Since the end of the war in April 2002 the official process of demobilisation and disarmament has come to a close and the emphasis shifted to the reintegration of almost 500,000 National Union for the Total Independence of Angola (UNITA) ex-combatants and family members. While the process is largely judged a success by the Angolan government and the international community – at least in having avoided predictions of widespread insecurity and broken the patterns of the past – its implementation reflects the government's military and political advantage and has failed at times to pay sufficient attention to the needs of the ex-combatants themselves. It has given insufficient attention to the 'reintegration' component of the Demobilisation, Disarmament and Reintegration (DDR) process, the absorption of ex-combatants into society, and their transformation into civilians. It is this process that will be crucial for the consolidation of peace over the longer term.

Although the Luena Memorandum was not a new peace accord but a revival of the Lusaka Protocol (itself built on the Bicesse Accords), demobilisation and disarmament in 2002-03 differed significantly from both Bicesse and Lusaka. The new framework again contained provision for the quartering and demobilisation of UNITA's military forces, the integration of a number of UNITA personnel into the FAA (Angolan Armed Forces) and Police, and the demobilisation and reintegration of the remaining combatants. The number of UNITA personnel for integration into the FAA was, however, limited to around 5,000, to be accommodated in accordance with existing structural vacancies. Whereas the Lusaka Protocol had stipulated a number "to be agreed between the Angolan Government and UNITA for the composition of FAA" and under Bicesse equal numbers of troops had been stipulated for both sides, Luena represented only the conclusion of the process of integration of the two armed forces from Bicesse onwards. This process allowed the government to consolidate its advantage; by prioritising those deemed a potential 'threat' to the peace process, senior generals and officers could be 'bought off' and the rank and file definitively separated from their leaders – an effective strategy on the part of the government.

In many ways, the DDR process under Luena thus reflected the nature of the ending of the war more than the original Lusaka Protocol – a victory by the Angolan government rather than a negotiated settlement. Whereas Lusaka had called for increased UN participation, on this occasion the Angolan government stressed its determination to carry out the DDR process alone. It took full responsibility for administering and funding the demobilisation and disarmament processes (US$187m by January 2004) with no provision for formal third party monitoring and verification, as there had been under the Lusaka Protocol, and notably no presence of the UN or other international bodies within the Quartering Areas. A new body, the Joint Military Commission was created to oversee the implementation of Luena, presided over by the government and comprising military representatives of the government and UNITA, with military observers from the UN and Troika permitted. A new UN mission was only authorised by the Security Council in August 2002, after formal demobilisation was already complete. Although a Joint Commission comprising representatives of the government, UNITA, UN and the Troika was subsequently created (or resumed) to oversee completion of outstanding issues of the Lusaka Protocol, the Angolan government pushed for this to be wound up as soon as possible. It was disbanded in October 2002, despite some UNITA complaints that its work was unfinished and fears this would reduce incentives for the government to ensure the proper completion of DDR processes.

UNITA's armed forces were demobilised at an astonishingly fast rate. Initial plans were based on a UNITA estimate of 50,000 combatants, but this proved to be less than half the number eventually reporting to the Quartering and Family Areas (QFAs). Indeed, people continued to arrive in the QFAs even after the formal conclusion of the demobilisation process, and right until their eventual closure. The numbers placed additional strain on logistical and supply capacities, and prolonged registration and demobilisation from 80 days to around four months, also delaying reintegration activities. It is widely believed that many reporting had not been active in the last stages of the conflict, an assertion that is to some extent borne out by the low number of light weapons handed in by ex-combatants, around 30,000.

Demobilisation formally took place on 2 August 2002, by first integrating all former UNITA soldiers into the FAA, and then demobilising them. Ex-combatants were to receive 5 months back payments of salaries according to military rank, a US$100 reintegration allowance and a 'kit' of basic household items and tools, as well as full identity and demobilisation documentation.

