World Bank Reprint Series: Number 477 R P477 Michael M. Cernea Social Integration and Population Displacement The Contribution of Social Science F>E ~CPY. Reprinted with permission from International Social Science Journal, vol. 143, no. 1. Copyright i United Nations Educational, Scientific and Cultural Organization (UNESCO) 1995. Michael M. Cernea Social Integration and Population Displacement The Contribution of Social Science Abstract Social integration and population translated by social scientists into the normative displacement: the contribution of policv of a large scale organization - the World social science .. Bank - did it become influential in practice. In Michael M. Cernea turn that organization and its policy became more effective in certain fields thanks to the absorption of the long-ignored knowledge. The discourse about social integration usuallv The article discusses how the cognitive disson- assumes that it is linked to development, and ance displayed by goverments and develop- that 'more' development induces higher degrees ment agencies vis-a-vis relevant research find- of social integration and reduces social conflicts. ings destroyed the 'normal' relationship between Yet in reality, development has multi-direc- supply and demand of knowledge. It also analy- tional effects, including integration, disinte- ses how the culture of a large bureaucratic gration, and reintegration processes. This article organization. and its habits of absorbing or analyses the class of processes usually labelled ignoring knowledge, can be changed. The key 'involuntary population displacement and resettlement' that are brought about worldwide methodological and epistemological issues of by development, and how disintegrative effects translating theory and research findings into can developmenti and hownteractegrathrough ef s policy prescriptions and germane implemen- can be mitigated and counteracted through poli- tation procedures are discussed. The article also cies and programmes enriched by knowledge reports the main findings of a recent studv. from social sciences. The bodv of concepts and based on a sample of 1992 development projects research findings on forced population displace- financed by the World Bank in 39 countries. ments. generated through decades of empirical that analysed the processes of development- field studies bv sociologists and anthropologists, cau sed tsa resettdements. had been left for a long time 'on the shelf,' without being used in re,levant policies and pro- grammes. Only after this body of knowledge was Social integration and population displacement: the contribution of social science Michael M. Cernea* Introduction A worldwide process: involuntary population resettlement 'Social integration' is a theme explicitly included During 1993-1994 one such worldwide process in the substantive agenda of the World Summit of social disintegration and reintegration was on Social Development, even though at the time studied under World Bank auspices. The social the decision was made to include it there was science aspects of this study are discussed in little clarity, let alone consensus, about what the present article. social integration is and how to define it. Since The study's subject was the forced displace- then, however, a number of social scientists ment and resettlement caused by certain devel- have returned to explore opment projects.2 Forced the concept, its content and Michael M. Cernea is the Senior Advisor population displacement is its coverage. The discussion for Sociology and Social Policy of the an ubiquitous process in of processes relevant to World Bank, Washington, D.C. 20433, that it accompanies devel- societal integration and dis- USA. His research focuses on the appli- opment in all countries, integration has been cation of sociology and anthropology to whether industrialized or iegratdion and will cer-tainly development policies and programmes, expanding and will certainly migration, population settlements, the developig, and to a large continue to expand after social organization of natural resource extent it is unavoidable. the summit. management, and public policy. He The uprooting of living One general assump- received the Solon T. Kimball prize communities, the imposed tion in the discourse about (1988) and the Bronislav Malinowski demise of functioning pro- social integration is that Award (1995). His recent publications tion system include Putting People First (1991) and duc s, and the integration is linked to Anthropology and Population Resettle- dismantling of informal development, and that ment (1993). social networks linking 'more' development would, many people are a painful and should, induce higher cost of some development degrees of social integration and reduce social projects, a cost paid in the currency of social conflicts.' Such discourse in itself is challenging disintegration of ongoing human activities and and prone to arouse controversy. It also suggests existing collectivities. Development is bound to some areas for intellectual reflection and for have multi-directional, rather than uni-direc- reinterpreting pre-existing research, undertaken tional influence on the degrees and forms of originally with different purposes, to clarify integration and sociability. This requires mul- issues of social integration and disintegration. tiple perspectives in exploring the conflicting processes relevant to social integration, and one such perspective is the uprooting and social reintegration of groups displaced by develop- ment. ISSI 143/1995 © UNESCO 1995. Published by Blackwefl Publishers, 108 Cowley Road, Oxford OX4 IJF, UK and 238 Main Street, Cambridge, MA 02142, USA. 92 Michael M. Cernea The present study has two objectives: first, production, and poverty alleviation, and benefit to analyse a certain type of situation in which large numbers of people and national economies development processes have adverse effects on in their entirety. Yet they inflict immediate social integration; this class of social processes losses and suffering on a significant fraction of is usually (and is in this article) labelled 'involun- the population. tary population displacement and resettlement'; Projections indicate that this process will second, to analyse how such disintegrative continue. In developing countries, the magni- effects can be mitigated and counteracted tude of development-related population dis- through policies enriched by knowledge from placement is likely to grow in the next decades, social science, aimed at reintegrating the dis- due to accelerated provisions of infrastructure. placed people. I will insist particularly on the Growing population densities in many countries second area of analysis, about which there is compound the problem, as infrastructural pro- still little information in the social science com- jects of similar size tend to result in increasingly munity, and will give a shorter description of larger numbers of displaced people. This makes the first one, which I dealt with in more empiri- development-related involuntary displacement cal detail in other studies (Cernea, 1988, 1991). and resettlement a problem of worldwide pro- The study mentioned at the outset, Resettlement portions and relevance. and Development, is available publicly (World Bank, 1994) and it contains a wealth of factual Social science and population material that supports, and complements the displacement argument contained in the present article. Contrary to a widespread misperception, The study of development-related displace- forced displacements are not rare and accidental ments is of interest also from another viewpoint: occurrences, happening only in the cases of namely as a class of development-related pro- major dams and affecting limited numbers of cesses on which social sciences - especially soci- people.3 Limited statistical information has ology and anthropology4 - have exercised abetted this misperception. In fact, forced popu- unusually strong and fruitful influence, parti- lation displacements are a historical companion cularly during the 1980s and 1990s. of development. Currently, they represent a This article analyses how the body of social process, and a social problem, of significant science concepts and research findings about worldwide magnitude. population displacement and resettlement, gen- Estimates documented with recent data and erated through decades of field study, was left worldwide projections (World Bank, 1994) indi- 'on the shelf for a long time, without being cate that every year a new cohort of at least 10 used in relevant policies and programmes. Only million people enter a cycle of administratively after this body of knowledge was incorporated imposed displacement and relocation required into the policy of a large scale organization - by 'right of way' for infrastructural programmes. the World Bank - did it become influential Indeed, each year an average of some 300 new politically and in the practice of development hydropower and irrigation high dams enter the programmes. In turn, that organization, the construction stage and entail an aggregate dis- World Bank, and its policies became more effec- placement of 4 million people. In parallel, urban tive in a given field thanks to the absorption of development and transportation infrastructure long ignored knowledge. projects started each year in the developing Significant lessons can be extracted from countries require the displacement of some 6 this reversal, lessons about both successes and million people. These two sectors alone have failures of development programmes. Such accounted for some 80-90 million displaced lessons give social scientists good grounds for people over the last decade. Additional involun- being, simultaneously, confident and humble. tary displacement takes place in other sectors The discussion below can also inform the on- as well, which is more difficult to quantify glo- going debate about effective strategies for bally. In most cases, the projects that exact this advancing other bodies of social knowledge in heavy displacement toll are indispensable for comparable policy domains. It is also relevant socioeconomic development, urbanization, food to the questions raised, and recommendations © UNESCO 1995. Social integration and population displacement 93 offered by scholars concerned to codify the developing countries, is that social research approaches apt to enhance the role and contri- findings will become effective guidance for butions of anthropology or sociology as 'policy future practice only if they result in the formu- sciences' (Weaver, 1985). lation and adoption of new or improved policies. The first section of this paper discusses Explicit social policies must guide - and by the relationship between social knowledge and 'guide' I mean both inspire and restrict - public social policy. The second section analyses the sector programmes that aim to induce develop- cognitive dissonance displayed by planners in ment. In practice, many activities and pro- governments and donor agencies vis-a-vis the grammes are planned in a policy vacuum, negative effects of development-caused dis- because no explicit policy has yet been formu- placement; this cognitive dissonance destroyed lated to address and govern that specific area the 'normal' relationship between supply and of activity. The implication is that social science demand of knowledge. The third section briefly analysis, tools and concepts, must be used not reconstructs the history, and synthesizes the just to evaluate programme results but