Introduction

This article examines the refugee experience in the Kigoma region of Tanzania2. It focuses in particular, on the challenges which refugees, the United Nations High Commissioner for Refugees (UNHCR) and the government of Tanzania have faced in this refugee programme in respect of maintaining the civilian and neutral character of camps. The study is based on the author's work experience as Head of UNHCR's Sub-Office in that region between October 1997 and September 1999.

At the very outset, it bears to be underlined that the "security and neutrality" issues facing UNHCR and the refugees in the Kigoma region have not reached the overwhelming proportions of the example in eastern Zaire already referred to. Kigoma is, in the final analysis, a fairly "normal" refugee situation, with typical problems, constraints and opportunities. Nevertheless, the relatively more modest scale of its problems granted, it is hoped that lessons that flow from the experience there should be of relevance elsewhere in Africa and indeed beyond.

Within the Tanzanian context, developments between 1994 and 1996 in the Ngara camps for Rwandan refugees3 clearly resulted in a general hardening of attitude vis-à-vis refugees. Long-tested response mechanisms and institutional relationships also came to be questioned. At the same time, it is fair to say that the dramatic ending of that refugee situation, through "induced mass repatriation",4 has left the government of Tanzania authorities, and indeed UNHCR as well, with an uneasy feeling of regret, if not of guilt. There has therefore been an eagerness to deal with the ongoing Kigoma refugee situation in such a way as to avert a repetition of the intractable dilemmas of the Ngara era.

Geo-political considerations also explain the government of Tanzania's greater tolerance towards the Burundian refugee caseload, and possibly a more ambivalent attitude towards the involvement of this caseload in "political" activities5. This attitude will become apparent in the course of our analysis. However, it is just as clear that we not talking of a situation of total or oblivious laissez-faire. As a responsible state and a haven of peace in a turbulent region, Tanzania is particularly keen to demonstrate to all parties that it does control the Burundian refugee situation in a manner compatible with both its international law obligations and its mediating role in the Burundi peace process. While this exercise appears at times acrobatic, it offers opportunities for UNHCR to posit its humanitarian arguments.

Background

At the end of 1996, which was also the time of militarily assisted repatriation from Ngara to Rwanda, the government of Tanzania's security concerns in relation to refugees were at a record high. The immediate reaction to a looming Burundian refugee emergency had been to close the border7. An incident of refoulement would result in the summary execution of 126 would-be refugees by the Burundian military in January 1997. Partly in reaction to this tragic mishap, the government of Tanzania left the border de facto open to newly arriving refugees and allowed UNHCR to set up a fully-fledged assistance programme in the Kigoma region. One large refugee camp, hosting Burundian as well as Rwandan refugees, most of the latter posing as Burundians for greater safety, remained in the Ngara district.

The Kigoma programme provides Burundian refugees with an environment of relative security. Several factors contribute, however, to a climate of tension within and outside camps and put the civilian character of these camps at risk. These factors are analysed below under the headings of "the government of Tanzania's refugee settlement policy", "the 'contaminating effect' of the Burundi conflict and "frictions between the Burundi and Tanzanian governments".

The Tanzania government's refugee settlement policy

On the face of it, this rationalisation is neither necessarily just glib nor even peculiar. However, as we shall see further below, it probably conceals more complex motives in the particular case of Tanzania.

An aggravating circumstance is the presence, in those border areas where camps have been set up, of a floating population of long-time refugees, who for the most part settled there spontaneously in the early 1970's. These "old caseload refugees" have maintained links with the Burundian refugee settlements in the interior of Tanzania, which are known to be more politicised. There is no doubt that the recent settlement policy of the government of Tanzania has resulted in strengthening physical as well as intellectual links between several segments of the Hutu diaspora. Obviously, this is a situation which the more radical elements within this diaspora can be expected to exploit.

The "contaminating effect" of the Burundi conflict

The fact that the refugee influx has continued uninterrupted - though with peaks and lulls - since the end of 1996, has been a major factor of instability in most camps. Newly arriving refugees are not only a bitter reminder of the continuing violence in Burundi, they are also rich material for anti-government, anti-Tutsi propaganda. At regular intervals, groups of defeated or stranded Hutu insurgents have retreated into Tanzanian territory and ended up in refugee camps. They may be unarmed, yet they remain without any indication systematically being established, for instance through a screening process, as to their future intentions. And, not to be ignored, the camps' proximity to the battlefield makes them ideal rest and recuperation destinations for rebel combatants.

