UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA
August 17, 2006
UNITED STATES OF AMERICA,
FRANCOIS KARAKE, ET AL., DEFENDANTS.
The opinion of the court was delivered by: Ellen Segal Huvelle United States District Judge
Defendants Francois Karake, Gregoire Nyaminani and Leonidas Bimenyimana face a four-count indictment relating to the March 1, 1999 killings of two American tourists in Bwindi Impenetrable National Forest ("Bwindi") in southwestern Uganda. The attack on Bwindi that resulted in the deaths of the Americans was carried out by the Liberation Army of Rwanda ("ALIR"). Several other tourists and one Ugandan national park guard were also killed as a result of ALIR's attack. The United States government asserted extraterritorial jurisdiction to prosecute defendants, all of whom are Rwandan nationals and former members of ALIR. The indictment charges defendants with two counts of murder, 18 U.S.C. § 2332(a), conspiracy to commit murder, 18 U.S.C. § 2332(b), and using a firearm during a crime of violence, 18 U.S.C. § 924(c) & (j). If convicted, defendants face the death penalty.
Defendants have moved to suppress the statements they made to Rwandan and American officials during the course of the investigation into the attack at Bwindi. The investigation spanned roughly four years and required the cooperation of law enforcement officials from at least four countries -- Uganda, Rwanda, the United Kingdom and the United States. The investigation produced a total of 29 statements that defendants seek to suppress: 15 by Nyaminani, 7 by Karake, and 7 by Bimenyimana. These many statements fall into two categories: (1) statements made to Rwandan officials out of the presence of any Americans; and (2) statements made in the presence of both American and Rwandan investigators. Except for Nyaminani's two earliest statements, all of the statements at issue were made while defendants were housed at what the Rwandans have referred to as a military "barracks," known as Kami, located outside the Rwandan capital of Kigali. (5/3 p.m. tr. at 65.) Kami Camp, approximately 200 acres in size, houses between 90 and 120 Rwandan soldiers. (Id. at 52.) According to Captain Alex Kibingo, who was in charge of Kami during the relevant time period, the camp was used to store military equipment such as uniforms, guns and bullets. (Id. at 51.) Kami also served as a detention center for Rwandan soldiers who were subject to disciplinary action (id. at 53-54), and captured ALIR soldiers prior to their transport to repatriation camps in Rwanda's Ruhengeri province.*fn1 (Id. at 11.) Most, but not all, of the statements obtained by Rwandan officials out of the presence of Americans were taken at Kami by Kibingo. Defendants were transported from Kami to the Rwandan National Police Headquarters at Kacyiru ("Police Headquarters" or "Kacyiru") for all of the interrogations in which American investigators participated.
Defendants advance two principal arguments. First, they argue that their statements were the product of physical and psychological coercion, resulting from both their conditions of confinement and their treatment while in Rwandan custody, and were therefore obtained in violation of the due process clause of the Fifth Amendment. Second, defendants claim that the Miranda warnings issued by American interrogators were inadequate to permit defendants to make a knowing and voluntary waiver of their Fifth Amendment rights against self-incrimination. Moreover, defendants advance a corollary argument that a joint venture existed between the United States and Rwandan governments, and therefore, they were entitled to the protections provided by Miranda v. Arizona, 384 U.S. 436 (1966), with respect to the interrogations conducted outside the presence of any American officials. See, e.g., Reid v. Covert, 354 U.S. 1, 5-7 (1957); United States v. Covington, 783 F.2d 1052, 1056 (9th Cir. 1986). Defendants' motion therefore presents several legal questions: 1) Has the government demonstrated by a preponderance of the evidence that each confession was voluntary within the meaning of the Fifth Amendment? 2) Was there a knowing, voluntary and intelligent waiver of Miranda during the joint interrogations conducted by Americans and Rwandans?*fn2 3) Did a joint venture exist between the American and Rwandan governments, and, if so, what was the scope and legal effect of that joint venture? As all parties agree, the resolution of these issues is intensely fact-specific. See, e.g., Haynes v. Washington, 373 U.S. 503, 513-14 (1963) (voluntariness determination requires examination of the "totality of the circumstances"); Covington, 783 F.2d at 1056 ("Whether or not United States officials are substantially involved, or foreigners are acting as their agents or employees, is a question of fact to be resolved in each case."). But, in determining the answers to these questions, it is not the Court's role to decide the issue of guilt or the truthfulness of the many confessions. See United States v. Bowie, 198 F.3d 905, 912 (D.C. Cir. 1999) ("Suppression hearings do not determine a defendant's guilt or punishment.").
The Court conducted a five-week evidentiary hearing to determine the suppression issues. Over the course of 22 days of testimony and one day of oral argument, the Court heard from 19 witnesses (11 defense, 8 prosecution), including 11 Rwandan nationals, 8 of whom testified in the Rwandan language Kinyarwanda. The Court received written reports and heard testimony regarding defendants' physical and mental condition from six expert witnesses -- two psychiatrists, two forensic pathologists, a dermatologist and an internist with expertise in the treatment of torture survivors. Two former cabinet ministers from the current Rwandan government testified. Two Rwandan witnesses were permitted to use a pseudonym and to testify under seal regarding their treatment as detainees at Kami Camp. The transcript from the hearing consists of 3913 pages, and the record includes more than 110 exhibits, including photographs and a video of Kami Camp, photographs of defendants and the victims, defendants' written statements, as well as the notes, reports and correspondence of both American and Rwandan investigators. In addition to Defendants' Joint Motion to Suppress ("Defs.' Mot.") and the government's opposition thereto, the Court received briefing from the parties on the permissible scope of cross-examination of defendants, the application of Fed. R. Crim. P. 12.2 as it relates to the government's ability to use a defendant's statements to his own expert for impeachment purposes, matters related to discovery of classified information under CIPA, and supplemental briefing on legal issues identified by the Court as critical to the resolution of the motion.
Based on the record and the arguments of the parties, the Court makes the following findings of fact and conclusions of law, and holds that the statements defendants gave to both the Rwandan and American interrogators starting in December 2001 are the product of coercion and therefore inadmissible.
II. FACTUAL BACKGROUND
A. Historical Context
Explaining the causes of the well-known and turbulent history of Rwanda is a task that need not be attempted here. Nevertheless, this history provides the backdrop to the events which are before the Court and is important to an understanding of the complicated factual setting presented by this case.
While the origin of the classification is a matter of some dispute, it is clear that for much of the second half of the twentieth century Rwandan society was divided largely along ethnic lines between Hutu, Tutsi and Twa.*fn3 The Hutu majority, which was thought to have been disfavored under Belgian colonial rule, assumed power in a series of elections in 1960-61 known as the "Hutu Revolution." With the establishment of a Hutu-dominated government, many members of the previously ascendant Tutsi minority were displaced or fled, creating a substantial refugee population. In the mid-1980s, this refugee population spawned the Rwandan Patriotic Front ("RPF"), a group made up primarily of Tutsi refugees who had grown up in neighboring countries and advocated the overthrow of Rwanda's Hutu regime. The military wing of the RPF, known as the Rwandan Patriotic Army ("RPA"), invaded Rwanda in 1990. The invasion was repelled, however, by the Armed Forces of Rwanda (known by its French acronym "FAR"), with the assistance of several foreign nations. Despite the failed 1990 invasion, the RPA continued to engage in guerrilla warfare over the next several years. From 1990 to 1994, divisions continued to grow within Rwanda between hard-line proponents of Hutu solidarity, on the one hand, and the Tutsi minority and moderate Hutus, on the other.
On April 6, 1994, the airplane carrying the long-time President of Rwanda, Juvenal Habyarimana, was shot down outside Kigali. (5/22 a.m. tr. at 21.) Responsibility for the assassination has never been confirmed. Habyarimana's death touched off a slaughter of the minority Tutsi population and many Hutus who were considered to be sympathetic to the Tutsis. The extent of the violence that followed is astonishing. Within the first five days after Habyarimana's plane went down, an estimated 20,000 Rwandans were killed, primarily Tutsis. Additionally, many prominent moderate Hutus were murdered, incluing Prime Minister Agathe Uwilingiyimana. Over a period of three months more than 500,000 Rwandans were killed in the genocide. Death estimates range from 800,000 to more than a million, or perhaps three-quarters of Rwanda's pre-genocide Tutsi population. The slaughter ended only with the defeat of the Hutu government by the RPA in July 1994.
The end of the genocide and the transfer of power to the RPF did not, however, result in the cessation of violence in Rwanda. The FAR fled to the Democratic Republic of Congo ("DROC") and became the ex-FAR. (5/3 a.m. tr. at 46-48.) The ex-FAR, along with genocidal civilian gangs known as the Interahamwe and other Hutu refugees, eventually formed a rebel force called the Army for the Liberation of Rwanda ("ALIR"). (Id.) Thus, much like the period from 1990-1994, warfare continued with ALIR making repeated guerrilla-style incursions into Rwanda until at least 2001. (5/3 a.m. tr. at 49, 55.)
The history of these warring factions is inextricably intertwined with the lives of the Rwandan witnesses who testified during the suppression hearing. For example, Alfred Ndabarasa, one of the Rwandans leading the investigation into the events at Bwindi, was raised by Rwandan refugees in Uganda and joined the RPA in 1989. (5/2 a.m. tr. at 24-26.) He participated in the RPA's October 1990 invasion of Rwanda. (Id. at 26.) Ndabarasa today serves as the Acting Director of Operations for Rwanda's foreign intelligence agency, the External Security Service. (Id. at 27-28). Similarly, Captain Alex Kibingo, who was the commander in charge of Kami Camp from 1999 until 2004, was raised in the Congo by Rwandese parents and joined the RPA in 1991. (5/3 a.m. tr. at 39-40.) Kibingo suffered multiple injuries during his service in the RPA. During the 1994 invasion, Kibingo sustained head injuries from shrapnel in a battle with the FAR (id. at 45), and in 1999, he was shot in the stomach while fighting ALIR. (Id.) As will be discussed further herein, defendants were members of ALIR and fought the RPA. In addition, two Hutus who had served as cabinet ministers in the RPF government testified. Theobald Gakwaya, the Minister of the Interior from March 2000 to April 2001 (5/17 a.m. tr. at 60), and Emmanuel Habyarimana, the Minister of Defense from May 2000 until early 2003 (5/17 p.m. tr. at 46), testified about their tenure in government service before they fled their country to seek asylum. Thus, while the events critical to resolving defendants' motion take place primarily between 1999 and 2003, the sins of the past and the long history of civil war in Rwanda provide important context to defendants' motion.
B. Background of the Defendants
1. Francois Karake
The first defendant named in the indictment, Francois Karake, was born in June 1964 in the Byumba Province of Rwanda. (5/22 a.m. tr. at 15.) Karake is a Hutu. (Id. at 14.) He received six years of primary schooling, but did not attend secondary school. (Id. at 17.) After completing this basic education, Karake worked with his parents doing carpentry work and farming. (Id.) In 1981, Karake moved to Kigali to work with one of his brothers, who had a business trading agricultural products. (Id. at 18.) That brother died, however, between 1987 and 1989. (Id.) At the time the 1990 war broke out, Karake testified that he was doing "different kinds of work" in Kigali, including "bricklaying." (Id. at 19.) According to Karake, when he heard on the radio about the RPA invasion, he returned home because the war was close to where his parents lived. (Id. at 19-20.) He eventually returned to Kigali, however, and resumed working in construction. (Id. at 20.) Four years later, with the assassination of President Habyarimana, the war came to Kigali. According to Karake, the day after the President's plane went down, RPA soldiers, who were stationed in Kigali to conduct peace talks, directed gunfire toward the area where Karake lived. (Id. at 22.) Karake testified that he fled Kigali within a week to escape the violence. (Id. at 22-23.) He never reached his parents in Byumba because the RPA surrounded the refugees and took them to a camp in Muhondo where, according to Karake, there was inadequate food and housing. (Id.at 27.) Karake fled this camp as well, spending time over the next several months in Rwandan refugee camps operated by the remnants of the Habyarimana government. (Id. at 30-34.) Eventually, however, the RPA attacked the camps where he had been living, forcing him to flee to the DROC.
Karake settled at a refugee camp in Kibumba, DROC. (Id. at 34.) In Kibumba, where the residents were dependent on the assistance of non-governmental organizations for food, Karake married and had a child. (Id. at 36-37.) He spent about two years in Kibumba before the RPA forces attacked the camp in 1996. (Id. at 38-39.) He and his family fled to yet another camp called Mwisake, but were there no more than a month before the RPA again attacked. (Id. at 42.) Karake was away from Mwisake the day it was attacked, however, and as a result was separated from his wife and child. (Id. at 43-44.) He spent the better part of the next year moving from camp to camp in the Congo trying to avoid RPA attacks. (Id. at 44-53.) In 1997, Karake met some ex-FAR soldiers who had come to the refugee camp at Kirumbo Bisumba to train the refugees to defend themselves against the RPA. (Id. at 51.) The training he received lasted about three months. (Id. at 52.)
Karake described numerous horrors he witnessed committed by RPA soldiers. He testified to seeing refugees jumping into rivers to escape RPA gunfire. (Id. at 42-43.) He recounted bombs or mortars falling on the refugee camps. (Id. at 46.) He told of being shot in the leg by an RPA soldier as he stooped to rescue a child who was attempting to breast-feed from his mother, who was already dead of a gunshot wound. (Id. at 46-47.) He claimed that the RPA destroyed a bridge as refugees were attempting to flee across it, causing many to fall into the river. (Id. at 48-49.) In 1998, Karake attempted to return home to Rwanda after hearing radio announcements indicating that the refugees would be welcomed back, but instead the group he was with was attacked by the RPA when they arrived in Ruhengeri. (Id. at 54-55.) It was at this point, Karake testified, that he sought out the ex-FAR soldiers and joined ALIR with the intent to overthrow the RPF government of Rwanda. (Id. at 55-56.) Karake was assigned to ALIR's Irondelle Company. (Id. at 57.) Karake remained a member of ALIR and Irondelle Company, serving as a "first soldier"*fn4 (5/23 a.m. tr. at 23-24)until 2000, when he returned to his family in Ruhengeri. (5/22 a.m. tr. at 70.) He was captured by RPA soldiers in August of that year. (Id. at 70-71.)
After his capture, Karake was sent to an RPA camp at Mweso, DROC, where he spent between six and seven months. (5/22 p.m. tr. at 5.) While at Mweso, Karake was held in a room he described as infested with fleas. When he first arrived there were two other people also being held in his room, but on the first day he was there one was taken away. Shortly thereafter, Karake heard gunshots. (Id. at 4.) The same scenario played out the following day when the second man was led away and soon thereafter Karake heard gunshots. Karake never saw either man again. He testified to "waiting" for the "day that I would be taken . . . [to] be killed." (Id. at 4-5.) Karake claims to have been deprived of food, water and clothes while at Mweso. (Id. at 7.) He was also questioned regarding ALIR's operations and the 1994 genocide. (Id.) Karake was beaten on multiple occasions with sticks and firewood; he was kicked and cut with barbed wire on other occasions, leaving visible scars. (Id. at 8-12.) During this time he wrote approximately three statements regarding his knowledge of ALIR and the genocide. (Id. at 14.) In March 2001, Karake was transferred to a repatriation camp in Gisengi, from which he was eventually released and permitted to return home to Rwambona in April 2001. He remained in Rwambona with wife and two children until he was taken into custody by the RPA as part of the investigation into the Bwindi attacks. (Id. at 14-15.)
2. Gregoire Nyaminani
Like Karake, Gregoire Nyaminani is a Hutu who grew up in the Byumba province in northern Rwanda. (5/23 p.m. tr. at 70.) Born in 1970, Nyaminani attended seven years of primary school before leaving school to work on his family's farm. Though his education was primarily conducted in Kinyarwanda, Nyaminani also learned to speak some French, though not enough to carry on a conversation. (Id. at 69-74.) Nyaminani remained in Byumba until after the defeat of the RPA invasion in 1990. (Id. at 76-77.) After seeing the consequences of the invasion for the people in his region, including physical injury and displacement from their homes, Nyaminani decided to join FAR in 1992. (Id. at 77-79.)
After receiving three months of training, Nyaminani was sent to the front to fight RPA insurgents at Mutara Gabiro. (Id. at 81.) During his 10 months at the front, during which time he was injured and his unit suffered multiple casualties, Nyaminani witnessed numerous atrocities committed by the RPA against civilians. One civilian was tortured (the RPA "pierced his mouth") because he had shared information with FAR soldiers. (Id. at 82-83.) Another was stabbed "in the ear" with a knife or bayonet such that "it came through the other ear." (Id. at 83.) He saw people whose arms or legs had been cut off, others whose eyes had been "removed," and still others who had been hanged in their homes by RPA soldiers. (Id. at 83-84.) After completing his tour of duty at the front, Nyaminani returned to Kigali in late 1993 and was assigned to a military police unit that guarded the country's crop storage site. (Id. at 85-86.) He continued this assignment until fleeing to escape the advancing RPA in 1994. (Id. at 86-87.)
Nyaminani settled at a refugee camp in Goma, DROC, which was run by former leaders of the Habyarimana government. (5/24 a.m. tr. at 10-11.) In 1996, the RPA destroyed the camp by setting up "big guns" on a hill overlooking the camp and bombing it from that position. (Id. at 13.) Nyaminani fled into the Congolese jungle, where he and other refugees were hunted by the RPA. (Id. at 14-15.) During this time Nyaminani claims to have witnessed RPA soldiers capture a group of refugees, kill them by hitting them in the head with a "small hoe," and then burn the bodies. (Id. at 15-16.) During his time in the Congolese jungle Nyaminani saw evidence of other mass killings by the RPA. (Id. at 18.)
Between 1996 and 1998 Nyaminani was captured twice by the RPA. (Id. at 19.) The first time he was captured, in 1997, Nyaminani was taken to Kwagasuko, DROC, where he was imprisoned in what he described as "a toilet." (Id. at 20.) It was an enclosure approximately 1 meter by 1.2 meters where he was kept with two other people and a goat. (Id. at 20-21.) Nyaminani described being tied up and beaten by RPA soldiers while at Kwagasuko. (Id. at 21-22.) One of the other prisoners died from this mistreatment, but Nyaminani and the other managed to escape. (Id.) In 1998, the RPA began announcing that refugees should return to Rwanda. (Id. at 22.) Nyaminani and other refugees arrived at a camp called Rubare, but Nyaminani received word that the RPA planned to kill the refugees at Rubare. (Id. at 23-24.) He began to tell other refugees, but was locked up after RPA soldiers learned what he was doing. (Id. at 24-25.) Nyaminani was able to escape yet again, however, after some colleagues with a gun scared off the guards by firing at them. (Id.)
Soon after his second escape, Nyaminani witnessed the killing of still more refugees at the Ka Rumangabo refugee camp. (Id. at 28-29.) It was after this experience that Nyaminani joined ALIR, in January 1998. (Id. at 28-30.) Nyaminani was assigned to ALIR's Irondelle Company. (Id. at 31.) While fighting as a member of ALIR, Nyaminani sustained a minor head injury from a grenade blast. (Id. at 33.) He also watched as the RPA trapped civilians on the bank of a large river called Ruhoro, shooting some and forcing the others to jump in to avoid being killed. (Id. at 34.) Nyaminani remained a member of ALIR, attaining the rank of First Sergeant (5/26 a.m. tr. at 92), until his capture by the RPA in battle on June 15, 2001. (5/24 a.m. tr. at 32.)
3. Leonidas Bimenyimana
The third defendant, Leonidas Bimenyimana, who also goes by the nickname or pseudonym "Zappy Gadi," is a Hutu who was born in the Ruhengeri Province in northern Rwanda. (5/30 a.m. tr. at 58.) He is 38 years old. (Id.) Bimenyimana's parents were farmers and raised eight children: four boys and four girls. Though three of his sisters are still alive, Bimenyimana is the only surviving son. (Id. at 60.) Bimenyimana received substantially more schooling than did his co-defendants. He attended five years of secondary school in the DROC (6/1 a.m. tr. at 9), in addition to eight years at a Rwandese primary school. (Id.) While in school he learned to read and write French, though not fluently, and was taught some English. (5/30 a.m. tr. at 59.) Bimenyimana's wife and three children currently live with his parents on their farm. (Id. at 60-61.)
Bimenyimana joined the FAR in 1988 and spent one year studying at ESO, which was the FAR's military academy. (6/1 a.m. tr. at 9.) Bimenyimana attained the rank of sergeant, and during the 1990 war, commanded a section of 11 or 12 soldiers. (Id. at 15-16.) After the war, Bimenyimana was promoted to the rank of "first sergeant." (Id. at 28.) In 1992, Bimenyimana attended another military school known as Egena, where he trained to become a judicial police officer. (Id. at 25.) After graduating in 1992, he returned to Ruhengeri to serve in that position.
(Id.) He remained in this position, assisting with crime scene investigations and presenting evidence in court in criminal cases, until April 6, 1994, when President Habyarimana's plane was shot down. (Id. at 26.) Bimenyimana fought with the FAR until the RPF's victory, at which point he fled the country. (Id. at 27.)
By 1997, however, Bimenyimana had returned to his family home in Ruhengeri. (5/30 a.m. tr. at 66.) According to Bimenyimana, the RPA began to harass his family in an attempt to capture him, and eventually killed 11 members of his family. Bimenyimana described the deaths of three of his family members at the hands of the RPA, all in 1997. (Id. at 65-66.) The first was his younger brother, who was taken from the family home and imprisoned at a military camp called Nyamilima, where he was beaten to death. (Id. at 65.) The RPA returned later that year and took Bimenyimana's cousin from their home, tied him up, placed his head on a stone on the ground and then beat his head with another stone until he died. (Id.) The RPA returned twice more, first to burn his father's house to the ground, and the next time to burn the home of his grandfather, who was killed in the fire. (Id. at 65-66.)
That same year, 1997, Bimenyimana joined ALIR and assumed the rank of first sergeant. (6/1 a.m. tr. at 28.) He was assigned to lead the third platoon of Irondelle Company, which was under the command of an individual by the name of Ntabwoba. (Id. at 28-29.) In the second half of 1999, having been promoted to chief adjutant, Bimenyimana left Irondelle Company to study for six months at an ALIR military academy called ESM in Masisi, DROC. (Id. at 33.) Bimenyimana received a further promotion to second lieutenant in 2001. (Id. at 35.) He remained a member of ALIR until his capture by the RPA in June 2002. (Id. at 34-35.)
C. The Bwindi Attack
Early on the morning of March 1, 1999, ALIR forces attacked Bwindi Impenetrable National Forest in Uganda. (5/22 a.m. tr. at 57.) Though the motivations for this attack remain disputed, it is clear that the ALIR unit responsible for the attack that day was Irondelle Company, under the command of Ntabwoba. (5/24 a.m. tr. at 38.) Irondelle Company consisted of three platoons, one of which was headed by Bimenyimana. (Id. at 37.) Nyaminani was one of four section leaders within Bimenyimana's platoon. (6/1 a.m. tr. at 31-32.) The other three sections in that platoon were headed by Jean-Bosco Hategekimana, Rukundo and Tindo. (Id. at 32.)
On the morning of the attack, several groups of tourists and approximately five Ugandan park rangers were staying at various campsites within the park. One park ranger, Paul Ross Wagaba, was killed during the initial raid on Bwindi. (Gov't Ex. 99, tab 22 at 1.) Though a few tourists managed to evade detection by the ALIR soldiers (Gov't Ex. 99, tab 1 at 1), most were gathered together by the ALIR soldiers at the Abercrombie and Kent ("A&K") campsite. (Id.) Various reports intimate that a female French tourist assisted ALIR soldiers in dividing the tourists between those who spoke English and those who spoke French. (Gov't Ex. 99, tab 1 at 1; Gov't Ex. 99, tab 18 at 2). The English-speaking tourists were then taken hostage while the French speakers were permitted to remain at Bwindi. (Gov't Ex. 99, tab18 at 2; 5/15 p.m. tr. at 17.) Of the group gathered at A&K, seventeen tourists were taken hostage, including four Americans, six British, three New Zealanders, and one citizen each from Australia, Canada, Switzerland and Uganda. (Gov't Ex. 99, tab18 at 2.)
The seventeen hostages were forced to march with the ALIR soldiers out of A&K and toward the DROC. (Id. at 2.) Two tourists were released soon after leaving the camp when an American woman, Linda Adams, feigned an asthma attack. (Id.) She and Fiona Morley, a British citizen, were left behind. (FBI 18 at 2.) Of the remaining 15 hostages, seven survived, including one American. (Id.) The eight who perished appear to have been killed in three separate incidents. (5/16 a.m. tr. at 7-8.) In one incident, three men were killed together, including American Robert Haubner. (Id. at 8.) Haubner was identified, at least in part, by a tattoo on his "left shoulder." (Def. Ex. K-21; K-4; 5/16 a.m. tr. at 8.) A man and woman were killed in another location. (5/16 a.m. tr. at 8.) The third set involved three women, including an American, Susan Miller, who was married to Haubner.*fn5 (Id.) Two of the victims were found with handwritten notes on or near their bodies. (Gov't Ex. 99, tab 22 at 3.) One read: "Here lies the Anglo-Saxon who betrayed us, favoring the Nilotics to the detriment of the Bantu cultivator farmers. If you do not learn these lessons, it is because you do not understand. You will now understand by the forces of nature." (Id.) The second read: "This is the punishment of the Anglo-Saxon who sold us. You protect the minority and oppress the majority." (Id. at 4.)
The remaining seven hostages survived, among them American Mark Ross. Ross was given a note by the ALIR leader, Commander Ntabwoba, with instructions to deliver it to the United States Ambassador. (Id. at 2-3; 5/22 a.m. tr. at 66.) Similar to the warnings found with the murdered victims, the note read:
People cannot ignore our problem. You have supported the Tutsi minority in Rwanda in oppressing and massacring the Hutus without constraint. You have looked on as they have killed the Bantus in the DRC. You have encouraged this without same. The Nilotics can never colonize the whole of Africa. All Africans know your imperialist secret. . . . We are addressing this to the westerners, above all Americans and Anglo-Saxon. (Id. at 3.) After agreeing to deliver the note as instructed, Ross and the other hostages were released.
III. THE INVESTIGATION
A. The Initial Investigation (March 1999 -- October 2000)
Two days after the attack, on March 3, 1999, a team of FBI investigators departed Washington, D.C. for Kampala, Uganda. (5/15 p.m. tr. at 17.) Among the investigators was FBI Special Agent Jennifer Snell Dent ("Dent"), who would later become the lead agent on the case and one of the government's primary witnesses at the suppression hearing. (Id. at 18.) Upon arriving in Kampala, the FBI met with U.S. embassy personnel, Ugandan military and police officials, and various witnesses. (Id. at 17.) Among the Ugandan officials interviewed were the Ugandan solicitor general, the director and deputy director of the directorate of prosecution, and the director of the criminal investigative department. (5/19 p.m. tr. at 7.) Together with the Ugandan officials and British investigators who had also arrived in Uganda, the FBI agents visited the crime scene at Bwindi. (5/15 p.m. tr. at 17.) By the time the FBI arrived, the Ugandan authorities had already completed an autopsy of Robert Haubner, which indicates the cause of death as a "brain laceration," and notes the existence of a tattoo on Haubner's left shoulder.*fn6 (Def. Ex. K-21.) Agent Dent and her colleagues remained in Uganda for approximately 10 days before returning to the United States. (5/15 p.m. tr. at 18.)
There were few developments in the investigation for nearly a year. Coinciding with the one year anniversary of the attack, in March 2000, the State Department launched the Rewards for Justice campaign, which offered monetary rewards to any person who provided information leading to the arrest and prosecution of those responsible for the Bwindi attack. (Id. at 19.) The Rewards for Justice posters were issued in both English and French.(Gov't Ex. 2 & 2a.) The FBI requested that they be distributed in Kinyarwanda and Swahili as well. (5/15 p.m. tr. at 20.) Special Agent Dent also returned to Uganda to conduct interviews of people who "walked into the embassy in response to the program." (Id. at 21.) It does not appear that these interviews provided the FBI with any significant information. Indeed, no money was ever paid out under the Rewards for Justice campaign. (5/11 p.m. tr. at 17.)
The first real progress in the case occurred when an ALIR soldier, Jean Paul Bizimana, decided to defect to Uganda. (Gov't Ex. 29, tab 1 at 2.) Bizimana, a former member of the Interahamwe and ex-FAR, first approached the Ugandan police in December 1999 and was taken into custody at that time. (Id.) In October 2000, Dent returned to Kampala to interview Bizimana. (5/15 p.m. tr. at 22.) The interview took place on October 8, 2000. (Id.) Bizimana told the FBI that he was in Bwito, DROC at the time of the Bwindi attack. (Gov't Ex. 29, tab 1 at 7.) Approximately two weeks after the attack, however, he witnessed a group of soldiers led by Captain Ntabwoba march through Bwito on their way to Massese, DROC. (Id.) These soldiers were carrying a number of unusual valuables, including luggage, radios, cameras and binoculars. (Id.) Bizimana provided a physical description of several ALIR soldiers who were of interest to the FBI, including Ntabwoba and a Colonel Bemera, who was one of the highest ranking officers in ALIR. (Id. at 8-10.)