to craft content, of the World Bank's policy regarding policies. Only policies have compelling authority involuntary population resettlement. Section over planning. four then highlights some of the methodological The class of processes discussed in this and epistemological lessons we derived about paper shows that planning criteria, and planning converting social theory and knowledge on routines as such, tend to resist new knowledge, development into policy, given the institutional and even tend to resist new policies. But if transactions intrinsic to development bureau- anything can break the back of such entrenched cracies. The final section reviews some of the routine and narrow planning it is only a shift in actual effects of the new resettlement policy policy, a'policy commandment' to start planning at two levels: policy replication and project differently, to pursue different goals with planning and execution. adequate means. In those areas that can be characterized as suffering from a policy vacuum, development 1. Evolving social policy from planning can be little more than an exercise in social knowledge guided administration. At its best, planning is a technique for resource allocation and work The concept of induced development defines sequencing. Neither the planners nor the plan- development stimulated by a deliberate pro- ning requirement are bound to obey the rec- gramme, typically initiated by governments, ommendations of impact evaluation studies. which uses public financial resources to create What planners must adhere to, however, are the new infrastructure or other economic assets policies and legal frameworks which planning, as (Cernea, 1991). Government-sponsored sector a tool for administering, is expected to translate or area programmes and plans aimed at inducing into action. Therefore, bringing sociological or accelerating development include numerous knowledge to bear upon the formulation of social development projects - i.e. discrete investments policies as frameworks for action is the most such as dams, highways, or irrigation systems. substantive way of compelling planning and planners. Policy vacuums How can such policy-relevant knowledge be generated? This flow of social knowledge Ex-post impact evaluation studies on such pro- must come from at least two major sources: jects are often carried out by social scientists. knowledge from operational evaluation research It is, therefore, important to examine the actual must be paralleled by knowledge obtained influence of the knowledge generated through through basic academic research. Ideally, if such evaluations: does the knowledge derived these two knowledge flows complement and from impact assessments modify subsequent reinforce each other, they can more effectively programmes for inducing development?5 converge to influence policy. My conviction, relying on experience inside a major development agency and in various © UNESCO 1995. 94 Michael M. Cernea Piecemeal studies do not substitute assumed personal insensitivity of planners that for policy permits the recurrence of poor social or environ- mental planning. Many researchers who have carried out social What guides planning is policy or strategy. impact assessments (SIAs) and environmental Policies provide the overall definition of objec- impact assessments (EIAs) have indisputably tives that underlie planning. I am referring to contributed to discovering the adverse effects a vast array of policies - national, sectoral, or of development projects on the environment even 'company policies' in the private sector. or on certain population segments. The usual If policies remain oblivious to past errors, and vehicles are case studies on individual projects. do not change or adjust when failures are ident- However, this type of contribution - the stan- ified, then planners do not receive any new and dard SIA or EIA case-report - has built-in compelling message. Hence, the perpetuation limitations. These limitations have not been of routine planning. Hence the exasperating sufficiently recognized and counter-balanced. recurrence of (more or less) the same mistakes. Empirical case studies are a prerequisite To formulate policies means to create for policy recommendations. But piecemeal frameworks for action which become compelling studies do not necessarily yield recommen- to the very bodies that issue such policies, and dations for policy reform, and usually SIAs or to related agencies and implementors.' Such EIAs tend to stop short of making recommend- bodies may be national governments, multi- ations that extend beyond the case at hand. national development agencies such as the Impact evaluators usually carry out studies World Bank or other bilateral donor agencies whose goal is to mitigate the impacts of a parti- (e.g. ODA - England, OECF'- Japan, CFD - cular project. They seldom formulate broader, France, GTZ - Germany, USAID - United forward-looking policies. This is often legitimate States, etc.). when only one individual's case research is Social research on a case-by-case basis is just reported. But piecemeal case assessments are not enough. Unless policy processes are opened insufficient, even when they are numerous, to up to research and evaluation, and unless social achieve paradigmatic and direct influence over scientists and impact analysts set their sights much programmes for inducing development. Inside higher than piecemeal mitigation, evaluation stud- the World Bank, for instance, there have been ies will remain a game with rather marginal useful- many evaluation studies which have signalled, ness. Time and location-specific studies that do time and again, some fallacies of project concept not look beyond their own 'nose', and which or planning. Yet these studies did not signifi- discuss only individual instances, can hardly cantly change subsequent Bank operations until trickle-up messages for broader purposes. Some- the Bank's policies were re-articulated. Piece- body needs to aggregate the case-by-case work meal evaluation studies have not been, and in order to distil broader policy and strategy intrinsically cannot be, a substitute for policy lessons, and introduce them into the normative reform.6 frameworks of development institutions. Policy reform must be initiated and carried The need to fill the policy vacuum out by the relevant governments or agencies. It must be recognized that some agencies or Why is it that knowledge gained through earlier governments, on occasion, prefer to maintain analyses does not become compelling 'do's' or a policy vacuum rather than issue binding nor- 'don'ts' in subsequent planning exercises? What mative guidelines and legal structures. Avoiding accounts for the recurrence of planning mistakes formal policy commitments leaves more oper- and biases identical to those revealed by prior ational flexibility in the short term, but often evaluation studies? Why so often don't planners at the expense of higher long-term costs, exter- hear what research is saying? nalized to others. Yet some government agen- To think that fault is rooted only in the cies are stubbornly reluctant to formulate or deafness of planners and managers is naive. accept public sector guidelines for activities that This is not where the ultimate responsibility they know are going to be problematic, difficult, resides. There is something more than the or controversial. © UNESCO 1995. Social integration and population displacement 95 The net result of such an anachronistic tive dissonance is just a drop in which a whole posture and mindset is that the interests of ocean is reflected: indeed, the drop mirrors a the displaced people, and of development in a sea of cognitive dissonance that engulfs many broader sense, are negatively affected. Low large scale development programmes informed level policy responses are an enduring cause of by economic knowledge alone. The authors and poor performance. managers of such programmes allow themselves Let us consider a concrete example. During to ignore the intrinsic social dimensions of devel- the 1960s and 1970s population displacement opment and the available knowledge about occurred in unsatisfactory ways in many coun- them. tries even under World Bank-assisted develop- ment projects. Yet, despite the findings of vari- Social disarticulation ous impact reports there was little improvement in subsequent projects in the manner in which Forced population displacement is always crisis- resettlement was planned until the Bank - as prone, even when necessary as part of broad we shall see further on - adopted a policy for and beneficial development programmes. It is a resettlement operations in 1980. profound socioeconomic and cultural disruption A vacuum in public policy allows detrimen- for those affected. Dislocation breaks up living tal practices to happen without checks or penal- patterns and social continuity. It dismantles ties. In fact, such a policy vacuum amplifies existing modes of production, disrupts social the risks of adverse consequences since legal networks, causes the impoverishment of many of safeguards for preventing adverse consequences those uprooted, threatens their cultural identity, are not institutionalized. In a context that lacks and increases the risk of epidemics and health policy and legally restraining norms, it cannot problems (Cernea, 1990, 1993). State agencies be assumed that individual planners or managers initiating displacement programmes have most will consistently listen to the sporadic whispers often failed to implement effective plans for of fragmented EIA and SIA reports. counterbalancing such adverse impacts. The disintegration of social support net- works that exist in communities subject to dis- 11. Cognitive dissonance: placement has far-reaching consequences. It knowledge supply and compounds individual losses with a loss of social demand capital: dismantled patterns of social organiza- tion, able to mobilize people for actions of The academic and applied research literature common interests and for meeting immediate has long documented the adverse impact of family needs are difficult to rebuild. Such loss population displacement on social welfare and is higher in projects that relocate people in a social integration. Yet for years social science dispersed manner rather than in groups and studies on forced displacement have remained social units. Field studies have documented that largely ignored by policy-makers, planners, such 'elusive' disarticulation processes under- economists, and engineers. The record shows a mine livelihoods in ways uncounted and major gap between practice and knowledge. On unrecognized by planners, and are part of the the one side is a long odyssey of disastrous complex causes of impoverishment.8 In the displacement operations that recur as virtual Rengali dam project in India, not Bank- carbon copies of one another; on the other side financed, a sociological study found various there is a growing, but largely uninfluential body manifestations of social disarticulation, such as of social science knowledge about resettlement, growing alienation and anomie, the loosening that demonstrates the mistaken assumptions and of kinship bonds, the weakening of control on inadequate procedures used by displacement interpersonal behaviour, and lower cohesion planners oblivious to lessons from previous dis- in family structures. Marriages were deferred asters. Students of cognitive dissonance can because dowries, feasts, and gifts became unaf- hardly find a starker illustration for this syn- fordable. Resettlers' obligations towards and drome than the case of forced population dis- relationships with non-displaced kinsmen were placements. In fact, this specific case of cogni- eroded and interaction between individual (C) UNESCO 1995. 