Demobilisation support was made available only to ex-combatants, however, with women eligible to receive only humanitarian support as civilians (only around 0.4 per cent of the total number of ex-combatants registered were women). Child soldiers were not generally registered as combatants but mainly transferred to Family Areas, and as such are not eligible for official reintegration programmes. While separating them from adult combatants was in their interests, it made them particularly vulnerable if they did not have family structures to assist them.

The delivery of humanitarian aid created further complications for the reintegration of child soldiers and children separated from their families. Aid was distributed by family, according to the number of family members (by international agencies at least, government distributions seem to have been less orderly), meaning it was in a family's interest to take in any 'stray' children such as child soldiers or orphans. Once out of the Gathering Areas (GAs), however, this rationale no longer held, leading to a number of unaccompanied children being 'lost' from the system. NGOs and child protection agencies are working with the Ministry for Social Assistance and Reinsertion, a civilian body, to try and correct this problem caused by a lack of long-term planning in what was essentially a military process.

The Quartering Areas themselves were established in broadly the same locations as under the Lusaka Protocol, with houses, meeting centres, schools and hospitals built by the ex-combatants themselves. Initially 27 were planned, increasing to 35, with an extra 7 satellite areas. They were generally divided into three sections: the Quartering Area, where ex-combatants were located; a separate though usually adjoining Family Area for women and dependents; and a further area housing primarily disabled ex-combatants and older people. Initially conditions were poor and levels of malnutrition frequently critical and even reaching famine levels in some areas. Assistance to combatants was the sole responsibility of the FAA and Angolan government, with the UN humanitarian agencies able to assist only families and dependents. Before providing any relief, however, the UN Office for the Coordination of Humanitarian Affairs (OCHA) went through a long process of negotiating access to the QFAs, leading to accusations of negligence against both OCHA and the government. Presence was eventually established as near as possible to the QFAs and in most places conditions soon stabilised. Indeed, visitors to the QFAs were frequently surprised by the level of order and tidiness, reflecting UNITA's renowned military discipline. Although rarely reported to the outside world, tensions within the QFAs were certainly present, fuelled by long delays between arrivals of food and supplies, confusions and irregularities in registration and demobilisation, frequent 'false alarms' of camp closure and general feelings of insecurity. Similarly there were reports of resentment from surrounding communities in some areas at the level of support that UNITA ex-combatants were perceived to be receiving.

The QFAs were renamed Gathering Areas in October 2002, to reflect the completion of the demobilisation process and the civilian status of inhabitants. Although in some GAs management of the military and non-military areas was separate, in practice the areas were not closely delineated and movement between them was common. Increasingly, they became settlements in their own right, with functioning markets, schools, (very basic) hospitals, and new arrivals as UNITA ex-combatants from other areas sought to locate their family members. This process was also encouraged by the distribution of seeds and tools for subsistence agriculture by some agencies and churches, a policy heavily debated for this very reason, with some focusing on short-term emergency needs, while others looked to the longer-term political and social reintegration of the country. It was feared that ex-combatants would be less motivated to return to their areas of origin and that mini 'UNITA enclaves' would be created. This, largely speaking, has not proved to be the case, although the continued cultivation of crops has slowed the return of ex-combatants and internally displaced people (IDPs) in some areas.

This was also a major fear of the government, and dates were set for the closure of the GAs, from October 2002 onwards. Persistent administrative delays made these deadlines impossible to meet, with the effect, if not the intention, of demoralising camp residents, who reported their frustration and powerlessness "in the hands of the government". Those who had received seeds and tools did not know whether to plant them in the GAs or wait until they had returned home. Those without identity documents could not leave even if they wanted to attempt the journey independently, and even short journeys to local markets could result in police harassment.