B. The Rwandans Join the Investigation (November 2000 -- November 2001)
On this same trip in October 2000, Dent also traveled to Kigali to seek the assistance of Rwandan authorities. (5/15 p.m. tr. at 23-34.) She met with Frank Mugambage, the Commissioner General of the Rwandan National Police, to inform him of the Rewards for Justice campaign and to indicate the FBI's interest in speaking with anyone who might have knowledge of the Bwindi attack. (Id.) Dent provided Mugambage with a list of the items stolen during the attack, two composite sketches produced by Scotland Yard, and a list of names and descriptions of the individuals who were of interest to the FBI. (Gov't Ex. 99, tab 5.) Dent provided the same information to the U.S. embassy's Regional Security Officer ("RSO").*fn7 (Id.)
After meeting with Dent, Mugambage issued a memo to the Army Chief of Staff informing him that the FBI "requested" assistance from "the Rwandan Police" in investigating the Bwindi attack. (Def. Ex. K-4.) The memo notes that the FBI "provided information they have so far gathered," and suggests that "the criminal investigations department . . . get together and work out how to handle this matter." (Id.) Three different agencies within the Rwandan government were included in the "criminal investigations department" -- the National Police, the Department of Military Intelligence ("DMI"), and the External Security Service ("ESS"). Mugambage sent copies of his memo to Jack Nziza, head of DMI, and Patrick Karegeya, head of ESS.*fn8 (Gov't Ex. 65; Def. Ex. K-4.)
The National Police, ESS and DMI took collective responsibility for conducting the Rwandan government's investigation into the events at Bwindi.*fn9 Still, little progress was made for several months. In the summer of 2001, however, the investigation began in earnest. In May and June of that year, a "major offensive" was undertaken by ALIR, resulting in the RPA's capture of upwards of 2,000 ALIR members. (5/2 a.m. tr. at 33-35.) Aware of the ongoing border clashes, the FBI sent word through Legat Paris to the RSO at the U.S. embassy in Kigali instructing him to ask the Rwandans to interrogate "those recently apprehended" for information regarding Bwindi. (Gov't Ex. 99, tab 2 at 2.) In July, Ndabarasa was instructed by the head of ESS, Patrick Karegeya, to meet with Police Commissioner General Mugambage regarding an assignment. (5/2 a.m. tr. at 29.) At this meeting, Ndabarasa was briefed on the events at Bwindi and told to travel to the northern province of Ruhengeri, where most of the captured members of ALIR were being held, to gather information regarding the Bwindi attack. (Id. at 32-36.) Ndabarasa was accompanied to Ruhengeri by Eric Kayiranga, a police officer. (Id. at 37-38; 5/12 a.m. tr. at 44.)
Among those captured in the spring of 2001 was Nyaminani, who had been taken prisoner in a battle on June 15. (5/24 p.m. tr. at 3.) Nyaminani spent roughly two weeks at two different RPA military installations before being transferred to Mudende repatriation camp in Ruhengeri. (Id. at 3-4.) Though Nyaminani alleges that he was subjected to verbal abuse from RPA soldiers while at Mudende and claims to have feared for his life due to his status as a prisoner of war, he does not allege any physical mistreatment. (Id. at 6-9.) While at Mudende, he was questioned twice by Kayiranga and Ndabarasa. (Id. at 5.) Ndabarasa learned of Nyaminani through an earlier interview with a captured ALIR soldier named Desire Hategekimana. (5/12 a.m. tr. at 46-49.) The exact dates of Ndabarasa's interviews with Nyaminani are unknown, and no notes from the meetings were introduced into evidence. In a report to Commissioner Mugambage dated July 22, 2001, however, Ndabarasa summarized the results of his investigation to that point.*fn10 (Gov't Ex. 43.) The report refers to Nyaminani as a "POW" and explains that, despite the fact that Nyaminani "vehemently denie[d] participation," he was still considered a suspect because he served in Bimenyimana's platoon, which was believed "to have been the point force in the attack." (Id.) During the Mudende interviews, Nyaminani admitted that he knew of the attack and named several ALIR members who participated, including Bimenyimana (id.), but claimed to have stayed at the edge of the park rather than entering the area where hostages were taken. (5/12 a.m. tr. at 53-54, 58.) This account was echoed by Desire Hategekimana. (Id. at 54.) Nevertheless, as Ndabarasa put it during his testimony, he was "not satisfied" with Nyaminani's responses. (Id. at 53.)
Ndabarasa's report refers to the fact that two of the nine victims at Bwindi were Americans, but makes no mention of any other victim's nationality. (Gov't Ex. 43.) He also states that the captured ALIR Chief of Staff, Pierre Habimana (also known as Bemera), "accepts that it was" part of ALIR's purpose to kill "RPA allies including American citizens." (Id.) Upon receiving the report, Ndabarasa's superiors informed him that it "was not enough," and he was instructed "to do more." (5/12 a.m. tr. at 56.)
Ndabarasa therefore returned to Ruhengeri to conduct further interviews. Sometime "towards the end of July," after his initial interviews at Mudende, Nyaminani was transferred to another repatriation camp called Nkumba. (5/24 p.m. tr. at 7.) At Nkumba, Nyaminani was subjected to further questioning regarding Bwindi. (Id. at 10.) In addition to Ndabarasa and Kayiranga, Nyaminani was interrogated by Major Jean Bosco Kabanda, a high-ranking DMI soldier, and four other soldiers serving at Nkumba Camp. (Id. at 10.) Nyaminani was also questioned by Kalisa Karangwa, a police officer who was assigned to assist Eric Kayiranga. (5/2 p.m. tr. at 26.) These interviews were conducted outdoors on the grounds of Nkumba Camp. (5/12 a.m. tr. at 58; 5/2 p.m. tr. at 32.) As was the case with Ndabarasa at Mudende Camp, Karangwa was steered to Nyaminani by Desire Hategekimana. (5/2 p.m. tr. at 31; Gov't Ex. 51 & 51a.) Karangwa's interview resulted in a written statement from Nyaminani regarding his role in the attack.*fn11 (Gov't Ex. 50 & 50a.) This statement, dated July 23, 2001, reflects Nyaminani's admission that he was at Bwindi on March 1, 1999, but that he "did not reach exactly to Bwindi." (Gov't Ex. 50a at 1.) He admitted to seeing other soldiers return from Bwindi with "bags" that "contained [a] camera" and other goods, but denied any knowledge of how the whites had died. (Id. at 2.) In reviewing the statement, which was in fact written by Karangwa, Nyaminani specifically asked for the word "death" to be scratched out of the statement in his response to a question regarding "those who killed" at Bwindi. (5/2 p.m. tr. at 37-38; Gov't Ex. 50 at 2.) This statement, according to Ndabarasa, was also unsatisfactory, for it was "not enough." (5/12 a.m. tr. at 60.) All told, Ndabarasa estimated that he and Eric Kayiranga interviewed Nyaminani three to four times in July 2001. (Def. Ex. B-4 at 2.)
Defendant Nyaminani testified that for the duration of his stay at Nkumba Camp after giving his statement to Karangwa, he was subjected to various forms of mistreatment. (5/24 p.m. tr. at 12.) He claims that one of the soldiers in charge of Nkumba, an individual named Gitega, told him that he would be killed if he didn't "do what they wanted." (Id.) He was forced to "crawl" through the mud in front of other prisoners at the camp. (Id. at 13.) He was singled out for other "demean[ing]" treatment or harassment in front of his colleagues, particularly by Major Kabanda, and was subjected to what he viewed as threats, including references to "cockroaches . . . know[ing] their mother," if he failed to tell them more about Bwindi.*fn12 (Id. at 13-15.)
Nyaminani also testified that people were sometimes taken from Nkumba in cars at night and never returned, providing further reason for him to fear for his life. (Id. at 15-16.)
During the months of August and September, U.S. defense attache Major Richard Skow, who had been assisting FBI investigators (Gov't Ex. 99, tab 26 at 4, tab 32 at 7-8), conducted interviews of some Bwindi suspects. (Gov't Ex. 99, tab 28 at 5.) Around August 7, 2001, Skow interviewed Jean Bosco Hategekimana and cabled Dent to inform her that Hategekimana had "indicated that he was a participant in the attack at Bwindi." (5/15 p.m. tr. at 28, 45.) In early September, Skow interviewed Colonel Bemera regarding the events at Bwindi. (5/3 p.m. tr. at 27-30.) Also in August 2001, Bryan Bachmann arrived in Kigali to serve as the State Department's RSO for Rwanda. (5/10 p.m. tr. at 66.) Bachmann began familiarizing himself with the U.S. embassy's case file regarding Bwindi soon after arriving at his post in Kigali, and in October he met with Commissioner General Mugambage to reiterate the U.S. government's request for assistance in solving the murders. (Id. at 66-68.) Bachmann stated that while the embassy's case file indicated that the U.S. government had turned over to the Rwandans "substantial information" derived from its investigation of the Bwindi attack (id. at 74), there was no indication in the file, nor did he ever acquire personal knowledge, that the Rwandans had undertaken any investigation into Bwindi prior to being asked for assistance by the U.S. (Id. at 72.)
In November, the FBI began planning another trip to Rwanda. (Gov't Ex. 99, tab 32.) The FBI, along with the U.S. prosecution team, was concerned that the suspects being housed at Mudende Camp were due to be repatriated and released "at any time." (Id. at 5.) The FBI hoped to meet with the head of DMI, Jack Nziza, as well as Commissioner General Mugambage. (Id. at 8.) The FBI's travel preparation memo noted that Jean Bosco Hategekimana had identified "the rebels that remained behind with the tourists at Bwindi." (Id. at 8-9.) This list of rebels included Bimenyimana, but did not mention either Nyaminani or Karake. (Id.) The memo also states that "Gregoire Minani" and "Desire Hakizimana" had "confessed to [their] participation in the Bwindi attack," but the memo did not specify any details regarding the extent of their "participation." (Id. at 9.)
On November 29, 2001, Nyaminani, Desire Hategekimana, another individual named Fulgence Nyibizi, and others were transported from Nkumba Camp to Kigali.*fn13 (Id. at 20.) During a stop in the town of Ruhengeri, Nyaminani witnessed an unidentified number of white persons speaking with some of the other people who had come from Nkumba, but Nyaminani did not speak with these people at that time. (Id. at 19.) After arriving in Kigali, a DMI soldier named Julien Rwangarabe transported the prisoners, including Nyaminani, to Kami Camp. (Id. at 22.) Nyaminani does not allege that he suffered any mistreatment this first night at Kami Camp. The following day, he was picked up at Kami by Rwangarabe and brought to Kacyiru National Police Headquarters, where he was questioned by a group of Rwandans that included Ndabarasa, Kayiranga, Rwangarabe and a senior police officer named John Gacinya. (Id. at 22.) During this questioning he was informed that the next day he would be meeting with "whites," and that he "had to explain about the whites that had been killed in Bwindi." (Id. at 23.)
C. The Interrogations (November 2001 -- February 2003)
In late November 2001, Dent returned to Rwanda with her investigative team to begin interviewing the Bwindi suspects who had been identified by Rwandan investigators. Upon arriving, the FBI team met with Bryan Bachmann and other embassy personnel before their meeting "with the Rwandan National Police and Rwandan military personnel to discuss [the FBI's] objectives and . . . interest in meeting with anyone who might know anything about the Bwindi attack." (5/15 p.m. tr. at 44.) The Americans met with Commissioner General Frank Mugambage and the head of DMI, Jack Nziza. (Gov't Ex. 99, tab 32 at 8.) According to Bachmann, by this time the relationship between the American and Rwandan investigative teams had developed into "a very good partnership." (5/10 p.m. tr. at 83.) Both sides shared investigative information freely. (Id. at 81-84.) The two teams were, as Dent put it, "working hand in hand." (5/16 p.m. tr. at 3.) The U.S. investigators found the Rwandans to be exceptionally accommodating. In describing this relationship to the grand jury, Dent testified that the Rwandans had "been great":
Every time, you know, we're sitting down to talk to someone and we learn of a couple more people that might have just a little bit more information, they . . . just brought the repatriation books in, and it's got the person's picture, their family history, and where they're presently located.
And I mean, we'll be sitting there mid-interview and, you know, one of the RPA guys will pick up the phone and . . . tell one of his colleagues, . . . go find this guy. And then we interview him the next day.
(5/16 p.m. tr. at 57-58.)
The close working relationship of the American and Rwandan investigative teams was also apparent in the interrogation process. Rwandan officials were a constant presence in the interview room at Kacyiru when the FBI was questioning the Bwindi suspects. (5/15 p.m. tr. at 63-64; Gov't Ex. 27.) Several Rwandan officials served as translators during these early interviews, including John Gacinya, Alfred Ndabarasa, Eric Kayiranga, and Egide Ruzigamanzi. (Id.; 5/2 a.m. tr. at 80.) Isidore Kayumba, a Rwandan national who worked for the State Department as an investigator under Bachmann, also assisted with translation. (Id.) Rwandan officials were also active participants in questioning the Bwindi suspects. (5/11 p.m. tr. at 41; 5/15 p.m. tr. at 59-60.) They acted, according to Ndabarasa, as "investigators." (5/2 a.m. tr. at 80.) Indeed, not every question that was asked by a Rwandan was translated into English for the benefit of the American investigators. (5/11 p.m. tr. at 41, 44.)
The Rwandans also assisted the FBI in administering an advice of rights, or modified Miranda warnings, to each suspect prior to the interrogation. (5/12 a.m. tr. at 85-86.) Because there was no advice of rights form available in Kinyarwanda (5/10 a.m. tr. at 47-48), the FBI used a form written in either English or French.*fn14 (5/15 p.m. tr. at 58.) With a few exceptions,*fn15 the advice of rights would be provided to the translator and read to the interviewee. (5/15 p.m. tr. at 67.) The FBI agents would not read the advice of rights aloud in English; the translator would simply take the form and translate it into Kinyarwanda. (5/19 p.m. tr. at 118-19.) At the conclusion of this exercise, the FBI would ask a series of questions designed to ensure that the interviewee understood his rights. (Id. at 119-20.) These questions asked whether the interviewee "[is] willing to make a statement and answer any questions," "do[es] not want a lawyer at this time," "understand[s] and know[s] what [he is] doing," and whether any "promises or threats have been made to [him, or] pressure or coercion of any kind has been used against [him]." (Id.; Gov't Ex. 31.) The questions would be posed one at a time and a translated response would be received before moving on to the next question. (5/15 p.m. tr. at 67.)
Every interview of a defendant that was conducted with American officials present took place at the National Police Headquarters at Kacyiru. (5/2 a.m. tr. at 77-78.) The room where the interviews were conducted was a "big conference hall" with "a long table." (Id. at 78; Gov't Ex. 3-7.) There was a small bathroom off the conference room that was used by interviewers and interviewees alike. (5/15 p.m. tr. at 42-43.) Food and water were provided to the interviewees during the interrogation. (Id. at 40, 43.) The exterior door to the conference room typically remained open, except during the rainy season. (Id. at 38-39.) This door led to an adjacent "open area" (id. at 42) or "courtyard." (5/16 p.m. tr. at 49; see also Gov't Ex. 3-7.)
Over the course of fifteen months from November 2001 to February 2003, more than 50 interviews of Bwindi suspects were conducted in this room at Kacyiru by the FBI, Bachmann and a host of Rwandan officials, including 11 separate interrogations of Nyaminani, 7 of Karake, and 5 of Bimenyimana. In addition, there were an unknown number of interrogations conducted by Kibingo at Kami Camp over this same period of time. While it is unknown how many written statements were obtained from the defendants by Kibingo, two written statements from Nyaminani (Gov't Ex. 53, 53a, 54 & 54a), and one from Karake were introduced into evidence. (Gov't Ex. 52 & 52a.) Though no written statements obtained from Bimenyimana during his incarceration at Kami Camp were proffered, Kibingo also interrogated Bimenyimana about his role at Bwindi during this time. (5/4 p.m. tr. at 48.) It is this fifteen-month sequence of interrogations to which the Court now turns.
1. December 2001
The first interview of any of the defendants*fn16 by the FBI was of Gregoire Nyaminani on December 1, 2001.*fn17
Present at the interview were FBI Agents Dent and David Paun, Assistant Legal Attache Ricky Chambers, Rwandan National Police Officers John Gacinya, Julien Rwangarabe, Eric Kayiranga and Egide Ruzigamanzi, U.S. embassy employee Isidore Kayumba. (Gov't Ex. 27; 29, tab 4 at 1.) Gacinya, the ranking Rwandan police officer, also served as the primary translator. At the outset of the interview, before any advice of rights was given, Nyaminani informed the FBI that Irondelle Company was made up of 3 platoons, one of which was led by Bimenyimana, and a "Special Section," of which he was a member. (Gov't Ex. 29, tab 4 at 1.) Nyaminani also defined the ALIR objectives as: (1) overthrowing the government of Rwanda; (2) releasing prisoners; and (3) targeting supporters of the government of Rwanda. (Id. at 2.) It was at this point in the interview that Nyaminani was orally advised of his rights. (Id.) The FBI apparently did not have an "unsigned or clean cop[y] of the advice of rights," so it was not possible for Nyaminani to acknowledge his consent in writing, but Agent Dent testified that he did verbally consent to the interview. (5/15 p.m. tr. at 65.)
After being advised of his rights, Nyaminani described the march to Bwindi and stated that the Special Section was "instructed to remain behind to fight off potential opponents." (Gov't Ex. 29, tab 4 at 3.) Nyaminani named 16 individuals whom he knew entered the park, including Commander Ntabwoba and Bimenyimana. (Id.) The FBI informed him that they did not believe his statements and terminated the interview after giving Nyaminani a pen and paper with instructions "to write down all that he could recall about the Bwindi operation." (Id. at 4.) At no point during the interview did Dent or any other investigator inquire as to Nyaminani's conditions of confinement. (5/16 p.m. tr. at 22-23.)
On December 2, the FBI conducted an interview of Jean-Bosco Hategekimana. Jean-Bosco informed investigators that Nyaminani had helped transport injured soldiers out of Bwindi, casting doubt on the version of events as described by Nyaminani on December 1. (5/15 p.m. tr. at 71-73.) Nyaminani was also identified to the FBI by Ernest Nkikabahizi on December 1 as a participant in the Bwindi attack. (Gov't Ex., tab 3 at 4.) Nkikabahizi, however, was not actually present at Bwindi on the day of the attack. (Id. at 1; 5/15 p.m. tr. at 69.) Based on the information received from Jean Bosco and Nkikabahizi, the FBI decided to interview Nyaminani a second time. (5/15 p.m. tr. at 73.)
Nyaminani was brought back to National Police Headquarters on December 5 for further questioning.*fn18 Present for the interview were FBI Agents Dent and Paun, Isidore Kayumba, ESS officer Alfred Ndabarasa, Rwandan police officers John Gacinya, Eric Kayiranga, Egide Ruzigamanzi, Julien Rwangarabe, and DMI soldier Jean Bosco Kabanda. Ndabarasa served as the primary translator. (Gov't Ex. 27.) Before the interview began, Nyaminani "provided two pages of handwritten notes in Kinyarwanda he'd prepared about the Bwindi operation" following his December 1 interview.*fn19 (Gov't Ex. 29, tab 7 at 1.) Nyaminani's written statement reiterates the story he told at his first interview, namely that his duty was to remain at the perimeter of the park "blocking the way," and that he remained there until the rest of Irondelle Company returned from the attack. (Gov't Ex. 9 (1/10/2002 e-mail from Isidore Kayumba to Bachmann).) Nyaminani identified a watch and camera as among the items that were looted from Bwindi. (Id.)
In his interview on December 5,*fn20 however, Nyaminani changed his story slightly. He admitted being a member of Irondelle Company's Third Platoon -- under the command of Bimenyimana. (Gov't Ex. 29, tab 7 at 1.) He admitted that he had entered the park and described the attack, including that two members of his platoon were shot. (Id. at 2.) Only after giving a general overview of his knowledge of the attack and the looting was Nyaminani read his rights. (Id. at 2-3.) He then signed the advice of rights form. (Id. at 3; Gov't Ex. 8.) Nyaminani continued to deny any involvement in the killing of any tourists. Rather, he claimed that his platoon was ordered to move out carrying the wounded soldiers. (Gov't Ex. 29, tab 7 at 4.) He also described seeing Commander Ntabwoba hand notes to some of the tourists after the entire company had marched out of the park and reached a clearing where they rested. (Id. at 4-5.) Nyaminani then stated that most of the soldiers moved out, but "Murenzi's section" remained behind with the tourists and carried out the murders. According to Nyaminani, Karake was one of the members of Murenzi's section. (Id. at 5.) This was the first time Karake's name had surfaced. (5/15 p.m. tr. at 81.)
The FBI continued to conduct interviews of other Bwindi suspects through December 8, 2001, including a follow up interview of Jean Bosco Hategekimana (Gov't Ex. 29, tab 9 & 10), and a second trip to Ruhengeri to interrogate Silver Dukuzumuremyi. On the afternoon of December 8, after returning from Ruhengeri, the FBI conducted two more interviews, of Leonard Harerimana and Pierre Habimana (a.k.a. Colonel Bemera), at Kacyiru prior to departing Rwanda. The interviews of Harerimana and Bemera are relevant to the instant motion less for the information the FBI derived therefrom than for pinpointing the arrival of Karake at Kami Camp. Karake testified that he was taken from his home in Rwambona and driven first to Byumba before going on to Kigali, where he was taken directly to the National Police Headquarters at Kacyiru. Upon arriving at Kacyiru, Karake testified, he found Colonel Bemera and Leonard Harerimana standing in the courtyard area outside the conference room where the FBI was conducting interviews.*fn21 (5/22 p.m. tr. at 17-18.) Karake's testimony conflicts with that of Kibingo, who testified that Karake did not arrive at Kami until mid-January 2002. (5/3 p.m. tr. at 36.) However, the first written statement obtained from Karake at Kami by Kibingo bears a date stamp of December 28, 2001 (Gov't Ex. 52), which is consistent with Karake's testimony that it was approximately two and a half weeks after he arrived at Kami that he gave a written statement to Kibingo. (5/22 p.m. tr. at 18-32.) Kibingo testified that he delivered Karake's first statement to DMI Headquarters the morning after he obtained it. (5/4 a.m. tr. at 29.) It is reasonable to conclude that it would have been date stamped at that time, especially given the government's lack of any explanation for how or why the document might have been back-dated. Moreover, Dent testified that she learned of Karake from the Nyaminani interview on December 5 and informed the Rwandans soon thereafter of her desire to speak with him. (5/16 p.m. tr. at 5, 51, 57.) In light of Dent's testimony regarding the speed with which the Rwandans were able to track down suspects (id. at 57-58), the only fair inference from the record is that Karake actually arrived at Kami Camp by early December, soon after the FBI made its request to interview him. This is also consistent with Karake's version of events. (5/22 p.m. tr. at 15.) Therefore, the Court finds that Karake was taken into custody at Kami Camp on or about December 8, 2001, and gave his first written statement regarding Bwindi to Kibingo at Kami Camp sometime in mid- to late December.
According to Karake, the morning after he arrived at Kami Camp, Kibingo began interrogating him regarding Bwindi.*fn22 (Id. at 19.) The initial questioning lasted more than an hour and took place outside Kibingo's house. (Id. at 20; see also Gov't Ex. 58a-c.) At this time, Karake testified, "[t]here was no problem. [Captain Kibingo] treated me well." (5/22 p.m. tr. at 20.) "[B]ut," stated Karake,"what I told him, he did not believe." (Id.) Kibingo did not ask Karake to write a statement at this first meeting. (Id. at 21.) At a second meeting roughly one week later, Kibingo again questioned Karake outside his home, this time in front of Nyaminani and Leonard Harerimana. (Id. at 24-25.) Again, Karake denied involvement in the killings at Bwindi. (Id. at 25.) Kibingo "wasn't happy" with Karake's version of events and instructed all three men to leave. (Id.) As with the first meeting, no written statement was requested by Kibingo. (Id.) Kibingo denies that there were any conversations between him and Karake prior to the one that led to his first written statement. Instead, he claims that within a short time after professing his innocence, Karake returned and spontaneously confessed to three murders, but then wrote out a confession in which he confessed to killing only one tourist. (5/3 p.m. tr. at 65-70.)
Though the conditions under which Karake ultimately gave Kibingo a written confession are disputed and will be examined later, there is no question that Karake was subjected to questioning by Kibingo, and that the written statement was drafted soon thereafter. (5/3 p.m. tr. at 65-70; 5/22 p.m. tr. at 25-38.) Kibingo took no notes of his conversations with Karake, however, and no other witnesses were present. (5/3 p.m. tr. at 65; 68.) In his written statement, Karake admitted to participating in the attack at Bwindi and kidnapping white tourists.*fn23 (Gov't Ex. 52 & 52a.) Karake also inculpated himself in the killing of one white person ("a white (1)"), and named Habimana and Minani as others who killed at Bwindi. (Id.) The written statement does not identify the gender of the white person that he purportedly killed. (Id.) Immediately after receiving the statement, Kibingo informed Nziza by phone before delivering the statement to DMI headquarters the following morning. (5/4 a.m. tr. at 29.)
Karake's written statement is inconsistent with Kibingo's testimony of their conversation immediately prior to the drafting of the statement. Kibingo testified that Karake told him that he was ordered by Commander Ntabwoba to kill two individuals who were covered with bed sheets with an axe, and then kill a third person -- a man with a tattoo of an eagle on his arm.*fn24 (5/4 a.m. tr. at 21-23.) Despite reading the written statement immediately after Karake drafted it, Kibingo did not question him or even express any surprise regarding the inconsistencies, particularly the acceptance of responsibility for only one, rather than three, murders. (5/4 a.m. tr. at 27.) Instead, Kibingo claims that he merely "thanked him" for "a very good step taken." (Id.) But Kibingo's account is inconsistent with the information that was then passed on to the Americans. An internal FBI memo dated January 11, 2002, indicates that the initial information received from the Rwandans (presumably from Kibingo) was that Karake had killed a white woman.*fn25 (Gov't Ex. 99, tab 33 at 2.) There was no mention of two additional murders, nor was there any mention of a man with a tattoo. (Id.)
2. January 2002
The Rwandan police contacted RSO Bachmann's investigator, Isidore Kayumba, on January 9, 2002, to inform him that they had an individual willing to talk about Bwindi. (5/10 a.m. tr. at 27; see also Gov't Ex. 30, tab 1.) Based on the information provided by the Rwandan police, Bachmann scheduled an interview with Karake on January 12, 2002. This interview took place at Kacyiru, in the same conference room where the FBI had conducted the December interviews. (5/10 a.m. tr. at 29-31.) According to Karake, Kayumba picked him up at Kami and drove him to Kacyiru. (5/22 p.m. tr. at 43.) Present at the interview were Rwandan officials Egide Ruzigamanzi and Captain John Bosco Kabanda. (5/10 p.m. tr. at 45.) Karake testified that Kibingo and Alfred Ndabarasa were also present for at least part of the interview (5/22 p.m. tr. at 44, 47), though this is not confirmed by Bachmann's testimony. Isidore Kayumba served as the interpreter. (Id. at 50; 5/10 p.m. tr. at 35.) Karake was provided with bottled water and bathroom breaks during the interview. (5/10 p.m. tr. at 62.)
According to Bachmann, Karake was read his rights in Kinyarwanda at the outset of the interview by Kayumba, and he consented to the interrogation in writing. (Id. at 49-51; see also Gov't Ex. 31.) Karake claims, however, that he did not sign the advice of rights until after "two hours" of questioning.*fn26 (5/22 p.m. tr. at 46.) There is nothing on the advice of rights form (Gov't Ex. 31), in Bachmann's notes (Def. Ex. K-4), or in Bachmann's cable summarizing the interview (Gov't Ex. 30, tab 1) that indicates at what time or point in the interview the advice of rights was executed. At some point, Bachmann took a photograph of Karake that depicts only his head and upper body. (Gov't Ex. 62.) Bachmann also testified that near the conclusion of the interview he asked the Rwandans who were present (except Kayumba) to leave him alone with Karake, at which time he inquired, using Isidore Kayumba to translate, about Karake's treatment at Kami Camp. (5/10 a.m. tr. at 44-45, 60.) Karake expressed no complaints at that time. (Id.) Bachmann's inquiry into Karake's conditions of confinement was not memorialized in either his own notes (Def. Ex. K-4) or in his subsequent cable. (Gov't Ex. 30, tab 1.)