Troop movements are regularly reported on the Burundi side, and in the immediate vicinity, of the frontier. The Tanzanian military is also deployed along the borderline. Their exact role in controlling "subversive" activities against Burundi is the subject of much speculation. It is difficult for an impartial observer - let alone the authorities in Bujumbura - to understand the evolution of front lines inside Burundi without accepting the possibility that the rebels transit through Tanzanian territory with great ease and impunity. As a matter of fact, Tanzanian villagers and peace-loving refugees alike often complain about suspicious movements around the camps and in the entire border area, whereas the Tanzanian army never seems to cross the path of these "visitors".

Tensions in relations between Burundi and Tanzania

"the presence of refugees is a source of tension in the relations between Tanzania and Burundi and to a certain extent Rwanda, arising from suspicions that the refugees are regrouping and training in warfare for attacking the countries of origin"8

The strategic value of refugee camps in the border area is obvious from the standpoint of armed Burundi opposition groups. While the precise role of the camps in a possible Tanzanian strategy is harder to figure out, there can be no doubt that they play some role in such a strategy. The most widely accepted theory is that Tanzania, without providing direct support to the guerrillas, is happy to allow them to exert enough pressure on the Burundi regime as to force political concessions out of the latter. This passive complicity, delicately adjusting the rapport de forces on the ground, would complement, not contradict, Tanzania's genuine efforts to secure a peaceful settlement of the Burundi conflict through the Arusha process. However, humanitarian actors, primarily UNHCR, may be left to deal with one serious down-side of this strategy, namely, that it situates the refugees on the wrong side of the war-and-peace equation.

The Burundian authorities make no mystery of their own stance on this issue. Their official rhetoric assimilates refugees, en bloc, to trouble-makers at best, terrorists at worst. They never miss an opportunity to accuse the government of Tanzania of tolerating, if not aiding and abetting, the preparation of armed subversion in the refugee camps. Tanzania, in return, denounces Burundian attempts at destabilising its western provinces. On the whole, the verbal war has turned to the advantage of Burundian diplomacy, which by the end of 1997 came very close to obtaining the deployment of a United Nations observer mission on both sides of the border. While officially incensed by such arrogance, as it saw it, the government of Tanzania then realised that it could no longer credibly lock itself into a systematic denial of the problem.

The semantics of militarisation

Subsequent joint assessment missions in essence re-stated the same conclusions. In August 1999, the Burundian Minister of Defense, meeting his Tanzanian counterpart in Kigoma, declared himself satisfied that the government of Tanzania was not utilising refugee camps to provide support to Burundian rebel groups. This statement can be interpreted as a diplomatic concession aimed primarily at keeping the dialogue open in a climate of strained bilateral relations. It is unlikely that the government of Tanzania was able to produce much evidence to appease Burundi's security concerns in relation to the refugee situation. The concept of "militarised camps" elicits images of full-scale involvement of the refugee population in armed conflict, which cannot be reconciled with the peaceful appearance and general normalcy of the Kigoma camps. There are, however, sufficient signs of more subtle and hidden threats to the civilian character of these camps as to worry not only the Burundian government, but UNHCR as well.

The tip of the iceberg

Nothing is more difficult than to feel the pulse of a refugee population. It is possible, however, to detect trends and changes in atmosphere. It has, for instance, been observed that tension is mounting inside the camps, and "security incidents" proliferate whenever a new round of peace negotiations starts in Arusha. Likewise, recruitment of refugee youths became exceptionally visible when the Jean-Bosco faction of FDD9 was reported to link up with loyalist Congolese groups and Interahamwe in South Kivu. As noted above, radicalisation and unrest have also coincided with fresh influxes of refugees into the camps. This was particularly noticeable in the last months of 1997 when the government of Tanzania - in practice, the Tanzanian armed forces - carried out a massive sweep of "illegal aliens", in their majority spontaneously settled refugees, whom they transferred to refugee camps10.

