A unified approach to displacement and resettlement

The forced migration studies literature is rife with conceptual tension. It is precisely because of the field's many disciplinary and ideological fault-lines that foundational concepts and terminology must be revisited. While specialists acknowledge the centrality of power asymmetries and coercion, there are still fundamental disagreements over basic parameters: whether displacement and resettlement is voluntary or involuntary, when these processes begin and end and whether they are restricted to physical movement or include other non-spatial forms of dislocation. While this debate potentially signals a healthy intellectual state of affairs, lingering disputes over rudimentary concepts are a major obstacle to building theory, refining and habituating norms, and ultimately refining practice (Finnemore and Sikkink 1998). Consensus continues to be hindered by narrow agendas demanding standardisation together with epistemic fault-lines dividing the many disciplines that comprise the forced migration field.

This chapter considers the ways distinct policymakers and researchers coalesced around separate categories of internal displacement and resettlement. It finds that the specialisation of knowledge and expertise in each domain reinforced unnecessary barriers between groups of scholars and practitioners. It traces the evolution of academic-practitioner networks engaged on various categories and how they constituted more or less unitary epistemic communities sharing knowledge about cause-effect relationships in specific policy arenas and adopting shared sets of normative and principled beliefs (Haas 1992: 3). These communities include adversarial advocacy coalitions and more collaborative public-policy networks that transmit and promote normative and ontological axioms across policy arenas. While discourse and practice associated with development, conflict and natural disaster-induced internal displacement and resettlement are epistemologically and bureaucratically compartmentalised, a unified perspective reveals that they have more in common than often assumed.

A critical review of core concepts in forced migration studies is essential to establishing a coherent and consistent nomenclature for the remainder of this volume. Despite their shared characteristics, different categories of forced migration are conceptually segmented and seldom compared, either in Sri Lanka or elsewhere. The particular framing of specific categories inevitably leads to specific forms of policy and practice. Many of the basic assumptions underpinning these framings play a crucial role in shaping specific regimes for categories of internal displacement and resettlement. In revealing the basic tensions in fundamental concepts of forced migration, the chapter demonstrates how more nuanced and comparative study could lead to very different judgements and conclusions relating to protection and durable solutions.

Definitions and labels

Definitions and labels matter. The representation of conditions and experiences in certain ways serves specific interests and is an inescapable element of public policymaking, its bureaucracies and its discourse. How, by and for whom a concept or phenomenon is defined and labelled frames debates, the design and implementation of interventions and valuations of success or failure. Labelling is neither neutral nor benign. In the case of the refugee label, Zetter (2007: 188) traces out the ways in which it is 'formed, transformed and "normalised" in policy discourse by bureaucratic practices which seem necessary, appropriate and even benign'. But the highly politicised dynamics of these processes are concealed. Labels affect the balance of power as a result of their capacity to reinforce or deny identity and guarantee conditionality, differentiation, inclusion and exclusion. Labels are therefore an essential feature of stereotyping and social control.

There is a compelling theoretical and moral case for enhanced clarity and interrogation of labels in the forced migration field. Certain refugee studies specialists actively critique the political interests that inform the labelling of refugees and the terminological opacity of their field (Zetter 1991, 2007; Turton 2003). Hathaway (2006: 2) believes that forced migration scholars have 'an ethical responsibility not to adopt categorical distinctions which, while perhaps administratively convenient, fail to reflect true substantive differences'. If arbitrary labels are uncritically accepted and academic contributions reinforce the absorption of these labels into policy 'then we acquiesce in and perhaps even support that arbitrariness' (ibid.).

Labels are iteratively transmitted between theory and practice. In the case of forced migration studies, as in other sectors relating to development or international law, the divide between the academic and practitioner community is blurred. For example, many labels regularly deployed in contemporary academic discourse featured prominently in the development, humanitarian, post-conflict recovery and natural disaster management sectors since at least the mid-twentieth century (Zolberg et al. 1989; Wood 1985; Drabek and Boggs 1968; White, G. F. 1964). Likewise, a sizeable number of the anthropologists and sociologists making seminal contributions to the forced migration field in the 1970s and 1980s were simultaneously employed by organisations involved in 'managing' displacement and resettlement, thus hastening an iterative exchange between theoreticians and practitioners. Since the 1970s, labels were simultaneously refined and cognitively transmitted between academia and various policy arenas through specialist journals, public forums, training courses and universities (Dwivedi 2002).