During the course of the interview, Karake informed the investigators that he had killed an American tourist with an axe after having been ordered by his superiors to do so. (5/10 a.m. tr. at 54.) Everyone present at the interview, including Kayumba, was asking questions of Karake. (5/22 p.m. tr. at 50.) Among the "significant statements" that stuck out in Bachmann's mind from the interview was his detailed description of the man he killed.*fn27 (5/10 a.m. tr. at 54.) This description was of "a white male, rather large and a tattoo of an eagle." (Id. at 55; see also supra note 24.) At the conclusion of the interview, Bachmann asked Karake "if he wouldn't mind writing out what he just told" the investigators. (Id. at 66.) Karake did so.*fn28 (Gov't Ex. 10.) After Karake finished writing his statement in Kinyarwanda, Bachmann had Isidore Kayumba review the statement to see "if there was anything different than what [they] just heard." (Id. at 69-70.) Kayumba did not inform Bachmann of any discrepancies, even though there are significant omissions from Karake's written statement, particularly any mention of a tattoo on the male supposedly killed by Karake. (Gov't Ex. 10.) In fact, this written statement is completely non-specific with regard to even the gender of the person killed.*fn29 (Id.) Upon returning to the embassy, Kayumba did a written translation that was transmitted to the FBI. (Id. at 72.) Two days later, on January 14, 2002, Bachmann filed a cable with the State Department summarizing his interview with Karake, including the description of a "white male" with "a tattoo of an eagle on his arm." (Gov't Ex. 30, tab 1; Gov't Ex. 99, tab 33 at 2-3.) The cable, as well as Karake's written statement and translation (Gov't Ex. 10), was received by the FBI on January 16, 2002.
At about this time, in mid-January 2002, Nyaminani's living conditions underwent a dramatic change. (5/3 p.m. tr. at 61.) Kibingo had been informed by Desire Hategekimana in their very first conversation at Kami that Nyaminani was "not allowing [the other suspects] to tell the truth." (Id. at 57; 5/4 p.m. tr. at 11.) He was, in Desire's words, "moving amongst us and telling us that we should keep quiet and not talk." (5/3 p.m. tr. at 57.) According to Desire, this had been going on while they were in Ruhengeri, well before their arrival in Kami Camp. (5/4 p.m. tr. at 11.) "[A]fter some time," Kibingo made the decision to separate Nyaminani from the other suspects.*fn30 (Id. at 61.) On January 15, Nyaminani was placed in his own room "so that he [would] not be a problem to the investigators who [were] investigating what happened at Bwindi." (5/4 p.m. tr. at 11, 13.) Even though Nyaminani would be locked alone in his room at night (id. at 14), according to Kibingo, he was permitted to interact with the other suspects during the day. (Id. at 32.)
Also around this time, the FBI began planning a return trip to Rwanda. (Gov't Ex. 99, tab 31.) The FBI raised the possibility of videotaping the interviews (Gov't Ex. 99, tab 30 at 2-4), but Dent was informed by "FBI management" that videotaping is "not something that the FBI typically does." (5/19 a.m. tr. at 87-88.) By January 25, 2002, the FBI's itinerary was in place. (Gov't Ex. 99, tab 31 at 2-3.) Dent and her team were scheduled to arrive in Kigali on the morning of February 4. (Id. at 3.) Their first day was to be taken up in meetings with the U.S. ambassador and various Rwandan investigative authorities. (Id. at 2.) The next ten days were scheduled for interviews of numerous Bwindi suspects, including some "not yet identified." (Id. at 2-3.)
3. February 2002
Between February 5 and 15, the FBI conducted interviews of Karake, Nyaminani, Leonard Harerimana, Saidi Mbarushimana, Fabien Nkundabaza, Silver Dukuzemuremyi, Jean Bosco Hategekimana and Colonel Bemera. (FBI 29, tabs 14-24.)
The FBI's first interview during this trip was of Karake on February 5, 2002.*fn31 (Id., tab 14.) Special Agents Paun and Dent conducted the interview together with Ndabarasa, Kabanda, Ruzigamanzi, and Kayumba. (Id. at 1.) Ndabarasa, Ruzigamanzi and Kayumba all provided translation assistance at different points during the interview. (Id.) When he arrived at Kacyiru that day, Karake had a bandage on his left wrist and an infection on his forearm. (Id.) The FBI agents inquired about the nature of his injuries. (Id.; 5/15 p.m. tr. at 90-92.) Dent testified that Karake informed them that he had blisters that had become infected. (5/15 p.m. tr. at 92-93.) The FBI did not inquire as to the cause of the blisters. (Id. at 93.) At the end of the interview, the FBI photographed Karake's wrists and arms. (Gov't Ex. 29, tab 14 at 14; 5/16 a.m. tr. at 5; Gov't Ex. 12, 13 & 13a.) Before conducting any questioning of Karake, the FBI reminded him of the advice of rights he signed during his January 12 meeting with Bachmann, and Karake signed an English advice of rights form. (Gov't Ex. 11.)
During the course of the interview Karake informed investigators that ALIR's primary objective was to target the government of Rwanda and its secondary objective was to target friends of the Rwandan government. (Gov't Ex. 29, tab 14 at 3.) He went on to describe ALIR's organizational structure before discussing the specifics of the Bwindi operation. (Id. at 3-6.) Karake stated that he was ordered by Ntabwoba and Murenzi to kill three white men. (Id. at 9.) He was instructed not to use his gun, so instead he took an axe and struck a tall white man in the back of the head three times with the butt of the axe. (Id.) Karake described the man, who was lying face down on the ground, as over six feet tall, wearing khaki pants, a vest and a shortsleeved t-shirt.*fn32 (Id.) Despite the short sleeved t-shirt, Karake claimed to have noticed a "dark- colored tattoo of a bird with large claws on [the victim's] upper arm." (Id.) Karake could not remember which arm the tattoo was on or whether the tattoo was multi-colored. (Id.) Another member of Karake's unit, Corporal Habimana, took the axe and killed the two remaining men. (Id. at 10.) Soon after these killings, Karake claims he saw a fourth victim murdered. This "killing took place very quickly and at a distance," leaving Karake unsure whether the victim was male or female. (Id. at 11.) The victim, however, had "a lot of hair." (Id.) Karake heard from an unnamed source that "Minani" was the killer, but wasn't sure if Minani was Gregoire Nyaminani. (Id.) Karake then described the looting that went on, and explained that notes were given to certain tourists before they were released. (Id. at 12.) Before the interview ended, Karake was shown a series of photographic spreads of ALIR members and asked to identify anyone he recognized. (Id. at 13-14.) At the conclusion of the interview, Kayumba drove Karake back to Kami Camp. (5/22 p.m. tr. at 55-56.)
Based on this interview on February 5, Dent was of the opinion that Karake had been truthful with them. (5/16 a.m. tr. at 11.) In fact, because Karake had "been so forthcoming," the FBI brought him back to Kacyiru the following day in order to get more information "about the attack at large." (5/16 a.m. tr. at 5.) No advice of rights form was executed on this day, but according to Dent, Karake's rights were read to him and he consented to the interrogation.*fn33
(Gov't Ex. 29, tab 15; 5/16 a.m. tr. at 12.) Present at Karake's interview on February 6 were Special Agents Dent and Paun, Ndabarasa, Ruzigamanzi, Kabanda and Kayumba. (Gov't Ex. 29, tab 15 at 1.) Ndabarasa served as the translator. (Id.) During this interview, which was substantially shorter than the one the day before, Karake "confessed that he had nightmares" after Bwindi, and that a soldier named Gafaransa teased him about them, saying, "have nightmares for the people he (Gafaransa) killed also." (Id. at 2; see also 5/16 a.m. tr. at 13.)
On February 7, 2002, the FBI interviewed Nyaminani for a third time.*fn34 In addition to the two FBI agents, present at the interview were Ndabarasa and Ruzigamanzi. There is no record of Isidore Kayumba's attendance at this interview, though Nyaminani testified that it was Kayumba who transported him from Kami Camp to Kacyiru that day. (5/24 p.m. tr. at 59.) Ndabarasa served as the primary translator at this interview, and read Nyaminani his rights in Kinyarwanda. (Gov't Ex. 29, tab 17 at 1.) Nyaminani signed the advice of rights form. (Gov't Ex. 14.) At this meeting Nyaminani named numerous individuals who participated in the Bwindi attack, and confirmed that he was the commander of Section 1 in Irondelle Company's Third Platoon. (Gov't Ex. 29, tab 17 at 2.) Nyaminani described the attack on Bwindi, the gunfight, the looting that ensued, and the injuries suffered by ALIR soldiers. (Id. at 3-6.) Though he admitted to participating in the attack, Nyaminani again denied any participation in or knowledge of the murders of the white tourists. (Id. at 10.) Nyaminani noted that Karake would sometimes "talk in his sleep about the killings." (Id. at 8.)
Over the next several days, the FBI conducted a follow up interview with Jean Bosco Hategekimana (Gov't Ex. 29, tab 18) and two interviews of Saidi Mbarushimana. (Id., tab 19-20.) By this time, however, the Rwandan authorities decided that the FBI's method of conducting "interviews . . . was really getting tedious." (5/2 a.m. tr. at 99.) According to Ndabarasa, the Americans were "somewhat methodical;" they were "very particular about everything." (Id. at 100.) As a result, the process "took so much time," and, as Ndabarasa repeatedly testified, it was "tedious" and "systematic." (Id. at 99-100; 5/12 a.m. tr. at 91.) Ndabarasa's "colleagues from the police and from the army" wondered where the investigation was "going with this kind of conducting interviews." (5/2 a.m. tr. at 101.) The Rwandans "wanted to get to the root" (id. at 96) and concluded that "if [the Americans were] going to do things their way at least we should [know] what actually took place." (Id. at 101.) Therefore, on February 12 the Rwandans brought to Kacyiru all the Bwindi suspects, roughly ten in number, for a different "type of interview process." (Id. at 97, 100.) Whatever information the Rwandans gathered, however, was to be "shared" with the Americans. (Id. at 102.)
The FBI investigative team had been scheduled to conduct interviews on February 12 (Gov't Ex. 99, tab 31 at 3), but they did not participate in the Rwandans-only session at Kacyiru on that day.*fn35 (5/2 a.m. tr. at 101). Dent had no recollection at the hearing as to why the FBI's schedule was changed or what she and her team did instead. (5/19 a.m. tr. at 58-60.) Though it is not clear precisely which Rwandans participated in the February 12 interrogations, the group included at least Ndabarasa and Kibingo. (5/12 a.m. tr. at 110.) Ndabarasa testified that other likely participants included Jean Bosco Kabanda and Egide Ruzigamanzi, which is consistent with Nyaminani's testimony. (Id. 112-116; 5/25 a.m. tr. at 29, 56.) According to Nyaminani and Karake, Isidore Kayumba was also present and participated in the interrogation on February 12. (5/22 p.m. tr. at 62; 5/25 a.m. tr. at 31, 56.) Ndabarasa described the process employed by the Rwandans as follows: "Bring in one, send the other out, bring in another, send the other out." (5/2 a.m. tr. at 100.) The order in which the suspects were chosen to be interviewed was random. (Id. at 98.) This way, "each of them knew we were talking about them [to] the other." (Id. at 99.)
The Rwandan method apparently had dramatic effects. Ndabarasa testified that February 12 was "the first time . . . Nyaminani and Karake confessed and tell us how they did the actual killing of the tourists." (Id. at 95.) According to Ndabarasa, Karake admitted to killing "tourists." (Id. at 98.) Ndabarasa's notes from this interview indicate that Karake admitted that the "3rd group was killed by both Karake and Habimana."*fn36 (Id. at 3; 5/3 p.m. tr. at 19.) This is consistent with his statements to the FBI on February 5 and 6. (Gov't Ex. 29, tabs 14 & 15.) According to Ndabarasa, however, Karake also wrote and signed statements on February 12 "acknowledg[ing] his role in killing two additional people." (Def. Ex. B-4 at 4.) No such statements were proffered at the hearing. Moreover, there is no indication in Ndabarasa's notes that Karake implicated himself in any murders other than the group "killed by both Karake and Habimana." (Gov't Ex. 44 at 3.)
Ndabarasa also testified that Nyaminani "admitted to having killed a group of three women." (5/3 p.m. tr. at 19; 5/12 a.m. tr. at 96.) This testimony is also not corroborated by his notes from the February 12 interviews. (Gov't Ex. 44.) His notes indicate, rather, that Nyaminani identified six individuals as "the people he believe[d] killed the white people." (Id. at 2.) These were Karake, Gafaransa, Nkundabagenzi, Habumugisha, Mandeve and Manudi. (Id.; 5/2 p.m. tr. at 15-16.) While Ndabarasa's notes do indicate that Karake identified Nyaminani as one of the people who "participated in the killings" (5/2 a.m. tr. at 98), and that Karake was told by Mulenzi "that the white woman was killed by Minani" (Gov't Ex. 44 at 1), it is not clear whether Karake was attributing one or multiple murders to Nyaminani. At a minimum, Ndabarasa's notes of his interviews fail to corroborate his testimony that Nyaminani took any responsibility for any of the killings. (Gov't Ex. 44.)
Sometime on February 12 or 13 Special Agent Dent received a phone call from Ndabarasa indicating that Karake had "confessed to participating in two more murders at Bwindi." (5/16 a.m. tr. at 18.) Dent admitted to being "surprised to learn" of Karake's new admissions, but despite being "curious," she chose not to make any inquiry into "how it was that [Ndabarasa] came to learn this." (Id. at 19.) She explained that she "didn't ask" the Rwandans how they were able to extract information from the defendants more successfully than the Americans because "[n]one of [the suspects] were in U.S. custody. None of them were charged by the United States." (Id.)
She did, however, conduct a third interview of Karake on February 13.*fn37 (5/16 a.m. tr. at 20; Gov't Ex. 29, tab 21.) Also in attendance were FBI Agent Paun, Ndabarasa, Ruzigamanzi, Kabanda, and Kayumba. (Gov't Ex. 29, tab 21 at 1.) Ndabarasa served as the primary translator. (Id.) Karake was read his rights in "the same manner as he was on prior occasions" (5/16 a.m. tr. at 20-21), after which he signed an English advice of rights form. (Gov't Ex. 15.) Thereafter, the FBI informed Karake that "they had learned since their last meeting with Karake that he'd reported new information to the Rwandan authorities." (Gov't Ex. 29, tab 21 at 1.) Karake "apologized for wasting their time" and proceeded to describe two additional murders. (Id.) Karake confessed to using the butt of an axe to kill two people who were lying face-down on the ground and were covered with bed sheets.*fn38 (Id. at 1-2.) He could not identify the gender of either person. (Id. at 3.) These killings took place prior to the murder of the white male Karake had previously confessed to killing. (Id. at 2-4.) Karake indicated that Bimenyimana was present when he killed the two people. (Id. at 3.) He also heard Bimenyimana state at a later point that "he wanted to kill the rest of the white people." (Id. at 5.) Finally, Karake told the investigators that he suffered from a recurring nightmare involving a "big bird with big claws." (Id. at 7.) Before the interview concluded, the agents observed that the swelling previously visible on Karake's left wrist had subsided considerably. (Id.)
After conducting more interviews with other suspects over the next several days, the American investigative team departed Rwanda and did not return until May 2002.
4. March 2002
Interrogations of the defendants continued despite the FBI's absence. On March 5, the Rwandan National Police contacted Bachmann's investigator, Isidore Kayumba, to inform him that Nyaminani was willing to talk more about the Bwindi attack. (5/10 a.m. tr. at 72-74; Gov't Ex. 30, tab 2 at 1.) Though Bachmann had not participated in the FBI's prior interviews of Nyaminani, he was aware that they had occurred. (5/10 a.m. tr. at 75.) Bachmann called back the police and set up an interview at Kacyiru on March 7, 2002.*fn39 (Id. at 74.) At the outset of the interview, Bachmann orally advised Nyaminani of his rights, and was informed through Kayumba that he consented to the interrogation. (Id. at 76, 78.) There was, however, no advice of rights signed by Nyaminani. (Id. at 85-86.) While Bachmann testified that he either read the modified Miranda rights from an English form or used an ordinary Miranda advice of rights card that he carried in his wallet (id.; see also Def. Ex. N-1), the Court concludes from his testimony that he relied on the Miranda rights on his wallet-size card, which was the only form of rights available to him during the interview. (See supra note 15 and Section V(B)(1).) Also present at the interview were Ruzigamanzi, Kabanda, and possibly Ndabarasa. (Id. at 88-89.) Kayumba translated. (Id.)
In this interview on March 7, 2002, Nyaminani "revealed how he assisted with the murder of a white male and white female." (Gov't Ex. 30, tab 2 at 1; 5/10 a.m. tr. at 79.) After describing the attack, Nyaminani informed the investigators that he was ordered by Bimenyimana ("Zappi Gadi") to find Ntabwoba, who in turn ordered him to take the man and woman into the woods and have them killed. (Gov't Ex. 30, tab 2 at 1-2.) Based on the testimony and the documents memorializing Nyaminani's interrogation, this group of two tourists appears to be the same two victims Karake had confessed to murdering in his interview only a few weeks before on February 13. Nyaminani said that he stood watch with his gun to ensure the tourists didn't try to escape while Corporal Nkundabagenzi beat them to death with a wooden club. (Id.) Bachmann "didn't feel [Nyaminani's statements] were genuine," (5/10 a.m. tr. at 90), largely, it appears, because his "account of the story differs from all those that have previously confessed." (Gov't Ex. 30, tab 2 at 2.) Specifically, according to Bachmann's cable to the FBI agent, the "FBI, RSO and Rwandan police [had] a good idea what happened to the first two murdered groups [i.e., the three men and second group of two], however the last group of tourists where Nyaminani was seen going off with [was] still a mystery." (Id.) The "last group" refers to "the three women [who] were brutally raped and murdered." (Id. at 3.) These post-interview notations by Bachmann lend credence to Nyaminani's testimony that during the course of the interrogation, "the interpreter," Isidore Kayumba, "indicated to [him] that they had done investigations," and "that those who had killed the men had been found and the only thing that is left is those who killed the three women." (5/25 a.m. tr. at 65.) Nyaminani "was instructed to think hard about his situation and was taken away into military custody." (Gov't Ex. 30, tab 2 at 3.) At the conclusion of his cable, Bachmann noted, without explanation, that "RSO and Rwandan National Police believe he is near a confession." (Id.)
Defendant Nyaminani was returned to Kami Camp following his March 7 interview. The following day, Kibingo resumed questioning him, which resulted in a written statement that was drafted on the afternoon of March 8, 2002.*fn40 (Gov't Ex. 53 & 53a; 5/25 a.m. tr. at 72.) In his statement, Nyaminani apologized for having lied to investigators previously and admitted to raping one woman and overseeing the murder of that woman and two others by fellow soldiers Manudi, Havugimana and Nkundabagenzi. (Gov't Ex. 53a.) Nyaminani described the woman he raped as approximately "1.65 cm [sic] tall, . . . wearing tight pants . . . and a short sleeves [sic] T-shirt." (Id.) He also described her as "wearing underpants that looked like red inside."*fn41 (Id.) The statement then recounts that the other soldiers, who were under Nyaminani's command, murdered the three women using small clubs as he "was standing by." (Id.)
Not two weeks later, however, on March 20, 2002, Nyaminani drafted another written statement at the behest of Kibingo.*fn42 (Gov't Ex. 54 & 54a.) In his March 20 statement, Nyaminani retracted his prior confession of rape, and now denied any participation in a rape. He claimed to have "arrived" to find his men raping the women and that "when they finished, we killed them." ( Gov't Ex. 54a.) In the statement Nyaminani accuses the same three soldiers that he named on March 8 -- Manudi, Havugimana*fn43 and Nkundabagenzi -- of raping and murdering the women. (Id.) Nyaminani also stated that he and his soldiers left the women "dying," but not yet dead, because they were in a hurry to escape from the Ugandans. (Id.)
5. May 2002
In late May and continuing into early June, the investigation began moving forward rapidly. Around May 20, Rwandan investigators again contacted Kayumba to inform him that Nyaminani "had more information" and "was willing to talk." (5/10 a.m. tr. at 91-92.) On May 22, Nyaminani was brought back to Kacyiru for another interview with RSO Bachmann.*fn44
(Gov't Ex. 30, tab 3.) Present at the interview were Kayumba (5/10 a.m. tr. at 92), Kabanda (id. at 100), and Ruzigamanzi. (Gov't Ex. 32.) Bachmann began the interview by telling Nayaminani: "I'm informed that you would like to tell us more information today and this time I would like the truth." (5/10 a.m. tr. at 93.) Bachman claims that he was not informed prior to the interview that Nyaminani had confessed to participating in the rape and murder of three women. (Id. at 96, 100.) Bachman then had Kayumba translate a French advice of rights form to Nyaminani, who signed it. (Gov't Ex. 32.) In response to questioning, Nyaminani told a story similar to that contained in his March 20 statement, i.e., that he found his soldiers raping the women and ordered them to stop raping them and kill them.*fn45 (Compare 5/10 a.m. tr. at 98 with Gov't Ex. 54 & 54a.) Bachmann had Nyaminani draft another written statement reiterating the facts of that day's confession. (Gov't Ex. 33a.; see also Gov't Ex. 30, tab 3 at 4.) RSO Bachmann testified that, at the conclusion of the interview, he asked Nyaminani, "outside the presence of Rwandan law authorities," whether he had been subject to "any undue pressure or coercion" for him to confess. (5/10 a.m. tr. at 97.) Through his translator Kayumba, Bachmann was informed that Nyaminani was not under "any pressure or undue stress." (Id.) Bachmann sent a cable summarizing the interview to Dent in Washington, D.C. (Gov't Ex. 30, tab 3 at 4.) Though this cable indicates that Bachmann also forwarded Nyaminani's March 8, March 20 and May 22 written statements and translations (id.), Agent Dent claims not to have received those statements until after she returned from her trip to Rwanda in May and June of 2002. (5/16 a.m. tr. at 33, 52.)
On May 31, 2002, the FBI returned to Rwanda to resume the investigation. (Gov't Ex. 99, tab 29 at 2.) During their one week stay in Kigali, the agents met with the U.S. ambassador to discuss extradition of Karake, who was to be indicted soon. (Id.) They then met with Commissioner General Frank Mugambage and Colonel Jack Nziza to thank them for their cooperation and the assistance of their personnel. (Id.) Dent and her team also conducted several more interviews of Bwindi suspects, including five more meetings with Nyaminani. (Gov't Ex. 29, tabs 25-28 and 31.)
The first of these meetings did not involve extensive questioning regarding Bwindi.*fn46
(Gov't Ex. 29, tab 25.) On May 31, FBI investigators met with Nyaminani primarily to obtain consent to take a DNA sample. (5/16 a.m. tr. at 40.) Present at the interview were Special Agents Dent and Prouty, Bachmann, Kayumba, Kabanda, Ruzigamanzi, and Ndabarasa. (Gov't Ex. 29, tab 25.) Kayumba served as the translator. (Id.) Though Dent testified that there was no interrogation conducted on May 31 (5/16 a.m. tr. at 42), the FBI 302 summarizing the meeting indicates that, prior to any advice of rights, "Nyaminani was asked about recent statements made by him." (Gov't Ex. 29, tab 25.) Nyaminani was also "shown photographs of the victims" of the Bwindi attack.*fn47 (Id.) At this point, the FBI brought up the issue of DNA testing, and a DNA consent form was translated by Kayumba and signed by Nyaminani. (Id.; Gov't Ex. 17.) Nyaminani was then transported to a medical facility where the DNA sample was to be taken. (Gov't Ex. 29, tab 25.) After arriving, but before he entered the facility, Nyaminani was presented with the FBI's overseas advice of rights form, which he signed. (Id.; Gov't Ex. 16.) A DNA sample was taken from Nyaminani, but investigators were never able to find a DNA match between Nyaminani and any of the victims.*fn48
6. June 2002
On June 1, Nyaminani was brought back to Kacyiru for another meeting with Dent and Prouty.*fn49 Also present were Ruzigamanzi and Kabanda. Kayumba served as the translator. (Gov't Ex. 29, tab 26.) The FBI informed Nyaminani that they were aware of his confession to Bachmann on May 22 and explained that "they were not judging him; however, it was important for him to be truthful." (Id. at 1.) The modified Miranda warnings were then translated for him, and he consented to be interviewed. (Gov't Ex. 18; 5/16 a.m. tr. at 44-45; Gov't Ex. 29, tab 26 at 1.) Nyaminani provided essentially the same statement he had made earlier to Bachmann: Commander Ntabwoba ordered him to kill the three women; Nyaminani sent them into the woods with three of his soldiers (Manudi, Nkundabagenzi and Gafaransa) and when he caught up to them in the clearing, he found them raping the women. He ordered them to stop raping and start killing, and they departed after beating the women badly but before they died. (Id. at 2-3.) Agents then asked Nyaminani to draw a diagram of the location where the crime occurred and he complied. (Gov't Ex. 19 & 29, tab 26 at 4.) When questioned about the woman he described raping in his March 8 statement to Kibingo (Gov't Ex. 53 & 53a), Nyaminani stated that he had lied and that he didn't see anyone wearing red underwear. (Gov't Ex. 29, tab 26 at 5.) Nyaminani accepted responsibility for ordering the killings and witnessing the rapes. (Id. at 9.)
When asked to identify the objectives of ALIR, Nyaminani professed not to know what ALIR's objectives were. (Id. at 7.) When reminded that on December 1, 2001, he had indicated that one ALIR objective was to target supporters of the government of Rwanda, Nyaminani stated that Americans were the number one enemy of ALIR and British citizens were the number two. (Id.) He stated that ALIR soldiers were taught to kill all captured American and British citizens. (Id.) Nyaminani was also under the (incorrect) impression that all three female victims killed by his soldiers were American. (Id. at 9.)
At the outset of the interview, soon after Nyaminani signed the advice of rights form at 11:19 a.m., the agents asked Nyaminani to describe the three female victims. (Id. at 1.) At this point, Nyaminani was only able to provide a very general description: the women were all white, with hair just above the shoulders, and taller than 5'6" or 5'7". (Id.) He said that one woman was wearing pants. (Id.) Sometime later, prior to a break at 3:45 p.m., investigators showed Nyaminani a photograph of each of the victims (Gov't Ex. 21; 5/16 a.m. tr. at 54-55), but he was unable to determine who raped whom. (Id. at 5.) Nyaminani was also unable to remember if any of the women wore glasses. (Id.) Much later in the interview, however, sometime after returning from a break at 5:20 p.m., and a further review of the black and white photographs, Nyaminani indicated that Manudi had raped Joanne Cotton, but that she was not wearing a hat or glasses as in her photograph. (Id. at 10; see Gov't Ex. 19.) When asked whether she had a British accent, Nyaminani did not know. (Gov't Ex. 29, tab 26 at 11.) He also could not recall her hair color but said her hair reached just beyond shoulder length. (Id.) He also identified Rhonda Avis as the woman raped by Gafaransa and Susan Miller as having been raped by Nkundabagenzi. (Id. at 10.) Nyaminani said he did not know Susan Miller's hair length. (Id. at 12.) Nyaminani was asked to write on the back of each photograph the identity of the person who raped each woman, which he did. (Id.; Gov't Ex. 19.) The interview was concluded at 10:15 p.m., after over 10 hours of questioning, interrupted by only two breaks totaling 45 minutes. (Gov't Ex. 29, tab 26 at 12.)
Approximately 12 hours later, on June 2, 2002, Nyaminani returned to Kacyiru for further interrogation. The questioning commenced at 10:16 a.m. and was attended by the same group as was present the prior day.*fn50 (Gov't Ex. 29, tab 27 at 1.) Again, Kayumba translated. (Id.) Nyaminani was advised of his rights and consented to be interviewed.*fn51 (Gov't Ex. 20.) After a series of background questions regarding his previous military experience and prior ALIR operations, "Nyaminani asked investigators what they wanted. Agents said they wanted the truth about the rapes and murders of the three women." (Gov't Ex. 29, tab 27 at 2.) In response, Nyaminani "confessed that he witnessed actual penetration of the women." (Id.) The investigators proceeded to question Nyaminani about the details of the rapes and murders. He informed them that Rhonda Avis had cried the most, because Gafaransa had been rough with her. (Id. at 3-4.) Nyaminani also stated that as he and his men left the scene, they heard Susan Miller gasping for air; Gafaransa was sent back to "finish her off." (Id. at 4.)
