Governing Refugees

Downloaded by [University of Defence] at 21:28 09 May 2016 Governing Refugees

Refugee camps are imbued in the public imagination with assumptions of anarchy, danger and refugee passivity. Governing Refugees: Justice, Order and Legal Pluralism marshals empirical data and ethnographic detail to challenge such assumptions, arguing that refugee camps should be recognized as spaces where social capital can not only survive, but thrive. This book examines themes of community governance, order maintenance and legal pluralism in the context of refugee camps on the Thailand–Burma border. The nature of a refugee situation is such that multiple actors take a role in camp management, creating a complex governance environment which has a signifi cant impact on the lives of refugees. This situation also speaks to deeply important questions of legal and political scholarship, including the production of order beyond the state, justice as a contested site, and the infl uence of transnational human rights discourses on local justice practice. The book presents valuable new research into the subject of refugee camps as well as an original critical analysis. The interdisciplinary nature of McConnachie’s assessment means Governing Refugees will appeal across the fi elds of law, anthropology and criminology, as well as to those whose work directly relates to Refugee Studies.

Kirsten McConnachie is Joyce Pearce Junior Research Fellow at Lady Margaret Hall and the Refugee Studies Centre, University of Oxford. Her research continues to study self-reliance and self-governance strategies among refugees from Burma. Downloaded by [University of Defence] at 21:28 09 May 2016 Law, Development and Globalization Series Editor – Julio Faundez University of Warwick

During the past decades, a substantial transformation of law and legal institutions in developing countries has taken place. Whether prompted by market-based policies or the international human rights movement, by the relentless advance of the process of globalization or the successive waves of democratization, no area of law has been left untouched. The aim of this series is to promote cross-disciplinary dialogue and cooperation among scholars and development practitioners interested in understanding the theoretical and practical implications of the momentous legal changes taking place in developing countries.

Titles in the series:

State Violence and Human Rights: State Ofﬁ cials in the South Andrew M. Jefferson and Steffen Jensen (eds)

The Political Economy of Government Auditing: Financial Governance and the Rule of Law in Latin America and Beyond Carlos Santiso

Global Perspectives on the Rule of Law James J. Heckman, Robert L. Nelson and Lee Cabatingan (eds)

Marginalized Communities and Access to Justice Yash Ghai and Jill Cottrell (eds) Downloaded by [University of Defence] at 21:28 09 May 2016 Law in the Pursuit of Development: Principles into Practice? Amanda Perry-Kessaris (ed.)

Governance Through Development: Poverty Reduction Strategies and the Disciplining of Third World States Celine Tan

Lawyers and the Rule of Law in an Era of Globalization Yves Dezalay and Bryant Garth (eds) Policing and Human Rights: The Meaning of Violence and Justice in the Everyday Policing of Johannesburg Julia Hornberger

Lawyers and the Construction of Transnational Justice Yves Dezalay and Bryant Garth (eds)

Gender, Justice and Legal Pluralities: Latin American and African Perspectives Rachel Sieder and John McNeish

Social Movements, Law and the Politics of Land Reform George Meszaros

Land Law Reform in Eastern Africa: Traditional or Transformative? Patrick McAuslan

Justice and Security Reform: Development Agencies and Informal Institutions in Sierra Leone Lisa Denney

From the Global to the Local; How International Rights Reach Bangladesh’s Children Andrea Schapper

Justice Reform and Development: Rethinking Donor Assistance to Developing and Transitional Countries Linn Hammergren

Governing Refugees: Justice, Order and Legal Pluralism Kirsten McConnachie

Forthcoming titles in the series:

Multinational Integration, Cultural Identity and Regional Self-Government, Comparative Experiences for Tibet Roberto Toniatti and Jens Woelk

Law and Society in Latin America Downloaded by [University of Defence] at 21:28 09 May 2016 Cesar Rodriguez Garavito

Policing and the Politics of Order-Making Peter Albrecht and Helene Maria Kyed This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 Governing Refugees