It is not surprising that the forced migration field exhibits so many contested labels. The literature itself comprises a diverse collection of scholarly and policy writings on subjects ranging from refugee law, economic and labour migration to people-trafficking, transhumance, colonisation and frontier settlement. Given such intrinsic disciplinary heterogeneity, the proliferation of different labels was inevitable. Zetter (1991) describes these competing labels as akin to 'currencies' with fluctuating values and exchange rates. But without a fixed exchange rate, consensus on core concepts and definitions remained elusive. There are also a wide range of bureaucratic interests that condition the application of specific labels. Labels are harnessed by a public policy discourse, whether security-oriented, developmental or humanitarian and used to apportion out assistance and codify entitlements. Sending and Neumann (2006) and Rosenau (2002) describe the instrumentalist use of labels as an extension of governmental rationality or 'governmentality'. Voutira (1997) described the fusion of bureaucratic interests with labelling as a form of 'policy logic': a calculated set of discourses and practices designed to control and regulate populations. While they may project neutrality and impartiality, they are only seldom critically interrogated. Although bureaucratically expedient, they (unintentionally) contribute to the denial of difference, particularly the experiences and attributes of affected populations or acts of (legitimate) resistance.

Labels are instruments of power because they assign differences in beneficiary rights (Zetter 1991, 2007). How a population group is labelled and defined and the distinctions between those who qualify for a particular right and those who do not, informs state and non-governmental obligations, budgeting, intervention strategies and associated entitlements. Shacknove (1985) observed how for the wide array of forced migrants facing acute vulnerability 'refugee status' is a privileged position. In contrast to others who may be worse-off or destitute, refugees are entitled to many forms of international assistance. Both refugee and IDP labels can also unintentionally promote social and economic exclusion by reinforcing latent and revealed prejudices among those who are not similarly categorised. According to Zetter (2007: 184) 'these new, and often pejorative labels, are created and embedded in political discourse, policy and practice'. They can generate stereotypes and essentialist understandings, whereby being a 'refugee', 'IDP', 'illegal asylum seeker', 'trafficked migrant' or 'economic refugee' overshadows other axes of identity, whether gender, ethnicity, class or caste. The social anthropologist Catherine Brun (2003: 377) noted in her study of internally displaced Muslims in Puttalum, western Sri Lanka, that the IDP category ascribed benefits and stigmas that affected the behaviour of resettled populations and host communities: the label constituted 'a social category and identity on the ground ... [leading to] discrimination instead of preventing it'.

Defining displacement

Despite their many ideological and disciplinary differences, forced migration scholars agree that displacement exhibits a range of characteristics distinguishing it from migration. These characteristics inform state and non-governmental discourse, policy and the bureaucracies associated with promoting protection and durable solutions. Displacement is conventionally accorded three core features by academics and practitioners, including involuntariness, temporariness and physical dislocation. By way of contrast, migration is characterised as voluntary, even if it is temporally and spatially varied. When held up to closer scrutiny, even these three fundamental axioms of displacement are on shaky ground.

Voluntary or involuntary Migration is conventionally attributed to voluntary decision-making while displacement is regularly described as 'involuntary' (Castles and Miller 2003; Ghosh 2000; van Hear 1998). Displacement is generally conceived as dichotomous to economic or labour migration: intra-urban and rural-urban population migrations associated with regional imbalances in supply and demand for labour are seldom characterised as displacement. Likewise, mass displacements generated by reservoir development, internecine conflict and earthquakes are distinct from large-scale migrations. Goodwin-Gill (2000: 164) argued that migration should be considered broad enough to include irregular or undocumented migrants, but not 'refugees, exiles, or others compelled to leave their homes' (italics added). Though migration can and often does take place before and after displacement it is generally conceded that they are not synonymous.