After extensive further questioning, agents informed Nyaminani that they "didn't believe he . . . didn't participate" in the rapes and that they "were tired of documenting lies." (Id. at 11.) A short break was taken, and when questioning resumed, Nyaminani admitted to raping Susan Miller, whom he identified by her photograph. (Id.) Nyaminani then proceeded to recount the rape again, including further details regarding the appearance of the three women.*fn52 Nyaminani stated that he raped Susan Miller, that Gafaransa and Manudi raped Rhonda Avis, and that Nkundabagenzi raped Joanne Cotton. (Id. at 12-13.) To reflect these new admissions, Nyaminani changed the statements he drafted the prior day and indicated his name next to one of the female victims' photographs. (Id. at 14; Gov't Ex. 21.) The interview was nearly eight hours long, concluding at 6:45 p.m.*fn53 (Id. at 14.)
The FBI brought Nyaminani back for a fourth consecutive day on June 3.*fn54 (Gov't Ex. 29, tab 28.) The purpose of this interview was to "make sure that his story wasn't going to change." (5/16 a.m. tr. at 77.) Nyaminani was again advised of his rights, and consented to be interviewed. (Id.; Gov't Ex. 22 & 29, tab 28 at 1.) Nyaminani reiterated his statements from the previous day: that he participated in the rapes and ordered, but did not carry out, the murders. (5/16 a.m. tr. at 78-80; Gov't Ex. 29, tab 28.) According to Dent, his statements on June 2 and June 3 were "substantially consistent." (5/16 a.m. tr. at 80.)
On June 4 and 5, Dent and Prouty interviewed other Bwindi suspects. (Gov't Ex. 29, tabs 29 & 30.) On June 6 they requested yet another interview with Nyaminani.*fn55 (Id., tab 31.)
Present at this interview, at various times, were Ndabarasa, Ruzigamanzi and Kabanda. (Id.) Kayumba provided translation assistance. (Id.at 1.) During the interrogation on June 6, Nyaminani's story remained consistent with his statements on June 2 and 3. He also related that he and Leonard Harerimana had encouraged Karake to confess to his role in the killings at Bwindi, and that, after confessing, Karake had assured him that he needn't be afraid to confess; that "his situation had remained the same." (Id. at 2-3.) When asked about Kibingo, Nyaminani said that he "wasn't afraid of Kibingo and that he . . . felt free to speak with" him. (Id. at 4.) He then told investigators how, during a meeting with Kibingo in March (5/25 p.m. tr. at 15-20), he had denied killing anyone at Bwindi; Kibingo told him not to lie and then "slapped Nyaminani four to five times on the lower back with a styrafoam [sic] flip flop." (Gov't Ex. 29, tab 31 at 4.) Nyaminani explained that he raped but didn't kill and therefore didn't think that he was lying to Kibingo. (Id.) Kibingo then called together several other Bwindi suspects who were at Kami, and Nyaminani confessed in front of them while writing his statement. (Id.) Nyaminani told investigators that what he told Kibingo was the same thing he told them on June 2. (Id.) Near the conclusion of the interview, he told investigators that the reason he lied to them previously was "because he didn't believe the truth would be learned about what occurred that day." (Id. at 6.)*fn56
A little over one week later, on June 16, Bimenyimana was arrested in the Congo. Bimenyimana was taken first to Nyamilima and then, after only one day, to Rutshuru. (Id. at 66-67.) Both of these places were RPA military facilities located in the Congo. (Id.) Bimenyimana was not beaten at either of these facilities, though he was questioned at gunpoint regarding his role in ALIR. (Id. at 67-69.) From Rutshuru he was transferred to a DMI facility in Rwanda called Nu-Kizungu, where he faced continued questioning regarding ALIR's operations. (Id. at 69-72.) Bimenyimana claims to have heard people being beaten at Nu-Kizungu. (Id. at 72.) He remained there for approximately three weeks before being transferred to Kami Camp. (5/30 p.m. tr. at 52.)
7. July -- August 2002
On the evening of July 10, 2002, Bimenyimana was moved to Kami Camp. (Id. at 56-58.) Kibingo testified that at the time Bimenyimana arrived at Kami, the Rwandans did not know his true identity because he was using the alias "Tuyigire Mustapha." (5/4 p.m. tr. at 48.) Kibingo stated that he only learned that Bimenyimana was the man known as "Zappy Gadi" when the other Bwindi suspects pointed him out in August. (5/4 p.m. tr. at 49.) Bimenyimana admits that he was using that name at the time he was arrested (5/30 a.m. tr. at 62), but claims that he gave Rwandan authorities his real name as early as his detainment at Nyamilima on June 16 or 17. (5/30 a.m. tr. at 67; 5/30 p.m. tr. at 51.) According to Bimenyimana, the soldiers in charge of Kami knew full well who he was on the day he arrived. (5/30 p.m. tr. at 67.) Bimenyimana's version is consistent with information known to the FBI. On June 18, 2002, Dent drafted a memo regarding her recent investigative trip to Rwanda in which she notes: "Recent information derived from a credible source suggests that Zappy Gadi . . . may be in Rwandan custody." (Gov't Ex. 99, tab 29 at 6.) Dent speculated that "the Rwandans are withholding [his] whereabouts . . . for national security purposes to glean all available information about the locations of Interahamwe soldiers at large before notifying the FBI of [his] apprehension." (Id.) The Court, therefore, does not credit Kibingo's testimony that it was several weeks before he learned of Bimenyimana's identity, especially given the investigators' intense focus on Bimenyimana as an ALIR leader as early as November 2001. (Gov't Ex. 99, tab 32 at 9.)
According to Kibingo, Bimenyimana was questioned only once at Kami Camp regarding his involvement in the Bwindi attack. (5/4 p.m. tr. at 57-58.) It is not clear precisely when this interrogation is supposed to have taken place, but at some point after learning of Bimenyimana's identity from other Bwindi suspects held at Kami Camp, Kibingo asked him about Bwindi and was told, "we really attacked it."*fn57 (5/4 p.m. tr. at 52.) Bimenyimana claimed he "did not agree" with Ntabwoba's decision to kill the tourists.*fn58 (Id. at 52, 55). After hearing Bimenyimana's story, Kibingo "told him to be ready, that . . . he'll be asked about it." (Id. at 53.) Kibingo then informed his DMI supervisors, Nziza and Kabanda, that he had Zappy Gadi in custody. (Id. at 55.)
As late as August 13, the Rwandans had not informed U.S. officials of Bimenyimana's capture. (Gov't Ex. 99, tab 38 at 3-5.) In late August, however, RSO Bachmann "was invited by the Rwandan National Police to question Bimenyimana." (Gov't Ex. 30, tab 4 at 1.) As with Bachmann's previous interviews, the Rwandans first made contact with Kayumba approximately two days before the interview took place. (5/10 p.m. tr. at 13-14.) On August 30, Kayumba went to Kami and brought Bimenyimana to Kacyiru for an interview.*fn59 (5/31 a.m. tr. at 26-27.) According to Bachmann, probably present at this meeting were Ruzigamanzi, Kabanda and Ndabarasa. (5/10 p.m. tr. at 23.) Using Kayumba as his translator, RSO Bachmann advised Bimenyimana of his rights, using a French language form. (Id. at 16-17.) Bimenyimana signed the form. (Gov't Ex. 34.) Bimenyimana described the Bwindi operation in detail, stating that "his platoon gathered the tourists and sent them to Ntabwoba," and that on Ntabwoba's instructions, he ordered the burning of vehicles. (Gov't Ex. 30, tab 4 at 2.) Bimenyimana stated that a group of soldiers including Gregoire Nyaminani, Nkundabagenzi, and Gafaransa, "went into the forest with the women tourists." (Id. at 3.) He also explained that, after learning that Mandevu had killed a tourist, "he was shocked because that was not the mission." (Id.) Bachmann noted some inconsistencies in Bimenyimana's statements, and was particularly suspicious of his professed ignorance of some of Nyaminani's movements during the attack. (Id. at 4; 5/10 p.m. tr. at 21.) Bachmann photographed Bimenyimana before concluding the interview. (Gov't Ex. 63.) Bimenyimana was not asked to provide any written statement (5/11 p.m. tr. at 59), but he was given a list of nine questions that investigators wanted answers to at their next interview.*fn60 (Id. at 58; 5/31 a.m. tr. at 58-59, 66; 5/12 p.m. tr. at 55.)
8. September 2002
Bimenyimana returned to Kacyiru on September 2, 2002 for further questioning by Bachmann.*fn61 (Def. Ex. B-2.) No document was proffered reflecting that Bimenyimana was advised of or waived his rights prior to this interview, but Bimenyimana testified that he did in fact sign a rights waiver. (5/31 a.m. tr. at 76.) Despite having receiving a call from the Rwandan police "saying he's got more to tell you," (5/10 p.m. tr. at 33), there weren't "any major changes" from what he had said on August 30. (5/11 p.m. tr. at 63.) Though Bachmann took notes on September 2 (Def. Ex. B-2), he did not memorialize it in a cable, nor did he even reference the fact of the interview in the cable he sent on September 4, 2002 regarding the August 30 interview. (Gov't Ex. 30, tab 4.) This would be the last interview conducted by an American investigator until February 2003.
Also in early September, Thomas Murangira, who was at the time the Chief Inspector of the Rwandan National Police, was instructed by his superior, John Gacinya, to obtain written statements from the Bwindi suspects who were housed at Kami Camp. (5/15 a.m. tr. at 5.) These statements were "official statement[s]" that were for use by the judiciary police and prosecutors. (Id. at 17.) Prior to conducting the interviews, Murangira obtained Ruzigamanzi's notes regarding the case, as well as defendants' written statements. (Id. at 6, 9-10.) Murangira conducted interviews of ten individuals between the 4th and 10th of September (id. at 7), including interviews of Karake (Gov't Ex. 40), Nyaminani (Gov't Ex. 41), and Bimenyimana (Gov't Ex. 42).*fn62 (5/15 a.m. tr. at 7). The statements given by the defendants basically follow their most recent prior statements.*fn63 In addition to the defendants, Murangira interviewed Silver Dukuzumuremyi, Desire Hategekimana, Leonard Harerimana, Saidi Mbarushimana, Fabien Nkundabaza, Innocent Sebureteri, and Jean Bosco Hategekimana. (Id. at 19-20.) These interviews were conducted in Kinyarwanda with no one else, including any Americans, present. (Id. at 8-11.) No advice of rights was given because, under Rwandan law, none was required. (Id. at 38-41; Gov't Ex. 61.) All of Murangira's interviews were conducted at National Police Headquarters. (5/15 a.m. tr. at 7-8.) Murangira's method for obtaining an official statement was to ask questions and record both the question and the response in his own handwriting. (Id. at 12.) The interviewee was then given an opportunity at the conclusion of the interrogation to review the statement and make any changes or corrections that he desired. (Id. at 12-13.) Each defendant reviewed his statements for 10 to 20 minutes and none asked to make any corrections. (Id. at 13.) After completing the interviews, Murangira drafted a report regarding his findings for the prosecutor.*fn64 (Gov't Ex. 64.)
9. February 2003
The FBI returned to Rwanda in February 2003 to conduct further interviews of Bimenyimana.*fn65 (Gov't Ex. 29, tabs 37-39.) The first interview took place on February 21, 2003 at Kacyiru.*fn66 (Gov't Ex. 29, tab 37.) In addition to FBI Agents Dent and Prouty, Kayumba, Ruzigamanzi, and Ndabarasa were present. (Id. at 1.) All three Rwandans assisted with translation during the meeting. (Id.) Because Agent Dent had been informed that Bimenyimana spoke French, the investigators provided him with a French advice of rights, which he read to himself. (5/16 a.m. tr. at 90-91.) It was not translated into Kinyarwanda for him. (Id. at 91.) Bimenyimana verbally consented to be interviewed and signed the advice of rights form. (Id; Gov't Ex. 24.) When asked about his health, Bimenyimana indicated that he received medical treatment when necessary. (Gov't Ex. 29, tab 37 at 2.) At this interview, Bimenyimana provided some background information and described the Bwindi attack in detail. (Id. at 3-12.) He stated that one of ALIR's objectives was to capture Americans, but not to kill them. (Id. at 4.) He also informed the agents that he learned from Nyaminani that he and three other soldiers had taken three female tourists into the woods. (Id.) Later in the interview he stated that he learned from Nkundabagenzi that Nyaminani and the others had raped and murdered the three women. (Id. at 11.) Bimenyimana was informed by investigators that they did not believe he was telling the whole truth. (Id. at 12.) He then stated that "he was present for the killing of a man and woman together but that he would provide the details after he had some time to think." (Id.) The interview was then terminated. (Id.)
Bimenyimana was brought back to Kacyiru the following day, February 22.*fn67 (Gov't Ex. 29, tab 38.) Agents Dent and Prouty were joined by Kayumba, Ruzigamanzi and Ndabarasa. (Id. at 1.) Again, all three Rwandans provided translations. (Id.) On this day, Agent Dent sought to test Bimenyimana's French language ability by having him read the Rewards for Justice poster aloud. (Gov't Ex. 2a; 5/16 a.m. tr. at 100.) Nobody in the room quizzed him regarding his understanding of the words on the poster, but Bimenyimana indicated that he understood what he had just read. (5//16 a.m. tr. at 101.) Dent went through the extra step of having him read aloud because "[t]here was some doubt" as to whether he had "understood his rights completely" the day before. (5/19 p.m. tr. at 35.) Bimenyimana then read and signed a French language advice of rights form. (Gov't Ex. 25.)
The investigators began to query him regarding the two murders that, during the prior day's interview, he claimed to have witnessed. Bimenyimana described hearing Ntabwoba order Mandevu to kill the tourists and then observed Mandevu and Gafaransa take a man and woman to a location about 15 meters from he was standing. (Gov't Ex. 29, tab 38 at 2.) According to Bimenyimana, Mandevu forced the man to remove his shirt, at which time he noticed the tattoo. Concerned that the man might be in the military, Mandevu called them over. Bimenyimana and Karake went to the location and observed a "dark blue" tattoo of a "bird with wings" on the man's arm. (Id.) After initially denying that he witnessed the actual murders, Bimenyimana eventually stated that he was approximately 5 meters from the man and woman when they were killed. (Id.) He informed agents that Karake "struck the man in the back of the head with the back of an axe." (Id. at 3.) He described the man as "shoeless, wearing pants, [and] had blond hair." (Id.) Despite his proximity, Bimenyimana was unable to recall if the man bled when he was hit with the axe or what the woman did when the man was killed. (Id.) He described the woman as having "brown shoulder-length hair, was barefoot, and was wearing pants and a tee shirt." (Id. at 4.) Bimenyimana stated that the two victims were not covered with bed sheets at the time they were killed.*fn68 (Id.)
The remainder of the interview covered several topics. Bimenyimana identified the objectives of ALIR as: (1) overthrowing the government of Rwanda; (2) releasing prisoners who can assist ALIR; and (3) making supporters of the government of Rwanda understand the ALIR cause. (Id. at 5.) He also described other ALIR operations that resulted in taking Western hostages (id.), and the purpose of the notes given to several hostages (or left with the bodies) by Commander Ntabwoba. (Id. at 7-8.) Additionally, he described a thirty-minute window during which Nyaminani was called from his position in Bimenyimana's platoon to carry out a task for Ntabwoba. (Id. at 4-6.) Bimenyimana later heard that Nyaminani had killed an American. (Id. at 6-7.)
Bimenyimana was brought back for a final interview on February 23, 2003.*fn69 (Gov't Ex. 29, tab 39.) Present at the interview were Dent, Prouty, Kayumba, Ruzigamanzi, and Ndabarasa. (Id. at 1.) All three Rwandans assisted with translation. (Id.) After reading the advice of rights form, Bimenyimana inquired of the agents whether he could have an attorney in the United States. (Id.) According to Dent, the agents asked him whether he was seeking a lawyer right then and he told them no. (5/16 a.m. tr. at 108.) He was told he could have an attorney in the United States, after which he agreed to conduct the interview. (Gov't Ex. 25.) Bimenyimana reiterated his story from the previous day about having witnessed the murder of a man and woman by Karake and having released Nyaminani from his duties with his platoon to perform a task for Ntabwoba. (Gov't Ex. 29, tab 39 at 4-5.) He then provided a detailed list of items that he knew were looted from Bwindi. (Id. at 6.) The FBI learned "[n]othing new" from Bimenyimana's interview on February 23. (5/16 a.m. tr. at 108.)
The reason for the lengthy delay between Bachmann's initial interview of Bimenyimana in late August 2002 and the FBI's return in February 2003 was ongoing discussions between the United States and Rwandan governments regarding the extradition of the defendants. (5/16 a.m. tr. at 88-89.) Following the indictments of Karake on June 6, 2002 and Nyaminani on August 8, 2002, the U.S. government, through the State Department and the DOJ's Office of International Affairs, initiated discussions to effect the transfer of the defendants to U.S. custody for prosecution. (Id.) On September 5, 2002, the two sides met in Kigali at the Rwandan Ministry of Foreign Affairs to discuss the proposed extradition. (Def. Ex. K-5.) Representing the United States were Ambassador Margaret McMillion and RSO Bachmann, among others. (Id.) The Rwandans were represented by the Foreign Affairs Ministry's Secretary General Joseph Mutaboba and National Police Commissioner General Frank Mugambage. (Id. at 1-2.) The primary impediment to the proposed extradition was a provision of the Rwandan Criminal Code that specifically prohibited extradition of Rwandan nationals. (Id. at 2.) There may also have been a secondary hurdle: Ambassador McMillion expressed concern in June that the Rwandans might demand an "exchange for persons" in the United States "wanted by the government of Rwanda." (Gov't Ex. 99, tab 29 at 2.) As an alternative to extradition, Rwandan officials proposed a joint U.S./Rwandan prosecution on Rwandan soil. (Def. Ex. K-5 at 2.) Negotiations continued over the next several months, and in February 2003, the group of U.S. officials handling the extradition issue returned to Rwanda. (5/16 a.m. tr. at 89.) The two sides reached an agreement on February 26, 2003, and defendants departed Rwanda in the custody of U.S. officials, accompanied by Isidore Kayumba, early on the morning of February 28, 2003. (Def. Ex. K-6.)
IV. ADMISSIBILITY OF DEFENDANTS' STATEMENTS TO THE RWANDANS
Defendants have moved to suppress their statements, all of which were given while in Rwandan custody, relying on two legal theories. Defendants' primary argument is that their due process rights were violated because their statements were the product of psychological and physical coercion, including torture and inhumane conditions of confinement. Even if the American investigators were not aware of this coercion at the hands of the Rwandans, defendants argue that any statements to the Americans are inadmissible as the "fruit of the poisonous tree" within the meaning of Oregon v. Elstad, 470 U.S. 298 (1985), and United States v. Patane, 542 U.S. 630 (2004). In the alternative, defendants offer a second legal theory to suppress their statements based on the proposition that the American and Rwanda investigators were engaged in a joint venture, and as a result, the strictures of Miranda applied to the Rwandan interrogations.*fn70
Defendants argue that even if the statements to the Americans were not tainted by the coercion, under either theory the statements given to the Americans were obtained in violation of Miranda, since there was no voluntary, knowing and intelligent waiver, and thus their Fifth Amendment rights against self-incrimination were violated.*fn71
The Court will begin its analysis with the issue of voluntariness, since this was the primary focus of the evidentiary hearing and the Court's determination of this issue impacts all other inquiries.*fn72 And, as noted, while the governing legal principles are not complicated nor are they the subject of any significant dispute, their application to the evidence is a challenging task.
A. The Law Regarding Voluntariness
Defendants' central contention is that their twenty-nine statements to American and Rwandan interrogators were the product of coercion stemming from the conditions under which they were confined and the physical and psychological mistreatment by their Rwandan custodians. (Defs.' Mot. at 1-2.) When "a confession challenged as involuntary is sought to be used against a criminal defendant at his trial . . . the prosecution must prove by at least a preponderance of the evidence that the confession was voluntary." Lego v. Twomey, 404 U.S. 477, 489 (1972); see also Colorado v. Connelly, 479 U.S. 157, 168 (1986) (reaffirming Lego). To meet its burden, the government must demonstrate that each of defendants' confessions was "the product of an essentially free and unconstrained choice." Schneckloth v. Bustamonte, 412 U.S. 218, 225 (1973) (internal quotation marks omitted). If the defendant's "will has been overborne and his capacity for self-determination critically impaired, the use of his confession offends due process." Id. at 225-26; see also Lynumn v. Illinois, 372 U.S. 528, 534 (1963) (question is "whether the defendant's will was overborne at the time he confessed").
Whether a confession was "'made freely, voluntarily, and without compulsion or inducement of any sort,'"Haynes v. Washington, 373 U.S. 503, 513 (1963) (quoting Wilson v. United States, 162 U.S. 613, 623 (1896)), is distinct from the question of whether the confession is accurate or reliable. United States v. Raddatz, 447 U.S. 667, 678 (1980) ("the interests underlying a voluntariness hearing do not coincide with the criminal law objective of determining guilt or innocence"); Jackson v. Denno, 378 U.S. 368, 384-85 (1964) (the "reliability of a confession has nothing to do with its voluntariness"). Indeed, the "probable truth or falsity" of a confession "is not a permissible standard under the Due Process Clause" on the "question [of] whether the confession was voluntary and admissible." Hutcherson v. United States, 351 F.2d 748, 753 (D.C. Cir. 1965.) Rather, as a long line of Supreme Court precedent makes clear, "confessions which are involuntary, i.e., the product of coercion, either physical or psychological," are inadmissible not because such confessions are unlikely to be true but because the methods used to extract them offend an underlying principle in the enforcement of our criminal law: that ours is an accusatorial and not an inquisitorial system -- a system in which the State must establish guilt by evidence independently and freely secured and may not by coercion prove its charge against an accused out of his own mouth.
Rogers v. Richmond, 365 U.S. 534, 540-41 (1961) (citing cases). "A coerced confession is offensive to basic standards of justice, not because the victim has a legal grievance against the police, but because declarations procured by torture are not premises from which a civilized forum will infer guilt." Lyons, 322 U.S. at 605.
This is not to say that the exclusionary rule under the Fifth Amendment does not serve a judicial interest in assuring reliable evidence. Beyond the "general goal of deterring improper police conduct," Elstad, 470 U.S. at 308, the Fifth Amendment serves "to protect the fairness of the trial itself." Schneckloth, 412 U.S. at 240. "[T]he Fifth Amendment goal of assuring trustworthy evidence" exists independent of any deterrent effect the exclusionary rule might have under some circumstances. Elstad, 470 U.S. at 308. Thus, while a confession obtained by means of torture may be excluded on due process grounds as "[in]consistent with the fundamental principles of liberty and justice which lie at the base of all our civil and political institutions," Brown v. Mississippi, 297 U.S. 278, 286 (1936), another legitimate reason to suppress it is the " 'likelihood that the confession is untrue.' " Linkletter v. Walker, 381 U.S. 618, 638 (1965) (quoting Blackburn v. Alabama, 361 U.S. 199, 207 (1960)).
The Court's role when faced with an allegedly coerced confession is to "'enforce the strongly felt attitude of our society that important human values are sacrificed where an agency of the government, in the course of securing a conviction, wrings a confession out of an accused against his will.'" Schneckloth, 412 U.S. at 225 (quoting Blackburn, 361 U.S. at 206-07). To do this without unduly impairing "the acknowledged need for police questioning as a tool for effective enforcement of criminal laws," Schneckloth, 412 U.S. at 225, the Court must conduct "an examination of all the attendant circumstances" surrounding the confession. Haynes, 373 U.S. at 513. This "totality of circumstances" inquiry (id.) requires consideration of "both the characteristics of the accused and the details of the interrogation." Schneckloth, 412 U.S. at 226. This can include: "the youth of the accused; his lack of education; or his low intelligence; the lack of any advice to the accused of his constitutional rights; the length of detention; the repeated and prolonged nature of the questioning; and the use of physical punishment such as the deprivation of food or sleep." Id. (internal citations omitted).
At a minimum, statements that are "extracted by threats or violence" violate the Due Process Clause. Hutto v. Ross, 429 U.S. 28, 30 (1976). The use of torture to extract a statement clearly contravenes "principles of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental." Brown, 297 U.S. at 284 (citation omitted). For, as further observed by the Court in Brown, "the freedom of the state [is] limited by the requirement of due process of law. . . . The rack and torture chamber may not be substituted for the witness stand." Id. at 285-86.
"[A] finding of coercion," however, "need not depend upon actual violence by a government agent; a credible threat is sufficient." Fulminante, 499 U.S. at 287 (statement involuntary where incarcerated police informant promised the defendant "protection" from his fellow inmates if he told the informant the truth about his crime). For, as recognized by the Supreme Court, "the blood of the accused is not the only hallmark of an unconstitutional inquisition." Blackburn, 361 U.S. at 206. Thus, less traditional forms of coercion, including psychological torture, as well as the conditions of confinement have been considered by courts in their assessment of the voluntariness of the statements. See, e.g., Brooks v. Florida, 389 U.S. 413, 414-15 (1967) (confession was involuntary where the defendant was held in solitary for 14 days, "saw not one friendly face from outside the prison" and was "completely under the control and domination of his jailers"); Stidham v. Swanson, 506 F.2d 478 (8th Cir. 1974) (in finding statement involuntary, court considered suspect's imprisonment in solitary confinement for 18 months in subhuman conditions, including a bug-infested cell, lack of sufficient food, and denial of visits with family and friends); Arnett v. Lewis, 870 F. Supp. 1514, 1523-25, 1540 (D. Ariz. 1994) (confession found to be involuntary where defendant was incarcerated in "oppressive conditions," including the lack of adequate plumbing and heating, clean water, blankets and nutrition); Townsend v. Henderson, 405 F.2d 324, 326 (6th Cir. 1968) (statement found to be involuntary; defendant held in solitary confinement with inadequate food); Wainwright v. LaSalle, 414 F.2d 1235, 1237-39 (5th Cir. 1969) (court notes that defendant was in "continuous incommunicado custody for 12 hours" before confession was elicited for the first time and the ultimate confession followed prior denials). Both for purposes of assessing the voluntariness of a statement under principles of due process and for judging the adequacy of a waiver of one's Fifth Amendment privilege against self-incrimination under Miranda, a court must consider the circumstances surrounding the interrogation, as well as the length of the detention and the conditions of confinement. See Miranda, 384 U.S. at 476 ("whatever the testimony of the authorities as to the waiver of rights by an accused, the fact of lengthy interrogation or incommunicado incarceration before a statement is made is strong evidence that the accused did not validly waive his rights.").
While both parties agree that it is the government's burden to prove by a preponderance of the evidence that the statements were voluntary, Lego, 404 U.S. at 489, exactly what this burden entails has become somewhat muddled with respect to statements made abroad to foreign officials. Some circuit courts, relying on Bram v. United States, 168 U.S. 532 (1897), have adhered to a voluntariness standard. See, e.g., Brulay v. United States, 383 F.2d 345, 348-49 (9th Cir. 1967) ("We then turn to the problem: were the statements voluntary within Fifth Amendment standards which we believe applicable?"); United States v. Mundt, 508 F.2d 904, 906 (10th Cir. 1974) (analyzing admissibility in terms of "voluntariness"). Others have applied a "shock the conscience" standard. See, e.g., United States v. Heller, 628 F.2d 594, 599 (5th Cir. 1980) (applying "shock the conscience" exception to the admissibility of statements obtained by foreign officials conducting interrogations in their own nations); United States v. Cotroni, 527 F.2d 708, 712 n.12 (2d Cir. 1975) (admitted statement given to official but noting that "[w]ere the conduct of foreign police so reprehensible as to shock the conscience, a different result might obtain").*fn73
Fortunately, however, despite the lack of guiding precedent in this jurisdiction as to whether the Court should formulate its inquiry in terms of "voluntariness" or "shock the conscience,"*fn74 this issue need not be resolved. Rather, as clearly explained by the government:
[I]n our view, that's not an issue of import because our argument from the start was there was no behavior by the Rwandan officials at any point during their custody by the Rwandan officials that amounted to involuntariness or shock of conscience.
We've always been proceeding from the position, and that's why we put Captain Kibingo on the stand extensively, to demonstrate that there was no mistreatment in any way. And indeed all of the statements whether they be at the hands of the Rwandans or the Americans, were voluntary statements. (6/19 a.m. tr. at 5-6 (emphasis added).) Moreover, defendants allege mistreatment that would "shock the conscience," rendering any distinction between the two legal standards moot. (See supra note 73.)
It is therefore immaterial how the issue is framed, for the Court's task is the same: to determine whether defendants' confessions to Rwandan officials were the product of coercion in violation of the Due Process Clause. That task requires a fact-specific inquiry that depends almost entirely on an assessment of the credibility of the witnesses. (See Gov't Legal Memo. at 2.) In particular, this assessment, in large measure, boils down to determining whose testimony is more credible -- the defendants or Captain Kibingo, since he was the only witness to testify who is alleged to have actually tortured the defendants during their lengthy detention at Kami Camp. It is to this task and to the diametrically opposed version of the facts presented by the parties that the Court will now turn.