By mid 2003, the majority of the GAs were emptied. The first stage in the journey was generally to transit areas, which were often IDP camps that had sometimes been recently vacated or still had IDP populations living in them. Problems recurred here as ex-combatants were frequently forced to abandon belongings and goods they could not fit into the badly overcrowded planes and trucks. By early 2004, the majority were believed to have left these transit areas and returned to their 'areas of origin', or moved on to other destinations. Concentrations may exist around certain urban centres, and in temporary locations, either because of ongoing crop cultivation or the wait to rejoin family members once they are re-established. The number is uncertain however, and there may be future population movements following agricultural cycles. Furthermore, there may also be a partial reversal of this trend if ex-combatants perceive greater economic incentives and opportunities such as training programmes around urban areas.

While the number of arms handed in by UNITA ex-combatants was surprisingly low, this may, as already mentioned, reflect the severe depletion of UNITA's active fighting forces at the end of the war. It has been estimated that the weapons handed in represent around 90 per cent of the total possessed by UNITA, with the remaining 10 per cent in the hands of the civilian population. It was feared that the remaining weapons would quickly be turned to use in banditry and localised violence, but widespread insecurity has so far failed to materialise.

It should be remembered however that, while UNITA ex-combatants were fully disarmed, the civilian population remains highly armed, with the Angolan Police citing a figure of 3-4 million small arms and light weapons in the hands of civilians. To many people, possession of a weapon has become linked to personal security as well as being a sign of political allegiance. In March 2004, the government announced a plan for the disarmament of civilians, but with the exception of some civil society initiatives, little practical action has so far been taken. During the war, armed militia known as the 'civil defence' were formed in virtually any government-controlled area, armed and loosely directed by the FAA. Participation was virtually compulsory for young and middle-aged males, and refusal would be interpreted as lack of political support for the government, if not outright sympathy for UNITA. It is not clear what their role in peacetime will be. Demobilisation and reintegration plans do not provide any formal support for them, although they often feel themselves to have fought legitimately in defence of their homes and communities and thus equally deserving of assistance as the UNITA ex-combatants.

With demobilisation concluded, reintegration constitutes one of the biggest challenges to peacebuilding in Angola. Government plans for reintegration were not revealed until late in 2002, and then only partially. It had been hoped a World Bank funded 'Angola Demobilisation and Reintegration Programme' (ADRP) would become quickly operational, but negotiations stalled over its financial management. The ADRP was finally launched in April 2004, but disbursal of funds to implementing organisations is likely to be slow. Actual projects are unlikely to become active until later in 2004-05. This is expected to cover some 105,000 UNITA ex-combatants and 33,000 government soldiers (to be demobilised to reduce the size of the army) but excludes the 'old caseload' ex-combatants from Bicesse and Lusaka the government had initially hoped to include. In addition, other donors have committed to funding specific initiatives, including US$54m from USAID alone. There is a risk, however, of weak capacity to handle and coordinate these reintegration-focused projects, especially outside Luanda. Institutions and bodies established to handle resettlement, return and reintegration have a fairly weak presence in the provinces, and there is little effective coordination between the activities of the various national and international NGOs. Reintegration is also taking place in the context of the return of around 3-4 million IDPs and almost half a million refugees, often to areas with weak state administrative capacity, and serious constraints on access to areas of resettlement.

At present, it is all but impossible to know how the reintegration process is going. Although the general consensus is that the situation is relatively stable, and that 'worst case scenario' predictions of widespread banditry and insecurity have not emerged, NGOs such as Human Rights Watch have identified abuses and human rights violations, and the 'norms' on return and resettlement are rarely met. In addition, more than two years since Luena, which promised access to vocational training and economic support, few programmes have actually materialised and it appears that neither central nor provincial governments have the means or will to support these over the longer-term. Yet reintegration is of crucial importance, not only among ex-combatants themselves – who may or may not live up to their reputation as potential threats to peace – but to the effective consolidation of peace among communities and society more broadly. While UNITA ex-combatants should not be seen to benefit unduly from their participation in the war, neither should they become a burden on their families and communities and a source of economic and social tensions. Rather reintegration presupposes their integration into a community and the development of that community as a whole, not as individuals or 'target groups'. Balancing the two is certainly a challenge, but one which must be addressed.