ICTR-96-17-A (Judgement, 13 December 2004) (died soon after his release from prison upon completion of his 10-year sentence); Father Emmanuel Rukundo, see: The Prosecutor v. Emmanuel Rukundo ICTR-2001-70-T (Judgement, 27 February 2009) and on appeal, see: Emmanuel Rukundo, v. The Prosecutor ICTR-2001-70-A (Judgement, 20 October 2010); and Father Athanase Seromba, see The Prosecutor v. Athanase Seromba ICTR-2001-66-T (Judgement, 13 December 2006) and on appeal, see: Athanase Seromba v. The Prosecutor ICTR-2001-66-A (Judgement, 12 March 2008) both Roman Catholic priests were convicted. One other clergyman was transferred to France, under Rule 11bis for prosecution in that country. He is Father Wenceslas Munyeshyaka, see: The Prosecutor v Father Wenceslas Munyeshyaka, ICTR-05-87-I (Ruling on Prosecutor’s Request for Referral of Wenceslas Munyeshyaka’s indictment to France, 20 November 2007).
 Akayesu ICTR-96-4-T; Akayesu ICTR-96-4-A. After Akayesu, there were other prosecutions for rape and sexual violence. The most notorious cases are The Prosecutor v. Mikael Muhimana; see: The Prosecutor v Michael Muhimana ICTR-95-IB-T (Judgement, 28 April 2005), and on appeal see: Michael Muhimana v. The Prosecutor ICTR-95-IB-A (Judgement, 21 May 2007) and The Prosecutor v. Pauline Nyiramasuhuko, see: The Prosecutor v. Pauline Nyiramasuhuko, Arsene Shalom Ntahobali, Sylvain Nsabimana, Alphonse Nteziryanyo, Joseph Kanyabashi and Eli Ndayambaje ICTR-98-42-T (Judgement, 24 June 2011).
 Obote-Odora, A, ‘Case Selection and Prioritization Criteria at the International Criminal Tribunal’ in Bergsmo, M, (ed.), Criteria for Prioritizing and Selecting Core International Cases; Forum for International Criminal and Humanitarian Law (FICHL) (International Peace Research Institute PRIO 2009), 41-54, at 49.
 UN Doc. SC/RES/1503 (2003), para. 8.
 Ibid, para. 7. See Jallow H.B, ‘Challenges of International Criminal Justice: The ICTR Experience’ (unpublished, copy with the author).
 UN Doc. SC/RES/1534 (2004), paras. 4 and 5.
 Raab, D, ‘Evaluating the ICTY Completion Strategy’ (2005) 3 Journal of International Criminal Justice, 84.
 See: Report of the Secretary General on the Establishment of a Special Court for Sierra Leone, S/2000/915, 14 October 2000, para. 29.
 See: Letter dated 22 December 2000 from President of the Security Council addressed to the Secretary General S/2000/1234, para. 1; and letter dated 12 January 2001 from Secretary General to the President of the Security Council, S/2001/40, para. 2.
 Jallow 2005, supra note 39, 145.
 Cote, L, ‘International Criminal Justice Tightening Up The Rules of the Game’ (2006) 88 International Review of the Red Cross 133-144.
 Rule 50 of RPE.
 Rule 51 of RPE.
 Rule 49 of RPE.
 Johnson, L.D, ‘Closing an International Criminal Tribunal while maintaining International Human Rights Standard and Excluding Impunity’ (2005) 99 American Journal of International Law 158.
 Mundis, D.A, ‘The Judicial Effects of the Completion Strategy on the Ad Hoc International Criminal Tribunals’ (2005) 99 American Journal of International Law 142.
 Angermaier, C, ‘Case Selection and Prioritization Criteria in the Work of the International Criminal Tribunal for the Former Yugoslavia’ in Bergsmo M, (ed.), Criteria for Prioritizing and Selecting Core International Cases; Forum for International Criminal and Humanitarian Law (FICHL) (International Peace Research Institute PRIO 2009), 41-54, 29-39.
 Jallow 2005, supra note 39, 145-61. See also Annex to UN Doc. S/2004/341 (2004), Completion Strategy of the International Criminal Tribunal for Rwanda, para.14.
 Cote 2006, supra note 72, 140.
 Ibid. 140-141.
 Some organizations, such as the International Committee of the Red Cross (ICRC), are reluctant to provide information to the prosecutor or to testify as prosecution witnesses for fear that they may compromise their neutrality and put the life of their personnel on arms way, particularly during on-going armed conflicts. National Security Organizations of many States decline to cooperate with the prosecution because they do not want to compromise their sources some of whom may be double agents. Further, some UN agencies, for example, the UN peace keeping force may cooperate with the prosecution where there are specific agreement(s) between the prosecution and the organization.