B. Application of the Law Regarding Voluntariness
1. Defendants' Testimony
Defendants tell similar, horrifying stories regarding Kibingo's use of physical and psychological torture and their conditions of confinement at Kami Camp. During most of the relevant time period, defendants were housed in a building or group of buildings known as "Special." (5/22 p.m. tr. at 23; 5/24 p.m. tr. at 39; 5/31 p.m. tr. at 50.) Each defendant was housed in a small room by himself. (5/22 p.m. tr. at 18-19; 5/24 p.m. tr. at 39; 5/30 p.m. tr. at 58-59.) Nyaminani described his room as approximately two meters by two meters, with a cement floor. (5/24 p.m. tr. at 39.) Bimenyimana's room was slightly larger, roughly 9 feet by 12 feet, also with a cement floor. (5/30 p.m. tr. at 60.) None of the rooms had electricity. (5/22 p.m. tr. at 19; 5/24 p.m. tr. at 47; 5/30 p.m. tr. at 60). Nyaminani and Bimenyimana testified that they were not provided with any sort of bed or mattress on which to sleep. (5/24 p.m. tr. at 39; 5/30 p.m. tr. at 60.) The rooms each had a window but, according to Karake and Nyaminani, after Kibingo found their responses to questioning regarding Bwindi unsatisfactory, their windows were covered with "iron sheets" that kept out the light. (5/22 p.m. tr. at 27-28; 5/24 p.m. tr. at 46.) Bimenyimana's window had already been covered up by the time he arrived in July 2002. (5/30 p.m. tr. at 61-62.) Defendants were given a "jerry can" to urinate in while in their cells, which they emptied in the mornings. (5/22 p.m. tr. at 39; 5/24 p.m. tr. at 53; 5/30 p.m. tr. at 60, 70.) The rooms were dirty and often wet. (5/25 p.m. tr. at 18; 5/30 p.m. tr. at 59; 5/31 a.m. tr. at 10.) Defendants were fed a small cup of corn and beans once per day. (5/22 p.m. tr. at 40, 5/24 p.m. tr. at 40, 47-50; 5/30 p.m. tr. at 69.) Each defendant described this food as contaminated with "sand" or "stones." (5/22 p.m. tr. at 40; 5/24 p.m. tr. at 47-50; 5/30 p.m. tr. at 87.) The amount of food was such that Nyaminani "lost a lot of weight" and suffered from "severe headaches." (5/24 p.m. tr. at 49-50.) Bimenyimana "could not stand because of hunger," and suffered from "dizziness." (5/30 p.m. tr. at 87.) Bimenyimana also testified regarding the mosquitos in his room at Kami and the lice that infested his hair and clothes. (5/30 p.m tr. at 87.)
These conditions caused other ailments. Bimenyimana contracted malaria while housed in Special. (5/31 a.m. tr. at 10.) Nyaminani suffered from "stomach pain," "worms, parasites," and a dermatologic problem that caused his skin to itch and flake. (5/24 p.m. at 49.) He also contracted malaria three times. (5/25 p.m. tr. at 71-74.) The first time, in January 2002, Nyaminani received no medical care, except for an herb that he mixed with water. (Id. at 72-73.) In June 2002, Nyaminani contracted malaria again. (Id.) This time he received medical care from a doctor at Kami. (Id. at 73.) The third time, toward the end of 2002, Nyaminani spent two weeks at the military hospital at Kanombe. (Id. at 74.) Karake also had "problems with his stomach," including "nausea" and vomiting. (5/23 a.m. tr. at 12.) According to Nyaminani, defendants were not permitted to speak to other detainees at Kami, including the other Bwindi suspects. (5/24 p.m. tr. at 51.) Nyaminani and Bimenyimana both testified that their shoes were taken upon their arrival at Kami Camp. (5/24 p.m. tr. at 40; 5/31 a.m. tr. at 26-27.)
These conditions lasted for many months. In July 2002, Nyaminani was moved from Special to another location known as "Go-Down." (5/24 p.m. tr. at 52.) Go-Down was a series of underground bunkers at Kami where some inmates were housed. (5/25 p.m. tr. at 68.) Lined with concrete, no sunlight entered the Go-Downs. (5/22 p.m. tr. at 73.) Nyaminani actually inhabited three different bunkers between July 2002 and his rendition in March 2003. (5/25 p.m. tr. at 68.) Karake was moved to a Go-Down in early 2003. (5/22 p.m. tr. at 72.) Unlike Special, multiple inmates were housed together in these bunkers. There were approximately five persons in Karake's Go-Down (id.), and an unnamed number of "other people" in Nyaminani's. (5/25 p.m. tr. at 69.) According to Nyaminani, the Go-Down was actually an improvement over Special in part because he was now living with other people, and because he was given clothes and better food. (Id.) Bimenyimana remained in Special throughout his detention at Kami Camp. (5/31 p.m. tr. at 12.) During their time at Kami, no one was ever allowed to visit Nyaminani or Karake, nor were they permitted to contact any friends or family. (5/22 p.m. tr. at 73; 5/24 p.m. tr. at 52.) On one occasion, however, in October 2002, Bimenyimana's wife was allowed to visit him for roughly one hour. (5/31 p.m. tr. at 9-12.)
Arguably, this testimony regarding their many months spent in solitary confinement, deprived of light, adequate food and contact with the outside world is, by itself, sufficient to sustain a finding of involuntariness. But the above-described conditions constitute only the basic day-to-day experience of the defendants. Of greater significance is the detailed testimony each defendant provided regarding Kibingo's abusiveness and his use of torture to extract the information that was then communicated to the Americans.
Defendant Karake testified that within two weeks of arriving at Kami in December 2001, he had been questioned multiple times by Kibingo. Though at the first of these interrogations "[t]here was no problem" when Karake denied involvement in the killings at Bwindi, matters soon worsened. (5/22 p.m. tr. at 20-21, 24-25.) A second interrogation took place about a week later, in the presence of Nyaminani and Leonard Harerimana. (Id. at 24-25.) Karake again denied involvement in the killings. (Id.) The following morning, Karake was brought before Kibingo for a third interrogation, at which time Karake drafted a written statement denying any involvement. (Id. at 26.) Kibingo's response was to cover the windows in Karake's room so that he no longer had any light. (Id. at 27.) Kibingo's soldiers then handcuffed Karake, left wrist to right ankle and right wrist to left ankle, and left him in that position in the darkness of his cell until the following day. (Id. at 27-28.) The next evening, Karake was taken back to Kibingo's house, where Kibingo physically struck him, while telling Karake to admit that he had killed people at Bwindi and that Nyaminani had done so as well. (Id. at 30-32.) According to Karake, Kibingo used both hands to slap him on his ears so hard that he "saw something blinking . . . like sparks of fire." (Id. at 30-31.) Kibingo also kicked him after Karake fell out of his chair. (Id. at 31.) Though Karake says he agreed at this point to "tell [Kibingo] whatever [he] want[ed]," Kibingo continued to beat him, first with a stick and then with a brick. (Id. at 32-35.) Only then was Karake given paper and a pen so that he could write his statement. (Id. at 38.) Kibingo "told [him] what to write," and, according to Karake, these were the circumstances under which he wrote the statement dated December 28, 2001 (Gov't Ex. 52 & 52a), wherein he admits to killing one tourist. (5/22 p.m. tr. at 33.) After this, Karake was returned to his room and spent the night in handcuffs. (Id. at 39.)
Karake testified of repeated beatings in February, both with bricks and with the stick he mentioned before, which he described as made of wood, roughly as long as a man's arm, that was balled on one end. (Id. at 32, 25, 60.) He claimed to have been beaten with a brick prior to his February 5, 2002 interview with the FBI and that the bandage on his wrist that was observed by the FBI (Gov't Ex. 12, 13 & 13a), was to cover an injury he sustained from that beating. (5/22 p.m. tr. at 52.) Karake was also beaten on the day after his February 6 interview with the FBI, ostensibly because the Rwandans had received information that he had killed more than the one person whom he had identified on February 5. (Id. at 59.) After he denied this to Kabanda and Kabanda had left, Karake was taken by the soldiers to Kibingo's house, and Kibingo continued to beat him and told him that if he did not admit to the other murders, he "would continue hitting" or "would kill" Karake. (Id. at 59-61.) On February 12, the day the FBI agents inexplicably interrupted their planned interviews, Karake was questioned by Rwandan officials, including Kibingo, Ndabarasa, Isidore Kayumba, Egide Ruzigamanzi and Kabanda. (Id. at 62.) At this meeting, the Rwandans wanted him "to repeat what Kibingo had beaten [him] to memorize." (Id.)
He also alleges that he was beaten again by Kibingo in March after his interviews with the FBI were completed, apparently because he had failed to tell the FBI that Nyaminani had killed at Bwindi. (Id. at 64-69.) Karake was beaten with the stick for a period of time before Nyaminani was brought in. (Id. at 64-65.) Kibingo then began hitting Nyaminani. (Id. at 65.) Karake watched Kibingo use a flexible, roughly 12 inch piece of rubber to beat Nyaminani. (Id. at 66-69.) Kibingo made Nyaminani draft a written statement before calling the other Bwindi suspects together to announce what Nyaminani had admitted to. (Id. at 65-67.) Karake and Nyaminani were then returned to their cells in Special. (Id. at 68.)
Throughout this entire period, from December 2001 until April 2002, whenever Karake was in his cell in Special, he was handcuffed from both wrists to the opposite ankles. (Id. at 71.) Karake described that position as "how [he] lived." (5/22 p.m. tr. at 28.) The consequence, for Karake, was a "constant headache" and "pain in the neck . . . and . . . back" from constant "bending." (Id. at 71.) The handcuffing also left visible scars. (Id. at 29.)
Nyaminani's description of Kibingo's interrogation methods is similar. He tells of being stripped of his clothes and having his blanket taken away in early December. (5/24 p.m. tr. at 46.) Toward the end of December, Nyaminani testified that he was subjected to positional torture known as "kwasa kwasa," where his wrists were bound together with nylon rope, one over his shoulder and the other behind his back. (Id. at 53-54.) Nyaminani was kept in this position for approximately two weeks. (Id. at 56.) He testified that the nylon rope cut into his skin, leaving visible scars. (Id. at 57-58.) After he was released from this position, Nyaminani was then handcuffed. (Id. at 61-62.) At times the handcuffing was simply behind his back (5/25 p.m. tr. at 19); more often it was the same positional torture Karake described: "right leg to . . . left arm." (Id. at 38; see also id. at 21, 5/25 a.m. tr. at 60, 62.) On one occasion Kibingo threw him against a wall, slapped him with a slipper all over his body while he was handcuffed, then had his soldiers cuff him from wrist to opposite ankle. (Id. at 20-21.) On another occasion, his wrists and ankles were handcuffed in such a way as to cause "agonizing" pain, and he was left overnight in that position. (5/25 a.m. tr. at 19-25.) Nyaminani also described being beaten with a plastic (5/25 p.m. tr. at 33) or rubber (5/25 a.m. tr. at 70) stick roughly 10-12 inches long and 2 inches thick. (5/25 a.m. tr. at 70; 5/25 p.m. tr. at 33.) Another tactic used by the Rwandans at Kami, according to Nyaminani, was to spray his room full of water, forcing Nyaminani to sit on the wet concrete floor, without clothes or a blanket. (5/24 p.m. tr. at 46.)
Nyaminani testified to the same series of events as Karake that resulted in Kibingo announcing Nyaminani's confessions to a group of Bwindi suspects. (Compare 5/22 p.m. tr. at 64-69 with 5/25 a.m. tr. at 68-73.) Nyaminani stated that on March 8, 2002, he was brought to Kibingo's house and found Karake in the corner crying. (5/25 a.m. tr. at 68.) Kibingo began to question Nyaminani about Bwindi, and Nyaminani denied any knowledge regarding the murders. (Id. at 69.) In response, Nyaminani testified, Kibingo set down his papers and cell phone, told Nyaminani "this is your day," and began to beat him with the rubber stick described above. (Id. at 70.) After several minutes of this beating, Nyaminani was given paper, on which he wrote his March 8 statement. (Id. at 71-72; Gov't Ex. 53 & 53a.) Kibingo "was standing by" Nyaminani while he wrote the statement, and "sometimes would tell [him] what to write." (Id. at 73.) Kibingo then called Innocent Sebureteri, Silver Dukuzumuremyi, Saidi Mbarushimana, Fabien Nkundabaza and Leonard Harerimana to his house and announced that Nyaminani had confessed. (Id. at 72-73.) Nyaminani testified that he did not "say anything to them directly" (id. at 73); only Kibingo spoke, and he told the other suspects that "whenever they got the interviews," "they should be saying . . . what Minani did." (Id.) Nyaminani was then sent back to his cell. (Id.)
Nyaminani testified that Rwandan officials continued to beat him until June 2002. Nyaminani testified that after his June 1 interview with the FBI when he refused to admit to raping the American women, he was not taken back to Kami; rather, he was taken to a police station called Gikondo in Kigali, (5/25 p.m. tr. at 51) where he was subjected to more physical abuse. The Rwandans rubbed the skin on his back "until it was hot," and then they slapped him with their hands. (Id. at 52.) They then took two logs; a large one weighing approximately 30 kilograms was placed on his shoulders, and a smaller one was secured behind his knees so that his legs were slightly bent. (Id. at 52-57.) "It was so painful," he testified, that he thought "my nerves and my muscles were going to erupt." (Id. at 57.) The following day, Nyaminani admitted to the FBI agents that he had raped Susan Miller. (Id. at 52; Gov't Ex. 29, tab 27.)
Bimenyimana's testimony is similar. He described being kicked and beaten while handcuffed during interrogations at Kibingo's home. (5/30 p.m. tr. at 66.) Like Nyaminani, the soldiers at Kami occasionally spilled a basin full of water in his cell, forcing him to sit in the water because he had no chair, bed or mattress. (5/31 a.m. tr. at 10.) During interrogations, one of Kibingo's soldiers would hit the bottom of Bimenyimana's feet (5/30 p.m. tr. at 66), causing them to swell up. (5/31 a.m. tr. at 9; see also Gov't Ex. 63.) Bimenyimana also testified that Kibingo would sometimes strike him with one end of a pair of standard metal handcuffs. (5/30 p.m. tr. at 71.) On other occasions soldiers would take a "brick and put it in a sock" and use that to strike Bimenyimana in the chest and back. (Id. at 83.) Like Karake (5/22 p.m. tr. at 30-31), Kibingo or his soldiers would also slap Bimenyimana on his ears, saying "maybe [you're] not understanding."*fn75 (5/30 p.m. tr. at 67.) Bimenyimana described the same positional torture endured by Karake and Nyaminani: having one wrist shackled to the opposite ankle. (Id. at 68.) He also testified that his arms were bound with rope just above the elbows so that his elbows met in the back. (Id. at 80.) He was left in this position for "about an hour," resulting in a permanent marking of three to four striations on his skin above both elbows. (Id.)
Bimenyimana testified to having been beaten three times by Kibingo in July 2002. (5/30 p.m. tr. at 75-76.) At these interrogations, Kibingo beat him with handcuffs while asking him questions about Nyaminani and Karake. When Bimenyimana responded that he "did not know anything about it," Kibingo told him that he "will change [his] mind." (5/31 a.m. tr. at 15.) He also described in detail an interrogation by Kibingo on August 10 at which he was severely beaten. (5/30 p.m. tr. at 76-84.) Though he was handcuffed from the outset of the interrogation, Bimenyimana claims he was not beaten until "the middle of the questioning" when he was "not agreeing with them." (Id. at 79.) Kibingo asked Bimenyimana about a French female tourist at Bwindi who had pointed out the English speakers to ALIR soldiers. (5/31 a.m. tr. at 52-53.) During the course of the beating, which began on Kibingo's veranda, his handcuffs began to cut his wrists, so the soldier bound his arms behind his back at the elbows, and dragged him into Kibingo's house, injuring his knee and toe. (5/30 p.m. tr. at 79-80.) Once inside, he was beaten further with a brick that had been placed inside a sock. (Id. at 82-83.) At this point, Bimenyimana agreed to write a statement. (Id. at 83.) On August 17, Bimenyimana claims Kibingo asked him to write a second statement, and that he agreed to write it to prevent Kibingo from beating him again. (Id. at 85-86.) The purpose of this second statement, according to Bimenyimana, was to include what Kibingo felt he "had missed in the first writing." (Id.) This included the fact that the French woman at Bwindi had children with her, and that Nyaminani had killed a tourist. (5/31 a.m. tr. at 54-55.) Neither of these statements was proffered at the hearing.
The only written statement Bimenyimana gave to Rwandan officials without any Americans present that was introduced into evidence was the "official" statement obtained by Thomas Murangira. (Gov't Ex. 42.) Though Bimenyimana does not claim to have been physically abused by Kibingo on the day he gave his statement to Murangira, he testified that when he was uncuffed and taken out of his cell that day, he was so "dizzy because [he] was hungry" that he "fell down" in a "trench." (5/31 a.m. tr. at 82.) As he "was struggling to stand up," the "soldiers were kicking" him and "telling [him] to stand up." (Id.) Bimenyimana was taken to Kacyiru. (Id. at 83.) Also present for the interview was "another sergeant called Rukara." (Id. at 84.) Murangira asked Bimenyimana a series of questions and then wrote down his responses. Bimenyimana testified that, even though he signed the document drafted by Murangira, he neither read it nor was it read to him before he signed. (Id. at 86.) Bimenyimana was then taken back to Kami Camp. (Id.)
This systematic and repeated physical abuse was responsible, according to defendants, not just for the statements they gave to Kibingo and Murangira. Fear of further abuse also caused them to give statements to the American investigators and prevented them from alerting the FBI to the torture they were suffering. Bimenyimana testified that he didn't tell the Americans that he was mistreated by Kibingo because he "was very scared of his soldiers that were there," and because "they would . . . tell Kibingo, and then he would kill me." (5/31 p.m. tr. at 41.) This was the same reason that he spoke to the FBI in the first place: he "did not know what to do because . . . the soldiers of Kibingo . . . were there." (Id.) Karake stated that he didn't tell Bachmann on January 12, 2002, that Kibingo was beating him because Kibingo "would have beaten [him] again and he would have even killed [him]." (5/22 p.m. tr. at 50.) He gave the same reason for not telling the FBI in February: he "would have been beaten again." (Id. at 54.) Indeed, Nyaminani testified that he attempted in March to tell some unidentified white man, who was with Isidore Kayumba, that Kibingo had beaten him with a stick.*fn76 (5/25 p.m. tr. at 34.) He supposedly did this in front of Kayumba, Ndabarasa, Kabanda and Ruzigamanzi, but claims that the white man and Kayumba left immediately. (Id. at 32, 34.) When he returned to Kami, "Egide told Kibingo that [Nyaminani] had reported him." (5/25 p.m. tr. at 35.) Nyaminani claims that Kibingo then beat him with a slipper, and thereafter he wrote another statement. (Id. at 36-38; Gov't Ex. 54 & 54a.) When Nyaminani told the FBI in June that Kibingo had slapped him with a slipper in March, the FBI never asked Kibingo about it. (5/16 p.m. tr. at 38.) The FBI asked other Bwindi suspects about it, "and they all more or less laughed about the incident." (5/19 p.m. tr. at 114.)
The abuse to which defendants were subjected also forced them to give official statements to Thomas Murangira, despite the fact that no physical abuse took place during the course of Murangira's interviews. Bimenyimana testified that he told Murangira about events at Bwindi because he felt there was "no way [he] could not talk to him." (Id. at 85.) In his words: "Had I not talked to him, I would be punished by Kibingo." (Id.) According to Nyaminani, Ruzigamanzi told Murangira prior to the interview that Nyaminani was required to "tell you everything he told us." (5/25 p.m. tr. at 70.) Kibingo turned over the written statements Nyaminani had drafted at Kami Camp and told Murangira that Nyaminani "should go and repeat the same thing [he] had written." (Id. at 71.) Likewise, Karake was instructed by Kibingo to tell Murangira the same information that Karake had given him; i.e., Karake "should repeat what [Kibingo] told [him] to say." (5/22 p.m. tr. at 69.)
2. Kibingo's Testimony
Kibingo testified extensively regarding the conditions under which defendants were housed at Kami. As Kibingo described it, Kami was a pleasant place to live, if not bucolic. The camp was open to the public, with no security fence or barbed wire enclosing the camp grounds. (5/3 a.m. tr. at 51.) The public was permitted to come on to the grounds of Kami Camp to graze cattle and raise crops in the fields. (5/5 a.m. tr. at 49-50.) No one -- not the public, the family of the prisoners, government officials or human rights organizations -- was denied access to Kami. (Id. at 64-67.) Kibingo's soldiers (numbering between 90 and 120) who lived at Kami cooked for themselves, preparing rice, Irish potatoes, cassava and maize flour, beans, cabbage, and onions. (5/3 a.m. tr. at 32 and p.m. tr. at 4-5.) The soldiers slept on mattresses placed on the floor, with blankets and bed sheets. (Id. at 5-6.) Soldiers were also provided with mosquito nets that hung over the mattress to prevent the transmission of malaria. (Id. at 6.) These same living conditions, Kibingo testified, were shared by the Rwandan soldiers who were there for disciplinary action (id. at 7-10), and by the captured ALIR soldiers. (Id. at 19-22.) "There's no one that would escape," Kibingo testified, not because of tight security -- for Kami was "an open camp" -- but because the ALIR soldiers were "received well."*fn77 (Id. at 19.)
Kibingo's testimony flatly contradicts that of the defendants. According to Kibingo, the Bwindi suspects could receive visitors at Kami (id. at 20, 50); they could leave the camp to fetch water in order to wash their clothes or bathe (id. at 20-21); they received their own mattress to sleep on and a mosquito net (id. at 22; 49); they cooked their own food, which was identical to that eaten by Kibingo's soldiers (id. at 49-50); and the doors to their rooms were locked only at night. (Id. at 22.) Kibingo testified that the suspects all lived together (id. at 48) and that all the rooms at Kami had working electricity. (5/4 p.m. tr. at 47.) No prisoner at Kami was ever handcuffed because "[t]hey were good people" (id. at 24), and no one ever died at Kami. (5/9 p.m. tr. at 36-37.) No handcuffs were even present at Kami Camp during the time the defendants were there. (Id. at 56-57.) While acknowledging the existence of "Go-Downs," Kibingo claimed that no one, including the defendants, was ever housed there, but that the bunkers were used to store ammunition. (5/5 p.m. tr. at 30-32.) No defendant was ever denied food, water or sleep. (5/4 p.m. tr. at 9, 47, 59. ) They were never physically abused, whether through binding (5/4 p.m. tr. at 26) or beating. (5/4 p.m. tr. at 9, 47, 59.) They were never threatened. (5/4 p.m. tr. at 47, 59.)
Kibingo "treat[ed] them well," he claimed, "because I was happy that my relatives or my people have come back." (5/3 a.m. tr. at 19.) Kibingo testified that he would sit "in chairs" "on the veranda" of his house and talk with the Bwindi suspects, occasionally sharing meals or drinks. (See id. at 57, 65-66; 5/4 a.m. tr. at 47; 5/5 a.m. tr. at 38.) Kibingo even claimed to have played checkers with one inmate at Kami. (5/5 p.m. tr. at 39-40.) During these talks, Kibingo testified, defendants were never threatened, he never raised his voice, and they were never subject to force of any kind. (5/4 a.m. tr. at 28.) Kibingo, in fact, didn't even consider these meetings an interrogation; rather, he and the suspect "were just discussing" matters. (Id.) Karake, in particular, is described by Kibingo as an eager interlocutor. "He used to come to my house," because he "liked conversing. That . . . time we spent, the relationship we had in the camp until [he left for] the United States was really good." (Id. at 46-47.) According to Kibingo, Karake would come by to tell him about his home and family; he would cook and eat there, and even occasionally take naps there during the day. (Id. at 47.)
To buttress Kibingo's testimony, the government introduced several photographs of Kami Camp taken by the FBI in January 2004. (5/4 p.m. tr. at 61; see Gov't Ex. 56-58.) These photos depict rooms with concrete floors. (5/5 a.m. tr. at 21-22; Gov't Ex. 56n-p.) Some of the windows are partially or fully covered with either blue plastic or metal sheets. (Gov't Ex. 56a.) Kibingo explained that the coverings were intended to prevent rain from entering through the windows, which had been broken during the war. (5/5 a.m. tr. at 17-19.) He claimed that the coverings could be removed by any person inside the room. (Id.) The government also introduced a video that was shot in 2006 of the Kami grounds.*fn78 (Gov't Ex. 55; 5/5 a.m. tr. at 40.)
Having heard the testimony of the defendants and Captain Kibingo and having observed their demeanor over the course of many days, the Court concludes that Captain Kibingo's testimony regarding his treatment of the defendants is not credible. His portrayal of Kami a something akin to a summer camp is unbelievable. His testimony also suffered from numerous inconsistencies and was impeached by other more reliable evidence, including evidence that severely undermined his central claims regarding the lack of torture, the humane conditions and the cordial relations that prevailed at Kami Camp. And, while no Rwandan who testified was free from bias, given the history of hostilities between Hutus and Tutsis and between the RPA and ALIR, Kibingo exhibited a more intense hatred toward these defendants and ALIR than did any other witness. But most importantly, the defendants' testimony was neither incredible nor without corroboration. In fact, the corroboration of defendants' testimony was far more compelling than that presented by the government to bolster Kibingo's credibility. Two other witnesses testified about their personal experiences while at Kami in years prior to defendants' detention; former high-ranking officials of the RPF during the relevant time period provided information regarding the abuses at Kami; and State Department reports and other reports to U.S. government officials documented rampant human rights violations, including specific reports of torture at Kami. But most compelling was the evidence of scarring that still remains on defendants' bodies and the corroborating evidence provided by both the defense and government medical experts who were unanimous in their testimony that the scarring is consistent with the physical abuse described by the defendants.
The Court will now discuss the evidence that corroborates defendants' testimony and supports the Court's conclusion regarding their credibility. It will then turn to the problems with Kibingo's testimony, followed by an analysis of the weaknesses in the government's putative corroborating evidence and the reasons offered for not crediting defendants' testimony.
3. Evidence That Corroborates Defendants' Testimony
a. Physical Evidence and Medical Expert Testimony
The most persuasive evidence in support of defendants' allegations of torture and abuse at Kami Camp is the physical evidence of scarring found on their bodies, combined with the analysis of that scarring by the medical experts presented by both sides. Each defendant underwent a physical examination by both a defense and prosecution medical expert. The written report of each expert was produced to opposing counsel in advance of the suppression hearing and introduced into evidence at the hearing as that expert's direct testimony; opposing counsel was given the opportunity to cross-examine the witness regarding the content of his or her written report.
The defense expert who examined Karake was Dr. Sondra S. Crosby, a board certified specialist in Internal Medicine and an Assistant Professor of Medicine at the Boston University School of Medicine. (Def. Ex. K-25 ¶ 1.) Dr. Crosby also serves as the Director of Medical Services at the Boston Center for Refugee Health and Human Rights. (Id.) Dr. Crosby's clinical practice and teaching focuses on the evaluation and care of survivors of torture. (Id.) Dr. Crosby examined Karake twice at the jail, first on December 27, 2005, and then again on January 5, 2006, for a total of approximately five hours. (Id. ¶ 2.) She also met with Karake for approximately 40 minutes at the courthouse on May 4. Each of these interviews was conducted in Kinyarwanda with the assistance of an interpreter. (Id.) During these examinations, Karake described severe, repetitive torture during his confinement first at Mweso and later at Kami. Concerning Mweso, Karake described being "beaten with sticks all over his body," "severe beatings to the head," including one head wound that required sutures, and being cut on his side with barbed wire. (Def. Ex. K-25 ¶ 18.) Karake informed Dr. Crosby that while at Kami, he was beaten with "bricks and batons," was "slapped in the ears," and was "shackled right leg to left arm for up to a week at a time." (Id. ¶¶ 26-32.) This torture caused chronic "dizziness," "headaches," and "joint pain," as well as a "ringing or buzzing in [his] ears." (Id. ¶¶ 37-39.)
Dr. Crosby's physical exam of Karake identified 21 distinct lesions or clusters of lesions, some of which contained six or seven individual scars within the cluster. (Id. ¶ 40.) The lesions are visible in a series of photographs numbered 1-29 ("P #1 -- P #29") that are attached to Dr. Crosby's written report. (Def. Ex. K-25.) Dr. Crosby asked Karake about the source of each of these lesions. (See id. ¶¶ 43-61.) Importantly, Karake did not attribute each scar on his body to torture. Indeed, some of the most clearly visible scarring, on his chest and abdomen, Karake described as "deliberate cutting wounds . . . inflicted . . . as part of a traditional practice . . . unrelated to his experience of torture."*fn79 (Id. ¶ 46 & P # 7-10; see also 5/31 p.m. tr. at 47-52.) Other scars were identified by Karake as caused by "an injection" (id. ¶¶ 50, 61 & P #15, 28) or by "childhood injuries." (Id. ¶ 53 & P #18.) The cause of still others Karake simply could "not recall." (Id. ¶¶ 48, 52, 55, 57, 59 & P #13, 17, 21-22, 24-25.) In fact, out of the 21 lesion clusters identified by Dr. Crosby, Karake attributed 13 of them to causes other than torture by Rwandan officials. Of the remaining eight, though she admitted that it is "impossible to date scars when they are healed" (5/31 p.m. tr. at 72), Dr. Crosby found the physical characteristics of each to be consistent with Karake's explanations of how they were caused.
