Following conflict in 1972, and then a decade-long civil war beginning in the 1990s, over a million Burundians sought refuge in neighbouring countries, predominantly Tanzania. After the signing of peace accords in 2000, subsequent ceasefires and changing regional and global asylum policies, over 700,000 former refugees returned to Burundi between 2002 and 2009.
The Rural Integrated Villages (VRIs) programme in Burundi was designed to meet the immediate shelter and other humanitarian needs of returnees who were no longer able to access their land, or were now unsure of where it was, or who simply had no land. It was envisaged by policymakers as contributing to reconstruction, longer-term sustainable development, peacebuilding and social cohesion in a post-conflict context. Instead, it created a situation that many returnees experienced as continued displacement even in their country of citizenship and generated feelings of disappointment, abandonment, and social distance from family and Burundian society.
Return to a country of origin, like displacement itself, can differentially affect men and women, and affect and be affected by pre-existing gendered relationships, as individuals, households and communities re-negotiate and re-establish their lives in new places. While many other factors such as the lack of genuine peace at the national level are fundamental to the failure of these villages, important lessons about gender and transitions can be learned from the ways that gender and kinship relations were transformed by return and villagisation, making certain returnees more vulnerable.
The first ‘Peace Villages’ programme in 2004 provided housing but many of the villagers lacked adequate access to basic services. The subsequent VRI programme took a more holistic approach, providing land – although many households have yet to receive arable land – and including a range of supporting projects, with the expectation of long-term sustainable integration of returnees in a predominantly agrarian setting with limited land and livelihood opportunities. 
Village creation programmes are not new in this region of Africa and have often been criticised for the ways in which they changed resource use with detrimental effects for surrounding environments and gendered division of labour. For example, a higher concentration of people makes the everyday task of collecting firewood more difficult as all village residents need to walk longer distances to seek wood, compared with when they had lived in dispersed rural homesteads. Where firewood collection is considered women’s work, as in Burundi, this impact is clearly gendered.
Gender analysis in the VRI programme seems to have been largely limited to aiming for gender parity in activities. In a statement echoed by staff of several agencies, one UN agency employee stated simply, “We had transversal themes like gender; … in the identification of beneficiaries […] at least 50% had to be women.”
It is often supposed that there are more women than men in refugee populations but when counted there were nearly equal numbers of men and women Burundian refugees in Tanzania in the 1980s. This makes the results of counting women in this village creation programme even more striking, because there are many more women than men in the VRIs. This is largely because of the parity aims mentioned above, and the consequent inclusion of female-headed households. While male-headed households usually also have adult women present, most female-headed households do not include adult men.
Village residents and programme staff alike surmised that the reason for the high number of women in the villages was related to laws and practices around women’s land inheritance. Most women do not inherit land from their parents, widows do not inherit land from their spouses, and divorced women usually do not have a right to land from former husbands. The gendered effects of land inheritance did not only affect women but also determined the presence of many men, including those whose mothers were divorced, or those who had not been acknowledged by their fathers and therefore had no recognised right to inherit land.
It is true that the VRI programme did provide housing in Burundi to returnee women who may not otherwise have had access to housing of their own upon return. However, the fact that more women are affected by villagisation becomes problematic because of the disproportionate challenges to reintegration and the re-establishment of livelihoods faced by women in this resettlement programme.
The presence of large numbers of landless villagers increased the supply of agricultural labourers so that waged work opportunities in surrounding villages are scarce. Wage rates have decreased, and to meet their basic needs many households rely on the labour migration of family members to other regions of Burundi or back to Tanzania. As women are more likely to have responsibility for children, they are less able to migrate for work. Consequently female-headed households do not have access to sources of income which are vital to the survival of other families.
Returnee women highlighted the transition from livelihoods established over decades in Tanzania, where they had greater access to land and resources, or rights to the distributive humanitarian regime in refugee camps that assured the basic necessities of daily life. For villagised returnees the inability to consistently provide meals for their families was tied to gendered expectations of women’s role in the household, gendered labour, and the ways in which prior experiences shape perceptions and expectations of life in the village.
Related and important factors for many returnees were whether they considered the village to be in their region of origin, and whether they had extended family in the region to which they returned. Returnees for whom village resettlement is near to extended family highlighted the importance of family to having access to land to cultivate, opportunities for compensated labour or mutually shared labour in cultivating and building, childcare, assistance when ill, and participation in ceremonies. Sinarizi, a returnee in one of the villages, described being resettled far from his region of origin: “…living here we continue to live like refugees, even though there was a village near [our place of origin] where we could have easily had family… I’ve spent five years here and I’ve never seen anyone [of my family] come to visit. … The UN, together with the Tanzanian and Burundian governments, said that each person will return taking the direction of his province of origin. Which is why we found it interesting then, because we wanted to return, to see our home. … Because a person cannot be planted like the banana.”
The research on which this piece is based was funded through a Canadian Social Sciences and Humanities Research Council Masters-Level Canada Graduate Scholarship.
 Daley P (1991) ‘Gender, displacement and social reproduction: Settling Burundi refugees in Western Tanzania’, Journal of Refugee Studies, 4(3) http://jrs.oxfordjournals.org/content/4/3/248.abstract