Justice, Order and Legal Pluralism

Kirsten McConnachie Downloaded by [University of Defence] at 21:28 09 May 2016 First published 2014 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 a GlassHouse Book Routledge is an imprint of the Taylor & Francis Group, an informa business © 2014 Kirsten McConnachie The right of Kirsten McConnachie to be identiﬁ ed as author of this work has been asserted by her in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identiﬁ cation and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record has been requested for this book McConnachie, Kirsten. Governing refugees : justice, order, and legal pluralism in the refugee camp / Kirsten McConnachie. pages cm—(Law, development and globalization) Summary: “This book studies justice and governance practices among Karen refugees from Burma living in camps in Thailand”—Provided by publisher. Includes bibliographical references and index. ISBN 978-0-415-83400-1 (hardback)—ISBN 978-0-203-42677-7 (ebk) 1. Refugees—Legal status, laws, etc.—Thailand. 2. Refugees—Government policy—Thailand. 3. Karen (Southeast Asian people) 4. Refugees—Burma. 5. Human rights—Burma. 6. Burma—Politics and government. I. Title. KPT485.R43M33 2014 362.87'5608995—dc23 2013039316 ISBN: 978-0-415-83400-1 (hbk) ISBN: 978-0-203-42677-7 (ebk)

Downloaded by [University of Defence] at 21:28 09 May 2016 Typeset in Garamond by Keystroke, Station Road, Codsall, Wolverhampton Exile is strangely compelling to think about but terrible to experience. It is the unhealable rift forced between a human being and a native place, between the self and its true home: its essential sadness can never be surmounted. And while it is true that literature and history contain heroic, romantic, glorious, even triumphant episodes in an exile’s life, these are no more than efforts meant to overcome the crippling sorrow of estrangement. The achievements of exile are permanently undermined by the loss of something left behind for ever. (Said 1984: 137) Downloaded by [University of Defence] at 21:28 09 May 2016 This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 Contents

Acknowledgements xiii A note on terminology xvii Acronyms and abbreviations xix

1 Governing refugees 1 Camps on the Thai–Burma border 5 Agency, sovereignty and legal pluralism 8 Agency, culture and community 8 The state: de-territorialized sovereignty 10 The law: justice, order and legal pluralism 12 Methodology 15 Chapter overview 17

2 The Karen in Burma: conﬂ ict and displacement 21 Burma’s ethnic politics 22 Who are the Karen? 23 Christian missions and colonization 25 The birth of Karen nationalism 27 Insurgency 28 Displacement 31 Downloaded by [University of Defence] at 21:28 09 May 2016 3 The camp community 39 Camp communities 39 A situational community: ‘We are refugees’ 41 An ethno-national community: ‘We are Karen’ 42 A community of shared governance 44 Constructing a political community 46 Narratives of identity and past 46 Language 49 x Contents

Clothes 50 Dynamic communities 52 Community interrupted: resettlement for those who remain 53

4 The governance palimpsest: order maintenance in Southeast Burma 58 Order beyond the state 60 Village sovereignty 60 Moral narratives 62 Animism or ‘spirit-law’ 65 Christianity 69 The Karen National Union and Kawthoolei Law 71 ‘One husband, one wife’: perceptions of harm in policing adultery 73

5 Sovereigns and denizens: camp governance and ‘the refugee’ 80 National policies and the ‘discourse of danger’ 82 Spatial ordering and ownership 87 International protection and the refugee as victim 88 Promoting self-reliance 91 The de-legitimizing local sovereignty 94 Refugee governance: the role of ‘non-state actors’ 97

6 The struggle for ownership of justice 103 Camp justice 106 Process: the pursuit of order 108 Sanctions 109 Legitimacy 111 ‘We must go step by step’: sociological legitimacy 113 The ‘due process critique’: human rights as trumps 114 NGOs and the rule of law orthodoxy 117 Legal Assistance Centres 118 Downloaded by [University of Defence] at 21:28 09 May 2016 Harmonizing camp rules and Thai law 119 Referral to Thai courts 123 ‘One nation, one law’ 124 A tentative counter-critique 126

7 Enacting interlegality: human rights and local justice 132 Culture and rights 132 Women’s rights as human rights 135 Contents xi

Sexual violence in refugee situations 135 Deﬁ ning sexual harm: adultery, rape and domestic violence 137 Reporting and accountability 139 Translating ‘SGBV’ 141 Youth delinquency and intergenerational relations 144 Delinquents or victims? 145 Why now? Moral panic and an ‘othering’ of deviance 147 The impact of rights: claims-making and forum-shopping 149