96 Michael M. Cernea / . X ^ B-- -' L- J ? ''' ?~~~K A family of Tatars, displaced decades earlier, on return to Crimea. Stefano de LuigiEditing families was reduced. As a result, participation Is it enough to supply knowledge? in group action decreased; leaders became con- spicuously absent from settlements; post-harvest As development programmes causing displace- communal feasts and pilgrimages were discon- ment multiplied in both industrialized and tinued; daily informal social interaction was developing countries, social scientists began to severely curtailed; and common burial grounds show interest in researching such forced popu- became shapeless and disordered (Nayak, lation resettlements. 1986). Some of the studies on resettlement pub- The difficulties of displacing long settled lished in the 1960s and 1970s have become social groups are then compounded by the intri- classic works of social science literature. In the cate difficulties of relocating them on new sites, United States, Herbert Gans (1959, 1968) and often among reluctant host populations. Social other sociologists and urbanists (Dentler, 1969; reintegration of those displaced is not purpos- Anderson, 1965; Hartman, 1964, 1979; Heller, ively pursued by most initiators of such pro- 1982) initiated pioneering work in the late 1950s grammes. What makes things worse is that and the 1960s on urban displacement as part of people who suffer the pains of dislocation rarely planned urban renewal (see also Finsterbush, share in the gains generated by such develop- 1980; Burdge, 1981, 1989). In India, in the ment programmes. In sum, the worldwide his- 1960s, Roy Burman (1961) carried out a major tory of forced displacements is a record of social study on the displacement effects of building disruption and of little effort to mitigate the the Rourkela industrial complex, setting a easily predictable risks of impoverishment and research model that was later followed by other social disorganization. Indian anthropologists and sociologists on © UNESCO 1995. Social integration and population displacement 97 displacement caused by dams, strip mining, road over planning processes. If there was any aware- construction, etc. (Varma, 1985; Fernandes and ness among planners and decision-makers that Thukral, 1989; Mankodi, 1989; Thukral, 1992; such research knowledge existed, it quietly suc- see also MARG, 1987). The book on the Ako- cumbed to the peaceful yet guilty tranquility sombo reservoir displacement in Ghana by of cognitive dissonance. The knowledge-on-the Chambers, Butcher and associates (1970), and shelf about forced resettlement remained largely Colson's monograph on the displacement of unused for practical endeavours. the Gwembe Tonga from the Kariba Reservoir With hindsight, we should question whether (1971) were in-depth analyses of the cultural, at least part of the problem was not with the economic, and psychological effects of forced social scientists themselves or with the kind of relocation in developing countries. Thayer 'platter' on which they offered knowledge to Scudder, an anthropologist who has devoted practitioners. The 'research monograph' was most of his research to resettlement studies, has designed as a vehicle for conveying knowledge, analysed cross-cultural commonalities in but was not intended or used as a vehicle for people's response to forced displacement in sev- translating knowledge into operational or nor- eral river basins in Africa and Asia and com- mative policy recommendations. There were pared them with voluntary settlement (1973, few attempts by social scientists to convert the 1985; Brokensha and Scudder, 1968). In turn, resettlement research findings into systematic social geographers (Ackerman, White and proposals of new ends and new operational Associates, 1973; Adams, 1985) have made a means to achieve them, such as could have been valuable contribution to resettlement research used by willing decision-makers. Part of the from their disciplinary perspective. Some parti- explanation for this cognitive dissonance was the cularly large displacement processes have gener- fact that development-oriented social scientists ated an entire literature, such as the Aswan were rather few and they were not located High Dam (Fahim, 1981, 1983; Fernea, 1973; within development agencies in institutionalized Geiser, 1986), while Sorbo (1977) and Salem- positions propitious for translating research Murdock (1989) have focused on the adaptation findings into policy. As outsiders to develop- of the Aswan resettlers at their new sites. ment agencies, academic and applied re- Research and evaluation studies on displace- searchers do not have an institutional voice and ment and resettlement have gradually expanded remain more or less unable to go to battle for in many other countries, both developed and promoting their ideas within a decision-making developing (Rew and Driver, 1986; Suarez and forum. associates, 1984; Billson, 1990). This brief litera- In sum, the mere supply of social science ture review is far from exhaustive.9 knowledge-on-the-shelf was not sufficient to However, despite this gradual accumu- trigger 'supply driven' policy reform. lation of research knowledge, most government programmes causing displacement have long Failure to demand knowledge remained oblivious to the new findings. The 'enlightenment model' of social science influ- The reverse question inescapably arises: why ence on society (Janowitz, 1970) which con- was there no explicit 'demand' for knowledge tended that simply exposing social ills would on resettlement despite the fact that many lead to their correction, proved to be little more governments and donors had to deal with dis- than a well-intentioned illusion in the case of placement under the programmes they fin- forced displacement. The real influence exer- anced? cised in the United States by the body of socio- For decades, neither policy-makers and logical research on urban relocation upon legal decision-makers, nor planners or project man- regulations regarding expropriation and com- agers, had explicitly demanded such social pensatory payments (United States, 1970, 1987) expertise. The theoretical paradigms to which was a rare case of substantive impact on practice development intervention have listened were by resettlement researchers. In most cases, the silent about the basic sociocultural variables of growing body of sociological/anthropological change. Some major adverse effects of govern- knowledge on relocation failed to gain influence ment programmes were fallaciously belittled as () UNESCO 1995. 98 Michael M. Cernea tolerable side-effects. The population groups thereby compounding the evidence. Thus, vari- affected by displacement were politically too ous endogenous and exogenous premises for weak to make their voice heard and force the change have been gradually accumulating and adoption of better policies. At the legal level, converging. the governments of most developing countries did not institute explicit and rigorous norms about how to carry out involuntary displacement Ill. Brief history: thi and relocation. Infrastructure projects that were emergence of a resettlemen2 flawed by lack of social planning continued to poficy be financed by the UN, multilateral, or bilateral policy donor agencies in a business-as-usual manner. Engineering consulting firms, responsible for A significant turning point occurred during the technical design of many major infrastruc- 1979-1980 when, for the first time, a major ture projects worldwide, routinely displayed a development agency, the World Bank, decided stubborn obliviousness to the adverse social to adopt an explicit policy regarding the social implications of their proposed designs. Govern- issues involved in involuntary relocation. That ment agencies in charge of projects causing decision was the product of two sets of circum- displacement tended to belittle the estimates of stances: first, the slow but steady progress made dislocation losses and relocation costs. They in-house in using social science knowledge in passed down the organizational burden of Bank-assisted projects; and second, the trouble- executing relocation to unequipped and under- some feedback from some forced relocation staffed low-level bureaucracies, thus com- processes, particularly in the Bank's Sobradinho pounding, through poor execution, the losses Dam project in Brazil, which occurred soon and pains inflicted on those displaced. after similar problems exploded in the proposed Theoretical assumptions about social Chico River dams in the Philippines. change usually underpin existing policies pro- The history of this policy's enactment and moting development, implicitly or explicitly, evolution over the last 15 years reflects both But such policies do not exist in most developing (a) the increased use of social science knowl- countries. However, it can be said that the edge, and (b) the increased political recognition absence of policy is a policy by default. The given to the adverse impacts of development. fact that a number of governments in developing In its first formulation, the content of the countries have not adopted strict guidelines for World Bank's resettlement policy was grounded displacement reflects the assumption that either: in social science knowledge generated by pre- (a) there is no need for such a policy, or (b) existent research.10 During the years following involuntary resettlement should not be done its enactment in 1980, the policy went through differently from in the past. This betrays either several rounds of improvements based on both a limited level of knowledge about the perverse feedback from operational projects and findings effects of such processes, or a political bias from basic social research. In each round, against the poorest and most vulnerable people. improving and rewriting this policy was the Reactions to this state of affairs, however, work domain of the Bank's sociologists and have gradually increased in both frequency and anthropologists. In turn, each new formulation sharpness. Protest and opposition to forced maintained the policy's fundamental goal, resettlement by the affected populations have enriched its content, and strengthened its been gradually growing in many countries, implementation tools. reaching increasingly sharp political intensity The important milestones in the history of (Oliver-Smith, 1990). Untiring criticism by this policy were the following: environmental groups and NGOs about the dis- astrous outcomes of development-caused dis- 1. February 1980: The World Bank issues its placement, initially sporadic, has become better initial resettlement policy, prepared in 1979, documented. NGOs' criticism also is effective entitled Social Issues Associated with Involun- in mobilizing public opinion. Social science tary Resettlement in Bank-Financed Projects researchers kept generating empirical studies, (World Bank, 1980, OMS 2.33). © UNESCO 1995. Social integration and population displacement 99 2. 