INTELLIGENCE AT UN HEADQUARTERS?

The Information and Research Unit and the Intervention in Eastern Zaire 1996

Fortunately, the terms ‘intelligence’ and ‘intelligence-gathering’, once banned from the UN lexicon, are becoming acceptable, if not fashionable, in the organization. They are now often, if not formally, used in the UN peacekeeping operations and field offices, in the sense of processed or analysed information (both open and secret) relating to security, something clearly separate from cloak and dagger activities (‘covert action’) practised by some national intelligence agencies. The advertised ‘Intelligence Analyst’ position was in the Investigations Section of the Office of the Prosecutor of the International Criminal Tribunal for the Former Yugoslavia. The job description was to conduct ‘in-depth research and analysis regarding criminal investigations of the conflict with information obtained from multiple sources related to the activities of persons under investigation’ which required ‘specialized areas of analysis (military, police or federal/national intelligence analysis agencies)’. The UN has even created posts for ‘intelligence analysts’ at the International Criminal Tribunal for the Former Yugoslavia.²

UN intelligence has also grown rapidly as a scholarly field in recent years, though almost all of the studies are on peacekeeping intelligence (PKI). Since the publication of the first case study of intelligence in a UN peacekeeping operation by Dorn and Bell,³ there have been detailed examinations of PKI in Bosnia, Haiti, Lebanon and several other conflict areas.⁴ Despite the proliferation of papers, there remains a lacuna in the study of PKI: the literature still does not deal with the vital issue of intelligence at UN headquarters.

Indeed, the term ‘UN headquarters intelligence’ is viewed by UN field personnel as yet another oxymoron for yet another reason. The headquarters is often seen as a black hole into which information from the field regularly disappears, with little or no information returned or feedback given. It is also seen as a place where very little analysis is done to convert information into intelligence, certainly nothing comparable to the work of national intelligence agencies. A long-standing sense of these deficiencies led several recent Secretaries-General to attempt to do something about it. This paper will feature these efforts and describe the most advanced intelligence body yet created at UN headquarters: theInformation and Research (I&R) Unit that was part of the Situation Centre (SitCen) of the Department of Peacekeeping Operations (DPKO) from 1993 to 1999. The story begins earlier, however, with the establishment of a precursor of sorts, the Office for Research and the Collection of Information (ORCI).

AN EARLY EFFORT: ORCI

The pool of information available to the Secretary-General is wholly inadequate.
— Javier Pérez de Cuéllar⁵

At the end of his second term as UN Secretary-General in 1991, Pérez de Cuéllar remained frustrated by the lack of information that had inhibited him from being more proactive and from practising early warning and preventive diplomacy. He had tried unsuccessfully to solve this problem by creating in 1987 the Office for Research and the Collection of Information. In the cost-cutting environment of the day, ORCI’s establishment was justified as an effort to ‘streamline the Secretariat and cut duplication in functions relating to political information and analysis’.⁶ But its mandate was far more ambitious. It was to assess global trends, prepare ‘profiles’ of various countries, regions and conflicts, and give early warning of emerging ‘situations’ as well as monitor refugee flows and emergencies.

There was an immediate backlash from a group of conservative US politicians, including Senators Bob Dole and William Roth. They cosponsored a bill in 1987 to withhold American funding for ORCI. In a letter to the UN Secretary-General they charged that the new office could ‘provide a cover’ for Soviet espionage in the United States and that it would ‘gather information on the internal political situation of member states—a definite United Nations intrusion’ into domestic affairs.⁷ They did not allege that the office might challenge the views of US intelligence agencies, though this was probably on their minds. The US State Department managed, however, to convince the Senators that the office was not a spy base but a useful development. ORCI replaced the Political Information News Service that, as part of the Department of Political and Security Council Affairs, had been managed by a Soviet national. The State Department officials also argued that the new UN would save money since it was amalgamating several other UN sections. At its establishment and during its tenure, ORCI was placed under the control of a long-standing international civil servant, James Jonah of Sierra Leone.⁸ The organizational diagram of ORCI is given in Figure 1.

Even though the US Senators backed down on their bill, the office continued to face resistance. ORCI was branded undesirable by other governments, which also feared UN intrusion into sovereign affairs. In addition, ORCI suffered from a number of internal problems. With an initial staff of 20 members (less than half of which were in the professional category), at a time of a recruitment and funding freeze, and with little automation, it was understaffed, under-equipped and unable to hire new staff from outside the UN system. Jonah made light of all this by saying it was ‘better to be overworked than overstaffed’.⁹ The office was equipped with only primitive teletype machines to receive news from wire services. The UN only caught up with the computer/information age many years later. The lack of technical and human resources, combined with an incessant demand for speech writing from the Secretary-General and senior UN officials, meant that ORCI could not devote the time and effort needed for deeper analysis of pressing international issues. Though it had an ‘Early Warning Service’ it did not issue significant early warnings.¹⁰

Shortly after Boutros Boutros-Ghali arrived as Secretary-General in 1992 he created the Department of Peacekeeping Operations (DPKO) to manage the burgeoning practice of UN peacekeeping that was quickly becoming a centrepiece of the organization’s response to the many post-Cold War conflicts.¹¹ ORCI was disbanded and a new organization within DPKO was established.