Karake testified in court and told Dr. Crosby during his examination that he was tortured both at Mweso and at Kami Camp. At Mweso, Karake testified, he was hit in the head several times with a piece of wood. (5/22 p.m. tr. at 8-12.) Dr. Crosby identified three lesions on Karake's head that she determined were consisted with blunt trauma injury as described by Karake. (5/31 p.m. tr. at 47-55; Def. Ex. K-25 ¶ 40, 44 & P #3-6.) Karake also claimed that while "trying to ward off blows" during one of his beatings at Mweso, "he was cut with a piece of wood" that resulted in scarring on his left upper arm. (Id. at 57-58.) According to Dr. Crosby, the two "hyperpigmented lesions that measure 4 and 2 centimeters" visible in P #14 are perfectly consistent with Karake's description of the object used to inflict the injury and his description of having "raised his arm to protect himself. (Def. Ex. K-25 ¶ 49 & P #14.) Karake also identified a series of "six hyperpigmented lesions ranging from 4 to 6 centimeters" in length on his "left lateral side" as caused by "two people holding a piece of barbed wire and cutting him." (Id. at 56-57; see also Def. Ex. K-25 at P #11.) Dr. Crosby's examination indicated that these lesions were "consistent with scarring caused by cutting wounds inflicted with a sharp object, such as barbed wire." (Id. ¶ 46.)
Dr. Crosby also identified a series of scars that were consistent with Karake's description of his abuse at Kami Camp. Just above the ankle on his right leg, Karake has an "irregular, hyperpigmented" scar slightly less than 1 centimeter long. (5/31 p.m. tr. at 59; see also Def. Ex. K-25 ¶ 40 & P #19.) This scar takes on added significance in light of an additional "small linear scar" on Karake's left wrist. (5/31 p.m. tr. at 61; see also Def. Ex. K-25 at P #23.) Individually, both these lesions are "consistent with scarring caused by shackles." (Def. Ex. K-25 ¶¶ 54, 56.) Together, they provide substantial corroboration that "he had been shackled from his left arm to his right leg . . . for extended periods of time." (5/31 p.m. tr. at 59.) The doctor found another 2 centimeter scar on Karake's left forearm to be consistent with Karake's explanation that it resulted from being struck with a brick. (Def. Ex. K-25 ¶ 51 & P #16.) Dr. Crosby also testified that the scar visible on Karake's wrist in P #23, which is consistent with scarring caused by shackles, is found on the portion of Karake's wrist that is covered by a bandage in the photograph of Karake taken by the FBI on February 5, 2002.*fn80 (Id. at 64; cf. Gov't Ex. 12; see also Def. Ex. K-25 ¶ 64.) Also visible in one of the FBI's photos from that date is an injury to Karake's forearm that corresponds to the scar Karake identified as having been caused by a brick.*fn81 (5/31 p.m. tr. at 64-65; 85; Cf. Gov't Ex. 13-13a.) It is also obvious, based merely on the Court's observation of Gov't Ex. 13, that Karake's entire forearm, from wrist to elbow, was substantially swollen on February 5, 2002.
Dr. Crosby further testified that Karake's descriptions of having been "slapped hard on his ears while at Kami Camp, which resulted in . . . ringing and buzzing in his ears," was consistent with a form of torture known as "telefono." (5/31 p.m. tr. at 65-66.) The doctor also noted that "[m]ost lesions that result from beatings (contusions) will heal without scars in 6 to 8 weeks." (Def. Ex. K-25 ¶ 63.) Thus, the "lack of extensive scarring in no way rules out that multiple beatings occurred." (Id.) Dr. Crosby found that Karake's "physical injuries are all typical sequelae of torture . . . and are consistent with [his] description of how the injuries occurred." (Id. ¶ 66.) Moreover, Karake "was extremely forthcoming about scars that did not result from torture." (Id.) Thus, based upon her "physical findings" and "overall clinical presentation," Dr. Crosby concluded that Karake "suffered physical . . . torture" while he was detained at Mweso and Kami camps. (Id.)
Nyaminani was examined by his expert, Dr. Jonathan Arden, on May 23, 2005 and November 23, 2005.*fn82 (Def. Ex. N-12B at 2.) Dr. Arden is a medical consultant who practiced forensic pathology for more than 20 years, primarily as a government-employed medical examiner. (Id. at 1.) Dr. Arden's written report states that shackling or binding often results in patterned scarring. (Id. at 4.) Scarring from binding typically approximates the form of the ligature in both width and texture. (Id.) Thus, handcuffs, which typically are constructed of two parallel metal plates, will often "leave a pair of parallel linear imprints," and they "characteristically cause more prominent marks on the backs . . . of the wrists." (Id.) Dr. Arden identified numerous patterned scars on Nyaminani's extremities that are consistent with Nyaminani's explanation as to their cause.*fn83 Dr. Arden found transverse linear scars on Nyaminani's wrists/forearms and ankles/lower legs. (Id. at 2.) Specifically, he identified "two parallel linear scars" on the back of Nyaminani's left wrist/forearm. The more proximal was roughly 6.5 cm long by less than 0.1 cm wide, and the more distal was 5.5 cm long by roughly 0.1 cm wide. (Id.) Because of the thinness of the scarring, Dr. Arden opined that these scars are consistent with Nyaminani's allegation that he was bound at various times with thin, nylon rope or fishing line. (Id. at 4; see supra Section IV(B)(1).) Dr. Arden testified that a fishing line typically exerts a concentrated force on a very thin profile, often breaking the skin and leaving a narrow injury. (Def. Ex. N-12B at 4.) Nyaminani has a 3 by 0.2 cm transverse linear scar on his lateral left ankle, with an area of scarring approximately 3 by 2 cm extending upward from it. His anterior right ankle has a 3 by 0.4 cm transverse scar. (Id. at 3.) This scarring is "slightly thicker than those on the wrists, and . . . consistent with being cause by tight handcuffs or shackles." (Id. at 4.) Dr. Arden noted that the irregular scarring extending upwards from the linear injury on Nyaminani's left ankle is consistent with Nyaminani's explanation that his ankle had become infected after being cut by the shackle. (Id.) "Such infection would tend to spread locally, especially in a proximal direction (i.e., up the leg) following lymphatic drainage." (Id. at 4-5.)
Dr. Arden also identified a "series of short (0.3 -- 0.4 cm) parallel longitudinal scars on the . . . left wrist." (Id. at 4.) The pattern of scarring would require "multiple separate incisions" and could not be "the result of random impact or a single cut." (Id. at 5.) Dr. Arden's assessment comports with Nyaminani's explanation that his wrists had become swollen from the positional torture he endured and "as an attempt at therapy . . . multiple cuts were made to release the swelling." (Id.) Dr. Arden found "nothing about these scars or their patterns that suggested a more probable explanation" than that proffered by the defendant. (Id.) Thus, Dr. Arden concluded that Nyaminani's scars "are highly consistent with [the] specific torture scenarios that he described."
Bimenyimana was examined on two occasions by his expert, Dr. David Stein, who is a board-certified dermatologist and a specialist in preventive medicine and public health. (Def. Ex. B-7 at 1.) Dr. Stein serves on the faculty at both The Johns Hopkins School of Medicine and The Johns Hopkins Bloomberg School of Public Health. (Id.) Dr. Stein met with Bimenyimana for approximately 3 hours on December 29, 2005, and then again for some 2 hours on February 28, 2006. Bimenyimana "convincingly described" to Dr. Stein the manner in which "his arms [were] extremely tightly bound" during his incarceration at Kami. (Id. at 2.) Bimenyimana also informed the doctor that his legs had been "very tightly bound for days or weeks at a time," and that at times, "his arms were also bound tightly to his legs." (Id.) The consequence of this tight binding was that Bimenyimana was forced to "repeatedly drag himself along the rough surface of the ground to relieve himself, obtain drinking water[,] or for other essential reasons," resulting in "repeated trauma, both acute abrasions and tearing of his skin and repeated pressure to the bony prominences of his arms and legs at multiple sites and also along the sides of his hips." (Id.)
Dr. Stein identified numerous scars during his physical examinations of Bimenyimana that were "highly indicative of prolonged extremely tight binding of his arms and . . . legs," as well as other "injuries which tore the skin" that were consistent with Bimenyimana's vivid description of having to drag himself along the ground while bound.*fn84 (Id. (emphasis in original).) On Bimenyimana's left arm, Dr. Stein identified five "externally induced markings" indicative of "being bound in an extremely tight fashion over a period of days to weeks." (Id. at 3.) On the "upper aspect of the antecubital fossa" (or, in layman's terms, the fold of the elbow), Dr. Stein found a 6 by 0.2 cm curved, linear, hyper-pigmented marking, and two linear, hyper-pigmented "streaks, " approximately 7 by 0.3 cm and 3 by 0.3 cm in size. (Id.) On the left mid-antecubital fossa Bimenyimana has a 9 by 0.2 cm linear hyper-pigmented mark. (Id.) And on the back of his left hand "over digits II, IV and V are linear atrophic scars from a rope or similar object cutting into the skin."*fn85 (Id.) Similarly, on Bimenyimana's right arm, Dr. Stein identified a hyper-pigmented linear serpentine mark roughly 6 by 0.3 cm near the antecubital fossa, a linear hyper-pigmented mark approximately 5 by 0.2 cm on the central lower aspect of the antecubital fossa, and a 1 by 0.4 cm hyper-trophic linear scar on the distal medial aspect of the antecubital fossa. (Id.) Additionally, Bimenyimana has a 7 by 0.2 cm and an 8 by 0.2 cm linear serpentine hyper-pigmented mark on the upper aspect of his right arm, and a 0.2 by 0.2 superficial hyper-pigmented linear serpentine marking on his right wrist. (Id.) Dr. Stein describes the latter as "rope-like." In addition, Bimenyimana's legs exhibit "evidence of prolonged extremely tight binding." (Id.) This includes a 6.5 by 0.2 cm linear mark with slight hyper-pigmentation located medial and distal to his right tibia. (Id.) His left leg exhibits linear, slightly hyper-trophic scars that are "consistent with cuffs or rope cutting into his skin." (Id.) Dr. Stein described the "hyperpigmented marks" as "highly consistent" with tight binding. (5/30 p.m. tr. at 24.)
Bimenyimana's legs also exhibit markings that "are consistent with trauma which tore the skin," rather than cutting it. (Def. Ex. B-7 at 3.) These marks corroborate the scraping injuries Bimenyimana testified that he suffered from dragging himself on the ground while bound.
Bimenyimana has a 4 by 2.5 cm atrophic scar with hypertrophic borders that exhibits both hypoand hyper-pigmentation. This scar is located at the top of his left knee. (Id. at 4.) Dr. Stein also identified a 3 by 0.6 cm linear atrophic scar on the distal medial aspect (the lower middle part) of the knee, as well as a 1 by 1.7 cm atrophic scar on the medial proximate aspect (upper middle portion) of his left leg and a 1 by 1.2 cm atrophic scar on the medial distal (lower middle) aspect of the left leg. (Id.) Additionally, Bimenyimana has a 2.2 by 1.5 cm atrophic scar on the medial aspect of his right tibia and a 1 by 0.7 cm atrophic scar below the lateral malleolus of his right ankle.*fn86 (Id.) These injuries are likewise "indicative of scraping or cutting through the epidermis" in a manner consistent with Bimenyimana's testimony. (Id.) Bimenyimana's arms also exhibit evidence of trauma caused by scraping or tearing of the skin. Distal to his left elbow there are two atrophic scars, one measuring 1 by 1.1 cm and the other 1.2 by 1 cm. On the medial aspect of the elbow is a 0.5 by 0.4 hypertrophic scar. (Id. at 3.) There is also atrophic scarring on the back of his right hand over the metacarpophalangeal joints. (Id.)
Based on his physical examination of Bimenyimana, Dr. Stein concluded "to a reasonable degree of medical certainty that the serpentine and linear markings proximate to his elbows, wrists and legs are consistent with [exposure] to prolonged extremely tight binding." (Id. at 4.) Moreover, Dr. Stein found "compelling . . . physical evidence of a medical nature" that Bimenyimana "sustained further injury resulting in scarring" from being "so tightly bound that he was forced to drag the sides and extremities of his unprotected body across a rough abrasive surface." (Id.)
In contrast to the typical case, the conclusions of defendants' medical experts are buttressed, not contradicted, by the findings of the prosecution's medical expert, Dr. Jerry Spencer. Dr. Spencer is currently employed by the Office of the Armed Services Medical Examiner in Rockville, Maryland as a Deputy Medical Examiner. (Gov't Ex. 116 at 3; 6/2 a.m. tr. at 30.) Dr. Spencer has served as professor of pathology at the Texas Tech University School of Medicine and as an Associate Professorial Lecturer in the Department of Forensic Sciences at George Washington University. (Gov't Ex. 116 at 3-4.) Dr. Spencer examined all three defendants; he met with Nyaminani and Bimenyimana on March 29, 2006, and with Karake on April 11, 2006, for approximately one hour each. (Gov't Ex. 112-115.) Dr. Spencer submitted written reports with respect to each defendant (Gov't Ex. 113 (Bimenyimana), 114 (Nyaminani) & 115 (Karake)), as well as a memorandum summarizing his conclusions as to all three. (Gov't Ex. 112.) Additionally, the government introduced two compact discs containing photographs taken during Dr. Spencer's examinations of Bimenyimana and Nyaminani (Gov't Ex. 120) and Karake that clearly portray the scarring that was found by all experts. (Gov't Ex. 121.)
Dr. Spencer's methodology was similar to that of the defense experts: after a scar was identified, the defendant's "recollection of the injury that result[ed] in the scar was obtained and recorded. Their explanations of the source of the scars were identified for reference purposes, and for formulating the [doctor's] opinions." (Gov't Ex. 112 at 2.) As a general matter, Dr. Spencer found the defendants' explanations "consistent with the scars identified." (Id.) Thus, in evaluating Karake, Dr. Spencer concluded that the "pattern of the scar of the left wrist . . . in combination with the scar on the right ankle . . . is consistent with the stated explanation . . . that his left wrist was handcuffed to his right ankle." (Gov't Ex. 112 at 4.) With respect to Nyaminani, Dr. Spencer found that the "single linear scar of the right wrist region . . . and the scars of the left palm and wrist . . . are consistent with scars resulting from injuries caused by binding or handcuffing of the wrists." (Id. at 3.) The doctor found that "the bilateral presence of the scars" provides additional support for concluding that the scars were "from binding or handcuffs." (Id.) In response to the Court's question regarding the obvious spotting that was visible on the pictures of Nyaminani's feet, Dr. Spencer testified that hyper-pigmented scars on the bottom of both of Nyaminani's feet were "not normal," and could be, consistent with what Nyaminani had told him, the result of being "struck in those areas while bound." (6/2 a.m. tr. at 51.) Additionally, the "pattern of scarring on [Nyaminani's] lower legs . . . is consistent with scars resulting from injuries made by shackles." (Id. at 3.) Again, "their appearance on both lower legs . . . [is] evidence consistent with shackling." (Id.) Dr. Spencer drew similar conclusions regarding the "pattern of scarring on the lower legs of [Bimenyimana]," stating that the "circumferential [completely around the lower leg] nature of these scars . . . and the fact that scars appear on both lower legs is evidence consistent with shackling." (Id.) Dr. Spencer also found patterns of scarring on Bimenyimana's right wrist and right and left arms that are "consistent with binding that caused injuries that penetrated through . . . the skin." (Id.)
Though Dr. Spencer basically concurred with the defense experts (6/2 a.m. tr. at 36-38), concluding that defendants' explanations were "consistent with" their scars, meaning that theirs was "one possible explanation" (6/2 a.m. tr. at 34), he resisted drawing any conclusions regarding the cause of the scarring "to a reasonable degree of medical certainty," because it is not usually possible to pinpoint the cause or timing of a scar. (Id.) Dr. Spencer could not, however, give any other credible explanation for the cause of the scars that was, in his view, either equally plausible or more plausible than those given by defendants. (Id. at 39.) In fact, Dr. Spencer essentially admitted that "the fact that these scars are on both hands [on] two of the individuals [and] on both legs of two individuals, increases the plausibility that they came from binding and shackling." (Id.)*fn87
The government attempts to belittle this testimony by its own expert, as well as the three defense experts, by arguing that "at the end of the day [the expert testimony] really doesn't do anything to assist the Court, because it all comes back to credibility . . . of the person reporting how they got the scar." (Id. at 55.) The Court cannot agree. There were many photographs introduced by the medical experts that clearly depict injuries as described by the defendants, as well as three taken contemporaneously by the FBI of Karake on February 5, 2002. (Gov't Ex. 12, 13, & 13a.) The expert testimony proffered by both sides unanimously concluded that this physical evidence is, at a minimum, "consistent with" the defendants' testimony, and even the government's expert could not provide any other plausible explanation of how three defendants, all of whom shared the same experience -- being housed at Kami for lengthy periods of time under Kibingo's custody -- could exhibit exactly the same telltale signs of abuse, including binding, shackling and handcuffing.
b. Non-Physical Evidence of Torture at Kami Camp
Defendants also rely on corroborating evidence from witnesses with first-hand knowledge of Kami and from reports to witnesses and to U.S. government agencies that document the abuse and inhumane conditions that prevailed at Kami.*fn88 For instance, defendants introduced the testimony of two witnesses who had experienced torture at Kami Camp. These two individuals, who testified under the pseudonyms of Mr. X and Mr. A, were detained and tortured at Kami by the same RPF government that defendants accuse of torture in this case. The testimony of Mr. A and Mr. X was originally provided under seal. In order to provide as complete a public record as possible, the Court ordered as much of the testimony unsealed as possible, redacting only such identifying information as necessary to ensure the safety of the witnesses and their families. (See August 11, 2006 and August 17, 2006 Orders unsealing testimony and pleadings.)
Mr. A served as a member of FAR during the Hutu-dominated Habyarimana regime. (5/17 a.m. tr. at 12-13) He was arrested shortly after the RPF took over the government (5/17 a.m. tr. at 13), and was taken to a nearby military facility where he was questioned. (5/17 a.m. tr. at 15-16.) Following his arrest, he was stripped of his clothing, had his arms tied behind him with a rope so "that the elbows could meet in the back," and then beaten over his "entire body," including his head. (Id. at 17-20.) After spending a little over a day being interrogated at two different facilities, Mr. A was transferred to Kami Camp, a place, he was told, "where . . . anything that he has not confessed, he will confess." (Id. at 20.) Mr. A was reminded when he arrived at Kami "to confess" because "whoever enters Kami doesn't get out." (Id. at 21.) While at Kami, Mr. A was subjected to further interrogations and beatings. Again Mr. A had his arms tied so that his elbows met behind his back.*fn89 While he was tied in this position, soldiers at Kami Camp drew another rope under his elbows and used it to hoist him so that he was hanging in the air by his arms. (Id. at 22.) While he was suspended in the air, he was beaten until he lost consciousness. (Id. at 23.)
When he first arrived at Kami, Mr. A was housed in a large Go-Down that contained as many as 200 people. (Id. at 23-24.) There was "no light whatsoever" in the Go-Down. (Id. at 23.) Once per day, prisoners were given a portion of food no larger than they could carry in their hands, and they had only a can to use as a bathroom. (Id. at 25-26.) During his detention it "was not possible" for Mr. A to contact any friends or family members. (Id. at 28.) Because of the frequency with which Mr. A was beaten into unconsciousness, he was unable to testify precisely how long he was kept in the Go-Down, but he estimated it at about six months. (Id. at 25.) After that time, Mr. A was moved first to another Go-Down known as "Sick Bay," in order to recover from his injuries. Sick Bay contained fewer prisoners (around 60), and Mr. A received slightly more food in a small bowl, rather than having to hold it in his hands. (Id. at 27.) Next, Mr. A was moved to "Special University," where he was assigned the task of disposing of the bodies of prisoners who died while at Kami. (Id. at 29-30.) Mr. A noted that, unlike Go-Down, which had "no windows," Special University was above ground. (Id. at 29.) Nevertheless, prisoners at Special University were likewise kept in "total darkness." (Id.) While at Kami, he saw handcuffs. (Id. at 49.) Mr. A was eventually moved back to the Go-Down where he was originally housed before being released from Kami after approximately two years of incarceration. (Id. at 31, 55.) Ultimately, he was granted political asylum abroad. (Id. at 35-36.) Mr. A's entire period of incarceration at Kami occurred several years prior to Kibingo's arrival as camp commander in 1999.
The second witness who was detained at Kami testified under the pseudonym Mr. X. Mr. X is a former soldier who joined the RPA prior to the overthrow of the Hutu government in 1994 and continued to serve in it thereafter. (5/18 a.m. tr. at 37-38.) After refusing to perform an assignment, however, Mr. X was arrested by DMI in January 1997 and taken to a military prison, where he was held for approximately three months. (Id. at 42-43.) While at the military prison, Mr. X was beaten (id. at 50) and accused of being untrustworthy (id. at 47), apparently not just because of his insubordination, but also because it was believed that Mr. X had shared classified information with unauthorized persons. (Id. at 51.) Mr. X was also subjected to electric shocks during interrogations. (Id. at 52.) Eventually, the Rwandans "stopped demanding" information and were simply "punishing" him by beating him on the legs, head and buttocks with sticks (id. at 53) and forcing him to perform "exercises" such as "jump[ing] up and down" with a heavy log on his shoulders. (Id. at 55-56.)
After three months, in April 1997, Mr. X was transferred to Kami. (Id. at 43, 56.) At Kami, Mr. X was forced to live in an outdoor pit that required a ladder to get into or out of. (Id. at 57.) The pit was wide enough at the bottom that two people could comfortably fit with a "big gap" between them. (Id.) Mr. X remained in this pit for all but two hours in the morning and two hours in the evening. (Id. at 58.) Another man, who was living in the pit when Mr. X arrived, died there during Mr. X's detention. (Id. at 59.) Mr. X stated that he was fed "the same food the soldiers were fed on[:] beans, rice, and corn seeds." (Id.) Mr. X does not, however, claim that he was beaten at Kami. (Id. at 95.) After three months at Kami, Mr. X was released by DMI, which told Mr. X that they had determined that he was innocent. (Id. at 60-61.) However, in 2004, Mr. X fled Rwanda to seek asylum elsewhere after learning that DMI was considering killing him because of his involvement with an opposition political party. (Id. at 61-63.)
The government urges the Court to disregard this testimony as out of time, since the events about which these individuals testified predated Kibingo's arrival at Kami in 1999. The government's argument is not persuasive. First, even though Kibingo had not yet taken charge of Kami, it was a DMI-controlled facility as early as 1994 and throughout the time period at issue. According to Kibingo, it was "the head of DMI decides who goes to Kami Camp and who leaves." (5/5 a.m. tr. at 72; 5/17 a.m. tr. at 15; 5/18 a.m. tr. at 50.) But more importantly, U.S. government reports lead credence to the inference that the practices and conditions that Mr. X and Mr. A endured continued throughout Kibingo's tenure. For instance, State Department Country Reports on Human Rights Practices from 2001 and 2003 (Def. Ex. K-2 & K-3) document "numerous serious" human rights "abuses" by the Rwandan government that date back several years, including "extra-judicial killings and other serious abuses," reports of torture and beatings by security forces, and "several credible reports of disappearances." (Def. Ex. K-2 at 1.) In particular, the 2003 State Department Country Report specifically identified Kami Camp:
On April 1 , police arrested RDF Major Felicien Ngirabtware,*fn90 interim commander of Nyakinama Military School, reportedly in connection with Brigadier General Habyarimana's self-imposed exile on March 30, and detained him at Kami military barracks in Kigali. There were credible reports that Major Ngirabatware was tortured, and his family spent more than 2 months without information about his whereabouts. At year's end, he was still in prison; however, his family was able to visit him. (Def. Ex. K-3 at 4.)
These State Department reports were further corroborated by other information received in February 2000 by U.S. officials. At that time it was reported by an asylum seeker who had been an intelligence chief for the RPA that he had "knowledge of a secret DMI jail at Kami Barracks where prisoners were tortured. He estimated that 400 persons were held at 'Kami go- down.' " (See Attachment 1 to the Court's May 9, 2006 Order Excluding From Discovery Certain Items of Classified Information and Authorizing the Substitution of a Summary.)
Further evidence of continuing abuse and torture was provided by two former Rwanda ministers, who learned about the serious problems at Kami in the 2001-2003 time frame. Theobald Gakwaya was the Minister of Internal Affairs for the Rwandan government from March 2000 until April 2001. (5/17 a.m. tr. at 60.) In that capacity, he was in charge of all civilian prisoners and the Rwandan police. (Id.) As part of his duties, Gakwaya visited all the civilian prisons in Rwanda. (Id. at 62.) Though he was not in charge of and did not visit them, Gakwaya was aware of three military prisons (Mulindi, Kibungo and Kinome) to which people had "access." (5/17 p.m. tr. at 19-20.) Kami Camp, however, was a "prison" that housed "soldiers and civilians" to which "nobody else had access." (Id. at 20.) Gakwaya attempted to visit Kami "several times" during his tenure as Minister of Internal Affairs but his requests were "denied."*fn91 (5/17 a.m. tr. at 62.) While he served as minister, Gakwaya reserved one day per week to meet with citizens and hear their complaints. (Id. at 63.) During these meetings, Gakwaya received several reports of abuse at Kami. These complaints continued until the time he left office in April 2001, when he fled the country amid fears for his personal security after having been terminated from his position. (Id. at 66; 5/17 p.m. tr. at 17.) Specifically, he heard from several people that "their relatives had disappeared [and] they were told that these people had been taken to Kami." (5/17 a.m. tr. at 63.) The central figure in many of these reports was Jack Nziza: "in all the reports of those that disappeared at Kami Camp, it was said that . . . they were sent by Jack Nziza."
Gakwaya also spoke with two individuals who had actually been imprisoned at Kami for a time. (Id.) One man showed him numerous scars from injuries he claimed he suffered while at Kami, including "scars that were caused by the ropes [that] had been tied on the arms," as well as scars on "his feet and on his private parts." (Id. at 64.) The man indicated to Gakwaya that he had been tied with his elbows behind his back and that he had been "beaten under the feet." (Id. at 64-65.) His genitalia appeared to have been burned; according to the man this occurred through the use of electric shocks. (Id.) This conversation took place sometime around June or July of 2000. (Id. at 64.)
Former Minister of Defense Emmanuel Habyarimana also testified. Habyarimana was a member of FAR prior to 1994, but joined the RPF government as part of its reconciliation efforts following the civil war. (Id. at 48.) Between 1994 and 2003 he held senior leadership positions in the Rwandan government, culminating in his service as Minister of Defense from May 2000 until March 2003. While he was the Minister of Defense, Habyarimana received a report from a woman that her husband, a man named Mugabo, had been detained at Kami Camp by Jack Nziza. (Id. at 40-42.) Though he knew of Kami Camp and had even visited it once in 1999, Habyarimana had not known that Kami was being used as a prison prior to receiving this report. (Id. at 35, 41) Habyarimana called Jack Nziza, who confirmed that Mugabo was being held at Kami. (Id. at 41.) Habyarimana complained to Nziza about imprisoning people in a place "not known as a prison" and "not known by the law," but Nziza contended that Mugabo was a threat to national security because of his political affiliation with former President Pasteur Bizimungu, a Hutu. (Id. at 42.) Habyarimana fled Rwanda in March 2003 because of fears for his own safety after reports linked him to an opposition political party. (Id. at 44.)
