The consequences of development-induced displacement
The consequences of DIDR depend largely on how resettlement is planned, negotiated, and carried out. In modern dam-building history, displacement strategies and resettlement schemes have ranged from positive to grim. Picciotto, Van Wicklin, and Rice (2001) point out that, in the cases of China's Shuikou and Yantan dam projects, displacees' incomes and living standards improved while satisfaction with resettlement was (reportedly) high. In contrast, the WFP report (1996) on Guatemala's Chixoy Dam Project in the late 1970s points to the massacring of hundreds of Maya Achi Indians by local civil patrols and the country's Armed Forces to make way for the dam's construction. In most projects, the conditions of displacement and resettlement have fallen somewhere between these two extremes, although it is rare to find examples of positive resettlement experiences.
The literature on DIDR is largely comprised of case studies. However, several theoretical frameworks have been proposed to explain the social consequences of forced relocation. Two models – Scudder and Colson's four-stage model and Cernea's Impoverishment Risks and Reconstruction ( IRR) model – are explained below. This is followed by discussions of the varying levels of risk that might exist for particular segments of a displaced population, and of the literature comparing and contrasting the experiences of development displacees and refugees.
In the early 1980s, building upon earlier approaches that dealt primarily with the processes of voluntary resettlement, Scudder and Colson proposed a four-stage model of how people and socio-cultural systems respond to resettlement. The stages were labelled recruitment, transition, potential development, and handing over or incorporation. In the recruitment phase, policy-makers and/or developers formulate development and resettlement plans, often without informing those to be displaced. During transition, people learn about their future displacement, which heightens the level of stress experienced. Potential development occurs after physical relocation has occurred. Displacees begin the process of rebuilding their economy and social networks. Handing over or incorporation refers to the handing over of local production systems and community leadership to a second generation of residents that identifies with and feels at home in the community. Once this stage has been achieved, resettlement is deemed a success.
The Scudder–Colson model focused on the different behavioural tendencies common to each of a series of stages through which resettlers passed. At first, the model was formulated to explain the stages of voluntary settlement, and was only later applied to some cases of involuntary resettlement (i.e., those "successful" cases that passed through all four stages). In the 1980s and 1990s, the mounting evidence of involuntary resettlement schemes that failed to pass through all four stages suggested that a new model was necessary to explain the consequences of involuntary relocation. In particular, it was recognized that a new theory was necessary to model what was increasingly seen as predictable impoverishment in forced resettlement schemes.
Cernea's Impoverishment Risks and Reconstruction ( IRR) model arose in the 1990s in response to this recognition. In contrast to the Scudder–Colson model, the IRR model does not attempt to identify different stages of relocation, but rather aims to identify the impoverishment risks intrinsic to forced resettlement and the processes necessary for reconstructing the livelihoods of displacees. In particular, it stresses that, unless specifically addressed by targeted policies, forced displacement can cause impoverishment among displacees by bringing about landlessness, joblessness, homelessness, marginalization, food insecurity, loss of access to common property resources, increased morbidity and mortality, and community disarticulation. To these risks, Downing and others have added: loss of access to public services, disruption of formal education activities, and loss of civil and human rights. The model also recognizes risks to the host population, which, while not identical to those of displacees, can also result in impoverishment. Not all of these processes necessarily occur in each case of forced resettlement and not all displaced households are necessarily affected in the same way by each process. Rather, the model notes that, when taken together, these processes capture the reasons behind many failed resettlement operations. Aside from distinguishing risks, the IRR model serves several other functions: as a predictor of impoverishment; as a guide for formulating research hypotheses and conducting theory-led field investigations research; and as a compass for risk reversal, advocating targeted resettlement policies, such as land-based (as opposed to mere cash-based) resettlement, job creation, health and nutritional safeguards, and social network rebuilding.
The IRR model has been used as a framework for a number of studies. Mahapatra (1996) uses the model to examine India's experience with involuntary resettlement from 1947-97, examining each of the IRR risks in turn. Thangaraj's chapter (1996) employs the model to analyse resettlement operations in two Indian projects – the Upper Indravati Hydroelectric Project and the Orissa Water Resources Consolidation Project. Lassailly-Jacob's chapter (1996) looks specifically at land-based resettlement strategies in African dam projects, arguing that such strategies must include not only land on which to resettle, but also common lands, adequate productive farmland, full title for lands (rather than tenant arrangements), and resettler-directed (rather than top-down imposed) development programs.
De Wet's article (2001) casts some doubt on our capacity to ever formulate a process that will ensure that all, or at least a large majority, of those affected by a project will benefit from it. While recognizing the thoroughness of the IRR model, he concludes that the model's assumption that resettlement problems can be erased by improvements in planning is overly optimistic. His article points to the importance of recognizing the complexities inherent in the resettlement process, such as "non-rational" political motivations and difficulties with financing and institutional capacity. De Wet advocates an open-ended, flexible approach to resettlement planning, which recognizes that projects rarely proceed according to plan.
