Skip to content
Voluntary repatriation and the participation of Mauritanian refugees

Popularly known as “the events”, the 1989 violence between Mauritania and Senegal was triggered by a dispute over grazing rights in the Senegal River valley which forms the border. In Senegal, where many shopkeepers were Mauritanian, shops were looted and most Mauritanians were expelled to Mauritania. Retaliation and riots targeting black Mauritanians in the river valley and white Moors in Dakar ensued throughout the following month. Mauritania’s south is heavily populated by black African Fula/Toucouleur, Wolof, Soninké and Bambara peoples, while the northern Moorish (Arabo-Berber) population had long dominated the politics of the country.  

To stamp out this violence, the Mauritanian and Senegalese governments organised  flights to repatriate their respective citizens which ended in the forced exile of about 70,000 Mauritanian southerners to Senegal, despite most of them having no links to the country. These Mauritanian refugees would slowly trickle back into the country during the following years but some 20–30,000 remained in Senegalese refugee camps.

The voluntary repatriation programme

In March 2007, the newly-elected Mauritanian president demonstrated his government’s political will to repatriate and “rehabilitate the rights of the black Mauritanians who suffered from acts of violence”. In collaboration with the Mauritanian and Senegalese governments, UNHCR launched an appeal for funding to repatriate and reintegrate 24,000 refugees between August 2007 and December 2008. 

In June 2007, a Mauritanian inter-ministerial committee visited the River Valley to sensitise and listen to the population’s views on the repatriation programme. A refugee committee was created in order to facilitate government and UN access to the population and so that the refugees could speak with a single, unified voice.

With its members elected by 167 of the 284 heads of refugee sites in the River Valley, the committee’s diverging opinions reflected the differences that exist between the refugee groups. Some refugees felt that it would be harder to respect the interests of the refugees if there were divergences among the group’s leaders – and therefore more difficult to ensure a return under good conditions. There were even accusationsof possible corruption among the refugee leaders who would potentially be able to take advantage of their position as mediators between the refugee population and the international community.

Some representatives of the refugee community were present at the signing of the tripartite agreement[1] in November 2007 between Senegal, Mauritania and UNHCR. This agreement set out the signatories’ responsibilities for implementation of the voluntary repatriation programme. While it was understood that the governments of Senegal and Mauritania, together with UNHCR, would manage the programme and thus be responsible for its financial aspects, several refugees expressed regret over the lack of refugee participation in drawing up the agreement and indicated that the refugees should also have responsibilities to uphold with regard to the repatriation programme. As the chairman of the committee said: “There should be four parties involved. The refugees should have some responsibilities. [Under this agreement] they are not responsible for anything…”

Other representatives, although they had been invited to attend the signing, refused to attend as they had not received the agreement and its contents prior to the signing.

Refugee conditions for repatriation

At first glance, the Mauritanian refugee population seems to have played an active role in the organisation of their return home. It appears, however, that refugee participation took place only in the actual implementation, not at the decision-making stage.

The refugee community stipulated eleven conditions for their return, two of which were “full involvement in all stages of the process of repatriation” and a complete census of the refugees. Their full list of conditions was accepted after lengthy discussion during the national ‘Consultation on the voluntary return’ which took place in November 2007. However, meeting these conditions proved to be another matter.   

In autumn 2007, UNCHR organised a ‘profiling’ of the refugee population in the River Valley. This census of the refugee population – one of the conditions stipulated by the refugees – enabled UNHCR to determine the number of individuals who wanted to return and established their preferred return location. Mauritanian authorities were firstly aiming to repatriate those individuals returning to regions with better infrastructure and more capacity to receive the returnees. The census thus allowed UNHCR to sort the population into categories so as to facilitate the management of the repatriation programme. The information collected during the census also enabled Mauritanian authorities to verify the nationality of those individuals who wished to return; the tripartite agreement indicated that the census would serve as an identity document until “the returnees receive state documents which should be delivered by the national authorities within three months of their return” (Article 16).

It is evident, therefore, that this census would play an important role. However in December 2007 two refugee representatives claimed that the profiling did not take place consistently throughout the population and that some families were omitted from the census. When several refugee heads of villages pressed UNHCR officials on the matter, the officials merely played down the importance of the census.