1985-1986: An in-house policy and oper- sociological analysis and advocacy consisted of ational study of how the new resettlement guide- sociological seminars on resettlement, analysis lines were applied makes additional recommen- of other projects, comparative reviews of per- dations, adopted by Bank management and formance in resettlement in various regions, issued formally as a new Operations Policy discussions in the Bank's Sociological Group"1 Note in Cctober 1986 (World Bank, 1986, OPN (Kardam, 1993) and an ongoing philosophical 10.08). This second policy statement rep- argument about the Bank's goals and means. resented a strengthening of the 1980 policy The process involved repeated consultations guidelines, by emphasizing that every project (sometimes confrontational) and 'bargaining' - causing displacement must develop a new pro- facts and past project evaluations in hand - with ductive basis for resettlers. a large number of Bank managers and staff from 3. 1988: Both policy documents (1980 OMS many departments about project objectives and 2.33 and 1986 OPN 10.08) are integrated into operational approaches. The explicit consul- one detailed policy-cum-technical Bank paper tations with various operational or policy-mak- entitled Involuntary Resettlement in Develop- ing units in the Bank were carried out over one ment Policy. Policy Guidelines in World Bank- year. In this context, the social guidelines for Financed Projects. For the first time, the Bank resettlement, prepared in-house at the initiative went public with its resettlement policy (Cernea, of the Bank's senior sociologist, were eventually 1988b). approved and instituted by management as a 4. 1990: Following internal Bank reorganiza- Bank Operational Manual Statement (OMS), tion, the resettlement policy was revised and equivalent to an internal policy document, under reissued as Operational Directive 4.30 on Invol- the title 'Social Issues Associated with Involun- untary Resettlement (World Bank, 1990). tary Resettlement in Bank-Financed Projects' 5. 1993194: A comprehensive study of all ongo- (World Bank, 1980). The OMS spelled out ing Bank-financed projects causing displace- principles and rules that were binding on Bank ment examined the consistency of actual oper- staff, as well as requirements that Bank bor- ations with policy, analysed performance and rowers were expected to meet, in operations weaknesses of resettlement operations on the involving displacement. ground, experience with resettlement inside and A systematic effort to implement the new outside Bank activities, and led to the Bank's approach, recruit knowledgeable social science adoption of important new measures to consultants able to advise staff and borrowers, strengthen the implementation of the Bank's and prepare and appraise resettlements differ- resettlement policy. ently from in the past, was initiated following the issuance of the 1980 guidelines. The emphasis of The initial 1980 policy guidelines the in-house sociological work shifted to policy implementation, monitoring resettlement per- The institutionalization of policy norms for formance critically, and codifying the lessons resettlement operations did not occur as a sud- from the application of the new policy. In- den edict-from-the-top, in a one-step act. It house, the new guidelines were not met by was the outcome of in-house analytical work, all with identical reactions. There was good knowledge dissemination and advocacy pro- institutional support, but there was also inertia, cesses which created important cultural premises as well as resistance among some staff and mid- inside the organization for adoption of a new level managers who remained unconvinced that approach to resettlement. the Bank should become too concerned about There were several factors which contrib- these so-called 'side-effects'. There was insuf- uted to this in-house discussion. The feedback ficient knowledge among project managers from projects was essential, and criticism from about how to shift gears to alternative oper- organizations outside the Bank reinforced the ational approaches, and the Bank itself was internal sense that change was necessary. Most not adequately staffed with social skills. Most important in this feedback was the disastrous important, the guidelines were not only innov- displacement and relocation in the Brazil-Sobra- ative but also shocking to many borrowing dinho project. Concomitantly, the in-house agencies, which turned out often to be not only © UNESCO 1995. 100 Michael M. Cernea unprepared, but also unwilling to change their and a need to be firmer and insist on borrowers' routine handling of expropriation practices and adherence to policy and legal loan agreements forced displacement procedures. signed at project negotiation. Furthermore, the study concluded that the impact of resettlement 1986: Analysing policy effectiveness on the host population and on the physical environment in receiving areas must be factored To ascertain the effects of the new policy and explicitly into the projects (Cernea, 1988a). the consistency of operations'with guidelines, a In February 1986, the Bank's senior manage- broad-based policy analysis study of resettle- ment discussed and adopted the study's policy, ment in World Bank-financed projects was car- operational, and staffing recommendations. ried out by the present author during 1984-1985. These recommendations were included in a new The study covered Bank-financed agriculture 'Operations Policy Note' (No. 10.08) written by and hydropower projects approved between the author of this paper and issued formally in 1979 and 1985, and found forced resettlement October 1986 (World Bank, 1986). The formal in at least some 40 projects in 27 countries. The statement not only reaffirmed and strengthened overall conclusion was that the introduction the initial 1980 policy, it also supplemented it of the Bank's resettlement policy had led to with additional policy elements and operational substantial improvement in the treatment of requirements to reduce the risk of impoverish- resettlement components of projects. ment. The new policy note prescribed that pro- At the same time, the study found that the jects should create a sound socioeconomic pro- Bank staff had not always applied the policy and ductive basis for those relocated, and explicitly its related operational procedures with adequate affirmed the concept that those displaced should rigour in all projects and sectors. During the share in the benefits made possible through the first five years, the 'consistency curve' between programme which caused their eviction. The projects and policy zigzagged. Consistency was policy also demanded more attention to the host higher in projects appraised in 1980 to 1982 population at the relocation site. Procedurally, than, by and large, in projects appraised in 1983 it instituted more demanding project planning and 1984, years for which the consistency curve and processing provisions regarding resettle- declined. It appeared that some erosion of policy ment, both in-house and for Bank borrowers. influence and weakening of operational com- Another immediate follow-up to the 1985 study pliance gradually set in. A well corroborated was the launching of a Bank-wide 'corrective finding was that inadequate legal frameworks actions' effort. Centrally monitored, these cor- and practices of various borrowing countries rective actions attempted to modify ongoing were frequently at the root of many difficulties projects found by the study to be inconsistent encountered by Bank staff in applying the guide- with the policy guidelines and likely to result lines and failures in resettlement implemen- in resettlement failure. tation. The study's findings pointed to the main The 1988 policy-cum-technical weaknesses in resettlement performance and to synthesis areas in which the policy guidelines and project processing procedures needed further strength- In 1988, the two prior policy statements issued ening. The findings were as follows: first, the internally - the 1980 OMS 2.33 and the 1986 quality of borrowers' preparation and detailed OPN 10.08 - were integrated into one single planning of resettlement components was found Bank paper for policy and technical guidance to need radical improvement; second, viable to resettlement operations (Cernea, 1988b). economic and social options for rebuilding the This was the first time the Bank published its productive capacity of displaced populations internal resettlement policy. were insufficiently specified and financed in The 1988 paper advanced the policy formu- project preparation and appraisal reports; third, lation process in two important ways. The first the Bank's supervision of resettlement im- was by combining the provisions of the earlier plementation was insufficient, requiring an statements on each one of the key points (e.g. increase in the use of professional social skills, objectives of relocation, compensation, etc.) © UNESCO 1995. Social integration and population displacement 101 and by clarifying some important issues such as jects with large-scale resettlement operations; the definition of re-establishment criteria based increasing local capacities for carrying out on 'with-and-without' project calculations. The resettlement operations adequately; improving second was by more specifically spelling out project design by reducing displacement and operational procedures germane to the policy incorporating production-based relocation stra- than in any prior Bank document. The paper tegies; providing increased allocation of finan- also introduced three new analytical tools for cial resources; and promoting greater consul- improving: (a) resettlement preparation; (b) tation and participation of affected people in resettlement economic and financial analysis; designing and implementing resettlement. and (c) resettlement monitoring and evaluation (see Annexes 1, 2, 3 to the 1988 paper). Further- The content of the Bank's more, wide distribution, and translation into resettlement policy Spanish, French, Chinese and Bahasa Indone- sian12 helped to disseminate the improved After this five-stage policy history, how can the approach to these difficult operations and core content of the Bank's resettlement policy increased the accountability of both the Bank be summarized? Which conclusions stemming and borrowing governments in this domain. from the prior decades of resettlement research have been explicitly absorbed into the content The 1990 Operational Directives of policy? The progression of this formulation and Following the Bank's reorganization, all pre- analysis process has produced a policy core that vious internal policy and operational guidelines can hardly be compressed here in a manner were updated and reissued as a set of new that will do it justice. A detailed description of documents called Operational Directives. In this this policy is available elsewhere (Cernea, 1988, process, the resettlement policy was revised and 1991; World Bank, 1990). The policy's key reissued as 'The Bank's Operational Directive points can be summarized as follows: on Involuntary Resettlement' (OD No. 4.30) in - Involuntary displacement should be avoided June 1990 (World Bank, 1990). The preparation or minimized whenever feasible, because of its of the 'Operational Directive' again prompted disruptive and impoverishing effects. a wide discussion inside the World Bank on the - Where displacement is unavoidable, the principles, concepts, institutional procedures objective of Bank policy is to assist displaced and outcomes of resettlement. persons in their efforts to improve, or at least restore, former living standards and earning The 1993/94 study capacity. The means of achieving this objective consist of the preparation and execution by the The review of resettlement performance under borrower of resettlement plans as development Bank-assisted projects active during 1986-1993 programmes. These resettlement plans are inte- covered 192 projects, of which 146 were still gral components of project designs. ongoing during 1993. Significant improvements - Displaced persons should be: (a) compen- were identified every time the policy prescrip- sated for their losses at replacement cost, (b) tions had been applied consistently; yet many given opportunities to share in project benefits, problems persist and performance has often and (c) assisted in the transfer and throughout been below expectations (see World Bank the transition period at the relocation site. 1994). - Moving people in groups can cushion disrup- The 1993/94 study led, in turn, to new and tions. Minimizing the distance between depar- important decisions. These contain new strategic ture and relocation sites can facilitate the orientations and immediate operational meas- resettlers' adaptation to the new socio-cultural ures in ongoing projects that demand corrective and natural environments. The tradeoffs actions. Among the strategy measures are: mak- between distance and economic opportunities ing agreement on policy with borrowing govern- must be balanced carefully. ments explicit and requiring the adoption of - Resettlers' and hosts' participation in plan- national policy and legal frameworks for pro- ning and implementing resettlement should be (C) UNESCO 1995. 