Another series of incidents involved Burundian combatants11 who had not yet reached refugee camps although it was extremely difficult to determine whether or not they had previously resided in a camp. They were apprehended upon entry into Tanzania, either through the land border or across Lake Tanganyika. In one particularly dramatic instance in September 1998, a group of war-wounded Burundian Hutus, some wearing pieces of military uniform, landed on the shores of Kigoma, seeking medical or surgical attention. They admitted to having fought in South Kivu, DRC, as part of an FDD contingent. Isolated incidents such as these may, here again, be symptomatic of a much larger phenomenon. In the absence of any screening mechanism however - Burundians seeking Tanzania's protection are recognised as refugees prima facie and sent to camps - it is impossible to assess the magnitude of the problem. It is indeed difficult, as we will see, to deal in a principled way with even the most obvious cases - the tip of the iceberg.

A response strategy

The implicit premise of these negotiations is that lack of resources, not lack of will, is the main obstacle to the government of Tanzania living up to its undisputed obligation to ensure the civilian and peaceful character of refugee camps. The development of a "security package" for camps in Western Tanzania can thus be seen as a case study of the "soft" type of option to which the High Commissioner for Refugees referred in her briefing to the UN Security Council in November 1998, and which she described as "the 'soft' option of providing training and support to build national law enforcement capacity"12.

The central thesis of this essay is, however, that, even in a functioning and peace-loving state such as Tanzania, the geo-political stakes inherent in a refugee situation are such that security of camps and border areas cannot be reduced to a mere question of law enforcement. This realisation makes it necessary for UNHCR and other international players to consider a much more complex array of "soft" - and possibly also tougher - measures to support and guide states towards compliance with international standards.

Control and repression

The package also includes, to a much lesser extent, additional support to the field-based staff of the Ministry of Home Affairs' Refugee Department and, crucially, the deployment of UNHCR Liaison Officers with a mandate to support and advice the police contingent in its security-related tasks. In practice however, only one UNHCR Liaison Officer has been deployed so far, whereas UNHCR estimates the minimum requirement to be four, considering the number and dispersion of camps along the Burundi border. It is important to note that, very much in the spirit of "soft" options, UNHCR has no place in the chain of command of the police force.14

The policemen deployed under this arrangement, an average of twenty per camp, have been given specific responsibilities in the "neutrality and security" area, in particular the control and repression of "subversive activities" among the refugee population. Their terms of reference encompass, however, all aspects of maintaining law and order in their respective duty stations.

In practice, the policemen lack time, resources and knowledge to do more than scratch the surface of "hard" security issues such as military recruitment, subversive propaganda, power struggle between rebel factions or infiltration of combatants. What is happening around the camps, in the highly volatile border area, is largely beyond their grasp - though it is theoretically within their purview.

Some of these constraints can probably be addressed through the injection of additional resources and a better organisation of work including through the assistance of UNHCR Liaison Officers. There are, however, inherent limitations in the current package's exclusive focus on policing the camp areas. For one thing, too many police may become a problem, rather than a solution. In addition, police work is relatively ineffective because it is carried out in isolation from equally crucial control systems. No linkage is foreseen, for instance, with the Tanzanian military, whose role it is to monitor security along the border. Perhaps more importantly, the police's efforts to detect subversion find no credible "outlet" in the areas of prosecution, punishment and separation. This critical set of issues is developed in the following section.

Separating out combatants and/or armed persons

Imaginative as they may be, criminal or immigration law approaches suffer from two major - indeed, irreparable - shortcomings. Firstly, and practically, penalisation of subversive activities leads, literally, nowhere, but to prison (for a while), then back to camps. The deterrent effect of police arrest and prosecution is nil if those found guilty of acts incompatible with their refugee status, to quote again from the joint assessment mission report, are eventually released back into refugee camps.

The second shortcoming is of a more conceptual nature, and goes to the heart of the "separation" issue. Penalisation of subversive activities does, in fact, miss the point, for the basic distinction to be made is not between law-abiding and delinquent refugees, but between refugees and non-refugees. An individual who shows and/or expresses a determination to participate actively in armed conflict does not fall within the scope of the obligations under which refugees are protected. When this individual is found on the territory of a non-belligerent, neutral state, he should not be treated according to refugee law standards. This does not mean that the receiving State cannot extend humanitarian protection. Applying International Humanitarian Law standards by analogy, ICRC has recommended that such cases be treated as combatants placed hors combat, i.e., that they be interned until hostilities come to an end in the neighbouring country.17 Needless to say, UNHCR insists that the place of internment should not be a refugee camp.