In addition to the short time allowed for the census and the fact that only one month was allotted to the processing of the documents, the Chairman of the refugee committee suggested that UNHCR’s objective – promoted by the government – of repatriating 7,000 people before the end of 2007 was unrealistic particularly given the nature of the conditions stipulated by the refugee community for a successful return. However, as voluntary repatriation often occurs in a highly politically charged atmosphere, the refugee community does not necessarily have much leverage in making choices for its own future.  As a former Mauritanian military official suggested, a full and real involvement may have entailed bringing all parties together: “…before doing this census, one should have the Mauritanians, the refugees together with UNHCR and human rights officials, sit around a table and develop the census questionnaire.”

Meetings organised by UNHCR and local NGO OFADEC[2] involved the refugee heads of villages in decisions over logistical aspects of their return. As part of their efforts to ensure that the repatriation took place ‘with dignity’, an OFADEC officer explained to the refugee population during preliminary meetings that those families returning home would come together the night prior to the departure to eat, sleep and be together. The refugee heads of villages were asked to decide upon a location and build an appropriate shelter for this evening. In this way, concrete tasks were delegated to the refugee community.

Over the course of the preparation meetings, several refugee heads of villages expressed concern and a reluctance to participate in the repatriation programme due to the lack of information that they had received on the repatriation conditions and their reintegration into Mauritanian society. The attempt to include the refugee population therefore appeared superficial and insufficient. Given this lack of information, delegating the responsibility of building regrouping centres was perhaps more token participation than the real participation that the refugees had stipulated in their conditions for a successful repatriation.

Consultation days

A series of ‘national consultation days’ was organised in order to examine practical aspects of the implementation of the voluntary repatriation programme. These took place in November 2007 and were attended by 724 members of the Mauritanian government, representatives from UNHCR and, according to the chairman of the Committee who was present at the consultation days, only 17 refugee representatives. The strong government representation and the weak refugee representation did not go unnoticed by the refugee community.

The attendees of these consultation days were divided into groups to discuss the repatriation conditions, their integration upon return, the possible sanctions for the perpetrators of the 1989 events and the rights of returnees. However, the chairman of the Committee pointed out that the 17 refugees in attendance had only received the 15-page report the night before the opening of the consultation days, making their full participation difficult.

The ‘Summary report of preliminary consultations’, written by the inter-Ministerial committee for the return of refugees, presented the results of preliminary consultations between political parties, NGOs, refugee associations and resource people, including senior political figures. It laid out the different structures and logistics for managing and facilitating the repatriation, and stipulated the need to create at least five committees for this purpose. For the chairman of the Committee, it was clear that refugees should be included in these committees to ensure that conditions they put forward were respected. However, the committee that was finally created in January 2008, ten days before the repatriation of the first group of refugees, was a tripartite commission consisting of representatives of the Senegalese and Mauritanian governments and UNHCR. This effectively eliminated the refugee community from participation.

According to UNHCR, by the end of April 2009 approximately 10,000 of the 24,000 refugees had been repatriated. However, although clearly expressed as an objective by all implementing partners, the full and real involvement of the refugee community in this voluntary repatriation programme has not been realised. Leaders have been appointed to facilitate the management of the programme, yet their participation remained superficial and symbolic throughout the process, leaving the community ambivalent about the voluntary repatriation programme. In March 2008, after the return of several hundred refugees, the refugee community expressed its discontent with the programme in a press release criticising UNHCR and the Mauritanian government’s decision to repatriate groups of refugees before they had been sufficiently informed of the conditions they would find upon their return.


In any repatriation programme, it is essential to fully involve the concerned population from the outset of programme planning. Although this is a time-consuming task, full inclusion of and consultation with the refugee population from the early stages of organisation are vital for providing important information on the community’s needs and desires. An in-depth assessment of the Mauritanian refugee community’s perceptions of the voluntary repatriation programme would have shed much light on the feasibility and usefulness of the programme and may have provided indications on how to implement it more effectively.

Symbolic or token participation of the concerned community is insufficient. Community participation should take place at the decision-making level; it should  not just entail implementing decisions made by others. While some may argue that the symbolic participation of the refugee community in the voluntary repatriation programme was beneficial, many of the refugees expressed their view that this kind of involvement created a feeling of ambiguity and disillusionment. This risks discouraging the population from adhering to the voluntary repatriation programme – and could indeed explain the limited number of returnees to date.


Leonora MacEwen ( is Assistant Programme Specialist with IIEP-UNESCO ( This article is based on fieldwork undertaken while pursuing her Master’s degree.

[2] Office Africain pour le Développement et la Coopération’ (OFADEC), a local NGO which helps UNHCR in its work (French)


This site is registered on as a development site. Switch to a production site key to remove this banner.