THE SITUATION CENTRE AND THE I&R UNIT

To manage the large set of new multidimensional field operations, the UN had to find middle ground between the concept of a headquarters ‘Operations Centre’ (which was not possible, given the meagre resources and the paucity of political will of member states) and a ‘Situation Room’ (which was found to be inadequate in scope). Hence a ‘Situation Centre’ was created within DPKO in April 1993.¹² This was to be more than the ‘cable room’ which had existed in the Office for Special Political Affairs, but not the nerve centre for command and control commonly found in national defence establishments. The ‘SitCen’ included a 24/7 Duty Room where knowledgeable officers could promptly refer peacekeepers to appropriate headquarters officials. This was done in part to respond to the blistering criticisms of peacekeepers like Major-General Lewis Mackenzie who accused headquarters staff of being unreachable – ‘at cocktail parties’, he said disparagingly – even as life-threatening crises were unfolding in places like Sarajevo, where Mackenzie has been the sector commander. The Duty Room, had three or four officers on duty 24 hours a day, dividing the world into three or four regions, as is commonly done in military headquarters. Each duty officer was responsible for communications with four to seven peacekeeping missions, including receiving daily situation reports (Sitreps) from them. The initial SitCen goal of ‘uninterrupted communications with all UN missions around the globe’ was achieved for all PKOs. At the height of PKO deployments in the mid-1990s, it was possible to boast that ‘the sun never sets on UN peacekeepers’. And in many places, the situation was tense, keeping headquarters busy.

Information from UN peacekeepers in the field was clearly not sufficient, however, if the UN wanted to engage in early warning and preventive diplomacy as Boutros-Ghali had advocated in 1992 in his landmark Agenda for Peace. To uncover the deeper forces underlying conflicts, DPKO needed a way to tap into the vast information networks of national governments with their numerous embassies and sophisticated intelligence agencies. But the major powers did not want to send information to the Secretariat without having someone linked to them (‘their man’) inside the UN to ‘handle’ and carefully disseminate sensitive information. The result was the creation within the SitCen in September 1993 of an Information and Research (I&R) Unit, composed of officers on secondment from the governments of four of the five permanent members of the Security Council: France, UK, Russia and USA.¹³ China declined to send an officer; apparently, it was unwilling to share information, a prerequisite for membership. Besides, it was not contributing substantially to peacekeeping and other members felt it could not be counted on to provide useful or unbiased information (which could be asked about the others as well!). The four I&R Unit staff were gratis officers, whose salaries were paid by their home governments but who still were subject to UN direction and, in theory, to ‘UN discipline’ (another oxymoron to some, especially concerning officers on secondment). These individuals maintained substantial links to the intelligence services of their home countries, most having come from these agencies. They were ‘the interface’ with these intelligence services. In return for the loan of these officers and the information they provided, the nations sought the UN’s coded cables (situation reports) from the field, some of which may have made their way back to national capitals, a prospect that displeased some UN Secretariat officials.

An international civil servant, Canadian Stan Carlson, headed the Situation Centre but the officers in the I&R Unit did not consider themselves responsible to the SitCen chief or under his authority. (It was under his initials, however, that many of the Eastern Zaire reports in 1996–97 were sent to the UN’s senior officials.) The officers were loosely accountable to both the UN (from which they could be expelled if found to be engaging in unsuitable activities) and to their home governments (to which most expected to return after their term ended). Perhaps most importantly, they kept a close eye on each other, making sure that national bias did not enter into their common reports. The Unit was nominally headed by a US Air Force intelligence officer for most of its existence but she acted more in the role of coordinator than boss. The organizational diagram for the SitCen in 1996, including its loosely connected I&R Unit, is shown in Figure 2.

In the first few years of DPKO and its SitCen, the US sought to build up peacekeeping capacity, identifying intelligence as a key deficiency. It provided substantial assistance for the development of the I&R Unit, though many in the UN Secretariat were uneasy about relying so heavily on US intelligence agencies. The US Defence Information Systems Agency (DISA) performed an ‘audit’ in 1994 and recommended that a relatively new US system, the Joint Deployable Intelligence Support System (JDISS), be provided to the Unit. This system had previously been installed in the UN compound in Mogadishu, Somalia, in 1992–93 during Operation Restore Hope, proving to be of some utility, though access to the terminal was denied during the subsequent trouble-plagued United Nations Mission to Somalia (UNOSOM II). JDISS was offered free of charge by the US but the UN insisted on purchasing it (for some $60,000). Thus, a JDISS terminal was installed in one corner of the room occupied by the I&R Unit on the 32nd floor of the UN’s Secretariat Building. This terminal was to help facilitate information transfer on a ‘need to know basis’ to the UN (i.e., when the US thought the UN needed to know). To the I&R officers, JDISS was a limited tool, useful only for certain types of information and often non-functional. The normal procedure would start with a request for information (RFI) from the Unit to the US Mission across the street through JDISS. Within the US Mission, an officer from the Defense Intelligence Agency/J2 (Joint Staff, Intelligence Directorate) would coordinate the response to the request. The Bureau of Intelligence and Research (INR) in the State Department was an important resource but it was the newly created ‘UN Support Desk’ in the National Military Joint Intelligence Centre (NMJIC) that was tasked to gain ‘properly sanitized’ responses from the US intelligence community. Though very sensitive information was hand delivered or orally briefed, most responses could be sent in standard formats using JDISS. Ambassador Madeleine Albright’s intelligence advisor later recalled that he could literally look across First Avenue from the UN Mission to the UN Secretariat building to see ‘my JDISS picking up the stuff’.¹⁴

FIGURE 2.ORGANIZATIONAL DIAGRAM OF THE SITUATION CENTRE IN 1996

Note: In the official charts of the day, the I&R Unit and the NATO Liaison Officer were not shown.

There were fears that the US could manipulate UN decision-making by providing selective and biased information, but officers in the I&R Unit claimed that this was not so. Certainly the intelligence officers from the other three countries kept a close eye on the information provided by their colleagues, providing a certain level of ‘checks and balances’. The papers and reports issued by the Unit were, as a rule, joint endeavours, requiring a consensus from all the officers covering a situation. Their ‘consolidated report’ was supposed to remove ‘national orientations’. The reports were then delivered to a number (typically 4–7) of high-ranking UN officials in the Secretary-General’s office, and the Department of Political Affairs (DPA), as well as in DPKO.¹⁵Still, Kofi Annan, as Under-Secretary-General for Peacekeeping, commented: ‘We have to be careful because the big powers only give us what they want us to know’.¹⁶ However, the regular feed of political and military information from them was deemed helpful. The US supplied substantial intelligence on conflicts in Angola, the Balkans, Burundi, Cambodia and Haiti. As we shall see, in the case of Zaire, US intelligence was problematic.