Certain scholars challenge the presumption that migration and displacement can be readily distinguished on the basis of a voluntary-involuntary dichotomy (Hathaway 2006; Castles 2003; Zolberg et al. 1989). They argue that the determination of 'involuntariness' depends on whether a choice to remain is in fact available. Discerning whether or not a choice is freely exercised is invariably subjective. Referring to the case of villagisation in Ethiopia, Gebre (2002: 270–1) problematised the binary classification and added new sub-categories or degrees of 'intent': voluntary, induced-voluntary, compulsory voluntary and involuntary. Critics of the voluntary-involuntary dichotomy contend that it presupposes a level (or lack) of agency: it artificially reduces displacement to situations where there are no viable alternatives other than self- or household-preservation. But who defines viable alternatives and self-preservation? Is choice exercised exclusively by an individual, by the family or a larger collective? What is the threshold that constitutes a threat to personal or household survival? These questions are inherently subjective and tricky to measure, as asylum claimants and immigration officials regularly attest.

A tipping point distinguishing the phenomenon of voluntary migration from involuntary displacement is real and implied coercion. Whether in the context of poverty or affluence, the practice of migration reflects a deliberate pursuit of new and better opportunities free of external encumbrance. An Intergovernmental Working Group of Experts established by the UN Commission on Human Rights (UN 1997a: 15) confirmed that migration occurs in 'all cases where the decision (to migrate) is taken freely by the individual concerned, for reason of personal convenience and without intervention of an external compelling factor' (italics added). By contrast, displacement is most likely to occur where choices are restricted and where an individual or an individual's family faces more physical and psychological risks than opportunities by staying in their place of residence (Muggah 2003; Hyndman 2000). While overt violence may trigger displacement, there are more subtle forms of coercion that also shape or restrict choices. As Penz (2003) observed, whatever the attributed real or perceived causes leading to population movement, 'if it is voluntary, it is not displacement'.

Even where displacement is assumed to be taking place, the label is often at odds with the actually occurring phenomena. As will be demonstrated in the case study sections of this volume, circumstances cast by policymakers and bureaucrats as voluntary migration (e.g. whether labour-related or as part of a colonisation scheme) can, on closer inspection, entail involuntariness such as forced evictions and evacuations. Likewise, interventions that generate displacement and resettlement, including dam and reservoir construction or the relocation of populations into emergency camps during wars, may trigger an influx of ostensibly voluntary encroachers, squatters, landless peasants and assisted migrants (Partridge 1993; Hansen and Oliver-Smith 1982, table 1). It is important to recall that settlement and resettlement programmes labelled as 'voluntary' or 'involuntary' may be accompanied by degrees of coercion or no implied threat at all.

Temporary or permanent The categorisation of different types of displacement converges around their attributed causes. For example, displacement is typically causally linked to a discernible external shock. These shocks range from urban renewal or village expansion to dams or mining projects, armed conflicts, ethnic cleansing and natural disasters such as tidal waves, cyclones, floods or mud-slides. With the attribution of a cause, it follows that displacement can be traced to a clearly demarcated starting point. But internal displacement can also begin well before the alleged triggering event(s). For example, the alienation and annexation of land that precedes urban renewal or a large-scale dam-building project may require planned 'evacuations' well before physical assets are appropriated, compensated for, or destroyed (Cernea 2003). Depending on informational asymmetries and the relative vulnerability of households, displacement may also begin long before a specific (life-threatening) attack or major climatic event – in which case population movement may, confusingly, be cast by outsiders as migration.

There is a persistent belief among forced migration scholars that displacement usually constitutes a temporary disequilibrium. As a short-term aberration, it follows that displacement can be 'normalised' with the introduction of 'protective measures' and ended with the provision of so-called durable solutions. There is an underlying compulsion among scholars and policymakers to redress displacement with the return of the displaced to either their original 'place' following their (voluntary) 'return' or 'resettlement' to a new place to resume an otherwise 'normal' existence. Pedersen (2003: 4) argues that this perspective is based on deeply embedded political and cultural values prevalent in the forced migration field (and others) that suppose a 'natural' link between people, identity and territory. Likewise, Malkki (1995) describes the instinctive attachment to sedentarism as part of the 'national order of things' (italics added). In practice, displacement is rarely temporally fixed, much less temporary. Depending on the situation, displacement ranges from nocturnal flight to protracted situations – and instances of varying periods of displacement can occur side by side as in northern Uganda, Colombia, Nepal or Sri Lanka (Muggah 2000a, 2005b, 2006).