It took UNHCR more than a year to convince Ministry of Home Affairs officials that separation out of refugee camps could only mean transfer of the "separated" elements to a separate internment facility. One such facility was eventually identified in Mwisa (Kagera region) and some forty Burundian combatants were sent there during the first half of 1999. A few months later, however, all but a few had absconded from the facility. As of preparing this paper, UNHCR is resuming negotiations with the government of Tanzania on the use of Mwisa and possibly other internment facilities.

Separation and internment, besides being the only legally correct response to the militarization of camps and camp areas, have considerable symbolic value in affirming the neutrality of the receiving state. For this reason, these measures are highly beneficial to refugees' security, even if they affect only a few hundred individuals, insofar as they contribute to restoring trust between neighbouring states and relieving international tension.

Attacking intermediate causes

To address this situation head-on is not easy, and probably not advisable, for a humanitarian organisation. Neither the refugees nor the government of Tanzania expect UNHCR to foster a debate in the camps about the comparative advantages of military action and political negotiation in effecting the desired changes in Burundi. UNHCR could, nonetheless, do more than it has done so far to get its humanitarian concerns through to those parties that are known to exercise the most militant influence on the refugee populations. With one notable exception, that is, the Jean-Bosco wing of FDD, those parties have established a respectable presence around the negotiating table at Arusha. This makes it easier for UNHCR, preferably in unison with the peace process facilitation team and/or the government of Tanzania, to voice its concerns and explain its principled position, inter alia on the issue of separation and internment.

It would be naïve to imagine that such a representation is guaranteed to cause radical Hutu opposition groups to leave the refugees alone. After all, the refugees are potentially their main power base. The humanitarian principles for which UNHCR stands can, however, retain or regain credibility through a transparent statement of the problem to those primarily responsible for it. On specific issues of particular concern, UNHCR may even hope to influence the armed opposition's practice.

A case in point is the recruitment of refugee boys and girls into the ranks of the guerrillas, of which there is sufficient evidence. Refugee youngsters are easy prey to political leaders proposing a military adventure. Post-primary education is not part of the regular assistance programme in refugee camps. Whatever activities exist in this critical area of adolescent development are organised by the community itself, with extremely limited means and no formal recognition by either Burundi or Tanzania. Non-governmental organisations offer some vocational training, support cultural and recreational activities, and assist a small fringe of resourceful refugees in setting up small businesses.

To most refugee adolescents, however, these opportunities appear as little more than make-up on the face of despair. The better years of their youth are irremediably wasted. Frustration builds up, leading to alcoholism, sexual and domestic violence, and, for some, a willingness to take up arms. The armed struggle is not only an outlet for their anger, it is also, in their eyes, the only way they can influence the course of their own destiny. It is clear to all agencies working in the Kigoma camps that much more attention must be paid to the needs of adolescent refugees - for all sorts of reasons, including the security risks entailed by idleness.

There are, unfortunately, very serious constraints to the development of "meaningful activities" for adolescents in a camp environment. To make things worse, those few initiatives that seem able to make a little difference have suffered from chronic under-funding.

Giving peace a chance

This is a tall order, but not "mission impossible". The Arusha process, in spite of its weaknesses, offers a window of opportunity, or at the very least a forum, for the refugees' voice to be heard. Since the middle of 1998, UNHCR has made it one of its main objectives to end the refugees' marginalisation vis-à-vis the peace process. The visit by the late Mwalimu Nyerere to the camps in April 199918 and the subsequent participation of refugee "resource persons" in the work of one of the Arusha committees19, symbolically were important steps in this direction. A positive side-effect has been a concession by Tanzanian authorities that refugee meetings centred around "peace" issues are not prohibited political activities - a significant contribution to a definition of subversion a contrario. The next challenge consists in organising a veritable dialogue on peace among the refugees, with structural links to both the internal and external peace processes. The facilitation of this multi-faceted dialogue should probably be entrusted to a Tanzanian or international non-governmental organisation with experience in conflict management and mediation.