The other countries also provided valuable and timely information. In particular, France gave its I&R officer intelligence on ‘trouble spots’ in Africa, for example: Eastern Zaire, where a refugee crisis had erupted after the 1994 Rwandan genocide; Burundi, which had Hutu/Tutsi ethnic divisions similar to Rwanda; Sudan, where there was an ongoing war between the ‘Muslim’ government in the north and ‘Christian’ rebel groups in the south; the Hanish Islands, which were claimed by both Yemen and Eritrea; and Angola, where rebel leader Joseph Savimbi was fighting to gain power after losing the 1993 election. The British officer took the lead for the Balkans, though he also contributed to the Zaire case and other hot spots as they arose. The Russian officer worked mostly on the Commonwealth of Independent States (CIS) region, especially in areas like Georgia where the UN had deployed peacekeepers. A non-commissioned intelligence officer from Belgium, with skills in geomatics, served as an ‘I&R assistant’. In a nearby office in the SitCen, NATO posted a liaison officer to DPKO, who gave ad hoc (generally twice monthly) briefings to SitCen and I&R Unit staff.

What quality of intelligence did the I&R Unit provide to the UN? Was its information accurate and its analysis unbiased? Did it improve the effectiveness of peacekeeping? To begin to answer these important questions, it is useful to undertake a case study. The crisis in Eastern Zaire in 1996 provides plentiful examples, insights and lessons. Fortunately, the uncensored I&R Unit reports on Zaire (1994–98) were made available to the author. A summary and analysis of these ‘information products’ from the Unit is provided below for the first time, after a description of the background situation.

THE EASTERN ZAIRE HOT SPOT

Eastern Zaire was the centre of world attention after the 1994 Rwandan genocide.¹⁷ Refugee camps were overflowing with over a million Hutus, including many former génocidaires who had committed tremendous atrocities against the Tutsi and their Hutu sympathizers. The map in Figure 3 of the Great Lakes region of Africa show the locations of Goma, Bukavu and Uvira, where the main camps were located. From these refugee camps,

members of the defeated Rwandan government forces (the Forces Armées Rwandaises, now called ex-FAR) staged attacks into Rwanda. They were in turn subject to attacks from the Tutsi-led Rwandan government and local Tutsi (Banyamulenge and Banyamasisi) allied to Rwanda. In late October 1996, the situation became so untenable that humanitarian agencies and the United Nations High Commissioner for Refugees (UNHCR) had to evacuate the camps and leave the refugees to their fate. The situation was dubbed a ‘genocide in the making’. The media over-reported sensationalistic NGO projections of thousands dying (later shown to be the result of hysteria in a fact-starved environment, leading to a loss in NGO/media credibility).¹⁸ At the time, there was an intense clamour for the UN to ‘do something’ to avoid ‘another Rwanda’.

In the UN Security Council, France and the US clashed. Neither wanted the other to intervene because these two permanent members of the Council supported opposing sides. France was sympathetic to the French-speaking ex-FAR and even had led a mission (Opération Turquoise) in 1994 to protect their withdrawal from Rwanda. The US had long-standing links with the predominantly English-speaking Tutsi Rwandan Patriotic Front (RPF) that had ousted the Hutus in 1994. Both powers deployed had large groups of special forces in the region. As international pressure for overt action mounted, the US turned to Canada to lead a mission.

In early November 1996, Canada secured promises of peacekeeping troops from about a dozen nations. On 15 November 1996, the Security Council mandated a Canadian-led Multinational Force (MNF) to help bring humanitarian aid to the camps and to facilitate the repatriation of refugees to Rwanda. But shortly after the UN decision, the Hutu camps were attacked outright by Tutsi forces, catalyzed by the news of the impending peacekeeping mission that might have provided protection to the camps.¹⁹ Over 600,000 refugees headed back to Rwanda but an unknown number (still subject to debate) fled deeper into Zaire. The UN-authorized mission was aborted a month or so later with the weak claim that the mission had successfully facilitated the return of hundreds of thousands of refugees. In fact, the mission had become untenable after the US withdrew its support, while at the same time providing contradictory information (see later).

Events had overtaken the mission, and not just in the camps. The US-backed governments of Rwanda and Uganda had invaded Zaire to hunt for their Hutu enemies. It later became apparent that these foreign military forces also assisted and guided the rebel Tutsi forces, including the one headed by Laurent-Désiré Kabila. Officially, the two neighbouring governments denied outright their presence but later Rwandan strongman Paul 20 Vice President Kagame told Nik Gowing of the BBC: ‘We used communication and information warfare better than anyone . . . our strength was to keep information from them [Western humanitarian organizations and the international media].’ Gowing (note 18), p.15.Kagame admitted it, boasting about his ability to engage in ‘information warfare’ with Western agencies, including humanitarian agencies, the media and the MNF.²⁰ Furthermore, Tutsi rebel forces carried out widespread massacres in Eastern Zaire, while denying investigators access to the sites of the atrocities.²¹

Events soon cascaded further, with Kabila’s rebels sweeping westward across Zaire to the capital Kinshasa. The authoritarian Zairian President, Mobutu Sese Seko, who had installed himself in mid-1960s with CIA backing, fled the country in May 1997, leaving Kabila to govern an impoverished, war-torn and deeply divided land. Soon Kabila and his Rwandan/Ugandan patrons became disaffected and the country was again divided by civil war in 1998. The newly named Democratic Republic of the Congo, far from ‘democratic’, became a vast battlefield for over a half dozen African nations who took opposing sides, making it a type of ‘continental war’. Other nations of the world had an interest in the region, not only for partisan or humanitarian reasons but also because of the lucrative mining contracts in the mineral-rich regions.²²