102 Michael M. Cernea K~~~~~~~~~~~~~~~~~~~~~~~~ t ~ ~ ~ ~~ - 7 Kud flen notcnIa, nca th urks odr pi 91 oleeUE il \!t0, - ' i :> 4 O% ; ' t =Co /.,, ,. ' ,: t _~~~ ~ ~ ~ ~ ~ ~ ~ .f Kurds fleeing northern Iraq, near the Turkish border, April 1991. Boyle/SCIecIIREA promoted. The existing social institutions of should not be grounds for denying such groups resettlers and their hosts should be relied upon compensation and rehabilitation. in conducting the transfer and re-establishment Tied to these policy provisions are pro- process. Providing information to, and consul- cedural requirements. The most important of tation of, those to be displaced about their these is that the Bank will not appraise and entitlements, options, moving schedules, etc., improve financing for a project that will cause should be ensured throughout the preparation displacement unless it has received and accepted and relocation process. a resettlement plan that shows how the borrower - New communities of resettlers should be will meet the policy's objectives, the plan's cost, designed as viable settlement systems equipped and timetable. with infrastructure and services, able to inte- If building policy on social science research grate into the regional socio-economic context. clarified the objectives of resettlement and the - Host communities that receive resettlers manner in which it should be carried out, linking should be assisted to overcome possible adverse the policy to project processing provided the social and environmental effects from increased teeth for its enforcement because it instituted population density. mechanisms for assessing compliance. - Indigenous people, ethnic minorities, pastor- The policy guidelines, which have been alists, and other groups that may have informal strengthened over the years, have created a new customary rights to the land or other resources framework for planning displacement taken for the project, must be provided with and resettlement operations, for allocating adequate land, infrastructure, and other com- resources, and for assisting those displaced to pensation. The absence of legal title to land re-establish themselves. The most fundamental (© UNESCO 1995. Social integration and population displacement 103 demand of this policy framework, indeed its Knowledge-testing in the field raison d'etre, is to counteract the poverty risks involved in forced displacement, prevent the Translating social science knowledge into pre- impoverishment of those displaced, and ensure scriptive policy is not just a desk-bound intellec- that their income and livelihood are restored tual exercise. It demands involvement in field through adequate resettlement (Cernea, 1990). operations. To carry weight, prescriptions must This framework cannot be compared to any be anchored in operational experience and previous one since formal rules simply did not analysis. previously exist. But if the 'practical policy' is The five identifiable 'milestones' in the his- considered, in other words if the previous rou- tory of the current resettlement policy described tine practice in resettlement is taken as a refer- in the prior section do not provide the full ence point, the difference is enormous. In fact, picture. These landmark moments were part of unregulated 'routine practice' continues to be a wider scenario, whose dominant feature was the current dominant and ruinous pattern found field work on actual resettlemerit projects - to in forced resettlement in many projects occur- prepare, plan, implement, monitor, or evaluate. ring outside World Bank assistance. In hundreds of operational field assignments, Bank staff and consultant social scientists, as well as local sociologists assigned by developing countries' agencies to those projects, applied and tested their knowledge against real life IV. A learning process in resettlement approaches, constraints, and actual applying social science project performance. Thus, the gradual refine- ment of the new policy has benefitted from the joint effort of a much larger group of social Beyond its 'physics', applied social science work scientists than the handful involved in actually has its own 'metaphysics'. Any impression that writing it. Other specialists (such as agronom- the policy formulation progression described ists, economists, lawyers, engineers, etc.) also was a smooth and linear process would be totally co-operated in this effort. The policy could false and misleading. On the contrary, it was not have survived and be enriched without the fraught with difficulties, some theoretical, some ongoing field work, which confirmed or falsified methodological, some institutional, some practi- one or another premise, and provided empirical cal. It entailed intellectual clashes with pro- substantiation for its legitimacy and refinement. ponents of other approaches, less sensitive to people and culture. It revealed gaps and grey Pushing the frontier of social science areas in the knowledge of social scientists them- selves, who believed they had a 'lock' on the Extracting policy from pre-existing research and resettlement theory. concepts was not a one-way street. It also The translation of conceptualized knowl- uncovered gaps in theoretical knowledge that edge into policy and planning approaches was were.not grasped by research, gaps that invited a theoretical and operational battle waged both new thinking. Thus, the translation of social in the office and in the field, in the project science knowledge into policy, and particularly implementation 'trenches'. Using social science its practical application, has also triggered gains knowledge turned out to be a major learning for social science itself. Such gains included: process for the social scientists who promoted considerable new research in quasi-experimental its application. situations; a vast body of empirical knowledge Dealing with the mixed collection of issues on displacement and relocation, conducive to and processes which I call the 'metaphysics' of new conceptualizations; the identification of converting social theory into policy, is not an new issues previously unperceived, or unstud- oft beaten path. It is important to reflect on ied; a much deeper understanding of the eco- some of the lessons acquired during this process. nomics of the post-displacement re-establish- ment process; an appreciation of practical tradeoffs, when alternative options become © UNESCO 1995. 104 Michael M. Cernea available; the crafting of new data-collection lished to date as a reflexive outcome of this tools and analytical procedures tailored to the development-oriented operational work, and pecularities of resettlement; and the develop- certainly more is to be expected. Nevertheless, ment of new solutions for resettlement arrange- the sociology and anthropology of resettlement ments, which broadened the practical inventory have become richer, stronger in their explanatory of applied social scientists. powers, and better suited to making informed However, only part of the prior concep- and judicious prescriptions. They have become tualizations generated by anthropological more sophisticated and are informed by the research could be used as theoretical 'building need to address, or react to, countless nuts blocks' for formulating the operational resettle- and bolts (social, economic, financial, legal, ment policy. Other such pre-existing concepts or administrative, institutional, technical) ques- theoretical constructs could not be used because tions. The span of theoretical challenges and they were not validated by the 'test' of oper- practical queries raised and answered by oper- ational circumstances. For instance, it became ational necessity has widened enormously, evident that the pre-existing Scudder-Colson incorporating a variety of new issues, from com- anthropological model of resettlement pro- pensation to mutual help; from subverted griev- cesses, which focused on the concept of stress, ance procedures to overall legal frameworks; was directing the researcher's (or practitioner's) from valuation of land and assets issues to attention to a derived psychological conse- consultation on relocation options; from quence - social and individual stress - rather entitlements of 'major sons' and those of other than centering on what is basic and primary: subgroups and categories of people, to coping the breakdown of the socioeconomic sustenance strategies for the reorganization of displacees' systems of those displaced. Though helpful for lives, and on and on. The new body of knowl- many ethnographers as a descriptive tool, the edge has been creatively developed by the Scudder-Colson (1982) model of resettlement, numerous social scientists who have partici- consisting of a four-stage temporal matrix of pated, in various capacities, in resettlement resettlement processes, could not be used for activities related to over 100 Bank-financed pro- planning purposes and proved insufficient as an jects in Asia, Africa and Latin America. A explanatory tool (Partridge, 1990). codification and synthesis of the new state-of- Another circumstance that did not help our the-art has certainly become a challenging and work at all is the lack of communication among timely task now. the researchers studying various types of dis- placed populations. A dichotomy persists in Methodological dilemmas the research literature itself between the social science studies of disaster-caused refugee flows Important lessons refer to the methodological and the studies of development-caused displace- difficulties that surface when one tries to convert ments. These two literatures virtually do not abstract theory or case material into prescriptive 'speak to each other' and their findings are policy. not compared and corroborated (for a detailed The first difficulty emerges when knowl- discussion of this dichotomy, see Cemea, edge exists in the form of site-specific or case- 1993a). This unfortunate circumstance is coun- specific research findings. Then conversion has ter-productive not only for research, but also to move from the particular or individual nature for the operational work with displaced popu- of field research findings to the tenure of