Tentative conclusions

Refugee camps may be regarded as convenient settings for the channelling of aid to large numbers of needy displaced people. At the same time, the closed environment they provide is particularly conducive to exploitation and manipulation. The case study of Burundian refugees in Kigoma demonstrates yet again how stakeholders in a political and military confrontation can take advantage of the more perverse side-effects of camp life. These include the refugees' sense of frustration, loss and resentment; their marginalisation vis-à-vis domestic political processes; and their tremendous dependency on the outside world, not only for economic survival, but also for information and empowerment.

International law is not, as some would probably suggest, blind to this reality. Existing legal standards are neither empty nor useless. They take due account of the risks involved in granting asylum, and of the even greater risks involved in letting the granting of asylum become a source of friction between states. UNHCR and other concerned actors should advocate more firmly the basic "security package" contained, in particular, in the 1969 OAU Convention20, including the "reasonable distance of camps from the border" standard. In Tanzania, as elsewhere in Africa, experience has shown that the OAU Convention principles are a consistent whole, from which states cannot just pick and choose without serious risk. International humanitarian law - the law of armed conflicts provides relevant complementary standards - including with regard to separation and internment of combatants.

Legal standards cannot, however, play any meaningful role in an environment in which the problem they seek to address is occulted. The operation of international agreements - indeed, of any agreement - is predicated on the good faith of contracting parties. Governments must be led to understand that it is not in their interest - much less in the refugees' interest - to deny security problems, or to present them as a failure of humanitarian relief programmes. On this class of questions, sensitive as they may be, governments should be more receptive to UNHCR's "friendly advice". Likewise, humanitarian arguments must be presented squarely and without fear to non-state entities as well.

The most obvious lesson one can derive from the Tanzanian experiment so far is that the lower steps of the "ladder of options" proposed by the High Commissioner for Refugees are arduous enough. The "soft option" is, in fact, a full set of options. While some measures belong in the traditional security/ law and order domain, it would be irresponsible to claim that repression and control alone can enforce neutrality among refugee populations that are subjected to strong contrary pressure from various sources. These measures must, therefore, be supplemented by serious efforts to attenuate intermediate causes of the refugees' inclination towards subversion. Most importantly, every opportunity must be seized to involve the refugees qua refugees in a constructive dialogue on peace21.