The American, British and French officers in the I&R Unit all provided substantial information on the situation in Eastern Zaire, but the French officer took the lead in preparing the I&R Unit reports. As the crisis grew to epidemic proportions in late 1996 and as the new mission was being launched, objective information was crucial for the UN. The media and NGOs, discredited to a large extent by their earlier scare-mongering, were considered unreliable. One Canadian MNF military planner said: ‘After about the first week working in the Great Lakes Crisis, I ignored the media reporting. It was too inaccurate and too irrelevant to the political and military decisions that I was involved in.’²³ I&R Unit officers regularly briefed the commander of the Multinational Force and his chief of staff, in addition to UN Secretariat officials. The Secretary-General maintained an open-door policy for I&R staff, though only he or his chief lieutenants were allowed to brief the Security Council on the evolving situation. The present paper analyses the reporting of the I&R Unit on Eastern Zaire, focusing mostly on the period of greatest crisis in late 1996. The case study is based on both written I&R reports²⁴ and interviews²⁵ at the time and afterwards.

EASTERN ZAIRE CASE STUDY

For most of 1996–97, the I&R Unit’s written reports on the situation in Eastern Zaire (Kivu provinces), were issued about once a month, though more frequently during the critical periods. In November–December 1996, when the MNF was being assembled, the written reports were supplemented by three or so oral reports a week at UN headquarters. Most written reports were classified as ‘UN Confidential’, though some were also ‘Eyes Only’ reports for designated UN officials. The memoranda usually carried standard section headings: Introduction, Event, Significance and Conclusion (all on one page) followed by a detailed analysis of the situation. Hand-annotated maps, disclaimers and confidentiality provisions were also often included.²⁶ In addition, the Unit made use of the open door policy of both Secretary-General Boutros Boutros-Ghali and Under-Secretary-General for Peacekeeping Kofi Annan (who was Secretary-General elect from 17 December 1996 until he took office on 1 January 1997).

One early event analysed in the I&R reports was President Mobutu’s agreement on 6 June 1996 that UN observers be stationed in Kivu to monitor the Security Council-mandated ban on arms exports to the defeated ex-FAR. The I&R Unit warned of the risks to UN officials who might be posted there on such a sensitive mission. Those selling, delivering or using the illegal arms would not want to be caught or have their illegal plots exposed. The report of 20 June 1996 provided both insightful analysis and a prediction (accurate as it turns out):

Despite President Mobutu’s insistence on declaring the FAZ [Force Armée du Zaire] as being neutral vis-a`-vis the ethnic groups in Kivu, the obvious pro-Hutu attitude of the Zairian military authorities in this area and their non-intervention policy would suggest that the bottom line would be the removal from Masisi [in North Kivu] of all the [local] Banyarwaranda/Tutsis in order to resettle the Hutus from the refugee camps. Should such a solution, covert or otherwise, be initiated the Rwandese authorities would react: they would definitely not tolerate such a potential threat in the form of a ‘Hutuland’ close to their borders.

The I&R Unit report went much further than most UN reports of that era: it made personality assessments of a head of state. President Mobutu had the ability ‘to skilfully play off the many discontents among opposition parties’ while attempting to ‘restore his reliability in the eyes of Western governments by performing a mediating role in regional African disputes’.

The reports give much information that was sensitive at the time and would have laid the UN open to criticism of ‘interference in internal matters’, however relevant the information to conflict management and resolution. But the details about many nefarious activities in Eastern Zaire would have been hard if not impossible for UN staff on the ground to learn by themselves. Such information necessitated intelligence-gathering methods in which the UN could not itself engage. As is usual in the intelligence world, I&R reports do not specify the sources or methods of information-gathering. They usually begin with an intentionally vague statement that a ‘variety of informed sources’ were used. It was understood that information usually came from the intelligence networks of one or more of the great powers represented in the Unit.²⁷

Among the many topics covered in the reports, four areas stand out for their potential importance to the UN. These are described in detail below:

1. Arms shipments;
2. Military (belligerent) activities;
3. Numbers and status of refugees;
4. Predictions and warnings.

In citing and quoting the Unit’s reports below, only the dates are given. Most of the reports were innocuously titled ‘Memorandum’ or ‘Information on Kivu (Zaire)’.

1. Arms Shipments

The UN had a responsibility to stop violations of the arms embargo it had imposed on the former Rwandan government forces (ex-FAR) in Eastern Zaire. Thus it needed first to identify and expose the ‘sanctions busters’ transferring weapons into the region. In theory such persons would then be arrested by the government of Zaire, which claimed to be strictly abiding by the embargo. But the I&R Unit implicated the government itself at the highest levels. Knowledge of illegal arms shipments was also important for the UN because weapons infusions posed a great danger to UN aid workers. They also served as early warning indicators of escalating violence. It was a highly combustible situation in a region with a long history of animosity and attacks, with many marauding tribes and militia of conflicting ethnicities (including Hutu, Tutsi and numerous Zairian ethnic groups), all made immediately flammable by the recent arrival of the desperate ex-FAR.

The I&R Unit revealed that illegal shipments were made to the Hutus by persons with close ties to the Mobutu regime, despite its claims of compliance with the embargo. The reports even implicated Mobutu’s own adventure-seeking sons, who piloted 727s filled with weapons. At the local level, a wealthy Zairian woman in Kivu, named in one report, channelled funds and arms to the Hutu rebel groups. She was the ‘front person’, as the nominal ‘General Director’ of a tobacco company, ‘whose hangars are used to transfer weapons, ammunition and other amenities’. She supported the extremist Hutu radio station, Radio Démocratie, located on the company premises and she chartered planes to medevac certain senior Zairian officers wounded during a ‘transfer operation’. She was ‘protected’ from authorities in the Zairian government by the Prime Minister himself. She was also described as the mistress of a radical Burundian Hutu commander (also named).