gener- lations. ally recommended prescriptions. The transition Generally, we can conclude that the fron- is from the descriptive to the prescriptive, and tiers of social science knowledge on population from problem analysis to problem solving. For displacement and project-related resettlement this, the policy architect must balance the sense have been pushed significantly forward during of expertise with humility, daring with prudence. both the policy development process and the Ethical restrictions are always important; but operational work that accompanied and fol- they loom even larger when the social scientist lowed it. Only part of this knowledge has been is not just an adviser to a policy maker or to distilled and systematized in the studies pub- a manager, but when he or she is the one ( UNESCO 1995. Social integration and population displacement 105 to develop policy, or to make an operational debates were devoted to the adequacy of com- decision loaded with consequences for many pensating the victims of displacement with cash. people. Proponents of hard-nosed market approaches Hard tradeoffs between desirable but some- argued that the Bank should care only about times competing principles must also be con- 'setting the prices right' for expropriated assets fronted under the pressure of immediate practi- but not get otherwise involved in the re-estab- cal needs. For instance, the principle of lishment of those displaced, while the anthropol- minimizing the physical and cultural distance of ogists produced waves of empirical data proving relocation, and the principle of maximizing the that specific cultural, political and economic economic development potential offered to the circumstances are certain to transform cash com- people relocated at the arrival sites are both, pensation into a recipe for quick impoverish- in theory, equally respectable and desirable ment and counter development. Other far- principles. It is indeed satisfying when propitious reaching controversies were ignited by issues circumstances allow for meeting both these prin- such as the entitlements of displaced landowners ciples simultaneously. But what is to be rec- vis-a-vis the status and rights of groups deemed ommended when reducing the physical distance by government agencies to be 'illegal squatters'. entails selecting sites with considerably less long- Many other examples of practical social di- term development potential than sites located lemmas and uncharted territories can be cited. two or three times further away? The reason for such conceptual controvers- Formulating ideal policy is always soul- ies resided in the fact that promoting a resettle- satisfying; but it also involves a high risk of ment policy implied, ipso facto, a change of becoming a futile exercise if the new policy ends some theoretical premises that underlie the up with an overload of ideal desiderata that model and goals of infrastructural development prove impractical or unenforceable. When the projects. In this case, the change meant broad- applied social scientist, much more than his ening the goal definition of infrastructural pro- academic peers, is confronted with the task of jects. It also means broadening the population formulating a strategy and charting a course of groups to be taken into account in project design action, he or she is forced - and must learn - and financing. For example, in building irri- to balance the desirable with the possible. gation dams the mandated goal became not the The metaphysics of applied work im- only downstream farmers' welfare, but included plies judgement calls on the opportunity, the protection of the upstream populations acceptability and feasibility of various practical (previously ignored and victimized). solutions, judgements that must be informed by knowledge about the political and cultural The modified vocabulary contexts within which projects are implemented, by the ability to predict, and by skills at building Reflecting the change in premise and policy, in safeguards. Policy prescriptions should be modifications appeared in the Bank's in-house justified and made acceptable to a large number vocabulary and public discourse about resettle- of countries with different traditional, legal, ment. Gradually, resettlement issues started to financial and administrative norms. Policy pre- be defined largely in anthropological/sociologi- scriptions must be ahead of current practice, cal concepts. The new vocabulary in which the yet connect with it and lift future performance policy itself was framed was about 'moving above the current entrenched routine. in groups', 'cultural identity' and 'settler-host interaction'; 'social networks', and 'kin-groups', Institutional clashes and change 'social integration', 'social uprooting' and 'social cohesion'; 'settler dependency', and 'social apa- As mentioned earlier, the formulation and pro- thy' and 'traditional authority systems'; about gression of the Bank's resettlement policy was 'community structure', 'lack of power' and not free of in-house disputes between those who 'alienation'; the cultural meaning of 'leaving supported tighter or looser policy norms. Social behind lands, deities and ancestors', and so on. scientists had to fight econocentric or engineer- The use of such terminology is not customary ing biases. For instance, repeated rounds of within agencies seeking economic returns and (© UNESCO 1995. 106 Michael M. Cernea engineering precision. Yet the new in-house For the social scientists involved in redesigning and public discourse arose precisely because institutional procedures, that means a good deal additional (social) variables began to claim oper- of administrative and organizational work, per- ational consideration. In turn, this new vemacu- haps not 'sociological' in a purist's view, but lar prodded the staff to think in new terms about definitely part and parcel of the applied sociol- what they were doing and to 'see' dimensions ogist's trade correctly understood. of 'their' projects which they had previously Obtaining the formal enactment of such overlooked. procedures within a large-scale organization is as important as consensus building around high Formulating organizational procedures policy principles. Only if principles and pro- cedures are substantively congruent, and are Last but not least, I would like to emphasize blended together normatively, is a policy able that formulating new policy means not only to become compelling and infuse discipline into setting resettlement principles, however im- the activities of agents and clients. portant these are. The social scientist engaged in development work is well advised to go a Alliances and interdisciplinarity step further, and also propose institutional and administrative procedures for implementing the To round up this analysis, I must point out the general policy principles. This is not an interdisciplinary nature of this policy-crafting eneral poand trivial exercise, as ivory tower intellectual endeavour, beyond its core social unworthy an rva xrie sioytwrcharacter. Economists, lawyers, engineers and dwellers may tend to assume from afar. It is cher Ec onomis t ers, poineel- another key part of the metaphysics - and real- other specialists also contributed to policy devel- politik - of converting social knowledge into opment- The social scientists developed infor- institutional policy. Without new procedures a mal 'alliances' within the institution with policy remains a simple statement of principles, exponents of other scientific disciplines or pro- Without new procedures germane to the policy's fessions. The sociological and anthropological Without thewpreoisdboud gme entropy t policy's focus on resettlement issues stimulated, in turn, content there is bound to be entropy in policy other kinds of intellectual work inside the Bank application. Without congruence between policy i eae ils hs otiuinwsids and procedures there also is no way to uniformly in related fields, whose contribution was indis- monitor policy execution. pensable for elaborating policy and operational In our case, formulating institutional pro- positions, but which probably would not have cedures meant designing the sequence of practi- happened without 'demand' from the sociologi- cal steps, and inserting them within the in-house cal side of the issues. For instance, innovative work processes, for addressing resettlement legal analysis and writing was contributed by issues during each stage of the project cyclem_ the Bank's General Counsel regarding national project identification, preparation, appraisal legal frameworks for resettlement, the protec- and supervision. Additional complexities arose tion of the human rights of resettlers, and the from the fact that the procedures had to pre- relationships between expropriation laws and scribe both what staff in the Bank must do, and development goals (Shihata, 1988, 1991); these what must be done by borrowing agencies in are pioneering contributions to legal thinking. their own countries. Similarly, economic writings contributed to It would be a mistake for applied social clarifying such issues as displacement costs and scientists to leave this translation task only to re-establishment benefits (Schuh, 1988). It others because of its 'technical' or mundane would be fair to say that more contributions to character. Of course it is 'technical', and more the economic and financial analysis of displace- often than not it i i ndeed mundane. Yet it is ment and resettlement would probably have too important to be left to others alone. The improved the policy's effectiveness. procedures themselves (not only the policy principles) ought to be sociologically informed, V. Policy diffusion which means designed with sensitivity to the nature of the given social process. They must The effects of the Bank's resettlement policy be germane to the policy which they facilitate. over the last 12-14 years have been important i) UNESCO 1995. Social integration and population displacement 107 and widespread. Policy implementation has not Bank, 1994). Social scientists or government been easy, however; resistance to the policy has officials have often emphasized that the Bank's also been substantial in various quarters, as policy has positively and significantly influenced could be expected, reducing the policy's bene- national, state, or sectoral policy frameworks ficial impacts. It is not the objective of the in various countries (see Nayak, 1989; Mougeot, present paper to analyse these effects in detail; 1988), while, on the other hand, in India and the interested reader will find such a detailed other countries social activists have deplored analysis in the study mentioned at the beginning, the fact that their states have 'no law on rehabili- which discusses the implementation of the tation . .. (and) for every project, compen- resettlement policy and its results over an eight- sation, awards and rehabilitation plans are made year period (see World Bank, 1994). A few piecemeal' (Dhagamwar, 1989). comments, however, are in order regarding pol- A rather sophisticated policy framework for icy diffusion. resettlement has been developed by Colombia's The resettlement policy adopted by the electricity sector; it focuses on restoring the Bank some 15 years ago influenced other devel- economic and social basis of the displaced popu- opment agencies and international donors to lation and it covers not only projects financed also adopt the basic tenets of the Bank's guide- by outside aid agencies, but all the domestically lines as their own policy (see IDB, 1990; Deruy- financed programmes in the entire sector. terre, 1992). In 1991 the development ministers Developed by an interagency working group of of all 23 OECD countries sanctioned and social scientists, planners, and company man- enacted uniform resettlement guidelines for agers, Colombia's 1990 resettlement policy is a their countries' aid agencies (OECD, 1992). major improvement over previous local Policy vacuums in developing countries, as approaches to resettlement (Guggenheim, pointed out earlier in this paper, have allowed 1993). for unregulated displacement practices, abusive It is to be expected that a social policy and directly ruinous to the people affected. The which deliberately sets standards considerably Bank has advocated changes in the very policies above current practice will be exposed to and legal provisions regulating population dis- countervailing factors, resistance, obstruction, placement and relocation in many of its bor- attempts to bypass it and other forms of oppo- rowing countries. As the Bank's General Coun- sition, and will have to contend for a while with sel has concluded, instances of low and sub-standard performance. But this is how the gap is being gradually nar- lessons derived from the Bank-assisted projects involving rowed. By adopting new standards, the policy resettlement [show] that in many countries the national legal framework of resettlement operations is incomplete sets in motion many resources able to support ... Resettlement legal issues [are treated] as a subset it. These are not only financial and economic of property and expropriation law. For various reasons, resources, but political ones as well. It also these national laws do not provide a fully adequate considerably empowers the affected people framework for development-oriented resettlement . . themselves to participate in the resolution of New legislation often must he introduced, or existing laws must be modified, in order to plan and carry out relocation problems, to defend their needs and involuntary resettlement adequately (Shihata, 1993). interests, to negotiate more effectively. The involvement of NGOs in resettlement pro- Consequently, internal factors in those grammes finds support now in the current policy. countries, particularly demands from NGOs, The growing resistance in many developing and the opposition by the affected people them- countries to the losses and social disintegration selves to inadequate resettlement, have caused by involuntary displacement results in reinforced the demand for better policies and an increasingly politicization of development laws. Indeed, significant changes have been programmes entailing resettlement. achieved in the policies of several developing Even under a carefully applied policy, countries, for instance, in the resettlement poli- involuntary resettlement is, and will always cies of China, Brazil and Mexico, and in several remain, a traumatic process in the life of the states in India (see Fernandes and Thukral, affected groups. Since such social disruptions 1989; Partridge, 1990; Huang, 1984; World will continue to accompany future technical and © UNESCO 1995. 108 Michael M. Cernea economic change, both urban and rural, further research and in knowledge utilization, remain improvements in relocation policies and legal imperative. frameworks, in implementation, in sociological Notes * The views, findings, and gigantic developmental benefits. from evaluation into 'planning' interpretation contained in this However, these projects also requirements. As a remedy, study are those of the author and entail major economic and cultural Armour proposed to shift the should not necessarily be losses, particularly due to emphasis from 'assessment attributed to the institutions with population dislocation. requirements' to 'planning which he is associated. requirements'. In my view, this 4. In this article I refer primarily remedy won't cure the ill. 1. See, for instance, the UNRISD to sociology and social Directing 'requirements' to paper prepared by Cynthia Hewitt anthropology; the terms will be planners would not achieve a de Alcantara (1994) which used interchangeably to refer to substantive change of actual explores several meanings of the both disciplines. For the purpose programmes, if programme goals term 'social integration' and the of this article, social science remain the same; it would only 'hidden assumptions' in furthering research on involuntary population pin hopes on another non-winning the goal of social integration. displacement and relocation refers strategy. A few years later this also to the research carried out on 'remedy', examined in hindsight, 2. Resettlement and Development. this topic by social geographers. will again fail to obtain a 'passing The World Bank, Environment grade'. Programme goals are Department, Washington, DC, 5. In fact, a similar question was defined by policy-makers; 1994. This study was prepared by asked publicly at the 1991 meeting therefore, knowledge, as well as a special World Bank Task Force, of the International Association requirements for programme led by the author of the present for Impact Assessment (IAIA) by changes, must be directed article. The study analyses a keynote speaker. The speaker primarily to policy-makers. worldwide experiences with questioned the audience: 'How are involuntary resettlement - we doing in integrating impact 6. Similarly, incorporating social primarily in projects receiving assessment and planning?', and knowledge into the design of World Bank financial assistance, gave a negative, disappointed individual projects, while and in domestically financed answer. She said: 'Regrettably, I definitely improving those projects as well. A very detailed have to say that I cannot projects, is not a substitute for analytic summary of this study is recommend a passing grade . . . incorporating such knowledge in available in French and Spanish (After) two decades of the formulation of overall translations. international experience with development policies and environmental impact assessment strategies: both are necessary, but 3. Development-caused . . . it is difficult not to feel more it is the latter that makes room displacements occur under many disappointment than satisfaction for the former. programmes of different types. with the way in which the field Installation of urban has developed . . . Progress has 7. Of course, confidence in the infrastructure, hydropower dams been excruciatingly slow' (Armour value of policy as an instrument and reservoirs, highway 1991). for change should be tempered by construction, the establishment of Despite the critical bent of the realistic caveat that firm industrial estates, building ports, this answer, neither the commitment to implementation open pit mining, the creation of explanation nor the remedy and policy enforcement is as national parks and biodiversity offered was convincing. In essential as the policy's content reserves, and many other projects Armour's explanation, the cause itself. - first occurring massively in of this slow progress resided in an countries of the North but over-preoccupation of researchers 8. The informal networks of recently expanding in developing with assessment procedures rather mutual help among households are countries - have generated than with including the knowledge essential in the daily economic life © UNESCO 1995. Social integration and population displacement 109 of the poor, but are barely visible incomes, compared to only 5 per 9. A recently published volume to the outsider's eye. During cent in the United States. The contains the largest published resettlement such networks are support can reach high levels. In collection of field social research, dismantled and dispersed, a net the Philippines, private transfers reported with ethnographic detail, loss to their members, but of among households in the lowest that documents the adverse effects course such loss is never counted quintile boost their income by of forced resettlement (see M. and compensated. For instance, more than 75 per cent. In Peru, Cernea and S. Guggenheim, household networks help cope the pre-transfer income of 1993). with poverty through informal households that are net givers of loans; exchanges of food, clothing transfers is 60 per cent higher 10. The resettlement policy and durable goods; mutual help than the income of recipient document was written by the with farming, building houses, and households. Such private transfers author of this paper, with caring for children. 'Household also function as informal credit assistance from David Butcher; networks pass around large arrangements and as mutual Debra Rubin and Lois Gram amounts of money, goods, and insurance mechanisms. Simulation provided assistance with social services, and may substitute for analysis shows that in Colombia research on Bank project public subsidies . . . But such transfers contribute up to 40 experiences with resettlement. recognition of the importance of per cent to stabilizing incomes in private transfers for economic households experiencing policy is relatively recent' (Cox unemployment (Cox and Jimenez, 11. An independent study of the and Jimenez, 1990). Such transfers 1990). internal impact of the Bank's flow from better-off to poorer The dismantling of such informal sociological group on households and help equalize the multifunctional, yet virtually Bank activities was carried out by distribution of income. Two 'invisible', social networks through an outside researcher. It found economists, measuring and displacement acts as one of the that the sociological group quantifying the contribution of 'hidden' but real causes of exercised considerable influence in such informal social networks, impoverishment through gradually increasing the in-house have documented what displacement. This is a loss of receptivity to social concerns and anthropologists and sociologists social capital. It is difficult, and it social analysis (see Kardam 1993). have long described in qualitative takes time, to reconstitute similar terms. Their research has found social structures and networks 12. The translation and reprinting that in developing countries 19 to among resettlers and their hosts, in Indonesia was done at the 47 per cent of people report capable of exercising similar initiative of a local NGO, while in recurrent transfers, representing as support functions at the new China it was at the initiative of a much as 20 per cent of household relocation sites. governmental organization. References ACKERMAN, W.; WHITE G. F.; ARMOUR, A., 1989. 'Integrating BROKENSHA, D.; SCUDDER, T., WORTHINGTON, E. B. (eds.), 1973. Impact Assessment in the 1968. 'Resettlement'. In Rubin, N. Manmade Lakes: Their Problems Planning Process. From Rhetoric and Warren, W. M. (eds.), Dams and Environmental Effects. to Reality'. Impact Assessment in Africa: An Interdisciplinary Washington, DC: American Bulletin, Vol. 9, No. 1(2). Study of Man-Made Lakes in Geophysical Union, Monograph Africa. London: Frank Cass. No. 17. ARMOUR, A., 1991. Impact Assessment and the Planning BURDGE, R. J., 1989. 'Utilizing ADAMS, W. M., 1985. 'The Process: A Status Report. Social Impact Assessment Downstream Impacts of Dam Presentation to the l1th Annual Variables in the Planning Model'. Construction: A Case Study from Meeting of the International Impact Assessment Bulletin, Vol. Nigeia'.Tranactins o theAssociation for Impact 8, No. 1(2). British Institute of Geographers, Assessment, June. pp. 292-302. BILLSON, J. M., 1990. BURDGE, R. J.; OPRYSZEK, P. 