4. Kibingo's Lack of Credibility
The strength of defendants' evidence regarding torture at Kami Camp must be measured against Kibingo's testimony denying that any such abuse took place. Kibingo's testimony is undermined by the bias he exhibited on the stand. Though the Court is well aware, as it has noted throughout this Memorandum Opinion, of the long-standing ethnic conflict in Rwanda and deep-seated animosity between those who fought for the RPA and those who fought for ALIR, no witness made his personal interest in the outcome of the proceedings more vividly apparent than Kibingo. Kibingo was injured twice while fighting FAR and ALIR; he was struck with shrapnel in a 1994 battle and then was shot in the stomach in 1999. For Kibingo, the war has not ended. He testified that after the FAR fled to the Congo, "they continuously came back to attack Rwanda and commit genocide." (5/9 a.m. tr. at 17.) The battle against ALIR continues "up to now." (Id. at 18.) Kibingo made it plain that, "wherever I'm going to be . . . I'm going to be fighting [ALIR]. " (Id.) In his view, the defendants "are working with ALIR," and, therefore, "[i]f they are punished . . . I will be happy. . . ." (Id. at 19, 21.) Kibingo stated that "[w]ars are fought in different ways," and agreed that he was "fighting ALIR" as he sat "in the courtroom." (Id. at 19-20.) In fact, Kibingo heatedly exhorted defendants' counsel that "even [you] should stand up and fight against ALIR." (Id. at 46.)*fn92
Beyond his palpable desire for revenge, Kibingo's credibility was damaged by the numerous matters about which his testimony was either impeached or simply not believable. As previously noted, there was considerable testimony contradicting Kibingo's testimony with respect to the conditions at Kami, the use of handcuffs, the access available to the outside world, the use of "Go-Downs" and his claim that no one died at Kami. There were also other significant inconsistencies that undercut large portions of his testimony. For instance, there is the matter of the date of Karake's arrival at Kami Camp. Kibingo testified that Karake did not arrive until January 2002.*fn93 (5/3 p.m. tr. at 37.) A January arrival was important to support Kibingo's story that Karake confessed virtually immediately without any encouragement or coercion (5/3 p.m. tr. at 65-69; 5/4 a.m. tr. at 19-29), since under Kibingo's scenario, there would have been no time for Karake to have been tortured into submission prior to his first statement. Yet, as the Court has already found, Karake's testimony that he was taken into custody around December 8 or 9, 2001 is far more credible. (See supra notes 21-22 and accompanying text.) Moreover, the fact that Karake's first statement bears a date stamp of "December 28, 2001" seriously undermines Kibingo's account regarding Karake's arrival at Kami. (Gov't Ex. 52.) Nor would a January arrival be consistent with Agent Dent's testimony that she alerted the Rwandans to her interest in Karake on December 5, 2001, and that the Rwandans were incredibly quick to track down suspects whom the FBI wished to talk to.*fn94 (5/16 p.m. tr. at 55-60.) Indeed, Kibingo inadvertently corroborated Karake's timeline when he testified that by the time he began interrogating him, Karake "had seen the people" conducting the interviews at Kacyiru and knew that they included "the FBI." (5/3 p.m. tr. at 69.) The only way Karake could have seen the FBI before February is for him to have been at Kacyiru in early December (as he so testified) before the FBI agents returned to the United States. Kibingo therefore cannot be believed when he claims that Karake did not arrive at Kami until mid-January 2002.
The Court likewise does not credit Kibingo's testimony regarding his lack of awareness of Bimenyimana's identity when he arrived at Kami Camp. As noted above, Kibingo testified that at the time Bimenyimana arrived at Kami on July 11, 2002, he did not know his true identity because he was using the alias "Tuyigire Mustapha." (5/4 p.m. tr. at 48.) Kibingo stated that he only learned that Bimenyimana was the man known as "Zappy Gadi" when the other Bwindi suspects pointed him out in August. (5/4 p.m. tr. at 49.) First, if this is true, it is difficult to understand why the other suspects would wait several weeks to point him out. Second, Bimenyimana admits that he was using that name at the time he was arrested (5/30 a.m. tr. at 62), but claims that he gave Rwandan authorities his real name as early as his detainment at Nyamilima on June 16 or 17. (5/30 a.m. tr. at 67; 5/30 p.m. tr. at 51.) According to Bimenyimana, the soldiers in charge of Kami knew full well who he was on the day he arrived. (5/30 p.m. tr. at 67.) Bimenyimana's testimony finds support in Dent's June 18, 2002 memo regarding her recent investigative trip to Rwanda in which she notes: "Recent information derived from a credible source suggests that Zappy Gadi . . . may be in Rwandan custody." (Gov't Ex. 99, tab 29 at 6.) Dent speculated that "the Rwandans are withholding [his] whereabouts . . . for national security purposes to glean all available information about the locations of Interahamwe soldiers at large before notifying the FBI of [his] apprehension." (Id.) Given Bimenyimana's notoriety, it is just not believable that U.S. investigative officials were aware of his capture, but those Rwandan officials who were responsible for housing and interrogating him were not. It is more plausible that, as with other high-ranking ALIR soldiers like Bemera, the Rwandans preferred to extract whatever information they could regarding ALIR before alerting the U.S. to Bimenyimana's availability for questioning regarding Bwindi. Kibingo's testimony, as with Karake, was more likely an attempt to undercut Bimenyimana's account of torture in July by eliminating the time gap between Bimenyimana's arrival at Kami in July and his statements to the Americans in August.
Kibingo's testimony regarding the timing of his interrogations of the Bwindi suspects is incredible in other material respects. Kibingo testified that he did not speak with a single Bwindi suspect about the attack until mid-January 2002. (5/3 p.m. tr. at 60.) That is flatly inconsistent with his repeated assertions that one of his most important duties was to find out what happened at Bwindi. (5/3 p.m. tr. at 53 ("The biggest responsibility was to guard them. And the other thing was to talk to them to know what they did in Bwindi."); ("It was in my responsibilities to talk to people who came to the camp."); ("It was one of my responsibilities to know about everybody who came in the camp."); at 54 ("[I]t was under my responsibility to talk to the people in the camp."); at 55 (instructed by General Nziza to "understand them and solve their problems."); 5/4 p.m. tr. at 26 ("It was in my duties to talk to him. And even others that were in the barracks had the right to talk to them.").) Given his singular focus with respect to the Bwindi suspects, the Court cannot credit Kibingo's explanation that he waited over a month to speak with any of the Bwindi suspects because he "saw that there was no problem," and just "wanted to know how their life was." (5/3 p.m. tr. at 61.) In light of his admission that he "wouldn't wait for a person to . . . instruct" him to interrogate the suspects because doing so "was [his] role" (5/4 a.m. tr. at 30), it is only reasonable to infer that he began the process of extracting information as soon as the suspects arrived at Kami.
Kibingo also attempted to minimize his role in the investigation in a manner that was highly suspect. First, Kibingo's claim that he was never instructed to interrogate the Bwindi suspects at Kami (5/4 a.m. tr. at 30; 5/4 p.m. tr. at 26) is belied by his testimony that he "was supposed to report" to his superiors at DMI whenever he obtained a written statement from one of them. (5/4 a.m. tr. at 29.) Second, Kibingo stated that he never had any conversations with Alfred Ndabarasa regarding the Bwindi suspects between December 2001 and March 2002. (5/4 p.m. tr. at 40.) He also claimed that he never participated in the interrogations at Kacyiru;*fn95 if he accompanied the defendants to Kacyiru, it was because he "was going to the office. [He] had no duty during these investigations." (5/5 p.m. tr. at 57.) He had no time to participate in interrogations at Kacyiru because of his "other duties." (5/9 p.m. tr. at 72.) If he ended up sitting in on an interview, "it's because [he] would be around just waiting for a ride." (Id.) These assertions are belied by Ndabarasa's testimony. Ndabarasa testified that Kibingo participated in the interrogation conducted by the Rwandans without the FBI on February 12, 2002.*fn96 (5/12 a.m. tr. at 110.) He also testified that the "morning just before the [February 12] interview," Kibingo met with investigators at Kacyiru and "revealed . . . details of how they conducted the killings." (5/12 a.m tr. at 97.) Kibingo also informed Ndabarasa that "he had made them write statements." (Id. at 98-99.) According to Ndabarasa, Kibingo was one of the people "facilitating . . . the investigation." (Id. at 112.) Ndabarasa's notes list Kibingo's name and phone number as someone "helping [him] in the investigation." (5/12 a.m. tr. at 111; Def. Ex. K-8 at 8.) He was "a valuable resource" (Def. Ex. B-4 at 4), because he "lived with [the suspects.] . . . He could talk to them without [Ndabarasa]," and then let them know if "he got to know something that was useful for the investigation." (5/12 p.m. tr. at 71.) This served the needs of the Rwandan investigators, as did his presence at interrogations, because it enabled the other investigators "to refer to him," and "say, look, . . . you told him this." (5/12 p.m. tr. at 85.) And even though Kibingo testified that he had no authority regarding when interviews were conducted ("I didn't decide anything. It was their program." (5/4 p.m. tr. at 7)), the evidence clearly demonstrated that whenever Kibingo informed his superiors that he had obtained new information from a defendant, the Rwandan would then contact Kayumba so that the Americans would return to conduct another interview. (See Gov't Ex. 30, tabs 1-4.)
Other parts of Kibingo's testimony simply made no sense. Kibingo testified that in his very first conversation with any of the Bwindi suspects, Desire Hategekimana told him that
Nyaminani was "telling [them] that [they] should keep quiet and not talk." (5/3 p.m. tr. at 57.) Though Kibingo initially testified that Desire told him this information in January after Karake arrived, on cross-examination he stated that his conversation with Desire occurred in December.*fn97 (5/5 p.m. tr. at 45.) After "some time," Kibingo "realized Nyaminani was a problem." (Id. at 57-58.) Kibingo confronted Nyaminani with this information, and he told Kibingo that the others were lying, i.e., that he was not instructing them to obstruct the investigation. (5/4 p.m. tr. at 13.) Nevertheless, on January 15, Kibingo "gave Nyaminani his own room" in order to ensure that Nyaminani did not "stop his friends" from talking about Bwindi. (Id. at 12-14.) Kibingo's response to this information is difficult to understand. First, if he viewed it as his duty "to know what they did in Bwindi" (5/3 p.m. tr. at 53), it is unclear why he waited several weeks to act on Desire's information that Nyaminani was sabotaging the investigation. Even more puzzling, though Nyaminani was locked in his room at night (id. at 14), Kibingo testified that he remained free to "stay with his colleagues . . . during the day." (Id. at 32, 33.) The purpose served by isolating him at night if he remained free to converse with his fellow suspects during the day is unclear. Kibingo stated that "[d]uring the day he would see them because they were even going to dig or to grow crop together. They went to fetch water together. They went together. They always stayed together during the day." (5/5 p.m. tr. at 46.) If Kibingo's description of defendants' freedom of movement around Kami during the day is accurate, then it made no sense to isolate Nyaminani at night. If defendants' movements were in fact more limited, or more carefully supervised, then it casts doubt on his assertions about the quality of life at Kami Camp. Either way, something is amiss.
Equally perplexing is Kibingo's explanation as to why Nyaminani was kept isolated from the other Bwindi suspects after March 8, 2002, when he first confessed to participating in the rape and murder of the three women. Kibingo initially responded that he was concerned that Nyaminani "would give them a plan for escaping." (5/4 p.m. tr. at 33.) Upon cross-examination, however, it turned out that what Kibingo actually meant was that he was concerned Nyaminani would "commit suicide" because he had no means of escape. (5/5 p.m. tr. at 17-18.) Kibingo's account of Nyaminani's attempted suicide is utterly bizarre and was contradicted by Nyaminani, who claimed that no such event occurred. (5/30 a.m. tr. at 40.) Kibingo testified that Nyaminani attempted to commit suicide by collecting bullets that were strewn around the Kami grounds, with the intention of ingesting the gunpowder "so that his stomach can get torn and then he dies." (5/4 p.m. tr. at 43.) Kibingo stated that when Desire told him of Nyaminani's plan, he counseled Nyaminani against suicide and forced him to pour the powder from the bullets out on the ground. (Id. at 44-45.) While this suicide story apparently was offered by Kibingo to justify the fact that he kept Nyaminani in isolation long after he confessed, it is completely nonsensical.
Lastly, the Court finds Kibingo's account of his relationship with the prisoners at Kami Camp totally implausible. As previously noted, Kibingo testified that he would sit "in chairs" "on the veranda" of his house and talk with the Bwindi suspects, occasionally sharing meals or drinks. (See id. at 57, 65-66; 5/4 a.m. tr. at 47; 5/5 a.m. tr. at 38.) Such an idyllic scene simply does not square with Kibingo's proclamation that "wherever I'm going to be . . . I'm going to be fighting [ALIR]. " (5/9 a.m. tr. at 18.)
5. Lack of Corroboration for Kibingo
The government attempts to deflect attention from its reliance on Kibingo by arguing that Kibingo's testimony was corroborated by U.S. investigators who claimed not to have observed any visible signs of torture during their interrogations. (6/19 p.m. tr. at 39, 62-63.) In the end, however, the government cannot prove a negative. First, as the government concedes, neither the FBI nor RSO Bachmann had any idea what Kibingo was doing at Kami Camp: "The fact is not everybody, either the Rwandan officials themselves . . . or the RSO and FBI, knew that Captain Kibingo was obtaining statements when he was obtaining them. . . . They just didn't know." (Id. at 74.) All the defendants' statements after December 2001 that were ultimately communicated to American investigators were initially obtained by Kibingo at Kami, and yet not a single witness, American or Rwandan, had any first-hand knowledge of the conditions of confinement or Kibingo's treatment of the defendants during their detention.*fn98 (5/4 p.m. tr. at 27.)
Second, the Americans had only limited contact with the defendants over the course of the 15-month interrogation period. No U.S. official visited or attempted to visit Kami while defendants were confined there. (5/5 a.m. tr. at 61; 5/10 p.m. tr. at 96; 5/16 p.m. tr. at 8.) U.S. officials saw Karake on four and Nyaminani on ten of the roughly 450 days that each was held at Kami; they saw Bimenyimana on five of the 240 days during his detention at Kami. Despite their knowledge (or assumption) that the Rwandan investigators were interrogating the suspects outside their presence (5/11 p.m. tr. at 93-94), when the Americans did have contact with the defendants, they made only cursory inquiries into the conditions of their confinement. In fact, Agent Dent conceded that it was indeed "possible" that defendants were tortured even though it did not occur to her at the time.*fn99 (5/16 p.m tr. at 19.) No U.S. investigator ever examined a defendant's body for signs of abuse other than what may have been visible to them. (5/16 p.m. tr. at 18 (Dent); 5/11 p.m. tr. at 68-69 (Bachmann).)*fn100 When Karake arrived on February 5 with a bandaged wrist and swollen arm, Agent Dent was told that he had a blister that had become infected, but she made no inquiry into how the injury had occurred in the first place. (5/15 p.m. tr. at 93.) The FBI never inquired about the conditions of Karake's confinement. (5/19 a.m. tr. at 32.) When Nyaminani informed the FBI that Kibingo had struck him with a flip flop because Kibingo did not believe that he was being truthful, the FBI did not make inquiry of Kibingo, but rather it disregarded Nyaminani's statements on the basis of reassurances from other Bwindi suspects, who "all more or less laughed about the incident." (5/19 p.m. tr. at 36-38; 5/19 p.m. tr. at 38, 111-14.) Importantly, none of these suspects admitted to actually seeing the alleged beating (Gov't Ex. 29, tabs 34, 36) and their statements that the environment at Kami was "open and friendly" (5/16 p.m. tr. at 17) could hardly be very reassuring given the substantial possibility of retribution at the hands of Kibingo.*fn101
Furthermore, while cursory inquires were made of each defendant about his treatment at Kami,*fn102 all inquiries were made through Isidore Kayumba, a Rwandan national who had fled Rwanda during the Hutu regime. (5/11 p.m. tr. at 90-91.) Though he worked as an investigator for the State Department (id.), defendants had reason to believe that he was closely associated with the Rwandan investigative efforts. He participated in the interviews not just as a translator, but also as an interrogator. (5/12 p.m. tr. at 50.) He often transported Bwindi suspects from Kami to Kacyiru in the company of Rwandan police. (5/2 p.m. tr. at 43; 5/31 a.m. tr. at 25-28.) And based on the defendants' testimony, he participated in the Rwandan interrogations on February 12 when no American was present, and he communicated either in person or by cell phone with Kibingo. (5/22 p.m. tr. at 62; 5/25 a.m. tr. at 13; 5/30 a.m. at 44-47; 5/31 a.m. tr. at 68.) Most importantly, it is apparent from Bachmann's cables that Kayumba served as the link between the Rwandan and the U.S. investigators. (Gov't Ex. 30; 5/10 a.m. tr at 27.) There was therefore every reason for the defendants to withhold information about their treatment from the Americans, believing that it would be conveyed by Kayumba either directly to Kibingo or indirectly via other Rwandan officials.
One reason for the American officials' reluctance to be more probing became crystal clear during the hearing -- the Americans did not want to tread on the Rwandans' sovereignty. They considered the defendants to be in the custody and care of the Rwandan government, which, in their view, limited their authority to inquire into the conditions of the suspects' confinement. (See, e.g., 5/11 p.m. tr. at 93; 5/16 p.m. tr. at 3-5.) No American official attempted to visit Kami during the time that defendants were incarcerated there. (5/19 a.m. tr. at 36.) This was a joint investigation in which the two sides were working "hand in hand," yet the Americans were completely dependent on the continued willingness of the Rwandans to locate potential suspects and to grant U.S. investigators access to them. (id.) Therefore, despite being "surprised" by the continuing flow of new information, the Americans made no inquiry of the Rwandans as to how they were able to obtain so much more information than U.S. investigators, or why suspects, whom the FBI believed to be forthcoming, suddenly revealed further inculpatory evidence when questioned by Rwandan officials.*fn103 While such studied restraint may have facilitated the investigation and strengthened the relationship between the two investigative teams, it provides no basis for the Court to conclude that either RSO Bachmann or Special Agent Dent has provided any meaningful corroboration for Kibigo's testimony.
6. Attacks on Defendants' Credibility
In addition to trying to bolster Kibingo's credibility based on the corroborating observations and testimony of the two American investigators, the government launches a frontal attack on defendants' credibility. While arguably the Court need not address this issue, since its findings regarding Captain Kibingo are more than sufficient to sustain its conclusion that the prosecution has failed to carry its burden of proof as to the voluntariness of defendants' statements to the Rwandans during their period of detention, several observations regarding the government's challenge are in order.
As argued in their closing, the prosecutors contend that the defendants' testimony must be rejected because they mitigated their involvement in the killings at Bwindi. (6/19 a.m. tr. at 7, 13, 19, 24.) In effect, according to the government, the voluntariness of the statements is shown by the fact that the defendants, even though they confessed, continued to minimize their role in the crimes. Under this theory, defendants lied about their involvement at Bwindi during the early interrogations and, when they later admitted the crimes, they claimed that they were "just following orders," thereby demonstrating their continuing willpower to disclaim responsibility in contradiction to their claim of torture. (6/19 am. tr. at 7, 19-20.)
This approach suffers from several flaws, both as a matter of logic and fact. First, while the Court's job is not to determine the truth or falsity of defendants' statements, the government's position is premised on an unproven assumption that defendants' early statements are false and that their later statements are at least partially true. Of course, it is equally plausible, as defendants contend, that their earlier denials were accurate and their later statements were fabrications designed to appease their abusers. Second, no fair inference of voluntariness can be drawn from any claim that they were following orders, since all three defendants were under the command of Ntabwoba, and it would therefore be reasonable to assume, given their place in the chain of command, that in fact they did follow orders with respect to their activities at Bwindi.
The government also contends that defendants' testimony was not credible because they claimed to be tortured even after they allegedly said what Kibingo wanted them to and their statements "were all over the place." (Id. at 14-15.) Though there were inconsistencies, confusing testimony, and a lack of satisfactory explanations with respect to some of defendants' accounts regarding why they said what they did during the interrogation sessions (see, e.g., 5/25 p.m. tr. at 11-22), their testimony regarding their conditions of confinement and torture was highly believable. Several details regarding their interrogations were also corroborated by documentary or other testimonial evidence. For example, as previously discussed, Karake's testimony that he was beaten with a brick, resulting in an injury to his wrist, is corroborated by the February 5 photographs that depict a bandaged wrist and substantial swelling of his entire forearm from his metacarpophalangeal joints to his elbow. (Gov't Ex. 12, 13 & 13a.) Bimenyimana's testimony that he was given nine questions to answer after his August 30 interview with Bachmann (6/1 p.m. tr. at 52) was basically confirmed by Alfred Ndabarasa. (5/12 p.m. tr. at 55-57.) Perhaps most telling is Nyaminani's testimony that at his March 7 meeting with Bachmann, when he confessed to killing a man and a woman at Bwindi (Gov't Ex. 30, tab 2 at 2), "the interpreter," Isidore Kayumba, "indicated to [him] that they had done investigations," and "that those who had killed the men had been found and the only thing that is left is those who killed the three women." (5/25 a.m. tr. at 65.) Strikingly similar commentary is found in Bachmann's cable to the FBI: the "FBI, RSO and Rwandan police [had] a good idea what happened to the first two murdered groups, however the last group of tourists where Nyaminani was seen going off with [was] still a mystery." (Gov't Ex. 30, tab 2 at 2.) Therefore, Nyaminani "was instructed to think hard about his situation and was taken away into military custody." (Id. at 3.) These sorts of details lend credibility to defendants' testimony. And as extensively discussed supra in Section IV(B)(3)(a), defendants' credibility is further bolstered by the testimony of the medical experts.
Nor is it persuasive to argue that the defendants must be lying because they included details in their statements that would have been unknown to their Rwandan captors. Specifically, the government points to Karake's description of the tattoo on one of the male victims' arms (6/19 a.m. tr. at 24), Nyaminani's reference to red underwear worn by the woman he allegedly raped (id. at 32-33), and Bimenyimana's knowledge that the male and female pair allegedly killed by Karake were not covered with a sheet at the time they were murdered (id. at 24-25), as indicators that they had personal knowledge of and involvement in the murders. First, no clothing was recovered from the crime scene, so the government has no way of knowing whether Susan Miller was in fact wearing red underwear. (Id. at 33.) The government argues that this means Nyaminani could not have been fed this information by Rwandan authorities. (Id.) However, without evidence that this information is true, as opposed to embellishment by either Nyaminani or the person telling him what to say, the government's point carries little weight.
Even Karake's testimony regarding the tattoo on his first victim's body creates substantial doubt as to the source of the information. Kibingo testified that Karake told him that his first victim had a tattoo "of an eagle on his arm." (5/4 a.m. tr. at 21-23.) Likewise, Bachmann testified that Karake described the tattoo as "an eagle." (5/10 a.m. tr. at 55.) But this critical detail was not included in either of the written statements that Karake authored in January. (See Gov't Ex. 10 & 52.) Moreover, as the autopsy photo shows and Dent readily admits, the tattoo was located on Haubner's shoulder, not his arm, and does not even resemble an eagle. (Def. Ex. K-12; 5/19 p.m. tr. at 16.) The incongruity between Karake's description of the tattoo, combined with his failure to include this critical fact in his two contemporaneous written statements (see Gov't Ex. 10 and 52), raises at least a colorable question as to whether Karake saw the tattoo himself or was fed that information. This is particularly true since, after Dent and Bachmann exchange communications regarding Karake's confession (Gov't Ex. 10 and 30, tab 1), the description of the tattoo begins to change so as to more closely resemble the actual tattoo. By the time the FBI arrives in February, Karake describes the tattoo not as an eagle, but as "a bird with large claws." (Gov't Ex. 29, tab 14 at 9.) The tattoo is no longer located just on the victim's arm, but rather closer to his shoulder on the "upper arm." (Id.) Nor is it even clear how Karake was able to see a tattoo that was on the victim's shoulder, in light of the fact that he also described the man as wearing a short-sleeve t-shirt and a vest. (5/19 p.m. tr. at 22.)
Likewise, the government asks the Court to credit Bimenyimana's written statement (which contradicts Karake's February 13 statement) that there was no sheet covering the male and female victims allegedly killed by Karake, since no such sheet was found at the murder scene (Gov't Ex. 29, tab 38 at 4). But it also wants the Court to ignore the portion of that same statement where Bimenyimana claims that the man with the tattoo was part of the male/female pairing. (Id. at 3.) There is no dispute that Robert Hauber, the American man with a tattoo on his shoulder, was killed alongside two other men, not a woman. (5/16 a.m. tr. at 8.) The Court does not find the government's position -- that a witness is credible when his or her testimony helps the government's case, but not when it doesn't -- to be persuasive.
On a more general level, the government's contention that such information could not have been known to the Rwandan investigators, and particularly to Kibingo, is not an accurate reflection of the record. Kibingo's sources of information were extensive. The killings were widely reported on the radio (5/3 p.m. tr. at 64) and in the newspapers, which reported that one or more of the female victims had been raped. (5/16 p.m. tr. at 31; Gov't Ex. 99, tab 37.) Kibingo was aware that a joint investigation was being conducted by the FBI, the Rwandan police and the Rwandan army. (5/4 p.m. tr. at 45.) He maintained close contact with Nziza and Kabanda, both of whom were well-informed regarding the details of the investigation. (5/3 p.m. tr. at 53-54.) Kibingo also learned information about the Bwindi attack from Bemera and the many Bwindi suspects housed at Kami. (5/9 a.m. tr. at 48-49.) Kibingo also often accompanied the defendants and various Rwandan officials on their trips to Kacyiru for interviews, and was in contact with other investigators, including Ndbarassa, Kayumba and Ruzigamanzi. (5/5 p.m. tr. at 57; 5/12 am. tr. at 97-99.) Thus, it is reasonable to infer that anything known to Rwandan authorities regarding Bwindi was also known to Kibingo.
In addition, the Rwandan authorities had multiple sources of information regarding the Bwindi attack. Agent Dent testified that she would not have been surprised if the Ugandan and Rwandan authorities were sharing information regarding the Bwindi attack. (5/19 p.m. tr. at 12.) Importantly, though Dent testified that she did not disclose the autopsy report (id. at 9), the Ugandan post-mortem report notes the existence of a tattoo and its location on the victim's left shoulder. (Def. Ex. K-21.) The FBI also provided information to Ugandan authorities that may have been shared with the Rwandans. (Gov't Ex. 99, tab 6.) The Rwandans also received periodic updates directly from the FBI as part of the joint investigation. (See Def. Ex. K-4; Gov't Ex. 99, tabs 5, 15, 42.)
Another fertile source of information for Rwandan authorities was Isidore Kayumba. The FBI often sent memos to the RSO simply keeping him informed (Gov't Ex. 99, tab 29), or asking him to inquire of Rwandan authorities about the "status of leads." (Id. at tab 11, 38.) Bachmann had access to "files and . . . correspondence" regarding Bwindi that had been maintained by his predecessor. (Id. at 27.) Dent also "enlightened" RSO Bachmann about the case at various points. (5/10 a.m. tr. at 27-28, 75, 83.) Kayumba was the key link between the American and Rwandan investigative teams. The embassy and the Rwandan investigators "called each other all the time back and forth." (5/11 p.m. tr. at 50.) As the U.S. embassy's investigator, it is fair to infer that Kayumba was privy to the information known by RSO Bachmann. He was also present at the majority of the interrogations, including, according to the unrebutted testimony of the defendants, the Rwandans-only interrogation on February 12. Kayumba served as the point of contact for Rwandan authorities when they invited the Americans to interrogate a Bwindi suspect. (5/10 a.m. tr. at 72-73.) He often traveled to Kami to pick up the suspects and transport them to Kacyiru. (5/5 p.m. tr. at 46-47.) During those trips, Kayumba would be accompanied by Rwandan soldiers because, as a civilian, he could not take custody of military prisoners. (5/4 p.m. tr. at 6-7.) One of the people who would accompany Kayumba was Egide Ruzigamanzi, with whom he certainly would have conversed on the 45 to 60 minute drive between Kacyiru and Kami. (5/19 p.m. tr. at 45.) There were occasions when Kayumba spoke directly with Kibingo either in person or on the telephone.*fn104 (5/4 a.m. tr. at 42; 5/5 p.m. tr. at 46; 5/31 a.m. tr. at 27.) Thus, given the multitude of potential sources of information flowing to Kibingo, little can be made of the sparse (and often contradictory) details of the killings that were provided by the defendants.
Nor can the government succeed by arguing the obvious truth that defendants, who face the death penalty, have a greater motivation to lie than the Rwandans. The Court is well aware of defendants' powerful motive to fabricate, and, as was the case with the Rwandan officials who testified, defendants' hatred toward the RPF and those who have fought with the RPA was clearly observable. However, it is not accurate for the government to argue that the Rwandans, and particularly Kibingo, lacked any motive to lie. (5/19 p.m. tr. at 63-64.) First, the RPF and its officials have a compelling and valid reason to want to avenge the Hutu genocide of their fellow Tutsis. They have been engaged in battle with the ex-FAR and ALIR for years, and as Ndbarassa explained, albeit in less vehement terms than did Kibingo, a high-ranking official of ALIR is "the enemy of his government." (5/12 a.m. tr. at 26; see also supra Section IV(B)(4).)*fn105
Second, the record is clear that the Rwandans, for whatever reasons, were anxious to solve the Bwindi killings in an effort to help, and perhaps curry favor with, the U.S. government. (See Attachment 1 to May 11, 2006 Supplemental Order Excluding From Discovery Two Items of Classified Information and Authorizing Revised Substitutions.) The focus of the Rwandans' investigation was on the American victims from the outset (Def. Ex. K-4), and, despite their extensive investigation, no other suspect has been prosecuted in Rwanda for the crimes committed at Bwindi. (5/11 a.m. tr. at 5.) As early as Ndabarasa's Ruhengeri interviews in the summer of 2001, he focused specifically on the American victims, rather than those from England, Uganda, and New Zealand. (Gov't Ex. 43.) According to Agent Dent, the Rwandans moved incredibly quickly to acquire custody of anyone whom U.S. investigators were interested in interviewing. (5/16 p.m. tr. at 6, 57-58.) The Rwandans' efforts to facilitate the investigation was consequently recognized at the highest levels of the United States government. (See, e.g., Def. Mot., Attachment B at 1.) But most importantly, it is clear that Kibingo saw his role as eliciting confessions to provide to his superiors, for use by the Americans.*fn106 These facts demonstrate that the Rwandans had a singular focus on solving the murders of the two Americans so as to accommodate the U.S. officials, which further undermines the government's contention that there was "no motive" for the Rwandans to lie. (6/19 p.m. tr. at 64.)
