Cover Image
close this bookMedical Assistance to Self-settled Refugees (Tropical Institute Antwerp, 1998)
close this folder3. The refugee-crisis: between self-reliance and pragmatic assistance
close this folderLate arrivals: the subsequent minor waves, 1992-95
View the documentCross-border movements in both directions
View the documentEarly versus late arrivals
View the documentNew agencies and changes in PARLS

Cross-border movements in both directions

Towards the end of 1992, there was still no progress in the peace process in Liberia. Internal strife within NPFL and ULIMO resulted in their further splitting up in different factions and new upsurges of fighting between them. In September 1994, this culminated in a ULIMO assault on the headquarters of NPFL in Gbarnga, and its subsequent recapture by NPFL, followed by a major offensive of NPFL in ULIMO-controlled territory. In Sierra Leone, the situation was hardly any better. The conflict spread over large areas, and forced many people to leave their homes, and either concentrate around cities, or flee to Guinea.

During this period, the level of violence was increasing, giving rise to some of the most horrifying human rights violations documented in recent history,4,11 during which, for example, whole groups of peasant women had both hands cut off.5,12 All factions widely practised a scorched earth strategy. This disrupted food security inside Liberia and Sierra Leone. At the same time, humanitarian agencies faced serious difficulties bringing relief and there were places with real famine.

The refugees who had arrived in Guinea in 1990-91 fared relatively well. There were no 'lacrimogenic' situations of extreme destitution nor famine. Whenever the situation allowed, refugees returned to their country of origin - temporarily or permanently. When the security situation improved, the movement went mainly towards the country of origin. When insecurity increased, it was the other way round. The feeling was that 'refugees were integrated'. There was little media presence and donors became less interested in these refugees. Moreover, lack of reliable refugee registration and fraud with food aid were perceived as major problems. Therefore, donors were pushing UNHCR and World Food Programme (WFP) to improve control mechanisms and to reduce food aid.

A SERIES OF MINOR WAVES. As the situation inside Liberia and Sierra Leone deteriorated, new refugees started trickling into Guinea.* The pattern was different from the previous waves of refugees. Now refugees arrived in smaller groups, as a series of minor waves. For those arriving at this late moment, fleeing to Guinea was not a first choice. They had tried to hang on as long as possible. When asked why they had not left their country earlier, they often answered that they lived far from the border, and first had tried to cope somewhere closer to home; that they had never been to and did not know anybody in Guinea, or that they did not know where to go in Guinea. Others stressed that fleeing outside the country would jeopardise their future inside their home country once the problems would be solved. The result was that they first tried to continue living in their villages, facing hardship, or moved around inside Liberia or Sierra Leone. It was only when these options became really impossible that they decided to flee to Guinea.

[* This document is focusing on refugee movements across international borders, but for the people involved in this conflict, these borders constituted only part of the picture. Each time one of the factions gained or lost control over a territory, the areas where people could reside in relative security changed. Invariably this resulted in within-country population movements.]

Most refugees arriving in 1992-95 had thus already been internally displaced. Some had moved several times. Most had suffered extreme hardship, close to forced labour or even slavery, inside their country. Consequently, more and more refugees were malnourished and sick upon arrival, without belongings, or even clothing. Family units were often split, with many families headed by women. Many of the refugees had no place to go to in Guinea, were exhausted on arrival and had little energy left to develop creative coping mechanisms. This fundamental difference between early and late arrivals is a clear illustration of Kunz' variety of 'refugee waves and vintages',13,14 and has also been documented in South Sudan15 and in East Sudan.16,17

One can distinguish more than ten different waves with some 150,000 new refugees during 1992-95 (Figure 6). Most arrived in the prefectures of Yomou, Macenta and Guu, which already hosted the highest number of refugees. When the conflict spread over larger areas of Sierra Leone in 1995, refugees also arrived in new areas in Guinea; some 9,000 refugees settled in Kissidougou and another 24,000 settled in Forriah, along the coast. The most important of these minor waves were those in Guu (June 1993), in Yomou (September 1994) and in Forriah (January 1995). These illustrate the changing nature of the refugee population, as well as the changing response of the relief system (Table 9).