'Opportunity or Tragedy: The 1981. Coping with Change: An ANDERSON, M., 1965. The Federal Impact of Canadian Resettlement Interdisciplinary Assessment of the Bulldozer: A Critical Analysis of Policy on Inuit Familes'. American Lake Shelbyville Reservoir. Urban Renewal: 1949-1962. Review of Canadian Studies, Vol. Urbana, Illinois: University of Cambridge, MA: M.I.T. Press. 20, No. 2, pp. 187-218. Illinois Press. ) UNESCO 1995. 110 Michael M. Cernea BUTCHER, D., 1971. An CERNEA, M. M., 1994b. 'Bridging and Rehabilitation. New Delhi: Organizational Manual for the Research Divide: Studying Indian Social Institute. Resettlement: A Systematic Refugees and Development Approach to the Resettlement Oustees'. In Allen, T. (ed.), In FERNEA, R., 1973. Nubians in Problem Created by Man-Made Search of the Cool Ground, Egypt. Austin: University of Texas Lakes, with Special Reference for forthcoming. Press. West Africa. Rome: Food and Agriculture Organization. CHAMBERS, R. (ed.), 1970. The FINSTERBUSH, K., 1980. Volta Resettlement Experience. Understanding Social Impacts: BUTCHER, D., 1990. Review of the London: Pall Mall Press. Assessing the Effects of Public Treatmnent of Environmental Projects. Beverley Hills and Aspects of Bank Energy Projects. Cox, D.; JIMENEZ, E., 1990. London: Sage Publications. World Bank Industry and Energy 'Achieving Social Objectives GANS, H. J., 1959. 'The Human Department, PRE Working Paper. through Private Transfers: A Implications of Current March. Washington, DC: The Review'. World Bank Research Redevelopment and Relocation World Bank. Observer, Vol. 5, 1990, p. 205. R l opmnan ocation Planning', Journal of the American CERNEA, M. M., 1988. Involuntary COLSON, E., 1971. The Social Institute of Planners, Vol. 25, No. Resettlement in Development Consequences of Resettlement: The 11, pp. 15-25. Projects. Policy Guidelines in Impact of the Kariba Resettlement GANS, H. J., 1968. People and World Bank-Financed Projects. upon the Gwembe Tonga. Plc,ris: Essays on Urban Problems World Bank Technical Paper No. Manchester: Manchester and Solutions. New York: Basic 80. Washington, DC: The World University Press. Books. Bank. COLSON, E., 1985. 'Using GEISER, P.' 1986. The Egyptian CERNEA, M. M., 1989. Anthropology in a World on the Nubian. Cairo, Egypt: The 'Anthropology, Policy and Move'. Human Organization, American University in Cairo Involuntary Resettlement'. British 44(3). Press. Association for Social DENTLER, R. A., 1969. Urban Anthropology in Policy and ProbleNm: Perspectives and GUGGENHEIM, S. E., 1990. Practice, Newsletter, No. 4, Problutios. Chic-ago, ILand 'Development and the Dynamics Autumn. Solutons. Chicago, IL: Rand of Displacement'. Paper presented McNally. for the Workshop on CERNEA, M. M., 1990. 'Poverty DERUYTERRE, A., 1992. 'Inter- Rehabilitation of Displaced Risks from Population American Development Bank's Persons sponsored by the Institute Displacement in Water Resources mef eve p en for Social and Economic Change Development'. Development Resettlementc Some Comments'a and MYRADA, Bangalore, India. Discussion Paper No. 355. Paper presented at the Conference GUGGENHEIM, S. E., 1993. Cambridgte, MA: Hnearvoard on Urban Relocation: Policy and 'Sectoral Planning for Development, Harvard University. Practice, Institute of Housing Resettlement in Colombia'. Studies, Rotterdam, February. Manuscript. CERNEA, M. M., 1991. DHAGAMWAR, V., 1989. HARTMAN, C., 1964. 'The Housing 'Involuntary Resettlement: Social 'Rehabilitation: Policy and of Relocated Families'. Journal of Research, Policy and Planning'. In Institutional Changes Required'. the American Institute of Planners, Cernea, M. M. (ed.), Puaing In Fernandes, W. and Thukral, Vol. 30, pp. 266-86. People First: Sociological Variables E. G. (eds.), Development, in Rural Development, 2nd edn, Displacement and Rehabilitation. HARTMAN, C., 1979. 'Comments New York and London: Oxford New Delhi: Indian Social on Neighborhood Revitalization University Press. Institute. and Displacement: A Review of the Evidence'. APA Journal, Vol. CERNEA, M. M.; GUGGENHEIM, S. FAHIM, H., 1981. Dams, People, 45, pp. 488-90. (eds.), 1993a. Anthropological and Development. New York: Approaches to Resettlement: Pergamon Press. HELLER, T., 1982. 'The Effects of Policy, Practice and Theory. Involuntary Residential Boulder, CO: Westview Press. FAHIM, H., 1983. Egyptian Relocation: A Review'. American Nubians: Resettlement and Years Journal of Community Psychology, CERNEA, M. M., 1993b. The of Coping. Salt Lake City: 10, pp. 471-92. Urban Environment and University of Utah Press. Population Relocation. Discussion HEwirr DE ALCANTARA, C., 1994. Paper No. 1-52, Washington, DC: FERNANDES, W.; THUKRAL, E.G., Social Integration, Approaches and The World Bank. 1989. Development, Displacement Issues. Geneva: UNRISD. © UNESCO 1995. Social integration and population displacement ill HUANG, Y. F., 1984. 'On the Agencies on Involuntary Analysis of Dislocated People'. In Reform of Resettlement Policies Displacement and Resettlement in Hansen, A. and Oliver-Smith, A. in China'. Chinese Journal of Development Projects. Paris: (eds.), Involuntary Migration and Water Conservation, No. 6. OECD. Resettlement, Boulder, CO: Westview Press. INTER-AMERIcAN DEVELOPMENT OLIVER-SMITH, A., 1991. BANK, 1990. Strategies and 'Involuntary Resettlement: SCUDDER, T., 1985. 'Sociological Procedures on Socio-cultural Issues Resistance and Political Framework for the Analysis of as Related to the Environment. Empowerment'. Journal of New Land Settlements'. In Washington, DC: Environmental Refugee Studies, 4(2), pp. 132-49. Cernea, M. M. (ed.), Putting Committee, Inter-American People First. Sociological Variables Development Bank. PARTRIDGE, W. L., 1990. in Development. New York and 'Involuntary Resettlement in London: Oxford University Press. KARDAM, N., 1993. 'Development Development Projects'. Journal of Approaches and the Role of Refugee Studies, 2(3), pp. 373-84. SHIHATA, 1. F., 1988. 'The World Policy Advocacy: The Case of the Bank and Human Rights: An World Bank'. World Development, RAJAGOPALAN, V., 1989. 'Policy Analysis of the Legal Issues and Vol. 21, No. 11, pp. 1773-86. and Planning in Involuntary the Record of Achievements'. Resettlement'. Bank's World, Denver Journal of International JANOWITZ, M., 1970. Political No. 9, September. Law & Politics, 17, p. 39 et seq. Conflict: Essays in Political Sociology. Chicago: Quadrangle REW, A. W.; DRIVER, P. A., SHIHATA, I. F., 1993. 'Some Legal Books. 1986. 'Initial Evaluation of the Aspects of Involuntary Population Social and Environmental Impact Displacement'. In Cernea, M. M. MANKODI, K., 1989. 'Displacement of the Victoria Dam Project'. and Guggenheim, S. E. (eds.), and Relocation: Problems and Overseas Development Evaluation Anthropological Approaches to Prospects'. In Fernandes, W. and Report. Evaluation of the Victoria Involuntary Resettlement: Policy, Thukral, E. G. (eds.), Dam Project in Sri Lanka. Vol. Practice, and Theory, Boulder, Development, Displacement and III. London: Overseas CO: Westview Press. Rehabilitation, New Delhi: Indian Development Administration, Institute, p. 150. Evaluation Department. SoRBo, G., 1977. How to Survive Development: The Story of New MARG (MULTIPLE ACTION Roy BURMAN, B. K., 1961. Social Halfa. Khartoum, Sudan: RESEARCH GROUP), 1987. Sardar Processes in the Industrialization Development Studies and Sarovar Oustees in Madhya of Rourkhela. New Delhi: Census Research Center. Pradesh: What Do They Know? of India. Vol. I-III. New Delhi: M.A.R.G. STATE COUNCIL OF THE PEOPLE'S SALEM-MURDOCK, M., 1989. Arabs REPUBLIC OF CHINA, 1991. Decree MOUGEOT, L. J. A., 1988. and Nubians in New Halfa. Salt No. 74. Regulations for Land Hydroelectric Development and Lake City: University of Utah Compensation and Resident Involuntary Resettlement in Press. Resettlement Necessitated by the Brazilian Amazonia: Planning and Conservation or Large- and Evaluation. Edinburgh: Cobham SCHUH, G. E., 1993. 'Involuntary Medium-Scaled Water Conservancy Resource Consultants. Resettlement, Human Capital, and and Hydroelectric Projects. Economic Development'. In Adopted February. NAYAK, J., 1989. 'Rehabilitating Cernea, M. M. and Guggenheim, the Displaced: A Critique of S. (eds.), Anthropological THUKRAL, E. G. (ed.), 1992. Big Policy for Upper Krishna Project'. Approaches to Involuntary Dams, Displaced People. New Paper presented at the seminar on Resettlement: Policy, Practice and Delhi: Sage Publications. Resettlement Planning and Theory, Boulder, CO: Westview Implementation, Jakarta, Press. SUAREZ, F. R.; COHEN, E., 1984. Indonesia, June. Efectos Sociales de las Grandes SCUDDER, T., 1973. 'The Human Represas en America Latina. NAYAK, P. K., 1986. Resettlement Ecology of Big Projects: River Buenos Aires, Argentina: CIDES- at Rengali Dam. Bhubaneshwar, Basin Development and ILPES. Orissa. Resettlement'. In Siegel, B. (ed.), Annual Review of Anthropology. UNITED STATES, LAWS OF 91ST OECD (ORGANIZATION FOR Palo Alto. CONGRESS, 1970. Uniform ECONOMIC COOPERATION AND Relocation Assistance and Real DEVELOPMENT), 1992. SCUDDER, T.; COLSON, E., 1982. Property Acquisition Policies Act Development Assistance 'From Welfare to Development: A of 1970. Public Law 91-646; Sec. Committee. Guidelines for Aid Conceptual Framework for 84. Stat 1894. © UNESCO 1995. 112 Michael M. Cernea UNITED STATES DEPARTMENT OF Narmada Basin. Bhopal: Operations Policy Issues in the HOUSING AND URBAN Government Central Press. Treatment of Involuntary DEVELOPMENT, 1981. Residential Resettlement. Displacement: An Update - Report WEAVER, T., 1985. 'Anthropology to Congress. Washington, DC: as a Policy Science: Part 1, A WORLD BANK, 1990. Operational Government Printing Office Highlight', Human Organization, Directive No. 4.30. Involuntary (GPO). vol. 44, 2; 'Part 11: Development Resettlement. and Training', Human UNITED STATES, LAWS OF 10im Organization, vol. 44, 3. WORLD BANK, 1994. Reseulement CONGRESS, 1987. Title [V - and Development. A Bankwide Uniformn Relocation Act. WORLD BANK, 1980. Operation Review of Projects Involving Amendments of 1987. Public Law Policy Note No. 2.33. Social Involuntary Resettlement, 100e17dmSents o319. Revsios Lw Issues Associated with Involuntary 1986-1993. Washington, DC: The 10017; Sec. 310. Revisions to Resettlement in Bank-Financed World Bank, Environment 1970 Uniform Relocation Act. Projects. Department (written by Michael M. Cernea and Scott VARMA, S. C., 1985. Human WORLD BANK, 1986. Operations Guggenheim). Resettlement in the Lower Policy Note No. 10.08. © UNESCO 1995. Headquarters 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. Telephone: (202) 477-1234 Facsimile: (202) 477-6391 Telex: MCI 64145 WORLDBANK MCI 248423 WORLDBANK Cable Address: INTBAFRAD WASH[INGTONDC World Wide Web: http: / /www.worldbank.org E-mail: books@worldbank.org European Office 66 avenue d'Iena 75116 Paris, France Telephone: (1) 40.69.30.00 Facsimile: (1) 40.69.30.66 Telex: 640651 Tokyo COfice Kokusai Building 1-1 Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan Telephone: (3) 3214-5001 Facsimile: (3) 3214-3657 Telex: 26838 The full range of World Bank publications, both free and for sale, is described in the annual Index of Publications, and of the continuing research program of the World Bank, in World Bank Research Program: Abstracts of Current Studies. The most recent edition of each is available without charge from: DISTRIBUTION UNIT, PUBLICATIONS OFFICE OF THE PUBLISHER THE WORLD BANK THE WORLD BANK 6 AVENUE dIENA 1818 H STREET, N.W. 75116, PARIS WASHINGTON, D.C. 20433 FRANCE U.S.A. ISSN 0253-2131