ENDNOTES

1. The views expressed in the paper are personal to the author and do not necessarily reflect the official position of UNHCR or the United Nations.
2. See text under the sub-heading "Background" below.
3. During 1994, north-western Tanzania became host to over
   600 000 Rwandan refugees who fled their country following the death of the Presidents of Rwanda and Burundi in Rwandan air space on 6 April 1994, the ensuing genocide and the advance of the Rwandan Patriotic Front. An estimated 170 000 refugees arrived in Ngara district on 28-29 April 1994 alone. Following the mass influx of Rwandan refugees and the continuing, parallel flow of asylum seekers from Burundi, the historically liberal sentiments of the Tanzanian policy makers, and to some extent of the Tanzanian public, towards refugees were gradually changing. The large presence of refugees was blamed for numerous negative impacts, chiefly the increase in crime rates in the refugee hosting areas and massive environmental degradation. Tanzania claimed that its resources had been over-stretched and the level of insecurity on its western frontier had become untenable. The government also declared publicly its intention to pursue the repatriation of all refugees within its borders. UNHCR assisted the Government in the formulation of a plan to promote voluntary return, however very few Rwandan refugees chose to return. Encouraged by massive repatriation from Zaire (mostly by force majeure), the Tanzanian authorities announced 31 December 1996 as the deadline for the repatriation of all Rwandan refugees. On 12 December, camp leaders in the Ngara area began to move refugees away from the border and further into Tanzania so as to maintain their control over the exiled population. In response, the Tanzanian army forced the refugees to turn around and "re-directed" them towards Rwanda. Hundreds of thousands of refugees were taken to the Rwandan border during the next few days. Those who managed to flee into the surrounding countryside were rounded up and trucked back to Rwanda under military escort. In all, an estimated 483 000 refugees were returned from Tanzania. See UNHCR, 1997. The State of the World's Refugees: A Humanitarian Agenda, OUP: London, p 20; Jeff Crisp/Elizabeth Tan, 1998. "The Refugee Crisis in the Great Lakes region of Africa", Refugee Survey Quarterly, Vol. 17, No. 2; Bonaventure Rutinwa, 1996. "The Tanzanian Government's Response to the Rwandan Emergency", Journal of Refugee Studies Vol. 9, No. 3, p. 295 ff.; UN Economic and Social Council, Report of the UN High Commissioner for Refugees, 2 May 1997; Amnesty International, 1997. "Democratic Republic of the Congo - Great Lakes Region: Still in need of protection: repatriation,refoulement and the safety of refugees and the internally displaced", AFR/02/07/97
4. See footnote 3, above.
5. See discussion below under "Tensions in the relations between Burundi and Tanzania".
6. As used in this paper, the "Hutu cause" refers to a complex set of political and ideological motivations, which are present in various degrees among the Burundian refugee population in Tanzania. In its more radical form, the "cause" includes the affirmation of Hutus as a nation and the legitimacy of violent means to ensure return of the exiled Hutu population to their homeland.
7. In what was a dramatic switch in policy, the government of Tanzania closed its borders on 31 March 1995, saying "enough is enough. Let us tell the refugees the time has come for them to return home and no more should come". See, for a more detailed analysis, Bonaventure Rutinwa, op.cit.,p. 291-295.
8. See Bonaventure Rutinwa, op cit note 7 at p. 297. See also the related study by J.C. Mwakasege, "The Impact of Refugees on Host Communities: The case of Kasulu, Ngara and Karagwe" a Report of an Oxfam-sponsored Research Project, presented at the International Workshop on Refugee Crisis in the Great Lakes Region, Arusha, Tanzania, 16-19 August 1995.
9. The National Council for the Defence of Democracy (CNDD) and its armed branch, the Forces for the Defence of Democracy (FDD) were formed in September 1994 by the former (Hutu) Interior Minister of Burundi, Leonard Nyangoma. The CNDD/FDD has come to the fore in the Kigoma region particularly since the 1996 arrival of rebels from their disbanded bases in eastern Zaire. FDD rebels have also been fighting on the side of DRC President Laurent-Desire Kabila since hostilities broke out again in the Kivus in August 1998. In February 1998, the FDD rejected Leonard Nyangoma as president and opened the way for the rise of the man in charge of military operations, Jean-Bosco Ndiyikngurukiye. See Howard Adelman/Astri Suhrke: The path of genocide -- the Rwanda crisis from Uganda to Zaire, Transaction Publishers New Brunswick (U.S.A.) and London (U.K.) 1999, p. 338; International Crisis Group (ICG): "Burundi's Peace Process: The Road from Arusha", 1998.
10. In October 1997, the Tanzanian government ordered the army to round up all foreigners living outside refugee camps, asserting that this was necessary to protect Tanzanian citizens and refugees living close to the border with Burundi. It also put forward the argument that if these foreigners were bona fide refugees that they should be held in refugee camps funded by the international community regardless of their length of stay in Tanzania, the level of integration in the Tanzanian village communities, or humanitarian factors. The refugees and migrants were given the "choice" of being forced back to their country of origin or relocated to the refugee camps. Although Congolese and Rwandans were also affected, the bulk of the refugees and migrants rounded up were Burundian, including a large number of "old caseload" Burundian refugees who were uprooted after having lived peacefully on settlements allocated to them by the Tanzanian government as far back as the 1970. For the most part, those rounded up were sent to the refugee camps; UNHCR estimates that of the approximately 18 000 refugees who were brought to refugee camps in the Kigoma region, roughly half were Burundian. For more details on the experience of Burundian refugees see "In the name of security - Forced Round-ups of Refugees in Tanzania", Human Rights Watch Vol. 11, No. 4 (A) July 1999, p. 15.