New refugees in Guu, June 1993

In June 1993, several groups of refugees arrived in Guu prefecture and settled down just across the border. They were in a bad shape, destitute and without any belongings. Soon after arrival, they received emergency food and medical care. UNHCR established several small camps, for between 1,000 and 3,000 people each, far away from the border (e.g. at Nyaedou, Fandouyema and Boodou*). UNHCR negotiated access to land for the refugees in Nyaedou and Boodou and assisted them with its exploitation. UNHCR made settlement in the new camps a condition for registration and further assistance. Despite this, many refugees preferred self-settlement without assistance, joining the ranks of previous refugees among who they judged having better chances of coping. Moreover, many refugees who officially registered in camps did not permanently reside there, but moved out in search of employment.

[* Boodou camp was created to settle former guerrilla fighters turned refugees, far from the border and isolated from the rest of the refugees, who were often hostile against them. Later, also 'civilian' refugees were settled in Boodou.]

From Gbarnga to Noonah, Yomou, September - December 1994

After the assault by ULIMO on the headquarters of NPFL at Gbarnga, some 27,000 new refugees arrived in Yomou prefecture in September 1994. Many, including adults, were severely malnourished. They constituted a very diverse group of former urban dwellers from different ethnic groups. At short notice, UNHCR and WFP made emergency food supplies available and constructed transit camps in DieckBignamou and Betha. These transit camps consisted of communal shelters of 15 by 7 metres to house up to 200 persons each, and were designed to be temporary. It was difficult to maintain an acceptable level of hygiene in such environment and soon a severe cholera epidemic broke out in the Dieckransit camp.

In November 1994, UNHCR tried to find a more suitable solution and designed a new camp in Noonah. A typical grid camp was laid out and men from the transit camps were moved by trucks to Noonah to clear the bush and to construct houses with local materials and plastic sheeting. The new camp remained almost empty until UNHCR made living there a precondition for registration. Nevertheless, many refugees refused to move (Box 1). This opting out from the relief system has also been observed in other refugee situations.15

From Kambia to Forriah, January 1995

In January 1995, an attack by the Revolutionary United Front (RUF) on Kambia forced some 24,000 Sierra Leoneans to flee to Forriah, near Conakry. Although most of the refugees were in good health, few carried any provisions. The refugees settled in areas inhabited by their kinsmen, with whom they had maintained close relations. This influx closely resembled the first and third waves of 1990, with rural refugees in good health and coping relatively well.

In June 1995 UNHCR encouraged these refugees in Forriah to move to camps. The refugees were told that food and medical assistance would only be provided there. By November 1995, most refugees had moved, although they had previously lived among the Guineans without major problems. This policy was thus in line with the more interventionist relief approach in the Forest Region, although the needs of these refugees did not require it.

Table 7: Overview of three minor waves of late arrivals, 1992-95

Provoking event

Estimated number and/or status at arrival

Settlement area in Guinea and mode of settling in

New refugees in Guu, June 1993

Fighting in Kailahun area (Sierra Leone) & Loffa county (Liberia)

In very poor condition, with many malnourished

Arrived in border areas of Guu; soon moved by UNHCR to small camps (e.g. Nyaedou and Fandouyema) with good access to land

From Gbarnga to Noonah, Yomou, September-December 1994

Attack by ULIMO on NPFL headquarters at Gbarnga

27,000 refugees, generally in poor condition

First in transit camps: DieckBignamou and Betha (Yomou); in January 1995, moved by UNHCR to Noonah with poor access to land (only some 8,000 accepted, rest self-settled, but unassisted)

From Kambia to Forriah, January 1995

Attack on Kambia by RUF

24,000 refugees, in good health

Forriah, self-settlement among kinsmen, later 'encouraged' by UNHCR to move to camps

Early versus late arrivals

The refugees who arrived in 1990-91 were fundamentally different from those who came in 1992-95. The needs of the late arrivals were more Important while the efficiency of their own coping mechanisms decreased. Tab compares the main characteristics of the early and the late arrivals.