The two ‘main routes’ for arms flows are described: in one route, planes leave Belgium, load weapons and ammunition in ‘Bulgaria, Egypt and even Sudan’ before landing in Kinshasa; in the other, cargoes depart South Africa, from ‘a secret facility run by former South African Special Forces affiliated with Executive Outcomes’ (a company that closed down in 1999 after South Africa regulated private military corporations along with the country’s expanding but disreputable mercenary export market). The reports also claim that ‘large cargo plane[s] from Uzbekistan have been monitored unloading arms and new military uniforms’ at the airport outside Kinshasa. The weapons were stored in hangars at Kinshasa airport before being shipped to Bukavu (Eastern Zaire), as well as to Angola and Cabinda to help rebel leader Joseph Savimbi. The Zairian Army Chief of Staff ‘facilitated’ the storage of the cargoes at Kinshasa airport. In Eastern Zaire, the main caches were located in two Parcs nationaux in Kivu and one training camp was located ‘somewhere in Kigonga, in the vicinity of Mugunga’ (just west of Goma). The weapons were paid for by the ex-FAR using funds they took out of Rwanda at the time of their retreat.

‘Politicians and militaries at the highest level in Kinshasa are involved in the FDD [Burundian armed rebel group] arms trafficking’, the I&R Unit alleges. For instance, a major arms broker was ‘Mr. Bemba’ (père), a ‘well known businessman’ who was married to the daughter of President Mobutu.

One of the Unit’s reports reveals an interesting incident: two officers of the Zairian Corps of Engineers were caught delivering arms to Hutu rebels in Uvira. ‘The local military system in Goma’, it was written, ‘was paid US$ 15,000 to cover up the case.’ (This suggests some close, if not inside sources of information for the report.) The memo concludes with the warning: ‘The trafficking activities are to be closely related with the inability of the Zairian government or authorities to pay even basic wages to their Armed Forces . . . any attempt to curtail such a source of living should be considered as life threatening.’

2. Military Activities

The reports give ‘order of battle’ information, including new ‘military assets’, like the arms mentioned above, of the warring factions. They often include maps showing the recent positions and dispositions of the fighting forces. The Hutu military in exile was estimated at ‘about 30,000 ex-FAR and Interahamwe members’ among the estimated 850,000 Hutu refugees. In Kivu three Hutu military training camps were supported covertly by the Zairian government.

In opposition, an estimated 3,000 Banyamulenge (Zairian Tutsi) had been ‘armed, equipped and trained in Rwanda’. The rebel force benefited from the arrival of mercenaries from South Africa, Angola (UNITA), Serbia, France and Belgium. (This is reminiscent of the UN’s problem with ‘soldiers of fortune’ in the 1960s when it had a mandate to identify, capture and repatriate mercenaries from the Congo.) Executive Outcomes, not generally known to have operated in the Congo, recruited 154 ‘highly skilled’ personnel to serve ‘the newly inducted military air assets’, which included MiG-21 fighters and Mi-25 attack helicopters (I&R Unit report of 19 December 1996).

At the time, the Western media had falsely assumed that ‘a ‘‘tin-pot’’ war [was] being waged by a ‘‘rag-tag’’ force of Tutsi rebels’ and only many months later was it widely recognized that ‘a carefully conceived military campaign’ had been planned and executed by Rwanda.²⁸ In addition, Kagame later said that these plans were ‘signalled explicitly to the United States as early as June 1996’ but that the US did not respond.²⁹ Kagame told US officials in June 1996 that the Hutu refugee camps ‘had to be dismantled, and that if the United Nations would not remove them, somebody would have to do it’. Kagame then said: ‘The US response was no response.’ Indeed, the US was a major trainer of Rwandan government forces and perhaps Zairian rebel forces as well.³⁰

The I&R Unit reports describe night raids by the Zairian rebel forces. In the town of Lemera, the rebels killed 20 Zairian soldiers and seized ‘200 small arms, 2681 mm mortars, 3 radio communication sets and a lot of ammunition’ (report of 24 October 1996, henceforth dates only are given). In an act of deception, the government of Rwanda stationed its forces ‘under the disguise of Banyamulenge’ in Zaire to protect hydroelectric plants that provide power to both Rwanda and Burundi (19 December 1996). In November 1996, the Rwandan forces even took control of Goma, an international seizure that was made quite apparent when the Rwandan flag was raised over the main administrative buildings of this Zairian town (6 February 1997).

Despite Rwandan government denials of involvement, its army provided ‘two battalions in direct support’ of the rebels and ‘also anti-aircraft artillery to secure airports in Bukava [sic] and Goma’. In one instance, Rwandan attack helicopters left more than 100 [Zairian] soldiers dead. Furthermore, Rwandan Special Forces troops spearheaded an attack on at least one of the refugee camps.

Uganda as well as Rwanda was now supporting the Zairian rebel forces, amalgamated under the leadership of long-time Mobutu opponent Laurent Kabila. His forces had also received long-standing support and training from Libya (Annex I, 1 January 1997). Like Rwanda, Uganda denied its support at the time, but later admitted it. All parties – Rwanda, the Zairian government and rebels – were carefully controlling access to the combat zones to keep Western eyes out, though Uganda (the least involved) was the most open. (The MNF headquarters was stationed in Entebbe, near Kampala, the capital of Uganda.)