8 Beyond encampment 155 Overcoming ambivalence and the limits of self-reliance 160 Nothing about refugees, without refugees 162

References 165 Index 191 Downloaded by [University of Defence] at 21:28 09 May 2016 This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 Acknowledgements

This book discusses the experience of refugees in Thailand, but its origins lie in a different refugee situation. In June 2007, I visited Dheisheh refugee camp in the Palestinian Territories of the West Bank. Created in 1949, Dheisheh is still home to more than ten thousand people. In my short time there I encountered a society that overturned my assumptions of refugeeness and catalyzed the questions that gave rise to this book: How do people cope with living in a refugee camp for many decades? Who governs? What happens? In the intervening years I have attempted to answer those questions in relation to Karen refugees in Thailand but I would also like to acknowledge the people of Dheisheh in their own continuing struggle to ‘govern exile’ with dignity, and to thank Trócaire for the opportunity to bear witness to it. Many people helped get this manuscript over the fi nishing line and I am grateful to them all. Foremost among these is my husband Michael Hamilton, who has helped at every step of the way, from the fi rst fi eldwork plans through to the last footnotes. His grace and humour has rarely wavered (despite the often intense provocation of my research and writing self-absorption) and his love and insight have enriched my work as they enrich my life. Also in the fi rst division of cheerleaders were my parents, Margaret and Sandy, who listened to more than their fair share of writing woes and kept up a steady stream of sympathetic encouragement. Kathryn, Ranald, Clare, Lorna and Frank all provided welcome support and kept me in touch with the world outside my study walls. Downloaded by [University of Defence] at 21:28 09 May 2016 At Queen’s University Belfast, Kieran McEvoy and Hastings Donnan were both generous mentors as well as inspiring scholars. Hastings was a meticulous and perceptive reader and the best possible guide for my foray into anthropological methods. Without Kieran’s belief in my work – and his example of engaged scholarship – I doubt that I would have considered an academic career. I owe him a great deal, and his infl uence is clear in this book and my continuing research interests. Many other people were generous enough to read the original thesis or resulting manuscript in whole or in part and their feedback was invaluable: Jane Abbey, Andy Aitchison, Sylvia Brown, xiv Acknowledgements

Jeff Crisp, Ron Dudai, Julio Faundez, Barbara Harrell-Bond, Shadd Maruna, Graham Mortimer and Jonathan Simon. Helpful last-minute reference checking and formatting assistance was provided by Rachel Hicks. At Routledge, Rebekah Jenkins and Colin Perrin have been patient and encouraging editors. The cover photograph was taken by Masaru Goto in Tham Hin refugee camp in 2004. I am deeply grateful to Masaru and to Sally Thompson/TBC for permission to use this image. Extensive, immersive fi eldwork was essential for this research and was made possible by the generous support of the Wenner Gren Foundation, the Emslie Horniman Fund of the Royal Anthropological Institute, the MacQuitty Trust and Edinburgh University School of Law. I owe particular thanks to Ockenden International for their continuing support of my work. It has been a privilege to hold the inaugural ‘Joyce Pearce’ fellowship at Lady Margaret Hall, and through my work to echo Ockenden’s own long-standing commitment to promoting self-reliance for refugees. Of the many people who have helped in my various fi eld trips to Thailand, I’m particularly grateful to Niamh (whose advice at the very beginning helped all of it happen), Beau Ler Moo, Dave, Ebenezer, Eh Htoo, Emelda, Friday Paw, Hannah, Hsa Klo Htoo, Hla Min, Htoo Thaw, Htoo Htee, Jane, Kaisa, K’nyaw Paw, La K’Pru, Mario, Mel, Newty, Paw M’wee, Paw Plah, Ta Wee Dee and Zion. K’Baw Say taught me a great deal about the Karen, the gawlawah and life in general. Many staff of organizations and agencies were generous with their time, advice and support including: ARC, Burma Issues, Burma Lawyers Council, COERR, DARE, IRC-LAC, KHRG, KWO, KYO, TBC, UNHCR and ZOA. Several Burma experts assisted my research at various points, including Paul Keenan, Graham and Harriett Mortimer, Alan Smith and Ashley South. I extend my thanks to them all for their advice and assistance, while recognizing that they may disagree with some of the conclusions I have reached. Needless to say, the usual disclaimers apply: the analysis herein is my own, as are any errors or inaccuracies. Much of the work in revising this book for publication took place during a postdoctoral fellowship at the University of Edinburgh School of Law. Thanks to all at the Centre for Law and Society (and especially to Anna Souhami) for their friendship over the year. I have been fortunate to fi nish the manuscript Downloaded by [University of Defence] at 21:28 09 May 2016 within the wonderful surroundings and supportive academic communities of Lady Margaret Hall and the Oxford Refugee Studies Centre. Parts of this work have been presented at many seminars, including at the University of Warwick, University of Edinburgh, Queen’s University Belfast, Oxford Transitional Justice Research, Central European University and the Refugee Studies Centre; and at the association conferences of the International Association for the Study of Forced Migration, the Law and Society Association and the Burma Studies Association. Acknowledgements xv