11. Some Congolese were also involved in this incident. However, their case is not discussed in this paper.
12. Reference is made here to a "ladder of options" which might be employed for the establishment of an international mechanism to assist host governments in maintaining the security and neutrality of refugee camps and settlements. Although the primary responsibility for ensuring the civilian and neutral character of refugee camps rests with the host states, they might be either unable or unwilling to take the necessary measures. However, together with the international community, they must ensure that a number of security-related tasks are undertaken, like maintaining law and order in refugee camps and disarming exiled groups who have access to arms. The ladder of option provides a framework within which the contribution of each actor can be identified and elaborated. Due to the great diversity in the nature and degree of the security problems, it recommends different measures, ranging from relatively modest to more robust options, that should not be regarded as mutually exclusive. As first steps, it is suggested to implement preventive measures and to build on existing national law enforcement mechanisms, such as training and support to national police and the strengthening of national judicial systems. The next options are successively the deployment of international observers, international support to national security forces, and the deployment of international police forces. The "hard" options include the deployment of regional military forces with or without UN endorsement, and the deployment of international military forces according to Chapter VI and VII, respectively, of the UN Charter. A good summary of the "ladder of options" is presented in: UNHCR, "The security, and civilian and humanitarian character of refugee camps and settlements", Executive Committee of the High Commissioner's Programme, Standing Committee (14th meeting), 14 January 1999.
13. See Memorandum of Understanding under the Programme of the United Nations High Commissioner for Refugees between the Ministry of Home Affairs (MHA) and the United Nations High Commissioner for Refugees, signed on 19 November 1998 by UNHCR and on 5 February 1999 by the MHA. An updated Memorandum of Understanding was signed on 28 September 1999.
14. Negotiations between the Government of Zaire and UNHCR, in contrast, led to the signature of an Aide-Mêmoire on 27 January 1995. In this document, they agreed to establish a security force aiming at improving public order in the refugee camps, preventing intimidation and violence designed to deter candidates for voluntary return; protecting humanitarian facilities and personnel and providing escorts from the camps as far as the Rwanda borders for convoys of refugees opting for voluntary repatriation. In order to attain these four objectives, the Government of Zaire undertook to deploy 1 500 security officers (police or military), 1 000 in northern and 500 in southern Kivu. For its part, UNHCR undertook to pay the Zairian personnel three dollars a day and to provide uniforms, the necessary resources for establishing two command centres and 10 camps, vehicles as well as radio and office equipment. In addition, it was agreed that UNHCR would provide a liaison group of security advisers, responsible for liaison between UNHCR and the Zairian contingent called the Zairian Camp Security Contingent (ZCSC) and provide support in training, logistics, telecommunications and administration. UNHCR-produced "Standard Operation Procedures" contained specific instructions and described in detail the practical modalities of the duties to be performed by the Zairian contingent.
15. See discussion above relating to "The semantics of militarisation".
16. See Section 12 of the Tanzania Refugees Control Act 1965 (Requirement to reside in reception area or refugee settlement): "(1) The competent authority may (a) by order, require any refugee to whom this section applies who is within his area to reside within a reception area or refugee settlement, whether such reception area of refugee settlement is within such competent authority's area or not; (b) require any refugee to whom this section applies who is within a reception area or refugee settlement within such competent authority's area to remove to and reside in some other place being a reception area or refugee settlement, whether such other place is within such competent authority's area or not. (2) Any refugee to whom an order made under this section applies who (a) fails to take steps forthwith to comply with such order; or (b) fails to move to or take up residence in a reception area or refugee settlement in accordance with such order with reasonable despatch; or (c) having arrived at a reception area or a refugee settlement in pursuance of such order, leaves or attempts to leave such area or settlement except in pursuance of some other order made under this section, shall, unless he is in possession of a permit issued in that behalf under subsection (3), be guilty of an offence against this Act. [...]". In December 1998, the National Assembly of