A more controversial allegation is that the United States ‘strongly supported and established the rebellion’ (19 December 1996). The I&R Unit boldly asserted that the Tutsi rebellion was backed by ‘American teams’ (6 February 1997). Despite official US support for the Canadian-led humanitarian mission in November–December 1996, the Unit alleges that the US sought to undermine the operation: ‘On the American request to deter the deployment of a UN-authorized Multi-National Force [MNF] led by a Canadian General, the FRP [Rwandan army] along with ADFL [rebel group] elements lured the ex-FAR and Interahamwe in a combat operation north of the Muganga camp (Zaire).’ If these allegations were true,³¹ it has a striking parallel with duplicity in the Congo mission in the 1960s. While UN forces were protecting the Congolese leader Patrice Lumumba as part of a UN operation explicitly backed (and partly paid for) by the US, CIA operatives in the Congo were trying to assassinate him and later backed the Army Chief Joseph Mobutu as he seized the Presidency.³² More likely, according to this author, the US at first wanted the MNF to succeed in stabilizing Eastern Zaire but as the situation deteriorated the US decided to fully back its long-time associate Paul Kagame and his rebel allies, who were by now winning many battles. Another motivating factor for the US, cited later by an I&R Unit officer in an interview, was that major American mining companies had shifted their support to Kabila in order to gain access to important strategic minerals in Kabila-controlled areas.³³

As a result of the attack on the camps, more than 800,000 refugees began the trek home. The not-yet-operational MNF was withdrawn ‘despite a remaining 400,000 scattered in Kivu’ (6 February 1997). This highlights one of the most contentious intelligence issues of the Zaire crisis: the number of refugees allegedly ‘abandoned’ by the international community.

3. Refugee Numbers and Status (The ‘Numbers Game’ and the ‘Intelligence Gap’)

How many refugees remained in Zaire after the coordinated attacks on their camps in mid-November 1996? The estimates vary from as little as 20,000 to over 400,000. The US claimed the lower figure, while aid groups and the I&R Unit alleged the number was 10 to 20 times higher.³⁴ The numbers game had important political ramifications. The UN and the MNF needed to know if a sizeable refugee group remained and needed assistance in Eastern Zaire. The US position after the attack was clear: it wanted the mission aborted. It saw the deepening conflict area as a quagmire and was now unofficially supporting the Tutsi forces. US Major-General Edward Smith stated categorically that aerial and satellite reconnaissance backed the US claim that almost all the refugees had returned to Rwanda. The US estimate of the number of remaining refugees was almost 400,000 less than the consensus figures used by the humanitarian community.³⁵ Lieutenant General Maurice Baril, the MNF Commander, was suspicious of US reports of numbers, which were too rapidly sliding downwards.³⁶ Members of the I&R Unit had briefed him on what they believed was disinformation.³⁷ Both the French and British officers in the Unit were tracking the numbers. France was providing figures from overflights with Mirage jets. The British officer was gaining information from UN agencies on the ground (e.g., UNHCR, UNICEF, WFP, etc.). They both concluded that the US numbers were far too low. In addition, Baril’s multinational force (mostly Canadian) engaged in a desperate attempt at reconnaissance and he himself flew over the territory for a first hand impression. But the counting was difficult because the frightened refugees, many fearful of aircraft, were dispersed over large areas, sometimes under thick jungle cover. The issue has never been resolved though General Baril believes that the actual number of remaining refugees was about 220,000.³⁸ Many thousands probably perished in the jungles of Zaire.³⁹In addition, one is left to wonder if a strong early UN intervention could have saved the Congo from the subsequent chaos and loss of over three million lives or at least have mitigated the human tragedy. The UN currently deploys some 17,000 peacekeepers in that country to keep a very strained peace.

4. Predictions and Warnings

In 1996/97, the I&R officers took on the difficult and potentially embarrassing challenge of predicting the future, something rarely done in the UN. In some instances, their warnings were prescient. For instance, months before Rwandan government forces and Zairian rebel allies attacked the Hutu camps in November 1996, the Unit made the following dire but rather general warning (20 June 1996): ‘Rwanda would be happy to put together a contingency plan to eradicate the ex-FAR (Forces Armées du Rwanda) troops and Interahamwe militias with the covert acceptance of Uganda.’

A more precise warning came in August but it was couched in cautious language, perhaps to hedge against an incorrect prediction: ‘No one can rule out a quick operation launched by Kigali [Rwanda] with the support of Kampala [Uganda] to terminate the ex-FAR and Interahamwe threat’ (26 August 1996).

In October, by contrast, a large-scale Rwandan operation was ruled out ‘during the absence of General Paul Kagame [Rwanda’s leader] who is currently visiting Israel and Italy’. The main attack on the camps occurred in mid-November.

As for the larger picture, the Unit postulated that Rwandan training of Banyamulenge (Zairian Tutsi) was ‘indicative of a larger design aimed at seizing part of Kivu and destroying the ex-FAR should President Mobutu die’. The Unit did not, at the time, foresee the ambition and ease with which the rebel forces would sweep across Zaire and seize Kinshasa, nor that the ex-FAR would be easily defeated (at least temporarily) in the attacks on the camps and would be dispersed so widely afterwards. But the British officer in the Unit did develop ‘scenario flow charts’ (12 November 1996).⁴⁰ In one of these scenarios, the rebels become the government of Zaire (which actually occurred in 1997). Another scenario is that the government of Zaire collapses and the country is de facto partitioned (which actually occurred in 1998). The alternative scenario was for a stable situation and the success of political efforts (hopefully where Zaire is heading today). Such model development showed that the Unit was aware of potential outcomes, though it was not predicting them.

Some warnings proved premature. On 4 November 1996, the Unit warned of a coup plot against Mobutu ‘very soon’, naming ‘three main plotters’ who were top Zairian Generals. The update of 19 December noted that one of these was dismissed while another was promoted. Again it warned that ‘a military coup should be expected soon’.

For the UN, warnings and risk assessments are an important part of any mission start-up. Thus I&R Unit warned that ‘aid workers could easily be caught up in the unrest or even targeted because of their relative wealth; four-wheel drive vehicles are prized and are often the first to be hijacked during a period of unrest’ (26 August 1996). In North Kivu, the Unit identified several other threats: landmines, ‘the ill disciplined behaviour of the FAZ [Forces Armées Zairoises] soldiers, the rampant crime, and the lack of adequate means of evacuation’ (26 August 1996). The Unit also warned that the UN could unwittingly assist the fighting factions. For instance: ‘World Food Programme’s supplies are being used to feed the military and militias, and to build up stocks in their ration depots.’ Furthermore, ‘UN agency transportation assets are also used to smuggle arms and ammunition’, in violation of the UN’s own arms embargo (20 June 1996)!