While the Court has thoroughly explained its reasons for rejecting Kibingo's testimony and for its finding that defendants' statements were the product of coercions, there is one further observation that convinces the Court of the correctness of its credibility determinations and its ultimate finding of involuntariness. Over the course of the defendants' detention at Kami, an unmistakable pattern continually repeated itself. Each defendant initially denied his involvement at Bwindi, but was then held incommunicado at Kami until Kibingo extracted statements that he believed were wanted either by his superiors or the Americans and were needed in order to solve the murders, close the investigation, or support a prosecution. With the exception of Nyaminani, the Rwandans did not even inform the U.S. investigators that they had a suspect in custody until they were informed that a defendant had confessed, and each initial confession was obtained only by Kibingo while no one else was present. He would then deliver the information to his superiors who, in turn, several days later, would contact Kayumba, at which point Bachmann would conduct an interrogation. If the defendant provided less information than was expected, as happened with respect to each defendant, he was returned to Kami and subjected to further interrogation and mistreatment. Only after Kibingo announced to the Rwandans that he had obtained further information, was the defendant returned to the Americans for more questioning.*fn107
While it is unclear whether the specific information that Kibingo sought from the defendants correctly reflected what the Americans needed or, alternatively, what the Rwandans believed was necessary to solve the murders, it is clear that they were relentless in their efforts to obtain this information. For instance, even though Karake confessed to the murder of Richard Haubner on February 5 and 6, the Rwandans (unlike the Americans) were not satisfied with the extent of his confession, so Karake was returned to Kami for several days and was interrogated by only the Rwandans on February 12 until he confessed to two additional killings.
With Nyaminani the pattern was even more pronounced. When Nyaminani admitted to killing a male and female tourist on March 7, all interrogators doubted his story, since they already had Karake's confession to these same murders, but they still had not solved the "mystery" regarding the three women. (Gov't Ex. 30, tab 2 at 3.) He was therefore sent back to Kami (see supra note 107) where over the course of March he executed two statements -- one on March 8, in which he admitted to raping Susan Miller (Gov't Ex. 53 & 53a), and one on March 20, in which he would only admit that he supervised the perpetrators of the rapes and ordered the women to be killed. (Gov't Ex. 54 & 54a.) Despite these conflicting stories from Nyaminani, two months later Bachmann, who was not informed of the two conflicting March statements, was told that Nyaminani had confessed to "assisting in the" rape and murder of three female tourists. (Gov't Ex. 30, tab 3, at 1.) However, by the time of the interview on May 22, Nyaminani denied his participation in the rapes and stuck by his March 20 statement, which Bachmann only learned about from Kabanda after the interview was well underway. (Id.) Again dissatisfied, the interrogators sent Nyaminani back to Kami, and it is not until he had been there several weeks and underwent two more grueling days of interrogation on June 1 and 2, totaling almost eighteen hours, that he confessed to the rape. Even then, his confession does not come until the afternoon after the Americans told him in no uncertain terms that they did not believe him and were "tired of documenting lies."*fn108 (Gov't Ex. 29, tab 27 at 11.)
The pattern with respect to Bimenyimana, while slightly different, shares some significant similarities. By the time of his capture, the supposed killers of the Americans had been identified and a different type of information was therefore either sought or needed from him. On August 30, he admitted that the attack was conceived by Ntabwoba to capture white tourists, that he gave several orders as the attack began, and that a French woman helped Ntabwoba to identify the Americans and the British. But he denied knowing anything about any murders or rapes until after the fact. (Gov't Ex. 30, tab 4.) Bimenyimana is then told that the interrogators do not believe him and he is returned to Kami to answer written questions that were provided to him by the interrogators. Thereafter, although Bachmann gets word from the Rwandan police that "he's got more to tell you" (5/10 p.m. tr. at 33), the September 2 interview did not produce anything new. (5/11 p.m. tr. at 63.) Following another six months of detention at Kami, he is subjected to a three-day ordeal on February 21-23 with American and Rwandan interrogators that totaled almost twenty hours of interrogation. During the first day, he defined one of ALIR's objectives as capturing Americans, and then on February 22, he further explained that the purpose for taking Americans and British as hostages was to give notice to governments that supported the RPF so that they could understand ALIR's cause. (Gov't Ex. 29, tab 38 at 5-8.) Nonetheless, at the end of the first day, he is told that the investigators do not believe that he is telling the truth. He then responded that "he was present for the killing of a man and woman together but that he would provide the details after he had some time to think." (Gov't Ex. 29, tab 37 at 12.) He is sent back to Kami for the night "to think," and the following day, on February 22, he provided the details of Karake's killing of a man and woman with an axe. (Gov't Ex. 29, tab 38 at 2-4.) While the details do not coincide with Karake's February 13 statement, since Bimenyimana said the couple was not covered with a sheet and the man had a tattoo, it appears that obtaining this admission from Bimenyimana was thought to be highly useful.*fn109
This striking similarity in the course of events, relating to each defendant's statements, just like the scarring on their bodies, cannot be chalked up to mere coincidence. Rather, the inescapable conclusion is that defendants' statements to the Rwandans were the product of coercion.*fn110 Indeed, based on the totality of circumstances, the Court finds that the conditions under which defendants were held at Kami and the abuse and mistreatment they endured while being interrogated shock the conscience and therefore render the statements involuntary and inadmissible.*fn111
V. ADMISSIBILITY OF DEFENDANTS' STATEMENTS TO THE AMERICANS
A. The Law Regarding Taint
Having found that the statements to the Rwandans must be suppressed on the grounds that they were the product of coercion, the Court must proceed to consider the effect of that finding on the admissibility of any statements made by the defendants to the American interrogators during the same time period. It is well-established that a "confession [obtained] under circumstances which preclude its use," does not "perpetually disable the confessor from making a usable one after those conditions have been removed." United States v. Bayer, 331 U.S. 532, 541 (1947); see also Holland v. McGinnis, 963 F.2d 1044, 1050 (7th Cir. 1992) (beating a confession out of suspect "'does not permanently disable him from giving a voluntary statement' at a later time" (quoting Wilson v. O'Leary, 895 F.2d 378, 385 (7th Cir. 1990)). Even in extreme cases, such as Lyons v. Oklahoma, 322 U.S. 596 (1944), "in which police forced a full confession from the accused through unconscionable methods of interrogation, the Court assumed that the coercive effect of the confession could, with time, be dissipated." Elstad, 470 U.S. at 311-12. "The Fourteenth Amendment does not protect one who has admitted his guilt because of forbidden inducements against the use at trial of his subsequent confessions under all possible circumstances. The admissibility of the later confession depends upon the same test -- is it voluntary." Lyons, 322 U.S. at 603 (emphasis added); see also United States v. Daniel, 932 F.2d 517, 519 (6th Cir. 1991) (Supreme Court looks to "voluntariness" of confession "following prior involuntary statements"). Thus, the ultimate inquiry into voluntariness is identical to that used for any confession: whether in making his statement, defendant's "will has been overborne and his capacity for self-determination critically impaired." Schneckloth, 412 U.S. at 225. This is a totality of the circumstances test. Id. at 226; see also Fulminante, 499 U.S. at 285-86; Elstad, 470 U.S. at 318 ("As in any [voluntariness] inquiry, the finder of fact must examine the surrounding circumstances and the entire course of police conduct with respect to the suspect" when considering the admissibility of a confession made after a prior inadmissible one.).
The Supreme Court has identified several factors that should guide a court's inquiry into whether a confession obtained in the wake of a coerced statement was sufficiently insulated from the prior constitutional violation as to be considered voluntary. These include: "the time that passes between confessions, the change in place of interrogations, and the change in identity of the interrogators." Id. at 310; see also Daniel, 932 F.2d at 519 ("[S]uch factors as intervening time, removal of the prisoner to a different place, and change in the identity of interrogators could make a second, warned statement voluntary, despite the prior involuntary statement."). Moreover, the court may take into consideration the continuing effect of the prior coercive techniques on the voluntariness of any subsequent confession. See Lyons, 322 U.S. at 603
("[T]he fact that the earlier statement was obtained from the prisoner by coercion is to be considered in appraising the character of the later confession. The effect of the earlier abuse may be so clear as to forbid any other inference than that it dominated the mind of the accused to such an extent that the later confession is involuntary."); see also Lisenba v. California, 314 U.S. 219, 240 (1941) ("The question of whether those confessions subsequently given are themselves voluntary depends on the inferences as to the continuing effect of the coercive practices which may fairly be drawn from the surrounding circumstances."). Whether effective Miranda warnings preceded the subsequent statements is also a relevant inquiry. Westover v. United States, decided together with Miranda, 384 U.S. at 496-97; see also Daniel, 932 F.2d at 519; Robinson v. Percy, 738 F.2d 214, 221 (7th Cir. 1984). The purpose of these factors is to ensure that there exists a "break in the stream of events . . . sufficient to insulate the statement from the effect of all that went before." Clewis v. State of Texas, 386 U.S. 707, 710 (1967).
B. Application of the Attenuation Factors
The insurmountable problem the government has in proving that, between the confessions obtained at Kami and those at Kacyiru, there was a "clean break" sufficient to dissipate the previous coercion, Daniel, 932 F.2d at 521, rests in the extensive involvement of Rwandan officials throughout the many months of interrogations. Though the Americans conducted their interrogations away from Kami, defendants remained in Rwandan custody at all times and were returned to Kami at the conclusion of each interrogation by the Americans. In addition, the Rwandans conducted interrogations of the Bwindi suspects at Kacyiru on at least one (5/25 a.m. tr. at 14) and possibly other occasions (5/25 p.m. tr. at 25; 5/25 a.m. tr. at 13) out of the presence of the Americans. What defendants said at Kacyiru was reported back to Kibingo, and some of their abuse at Kami is directly attributable to what they told (or failed to tell) the Americans at Kacyiru. Thus, at no time could the defendants have a reasonable expectation that the change in the location of the interrogation would result in an improvement in their living conditions.
Nor was there a change in the identity of the interrogators sufficient to alleviate the coercion defendants were experiencing at Kami. Four or more Rwandan officials were present at most, if not all, interrogations conducted by the Americans. (See Gov't Ex. 27.) Some of these officials were responsible for transporting defendants back to Kami at the conclusion of the interview. (5/5 p.m. tr. at 46-47; 5/19 p.m. tr. at 45.) These officials were active participants in the interrogation process, asking questions, translating and, according to defendants, sometimes suggesting answers. (5/22 p.m. tr. at 51 (Rwandans would "remind" Karake to include things if he forgot them); 5/23 p.m. tr. at 40 (Karake reminded about the bird tattoo).) Moreover, their primary tormentor, Kibingo, often accompanied them to Kacyiru, participated in the interrogations, and returned with them to Kami in the evening. Kibingo's boss at DMI was Jean-Bosco Kabanda, who reported directly to Jack Nziza, the head of DMI (Gov't Ex. 65; 5/5 p.m. tr. at 48), and was a frequent participant at the interrogations conducted by both the FBI and RSO Bachmann. (Gov't Ex. 27.) The Rwandan officials, who were part and parcel of the government apparatus responsible for defendants' mistreatment, were present at all times while the Americans conducted their interviews. (5/19 a.m. tr. at 33-34.) As Bimenyimana testified, he did not tell RSO Bachmann at their first interview that he was being mistreated at Kami because the interpreters were themselves involved in the mistreatment. (5/31 a.m. tr. at 80.) Therefore, from defendants' perspective, the identity of the interrogators had not changed in any significant way.
Though the length of time between the torture at Kami and the subsequent interview by an American official often varied, there was no corresponding change in defendants' circumstances that would cure the coercive nature of their confinement. Indeed, to the extent that defendants' conditions improved in the days prior to an interview with the FBI, it is clear that those changes were made to improve defendants' appearance for their meeting with the Americans. (5/22 p.m. tr. at 39-41.) Moreover, defendants testified to specific instances of coercion immediately prior to their meetings with Americans. For instance, Kibingo reviewed Karake's prior statements with him before his interview with Bachmann in January, and instructed him to "memorize them." (5/22 p.m. tr. at 41-44.) Karake also claims that Kibingo beat him with a brick prior to his February 5 interview with the FBI, causing visible injury to his wrist, because Karake "wasn't memorizing what" Kibingo was "telling [him] to say" to the "whites." (5/22 p.m. tr. at 51-52.) Prior to a February meeting with the FBI, Nyaminani was forced to dig a hole in the ground that he believed to be his own grave. (5/25 p.m. tr. at 65.) Moreover, Nyaminani was questioned by the Rwandan investigators prior to each of his interviews with American officials. (5/24 p.m. tr. at 25.) Bimenyimana likewise testified that he was prepped by Kibingo before his first meeting with Bachmann.*fn112 (5/31 a.m. tr. at 25-27). He also testified that he was beaten in between his meetings with Bachmann and the Rwandans on August 30 and September 2, 2002. (5/31 a.m. tr. at 72.)
The critical question with respect to attenuation is not the length of time between a previously coerced confession and the present confession, it is the length of time between the removal of the coercive circumstances and the present confession. Lyons, 322 U.S. at 597 (question is whether "the unlawful inducements which vitiated the prior confession" were alleviated prior to the second). Defendants were housed in unconscionable conditions for months on end and subject to repeated beatings which were linked specifically to their acceptance of responsibility for the crimes committed at Bwindi. Where, as here, the coercion was a product of both discrete beatings, as well as the general conditions of confinement, it is impossible for the Court to conclude that there was any meaningful relief from those conditions prior to the interrogations by American investigators. Indeed, the Court need not determine whether the Rwandans' "coercive practices" had a "continuing" or residual effect that rendered defendants' subsequent statements involuntary, Lisenba, 314 U.S. at 240, because those practices were ongoing throughout the Americans' interrogations through the continuing presence of a cast of Rwandan officials, which reasonably caused defendants to fear for their safety. (5/22 p.m. tr. at 51-52; 5/25 a.m. tr. at 68-73.)
1. Effectiveness of the Advice of Rights
The government also claims that "the subsequent administration of Miranda warnings . . . attenuates any taint of the failure to administer Miranda warnings" at previous interrogations. (6/19 p.m. tr. at 53.) It is important to note, first, that the government's argument assumes that there was no actual coercion at the prior Rwandan interrogations, but merely a failure to administer warnings. (Id.) Therefore, while it is certainly correct that "a careful and thorough administration of Miranda warnings serves to cure" the presumption of coercion that requires suppression of "an unwarned but clearly voluntary admission," Elstad, 470 U.S. at 310, not even the government contends that Miranda warnings alone can dissipate the taint of actual coercion. In the context of actual coercion, such as the unconscionable conditions and abuse found in this case, whether the defendant "knowingly and intelligently waived his right to remain silent and his right to consult with counsel prior to the time he made the statement," Westover, 384 U.S. at 494, is a factor to be considered, alongside the other factors listed in Elstad. 470 U.S. at 310. (See also Gov't Legal Mem. at 10.)
It remains the government's burden to prove by a preponderance of the evidence that defendant waived his rights under the Fifth Amendment. Connelly, 479 U.S. at 168. In order for a waiver of a suspect's Miranda rights to be effective, it must be "voluntary," i.e., not the product of "governmental coercion," id. at 169, as well as "knowing and intelligent," i.e., "made with a full awareness of both the nature of the right being abandoned and the consequences of the decision to abandon it." Moran v. Burbine, 475 U.S. 412, 421 (1986). These are separate inquiries: "Only if the 'totality of the circumstances surrounding the interrogation' reveal both an uncoerced choice and the requisite level of comprehension may a court properly conclude that the Miranda rights have been waived." Id. (quoting Fare v. Michael C., 442 U.S. 707, 725 (1979)).
The Court has already detailed the coercive circumstances that infected defendants' interviews with American investigators. It is clear that, under such conditions, any waiver of Miranda rights could not be voluntary. Defendants had been instructed by Kibingo prior to the interrogation to repeat what they had said to (or been told by) him regarding their role at Bwindi. (5/31 p.m. tr. at 17, 30-32 (Bimenyimana); 5/22 p.m. tr. at 41-44 (Karake); 5/26 p.m. tr. at 26 (Nyaminani).) At most interviews, defendants did not tell the Americans they were being beaten out of fear of what Kibingo might do to them. (5/25 p.m. tr. at 33 (Nyaminani); 5/31 a.m. tr. at 80 & 5/31 p.m. tr. at 34 (Bimenyimana); 5/22 p.m. tr. at 50 (Karake).) The evidence indicates that defendants were in no position to refuse to speak with the American investigators, regardless of any arguable comprehension of their Fifth Amendment rights. Bimenyimana in fact testified that, while he understood his rights, he signed the waiver purely out of fear of what Kibingo would do if he refused to talk. (5/31 p.m. tr. at 27.) Thus, the Court cannot conclude that the government has met its burden to prove the voluntariness of defendants' waiver of their Miranda rights.
The Court need not rely on voluntariness alone, however, because there are substantial reasons to doubt Karake's and Nyaminani's comprehension of their Fifth Amendment rights, especially given their limited education and rural upbringing. Moreover, as to some of the interviews, it is not even clear that warnings were given, or if they were given, whether the language of the warning was correct. The government introduced no evidence that Bimenyimana was advised of his rights during his September 2 meeting with Bachmann. In another instance, at his March 7 interview of Nyaminani, Bachman failed to have the witness sign a rights waiver. (5/10 a.m. tr. at 85.) Bachmann testified that he orally advised Nyaminani of his rights, either by using a wallet card that contained the standard domestic Miranda warnings (Def. Ex. N-1) or by reading from an English-language foreign Miranda form. (Id. at 86.) Bachmann could not remember which of those methods he used, but if he in fact had an advice of rights form with him, he certainly would have had Nyaminani sign it. (5/11 a.m. tr. at 70-71.) And, as Bachmann testified, he carried his Miranda wallet card "with [him] all the time." (Id. at 71.) The only reasonable inference is that Bachmann used his wallet card. (Id. at 74-76.) The problem, of course, is that the language used to issue Miranda warnings in the United States is not appropriate for overseas interrogations. See Bin Laden, 132 F. Supp. 2d at 187-89. The domestic warnings promise rights that the United States government cannot provide in Rwanda, such as the provision of legal counsel at government expense. (Def. Ex. N-1; 5/11 a.m. tr. at 74.) Obviously one cannot knowingly and intelligently waive rights if one has not been accurately informed as to what those rights are.
Similarly, Dent admitted that on February 6 she had no rights form available for Karake to sign but that he was orally advised of his rights (5/16 a.m. tr. at 11), and that on December 1, 2001, she had a rights form, but since she had no clean copy, Nyaminani never executed the form. (5/15 p.m. tr. at 65.) These lapses are hardly insignificant, especially with respect to February 6, since it is impossible to believe that the complicated and lengthy set of rights that were supposed to be communicated would have been accurately remembered or properly repeated. (See supra note 14.) Moreover, on December 5, 2001, agents questioned Nyaminani extensively about his role in the Bwindi attack before eventually reading him his rights. (5/24 p.m. tr. at 60; Gov't Ex. 29, tab 8 at 1-2 (two typed pages of substantive notes before Nyaminani was informed of his rights).) On February 6 in their second interview of Karake, the FBI informed him "that the rights he was advised of [the previous day] still applied and that his cooperation and willingness to speak with the FBI was voluntary," rather than reading him his rights in full. (Gov't Ex. 29, tab 15 at 1.) This came after spending only four minutes going over Karake's rights with him during his prior initial interview on February 5. (Id., tab 14 at 1.) Karake did not ask a single question of the agents present regarding the advice of rights form that was read to him and he signed on February 5 (5/19 a.m. tr. at 28; Gov't Ex. 11), nor did Agent Dent ask any follow-up questions to ensure that Karake had some minimal understanding of his rights. (5/19 a.m. tr. at 31.) For his part, Karake testified that no one informed him of what an American court is or what function a lawyer might perform for him.*fn113 (5/22 p.m. tr. at 53-54.) Moreover, Agent Dent permitted Bimenyimana to endure an entire day of questioning on February 21, 2003, despite harboring "some doubt" as to whether he had understood the rights waiver that he had executed. (5/19 p.m. tr. at 35.)
Nor did the FBI take even the basic step of providing an advice of rights form that was written in Kinyarwanda to assist the defendants in their understanding of the rights that they were being asked to waive. (5/19 a.m. tr. at 22, 24.) Agent Dent never even discussed the possibility with RSO Bachmann. (Id.) Indeed, despite the fact that each form signed by the defendants states that he "read" the form, neither Karake nor Nyaminani could read the document, whether it was in English or French. (5/24 p.m. tr. at 28.)
The efficacy of the Miranda rights is further undercut by the serious questions regarding the adequacy of the translations during these interviews. Nyaminani alleges that during his December 5 interview with the FBI, the Rwandan interpreter told him that he had "the right to talk to these people" because he was "not accused." (5/24 p.m. tr. at 29.) He was further told that he didn't "need a lawyer because [he was] not in the courts." (Id.) When Nyaminani returned for a subsequent interview in February, he claims he was similarly misinformed by a Rwandan that "this paper is about my rights, and that I should talk because I'm not accused, and you don't need a lawyer because you're not in court." (5/24 p.m. tr. at 61.) He was told it was the "usual thing," i.e., the same thing he was read in December. (Id.) Similarly, Karake testified that he was not asked whether he wanted a lawyer at his January 12, 2002 interview with Bachmann. (5/22 p.m. tr. at 48.) Tellingly, when asked what Karake said in response to being asked whether he needed a lawyer, Ndabarasa testified: "in Rwanda . . . it's not so easy to get lawyers, so they suggested that they would continue with the investigation without a lawyer." (5/12 p.m. tr. at 30 (emphasis added).) When the Court redirected Ndabarasa's focus to the Karake's response, he stated: "we would ask the witness, do you think you can go on with the questioning without a lawyer? If he said yes, then it would go on." (Id.)
The government also appears to misunderstand the relevance of the translation issue. It is not, as the government contends, just a question of competence. (6/19 a.m. tr. at 39-41.) Rather, it is a question of "whether the translated statements fairly should be considered the statements of the speaker." United States v. Nazemian, 948 F.2d 522, 527 (9th Cir. 1991). This requires consideration of several factors, including: "which party supplied the interpreter, whether the interpreter had any motive to mislead or distort, the interpreter's qualifications and language skill, and whether actions taken subsequent to the conversation were consistent with the statements as translated." Id. In this case, in all instances the interpreter was a Rwandan provided by one of the two governments conducting the investigation. Moreover, the interpreter was not only supplied by the investigating authority, he was also a member of the investigative team, creating bias issues that could undermine the integrity of their translations. John Gacinya, Alfred Ndabarasa, Eric Kayiranga and Egide Ruzigamanzi, all of whom provided assistance with translation at one time or another, were also active investigators and interrogators. (5/2 a.m. tr. at 80.) Even Dent acknowledged the possibility that Gacinya had a conflict acting both as an investigator and as a translator. (5/16 p.m. tr. at 25.) Though Ndabarasa testified that he read the Miranda form "as it was written" (5/12 p.m. tr. at 26), he also expressed frustration at the "tedious" and "methodical" manner in which American investigators conducted their interrogations. (5/2 a.m. tr. at 99-100.) Even Isidore Kayumba, the State Department employee who served as the primary interpreter for the majority of the interrogations, is not free from bias, given his role as an active investigator. (See supra Section IV(B)(6).) Notably, Kayumba, who is an employee of the United States government and an essential participant in the Rwandan/American investigation, was not called to testify, preventing the Court from assessing his credibility. Therefore, while the Court has no reason to doubt the proficiency of the individuals who served as interpreters throughout the investigation, their motives are certainly suspect.
Additionally, evidence in the record calls into question the accuracy or completeness of the interpretations given to U.S. officials with respect to the advice of rights, as well as other matters. Specifically, on February 23, 2003, U.S. officials were told that Bimenyimana "asked if he could have an attorney in the United States." (5/19 p.m. tr. at 41.) Agent Dent testified that she inquired whether "he was asking for an attorney in order to proceed with the interview," and then determined that he had "waived his right to an attorney." (Id. at 41-42.) Agent Dent's account raises serious questions as to either the accuracy of the translation or Bimenyimana's comprehension of his Fifth Amendment rights. Dent had not told Bimenyimana that he would be going to the United States for prosecution, therefore Bimenyimana should have had no reason to expect that he would need an attorney "in the United States." (Id.) Moreover, this came on the heels of an interview two days earlier about which Agent Dent expressed "some doubt" as to Bimenyimana's comprehension of his rights. (5/19 p.m. tr. at 35.) In either case, the Court is not convinced that Bimenyimana effectively waived his rights.
There is other testimony raising doubts as to the reliability of the translations. Bachmann's account of what he was told Karake said on January 12 is not corroborated in Karake's handwritten statement from that day. Bachmann testified that Karake described his victim as "a white male, rather large and a tattoo of an eagle." (5/10 a.m. at 55.) This was, in Bachmann's mind, a "significant statement." (Id. at 54.) Yet, Karake's written statement is completely devoid of any details regarding the physical appearance of his victim. (Gov't Ex. 10.) And when Bachmann had Kayumba review the statement to see "if there was anything different than what [they] just heard," Kayumba reported no discrepancies. (Id. at 69-70.) Moreover, the absence of a neutral interpreter often resulted in conversations taking place in Kinyarwanda that were not translated for the benefit of the American interrogators. Bachman conceded that not every question that was asked by a Rwandan was translated into English. (5/11 p.m. tr. at 41, 44.) There were times that he had to stop the Rwandans from speaking to each other in order to say, "what's going on here? What are you talking about?" (Id. at 44.) Even Kayumba, Bachmann's right-hand man, "[did] not translate every word." (Id. at 88.) As a result, "if a question was being asked in Kinyarwanda," Bachmann "would not have understood." (Id. at 41.) Dent also admitted that she didn't actually know what was said in Kinyarwanda to Karake regarding his rights. (5/19 a.m. tr. at 22.) Under these conditions, no one can say whether the Rwandan officials properly advised the defendants of their rights, or if they suggested answers to questions in a manner undetectable by the Americans.
The Court therefore finds that the government has not met its burden to demonstrate a "knowing and intelligent" waiver by defendants of their Miranda rights with respect to any of the interrogations conducted by the Americans. Moreover, as previously discussed, the Court also finds that the waivers were involuntary with respect to each interrogation conducted after December 5, by which point the conditions of confinement and systemic abuse at Kami Camp rendered defendants incapable of voluntarily waiving their rights.*fn114 While the Court is cognizant of U.S. law enforcement's need to be able to conduct investigations overseas with the assistance of foreign governments, the government here has failed to prove the validity of the translated advice of rights or defendants' comprehension of their rights. Nor did investigators employ techniques to ensure the comprehension of defendants, such as providing rights forms in Kinyarwanda, or to provide proof of what was said by all sides during the interrogations by use of an audio- or videotape.*fn115 Even though such measures might not be necessary under all circumstances, this is not the typical case. If the government expected to rely on an argument that defendants waived their Miranda rights, it should have recognized that the unusual circumstances of this investigation called for better safeguards.
The Court is painfully aware that two innocent American tourists were brutally killed at Bwindi on March 1, 1999. But that sentiment may not, under the law, dictate the result here. It is the government's burden to prove that defendants' multiple statements were each "the product of an essentially free and unconstrained choice." Schneckloth, 412 U.S. at 225. The government cannot, however, meet its burden where defendants' statements were extracted only after countless hours of repetitive questioning over a period of many months, during which time they were subjected to periods of solitary confinement, positional torture, and repeated physical abuse. It is a "fundamental . . . concept" of our constitutional system of criminal law that "neither the body nor the mind of an accused may be twisted until he breaks." Culombe v. Connecticut, 367 U.S. 568, 582 (1961). Because due process of law demands that guilt cannot be established through evidence secured by coercion, defendants' motion is GRANTED.