Varying levels of risk for indigenous peoples, women, and other groups
In aggregate terms, DIDR often affects the economically, politically, and socially most vulnerable and marginalized groups in a population. However, at the individual and community levels, impoverishment risks associated with resettlement can be felt more intensely by certain segments of the displaced population.
Colchester's paper (2000) provides an overview of the impact of dam projects throughout the world on indigenous populations and ethnic minorities. It highlights that these groups make up a disproportionately large percentage of those whose livelihoods are adversely affected by development projects – for example, despite constituting only 8 per cent of India's population, Adavasis (tribal peoples) are estimated to make up 40-50 per cent of those displaced by development projects in the country. Colchester's paper points out that the experience of indigenous peoples with dams has been characterized by cultural alienation, dispossession of land and resources, lack of consultation, insufficient or a complete lack of compensation, human rights abuses, and a lowering of living standards. The specific and strong cultural connection that many indigenous groups have with the land on which, and the environment in which, they live makes their physical dislocation potentially more harmful than is often the case for other groups. Cultural Survival Quarterly has published several issues focusing specifically on the displacement and resettlement of indigenous populations, all of which are listed in the bibliography below.
In general, the issue of gender disparities in resettlement operations has been ignored in the literature. A small number of studies have shown that women often experience the adverse consequences of forced resettlement more strongly than men. For example, compensation payments are usually paid to the heads of households, which can concentrate the cash value of family assets in male hands, leaving women and children at higher risk of deprivation. Agnihotri's chapter (1996) exposes another form of gender discrimination in compensation criteria in Orissa, where entitlement to land compensation for unmarried persons is set at age 18 for men and age 30 for women. Other research has shown that, in urban development projects, women can be harder hit by displacement because they are more likely to derive income from small businesses located at or near their residences. In rural areas, women can be more adversely affected because they are often more dependent than men on common property resources for income sources. Participatory methods of resettlement planning can also expose the ways in which women can be prevented from shaping and/or benefiting from projects. For example, Guggenheim's (1993) discussion of Mexico's Zimapan Dam Project highlights that, at first, women were common participants in community consultations because their husbands were working away from home for the agricultural harvest. The women's demands changed resettlement plans to include not only land compensation but also credit to open sewing and baking enterprises. However, once the consultations began producing tangible results, men began attending in place of their wives.
For children, Cernea (2000) notes that resettlement often interrupts schooling. In many households, owing to drops in income and living standards, children may never return to school, instead being drafted into the labour market earlier than might otherwise have occurred. Other groups, such as the elderly and the disabled, might also face higher risk intensities in the displacement and resettlement processes, although, as for the other groups, the conditions of the project, resettlement procedure, and resettlement site play a role in determining which groups, if any, experience different and more intense risks.
Comparing the experiences of development displacees and refugees
While few studies compare the experiences of development displacees with those of refugees and IDPs, there is a growing awareness that all types of displacement, whether by development projects or as a result of violence, persecution, or natural disaster, can lead to impoverishment. Cernea (1996) voices an interest in bridging the research and policy divide concerning both groups, and in encouraging a two-way transfer of knowledge on successful settlement experiences. While recognizing that differences exist between the two populations – for example, in the compensation for expropriated lands in most cases of DIDR – Cernea questions the divide between the bodies of knowledge, pointing out that both populations experience a major disruption in their patterns of social organisation and culture, and therefore face the same challenge of physically and culturally surviving this disruption by reorganising their economies and ways of life.
In contrast, Voutira and Harrell-Bond (2000) argue that such knowledge exchanges can only be limited because, while impoverishment serves as a methodological common denominator among the two groups, the causes of impoverishment and the definitions of success are not the same. Moreover, they note that, beyond conceptual barriers to knowledge exchange, institutional obstacles also exist. The two populations are dealt with by separate agencies that operate with separate budgets and a radical division of labour, that define their roles and relationships to host and donor states differently, that function under different legal and regulatory regimes, and that sometimes promote incompatible long-term objectives.
Muggah (2000) compares the experience of development displacees and those fleeing from conflict in the context of Colombia, discussing the validity of the IRR model in examining conflict-induced displacement ( CID). While highlighting that DIDR and CID share similarities, Muggah stresses that fundamental differences exist in terms of predictability, permanence, and purposive versus arbitrary targeting. His article finds that the IRR model's impoverishment risks feature prominently in CID, but that, in the context of CID, the model fails to address potentially destabilising structural issues preceding the displacing event and focuses solely on risks without recognizing the potential and real capabilities of IDPs for poverty avoidance.
The volume edited by Cernea and McDowell (2000) contains a number of essays comparing and contrasting the experiences of development displacees and refugees in different contexts, using Cernea's IRR model as a basis for comparison. Robinson (2000) briefly considers the similarities between development displacees on the one hand and refugees and IDPs on the other. Much of his report focuses on a discussion of DIDR, including several case studies, and a consideration of the international community's responsibility towards development displacees who have been forced to bear the costs of development projects while being denied a share of their benefits.