On 19 December 1996, the Unit described the offensive launched by government troops in Eastern Zaire and their subsequent retreat. It incorrectly predicted that the recent rebel successes, gained through surprise, would not continue in the intense fighting of the future.

Later reports correctly describe Zairian armed forces (FAZ) in total disarray, with soldiers deserting, joining the enemy, or engaging in rampages of ‘looting, raping and killing’ (6 February 1997). The rebel march progressed across the vast country and only ended after forces loyal to Laurent Kabila seized Kinshasa and control of the government. In an effort to protect his regime, Mobutu had called on other African governments for support, leading to Africa’s ‘continental war’. Ironically, President Mobutu, who had survived many assassination attempts and the overthrow of his government, died later that year of prostate cancer while in exile in Morocco.

CONCLUSION

We have learned, contrary to past hesitation, that intelligence is necessary and that we need to have solid political analysis to be able to, if not determine, then envision how the crisis is likely to develop and how we would act if it went in one direction or the other.
— Kofi Annan, 18 December 1996⁴¹

The reports of the I&R Unit were valued by UN officials like Kofi Annan, who at the time was Under-Secretary-General for peacekeeping and Secretary-General elect. Even if the UN could not deal directly with nefarious activities in conflict zones, it had to know about them in order not to unintentionally get caught up in them. The game in Zaire was for high stakes and powerful forces were at play, including long-standing ethnic hatreds, displaced armies and the economic interests of Western mineral resource companies.

To execute its mandate to assist refugees, the UN needed to know many things. The critical and priority information requirements (CIRs and PIRs) in the Zaire crisis included:

• Refugee numbers, dispositions and locations;
• External influences, including those of neighbouring states and major powers;
• Weapons flows and military preparations for attacks;
• Potential threats against UN personnel;
• Potential scenarios for the future.

It is clear from the I&R reports that the Unit provided much important information on these categories, though still insufficient overall. But even if it was incomplete and uncorroborated intelligence, at least UN officials had an idea of what was probably happening. UN officials could find opportunities to privately question governments, to seek further information on the ground and in some cases to expose misinformation and disinformation. For example, a growing awareness of illegal arms shipments into the Congo caused the UN to initiate several Commissions of Inquiry, resulting in several governments (Rwanda and Uganda in particular) and many Western businesses being later ‘named and shamed’.⁴²

The revelations in the I&R reports include: clear evidence of the powerful influence of money, weapons and illegal trafficking in the region; the complicity of governments in activities that they denied; and revelations of a planned coup d’état in Zaire. The conflict analysis included: estimates of the numbers and movements of troops (Zairian and foreign), rebels and mercenaries; tracing arms shipments violating a Security Council embargo; exposing diversions of UN humanitarian aid and corruption at the senior levels of Zairian society; providing early warnings about potential violence against UN workers and identifying military officers planning a coup against the Zairian President. The Unit also performed scenario-building and analysis of motivations of political and military leaders, practices which the UN has traditionally shunned.

Many of the claims made in the reports (including direct Rwandan involvement in attacks on the refugee camps; US support for Rwandan forces; illegal mineral exploitation and extensive arms smuggling) only became public knowledge much later. While the US had been training Rwandan forces since 1995, the full extent of the activity (including training in combat, counter-insurgency, counter-intelligence and psychological operations) was not publicly revealed until long after the attacks on the camps.⁴³

Were the reports biased? A French bias can be alleged but since the truth about many of the contentious statements is uncertain, it is difficult to prove. Several I&R reports presented the same view as the French government. In one report (6 February 1997), the danger of not establishing a multinational force was emphasized, an issue France was clamouring about. More significantly, the reports allege that the US trained and strongly supported the Zairian rebels, including their attacks on the refugee camps, and that the US disseminated disinformation on the number of refugees remaining after the attacks.

One I&R report questioned US analysis of the problems in the region. For instance, it criticized a ‘US security report’ saying it ‘does not address the key factors; i.e., the military and paramilitary balance in the area, the short and mid-term struggle of the main current political leaders, the control of economic assets (minerals, diamonds) and hard currency generating profits (weapons smuggling)’ (26 August 1996). The issue of the control of economic assets was a sensitive one, since US firms (e.g., American Mineral Fields) were later found to have cut lucrative deals with the rebels in exchange for ‘rights’ to the strategic minerals in the region.

Furthermore, the French government was known to be partial to the francophone Hutus and maintained long-standing connections to the Hutu leadership. It is possible that this influenced the Unit’s reports, which relied heavily on French intelligence. For instance, one report (6 February 1997) indirectly admitted French complicity with ex-FAR in helping the Hutus to cross into Zaire with most of their heavy equipment in 1994 during the French-led Opération Turquoise. But the reports provide no information about subsequent French assistance to the Hutus, though perhaps France was now keeping its distance from its former friends.

The lesson for the UN in Eastern Zaire is similar to the lesson learned by Canada, the lead nation in the MNF: the need for an independent information collection capability. ‘The CF [Canadian Forces] lacked an independent strategic intelligence capability causing reliance on allies for virtually all in-theatre collection. This brought with it the danger of data manipulation or distortion for political reasons as was evident during the debate over refugee numbers.’⁴⁴

To be truly independent, the UN would have to directly gather information about the nefarious, criminal and duplicitous world of warlords, rebels and dictators. For an organization dedicated to transparency and openness, this would be a great challenge. It cannot run covert intelligence-gathering operations since this would open the UN to criticisms of partiality and stain its credibility. But at least the UN should have channels to receive such information and have means to openly verify the information where possible. For the UN to cover its eyes, ears and mouth (‘see no evil, hear no evil, speak no evil’) would be to ignore the reality of the many evil atrocities committed in the war zones where it operates. (See Romeo Dallaire or A. Walter Dorn and J. Matloff for intelligence failures prior to the Rwandan genocide.⁴⁵ This case study shows that there are many important facts the UN should know if it is going to be effective in providing a peaceful alternative to war.