Box I: Noonah refugee camp & informed consent to non-assistance

Noonah is a small Guinean village. Most of the forest around is part of the Sacred Forest, where only people initiated through local rites may enter. UNHCR established Noonah camp to resettle the new refugees. Only 8,000 refugees accepted the move, as the camp offered hardly any access to farmland or labour opportunities. The remainder, some 19,000, refused to move to Noonah, in full knowledge of the fact that by doing so they would not be entitled to free food or free medical care. They became officially unregistered refugees. This is a case of 'informed consent to non-assistance' because the pre-conditions unilaterally imposed by UNHCR were not acceptable to the refugees. This refusal was partly due to the bad track record the relief system had in the eyes of many refugees. The refugees did not trust UNHCR and its implementing agencies when they promised that refugees would get full relief rations in Noonah camp. Previously, they had indeed not been capable to supply food as scheduled. Many refugees judged rightly that their chances to develop economic self-sufficiency in Noonah camp were slim, and preferred to opt out and rely on themselves rather than on the relief system.

UNHCR was so strict because it could not distinguish between 'real new refugees' and 'false new refugees' ('older' refugees trying to register twice or Guineans trying to get registered as refugees). The physical separation of the new refugees from the old ones was indeed a solution to this problem, as very few 'false new refugees' would ever accept to move to Noonah camp. However, it also excluded a large number of 'real new refugees' from registration.

The refugees who opted out were probably those who had better chances of coping and becoming self-sufficient. Many who settled in Noonah camp moved back to where they had first settled, or into the Guinean villages around Noonah; they would only return to the camp on the days of food distribution.

During 1995, the refugees who refused to move to Noonah camp were proved right. Although Noonah camp was prioritised for food distributions, these were insufficient and malnutrition became highly prevalent. In 1995, the situation deteriorated in many areas, but in Noonah camp the situation was worse than elsewhere, and took longer to redress.

Table 8: Characteristics of the refugees and their reception, 1990-95

Early arrivals, 1990-91
Four major waves

Late arrivals, 1992-95
Subsequent minor waves

Nature of the refugees

When hostilities reached their area of residence, people fled to Guinea as a first choice. They lived close to the border and had ethnic links, often family, in Guinea.

When hostilities reached their area of residence, people first struggled to remain within their country. Only when this became impossible, they fled to Guinea (internally displaced, then refugees). Sometimes they included refugees who had returned to their country but had to flee again (refugees, returnees, and then re-refugees).

General condition of the refugees

Refugees were generally in good condition. Refugees arrived in relatively homogeneous groups of people of the same ethnic group.

Many refugees arrived malnourished, and in poor health. Many families were split before arrival. Refugees arrived in heterogeneous groups of mixed ethnicity that did not originally live together.

Characteristics of the reception in the host area

The refugees arrived in areas inhabited by relatives and where no refugees had yet arrived. The reception of the refugees by the host population was in general very generous.

The host population was already supporting large numbers of refugees. Kinship relations between newly arrived refugees and their hosts were weak or non-existent. The hosts often considered areas of arrival 'saturated' with refugees.

The aid system was not yet in place in the area of arrival and no registration of refugees had taken place previously.

The relief system was already in place and earlier arrivals had been registered. 'Registration = food aid' - logic was already installed.

Most refugees did not expect relief, nor did their hosts.

As internally displaced, some refugees had already received food aid. On arrival in Guinea they expected the 'international community' to take care of them. The local population and the refugees already present counted on aid from the relief system for the new refugees.

Mode of settlement in the host area

Self-settlement among host population:
refugees mixed with existing population in the border areas. After some weeks or months, they constructed their own houses, often spatially integrated in existing villages, or in paired villages.

UNHCR tried to keep old and new refugees separate. It prepared camp sites at a distance from the border, and new refugees had to settle there to be registered. The populations of these camps were mixed and were supported by the relief system. After some time, these camps often became 'ghost camps'. Vulnerable refugees and dependants often remained in the camps. When food distributions took place, refugees returned to the camps.

New agencies and changes in PARLS

The three examples described above show that the results of the 'new relief approach' were quite different from one place to another. In Guu, the new relief approach was reasonably effective, though undoubtedly more expensive than the previous one. In Noonah in Yomou, it was a complete failure (Box 1). In Forriah, resettling refugees in camps was unnecessary and costly.