Finally, and most importantly, I am indebted to everyone who participated in this research – those who were interviewed and the very many more people who spent time with me along the way. A research topic that I anticipated would be full of sadness was instead full of warmth and laughter and for that I am profoundly grateful. This book falls sadly short of capturing the richness of the community that I encountered in Thailand and the intense admiration I have for those who have created and sustained it. Despite its limitations, I would like to dedicate it to the Karen people and to the refugees on the Thai–Burma border, who invited me into their lives with great generosity and hospitality – and in so doing, ensured that my own life would be forever altered. Downloaded by [University of Defence] at 21:28 09 May 2016 This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 A note on terminology

In 1989, the then-ruling military regime changed the country name of ‘Burma’ to ‘Myanmar’, and changed several names of districts and regions within the country, including that of Karen State to ‘Kayin State’ (and of the Karen people to ‘Kayin’). These name changes were resisted by opposition groups, including the National League for Democracy, on the grounds that the military junta was an illegitimate government with no authority to make such a change. In 2013, Myanmar was the name used by the United Nations, by the Association of South East Asian Nations, by most national governments (though not that of the UK) and increasingly by international media outlets and scholars. I have used the name ‘Burma’ because this continues – for now – to be the preference of the refugees and community organizations in Thailand with whom this book is concerned. Downloaded by [University of Defence] at 21:28 09 May 2016 This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 Acronyms and abbreviations

ABSDF All-Burma Students Democratic Front ARC American Refugee Committee CBO Community Based Organization CCSDPT Committee for Coordination of Services to Displaced Persons in Thailand DKBA Democratic Karen Benevolent Army (formerly Democratic Karen Buddhist Army) IDP Internally Displaced Person IRC International Rescue Committee KHRG Karen Human Rights Group KNU Karen National Union KNLA Karen National Liberation Army KNPP Karenni National Progressive Party KRC Karen Refugee Committee KWO Karen Women Organization KYO Karen Youth Organization LAC Legal Assistance Centres NGO Non-Governmental Organization SGBV Sexual and Gender Based Violence SPDC State Peace and Development Council TBC The Border Consortium (formerly Thai–Burma Border Consortium) UNHCR United Nations High Commissioner for Refugees Downloaded by [University of Defence] at 21:28 09 May 2016 USDP Union Solidarity and Development Party ZOA ZOA Refugee Care This page intentionally left blank Downloaded by [University of Defence] at 21:28 09 May 2016 Chapter 1 Governing refugees