In Zaire 1996, the UN did not know enough. It was therefore possible for the US to lead the UN away from engagement in late November 1996, one of the principal reasons the mission was forced to abort.⁴⁶ The I&R Unit gave UN officials and mission leaders a sense of what might be the US agenda but it was too late to stop the chain of events set off by the attack on the camps. The attacks brought about at least one positive development: hundreds of thousands of refugees returned to their homeland. Later some I&R Unit officers regretted that they did not expose the Rwandan and US action more fully, more forcefully and earlier, though they did plant questions and suspicions in the minds of the Multinational Force leaders, including its Commander, General Baril of Canada.

The UN’s Department of Peacekeeping Operations lost its I&R Unit inadvertently in February 1999 when all the gratis personnel working at the UN, mostly in its peacekeeping and humanitarian departments (some 500 personnel, including 129 military officers), were forced to leave because of a General Assembly resolution pushed by the developing world. Perceiving an unfair advantage to the developed world, these nations wanted positions to be opened up for their nationals. Since they could not afford to post their nationals to New York using meagre national budgets they sought new UN positions paid for by the UN’s regular budget. But new funds, posts and units for information analysis were not forthcoming.

The Situation Centre currently has no I&R Unit. The SitCen’s present information gathering and reporting draws almost exclusively from UN field missions and open sources (the media in particular). It does not have the human resources even to acquire all the open-source information needed for missions. Data analysis is conspicuously missing. The SitCen does no scenario-building and the links with government intelligence agencies (as well as UN agencies like the UN Development Programme) range from weak and ad hoc to non-existent. The UN’s Office for the Coordination of Humanitarian Affairs (OCHA, formerly the Department of Humanitarian Affairs or DHA) is only slightly better off, with an advanced computerized systems for collation of information from the media and the field. Its Integrated Regional Information Network (IRIN) and Reliefweb were both producing valuable information in the Eastern Zaire in 1996 and they remain useful today to follow development in many complex and natural disasters.⁴⁷ But they do not include a substantial in-house information analysis component.

The 2000 Brahimi report recognized the need for intelligence and boldly proposed the creation of an ‘Information and Assessment Secretariat’ but this proposal was resisted by many governments, despite pleas from the UN Secretariat. The August 2003 Baghdad bombing brought added urgency to the pleas for greater threat awareness and staff safety. A new Department for Safety and Security was set up in 2005 with a mandate to conduct threat and risk assessments with a planned analysis cell of eight to ten persons. If the mandate is interpreted broadly, the new department could develop and share important mission-related intelligence with member states as well as with the field missions, which are once again growing in number and size. (The Congo peacekeeping force will shortly surpass the Liberia operation as the UN’s largest mission, with over 18,000 peacekeepers.)

In the field, there have been advances in intelligence-gathering: Joint Mission Analysis Cells (JMACs) have become a standard feature in new peacekeeping operations. For instance, the present UN mission in the Congo (MONUC) provides ‘surveillance training’ (however rudimentary) for deployments at airports and border flashpoints. A UN report found that much more was needed, including ‘lake patrol and air surveillance capabilities, including appropriate nocturnal, satellite, radar and photographic assets’.⁴⁸ The UN technical surveillance capabilities (‘blue sensors for blue berets’) is an area for further study.⁴⁹ But at headquarters the lacuna of intelligence remains the greatest. Furthermore, any intelligence system needs to provide for robust data sharing with the field, that is, in both directions.

In delving into the murky world of intelligence agencies, the UN must determine the proper balance between secrecy and openness. Given the recent emphasis on the safety of UN personnel, the pendulum is now swinging towards secrecy. In order to uncover secret, nefarious plots to attack UN personnel or other aid workers, the UN will need to delve more deeply into the world of terrorists and spoilers who seek to sabotage the peace process. After the Baghdad bombing, with the loss of the chief of mission, Sergio Vieira de Mello,⁵⁰ the UN became galvanized in its quest for prevention. This involves the Situation Centre (whose chief is now also the focal point for security within DPKO) and a quest to find new links to national intelligence agencies.

Should a future intelligence unit be modelled after the I&R Unit to include only nationals from the permanent five members of the Security Council? To do so would raise the rancour of much of the rest of the world. But without providing guarantees of confidentiality to the major powers the flow of intelligence from them would dry up. A balanced system must therefore be created, based upon a strong confidentiality mechanism, expert analysis and an effective but limited distribution mechanism. To reduce national bias, the UN should rely on a larger number of countries for information. The former SitCen chief, Stan Carlson, recently wrote: ‘if there is a shared interest between the UN and a member state, information/intelligence can and will flow’.⁵¹

Should the UN enter into formal agreements with information/intelligence providers in various nations? Many believe this is desirable, particularly when UN missions are subject to threat. But to play in that game, the UN Secretariat must become intelligence savvy. Does it have the wherewithal to do so?

The unofficial motto of the Information and Research Unit was ‘Keeping an Eye on the World’. The I&R Unit proved that intelligence was extremely important to understand the fast-moving currents of the Eastern Zaire crisis. It proved to be a useful early example of what can be done within the intelligence-sharing sphere. It is clear that the UN must have an even more vigilant eye if it is to be effective in its ambitious task of peacekeeping in the Congo and elsewhere.

At the same time, the UN must determine the proper limits of its intelligence-gathering and sharing – politically, legally, morally and in terms of resources – the subject of contentious debate.⁵² In a previous paper, ‘The Cloak and the Blue Beret: The Limits of Intelligence-Gathering in Peacekeeping’,⁵³ the author sought to present an analytical framework, based on the UN’s historical and current experience, for determining such a balance and such limits. A better appreciation of intelligence, its uses and pitfalls, would benefit the UN. As the UN gains more experience and an institutional evolution takes place – as in nature, in a non-linear fashion – we can hope that the UN is becoming not only more intelligence-savvy but also more intelligent and, hence, more effective.