But most of the new refugees needed more assistance and the relief system was ready to deliver it. Food aid and medical care quickly became available. All children were vaccinated against measles, anthropometric surveys were conducted and, when indicated, feeding centres were started. Clean water was also made available in all new settlement sites. During 1995, however, the relief system failed overall to supply reasonable quantities of food.18 Consequently, food insecurity increased and malnutrition among the refugees rose considerably. As an answer to this problem, UNHCR reinforced programmes for vulnerable refugees. Malnourished children were relatively easy to identify, but this was more difficult for female-headed families and unaccompanied elders within a highly mobile population. In an attempt to decrease irregularities during food distribution, food basket monitoring was started. At every distribution, checks were made to determine what were the real quantities of food received by the refugees. It soon became clear that even at the end of the distribution channel problems existed. UNHCR and WFP decided to change the 50-beneficiaries-ration-card system to a distribution at household level. This increased the reliability of the distributions for the refugees. Also, support for income-generating activities, mainly rice production, was stepped up. In previous years, this had consisted of distribution of agricultural tools and seeds. Later, improving access to land, mainly through exploitation of new swamps became the target, with better results.

This more interventionist relief approach also brought new actors on the scene. Up to 1993 PARLS was carried by UNHCR, WFP and NGOs already working in the Forest Region in the context of development programmes. Only the Red Cross, the Adventist Development and Relief Agency (ADRA) and the International Rescue Committee (IRC) had joined PARLS as new NGOs. During 1993-95, however, additional NGOs started operating, such as Action Contre la Faim, Jesuit Refugee Service and Eglise Protestante Evangque. Early 1996, GTZ took over PARLS from MSF in Guu, and from Oxfam in Forriah. GTZ later also replaced ADRA to transport food aid. In less than two years, GTZ became one of the main actors of PARLS.

Not only did the nature of the refugees evolve over time, so did the attitude of the hosts and the preparedness of the relief system. The role of the relief system increased. In terms of refugee livelihood, there was a shift from self-supporting to relying on assistance from outsiders. To a certain extent, this change in approach was an understandable response to the changing conditions of the refugees. But the change in policy was not necessarily appropriate, as illustrated earlier. Table 9 compares the response of the hosts and the relief system to the early and late arrivals.

Table 9: Response to the refugee waves, 1990-95

Early arrivals, 1990-91
Four major waves

Late arrivals, 1992-95
Subsequent minor waves

Attitude of the hosts and the relief system towards

Collective wishful thinking and generosity. Most people thought: 'This is a short-term problem, we have to help these refugees through this difficult period of a few months, after which they will return home'. Resources were pouring in from the donors to help the refugees.

Management approach. Many people thought: 'This problem is lasting longer than anyone could have expected', and 'Many refugees are misusing the aid system'. Donors started imposing conditions on better use of food aid. 'Old' refugees who had arrived in 1990-93 were assumed integrated and self-sufficient.

General approach of the relief system

Low-key approach: a limited relief system that followed the refugees. During the first wave, UNHCR was not yet present in Guinea. MOH and MSF took the lead with a low-key approach. UNHCR became fully operational only after 4 to 5 months.

The existing relief system led by UNHCR took the initiative, as the main actor. The relief system was already fully operational upon arrival of the refugees.

The registration system was lax during the first and second waves; and still quite liberal during the third and fourth waves. Distributions, once started, were poorly targeted. Large quantities of food were misappropriated. Food prices on the local market plummeted to an 'all-time low'.

Different actors had highlighted adverse effects of food aid. Consequently, UNHCR started control on registration and fraud. Food aid was decreased for old refugees. When new refugees arrived in the same areas, they seriously interfered with control and verification.

Not only the general attitude and approach to refugee assistance changed, also the technical content of the assistance evolved over time. Where the early arrivals had received a low-key slimmed assistance package, the late arrivals received a more comprehensive package (Table 10).

Table 10: Content of PARLS, 1990-95

Early arrivals, 1990-91
Four major waves

Late arrivals, 1992-95
Subsequent minor waves

Medical relief

At the onset basic curative care, measles vaccination, disease surveillance and nutritional monitoring were considered to be an appropriate package of relief activities.

Therapeutic feeding centres, a supplementary feeding programme and services by health animators were added to the package of relief activities.

Food aid

Refugees lived several weeks, even months without food aid, without serious consequences on their health.

Refugees received food aid rations from the first days or weeks on.

Water Supply

Improvements in village water supply (e.g. protection of existing shallow wells) were first attempted. New wells and boreholes with hand pumps were installed months later.

New wells and boreholes were dug very early in the camps.


Initially no assistance was given for the construction of shelter. Months later some plastic sheeting was distributed.

Assistance in the lay-out of camps and the construction of shelter was given from the onset.