Tanzania passed a new refugee law, the "Refugees Act, 1998" which superseded the 1965 Refugee Control Act. It was signed by President Mkapa on 24 January 1998 and became effective on 15 April 1999. The section in question is now section 17 "(1) The competent authority in consultation with the Minister or the Director may by order, require any asylum seeker or refugee or group or category of refugees to whom this section applies who is within his area to reside within a designated area whether or not such area is within the jurisdiction of competent authority. (2) The competent authority in consultation with the Minister, or the Director may require any asylum seeker or refugee or group or category of refugees to whom this section applies who is within a designated area within such competent authority's area to move to or reside in any other designated area whether within such competent authority's area or not. (3) Any asylum seeker or refugee to whom an order made under this section applies who (a) fails to comply with such order; or (b) fails to move to or take up resident in a designated area in accordance with such order within reasonable time; or having arrived at a designated area, in pursuance of such order, leaves or attempts to leave such area, except in pursuance of some other order or permit made under this section, shall be guilty of an offence against this Act. [...] (5) (a) No asylum seeker or refugee shall be allowed to leave a designated area as directed under this section unless he has sought and obtained a permit from Director or Settlement Officer as the case may be, and, subject to such terms and conditions as the Director or a Settlement Officer may prescribe in the permit."
17. Authority for the State to intern individuals taking actively part in the armed conflict can be derived, by analogy, from Article 11 of the Hague Convention (V) respecting the Rights and Duties of Neutral Powers and Persons in Case of War on Land of 18 October 1907, which stipulates that "A neutral Power which receives on its territory troops belonging to the belligerent armies shall intern them, as far as possible, at a distance from the theatre of war. It may keep them in camps and even confine them in fortresses or in places set apart for this purpose. It shall decide whether officers can be left at liberty on giving their parole not to leave the neutral territory without permission." Needless to say, the State must not allow the forming or recruiting of corps of combatants on its territory (Articles 4 and 5).
18. Following the March session of the Arusha peace talks, Facilitator (and former President of Tanzania) Julius K. Nyerere visited Burundian refugee camps in the Kigoma and Kagera regions. He spoke directly with the refugees about his hopes for a lasting peace in Burundi, and he made a vibrant call for ethnic tolerance and forgiveness. Nyerere also voiced support for direct refugee contributions to the talks, and more specifically within Committee IV on Reconstruction and Development (See footnote 20. The Tanzanian Government and UNHCR permitted and assisted a refugee delegation to be present at the following sessions of peace negotiations in Arusha, so that they could report back on the progress of the talks to the larger refugee population. See ICG: "Burundi's Peace Process: The Road from Arusha", 1998; ICG: "Burundian Refugees in Tanzania: The Key Factor to the Burundi Peace Process", Central Africa Report No.12, November 1999.
19. Since June 1998, the parties to the Burundi conflict started peace talks in Arusha under the auspices of former Tanzanian President Julius Nyerere. At the end of a week of bilateral consultations with the mediating team in the first round of negotiations from 15 to 21 June, the parties signed a joint declaration and listed the five committees to be set up and the themes on which each would negotiate: Committee I dealing with "the nature of the conflict" and covering the problems of genocide and exclusion; Committee II on "democracy and good governance" dealing with the transitional institutions, the institutional, constitutional, judiciary and administrative systems and questions regarding justice and impunity; Committee III on "peace and security for all" covering the cessation of hostilities, the terms of a permanent cease-fire and the question of security forces; Committee IV on "reconstruction and economic development" dealing also with refugees and displaced persons in that it was specifically tasked to find a solution to the refugee plight and allow them to return safely and to be restored in their citizen rights; finally, Committee V dealing with the "guarantees for the application of a peace agreement". See ICG: "Burundi's Peace Process: The Road from Arusha", 1998. Following J. Nyerere's death in November 1999, the role of Facilitator of the Burundi peace negotiations was vested upon Dr. Nelson Mandela.
20. OAU Convention Governing the Specific Aspects of Refugee Problems in Africa, adopted in Addis Ababa on 10 September 1969, UNTS No.14691. Articles II ("Asylum") and III ("Prohibition of subversive activities") are particularly relevant to considerations of State, as well as refugee, security.
21. On this last aspect, which is more a promise than a reality as yet in the Burundian context, important lessons can be learned from refugee and returnee programmes in other parts of the world. Guatemalan refugees in Mexico in the late 1980's "didn't wait for peace, they helped forge it". See Paula Worby, "Lessons learned from UNHCR's involvement in the Guatemala refugee repatriation and reintegration programme (1987-1999)", UNHCR Regional Bureau for the Americas and Evaluation and Policy Analysis Unit, December 1999.