Living in a refugee camp presents many challenges. Some of these are obvious – poverty, dispossession, dependence on external support for the most basic needs of food and shelter. Others have become increasingly well understood, such as the power imbalances that refugees face vis-à-vis a host society, host state authorities and in engagement with international humanitarian agencies. A further set of challenges has thus far been largely overlooked: those related to political governance and the administration of justice. More than ten million people around the world currently live in camps and settlements.1 For refugee-hosting states, camps are a means to contain a refugee ‘problem’ and to ensure that international actors take some responsibility for refugees’ support, while for international agencies camps offer an administratively efﬁ cient model of service delivery. However, the average duration of refugee displacement is now close to twenty years, more than double the average in the early 1990s (Milner and Loescher 2011: 3). Many encampment situations endure far longer, such as Palestinian refugees in Lebanon and the West Bank (more than sixty years), or refugees from Burma in Thailand (more than thirty years). The quasi-permanent character of these situations is fundamentally different from the policy intention of camps as short-term crisis relief – and it creates fundamentally different concerns for camp residents. Protracted encampment has been challenged as ‘warehousing’ which forces refugees into conditions of dependency, stunts capacities, creates new vulnerabilities and violates rights to freedom of movement and choice of Downloaded by [University of Defence] at 21:28 09 May 2016 residence (Smith 2005; Verdirame and Harrell-Bond 2005: 15–16; Verdirame 2011: 281–298). While acknowledging these important arguments, my concern here is somewhat different: not with whether refugee camps should exist but with what happens inside a camp. Living in a refugee camp is widely perceived to erode individual, communal and societal coping strategies and give rise to ‘a breakdown in law and order’ (Norwegian Refugee Committee 2007: 315; UNHCR 1995: 2.1b). This perception is reﬂ ected in descriptions of camps as places where ‘refugees invariably live in conditions of insecurity and deprivation’ (Wilde 1998: 109–10) 2 Governing refugees

and as ‘harsh, dangerous and characterized by social chaos and lack of normal social structures’ (Bartolomei et al. 2003: 87). Agamben (1995: 166–180) has adopted the camp as a metaphor for absolute power over a population stripped bare of identity and agency, describing it as the ‘paradigmatic state of exception’, where residents exist wholly outside the protection of law and thus devoid of political identity. This emphasis on insecurity, danger and vulnerability presents refugee camps as exceptional spaces beyond the realm of normal experience. However, while refugee camps originate in trauma and loss – and while many do experience considerable insecurity – this is not the full extent of their experience. From the very beginning of a refugee situation an infrastructure of schools, clinics, churches, mosques and community organizations will emerge. A unique economic life develops, built around external aid, remittances from family in other countries and small-scale entrepreneurship. Social and political organizations are formed. Over time, the camp becomes a site of deep-rooted social networks and of a vibrant social, cultural and political life. As Agier (2011: 86) writes, ‘Beyond the legal and political exceptionality, real life is constructed in the camps as a social life that is largely resilient and, in its own way, also transformative.’ Whether construed as resistance, adaptation or simply the practice of everyday life, this is a very different image from the portrayal of refugee camps as sites of anarchy and the destruction of communal and individual capacities. Representations of refugee camps as disordered and devoid of political life echo the ‘pernicious premises’ that Wacquant (1997) argues have characterized depictions of the American ghetto; premises which de-politicize and de- historicize the reasons for the existence of ghetto areas, which focus on defi ciencies and fail to acknowledge internal ordering and modes of governance, and which ‘exoticize’ ghetto life but fail to recognize its normality as viewed from the standpoint of residents. The pernicious effect of these premises is apparent in research agendas and policy development and can only be reversed, Wacquant argues, by recognizing that despite superfi cial appearances of disorder ghettoes may possess a unique normative world that is created by internal political organization as well as wider relationships of power. This book challenges the ‘pernicious premises’ that prevail in analyses of encampment by examining the normative world of Karen refugees living Downloaded by [University of Defence] at 21:28 09 May 2016 on the Thailand-Burma border. The fi rst refugee camps on the Thai–Burma border were built in 1984 to provide temporary asylum for people fl eeing confl ict and oppression in Burma’s Southeastern borderlands. Thirty years later, despite recent political liberalization inside Burma, camps remain a necessary home for more than 130,000 people. Refugees have very restricted freedom of movement and there is almost total aid dependence; but contrary to the received wisdom that protracted encampment leads to an escalation of insecurity, these camps remain broadly safe and secure.2 In other dimensions, Governing refugees 3

too, these camps challenge premises of anarchy and anomie. Community structures have not disintegrated but play a central role in all aspects of camp organization, administration and cultural life. Internal ordering and modes of governance exist and are derived from pre-camp identities, historical practices and a wide-ranging network of political relationships. Refugees are constrained by and vulnerable to external forces (including national politics in Thailand and Burma, and the policies of international humanitarian agencies) but they are not entirely the subjects of power imposed from above. Through refugee-led camp committees, community- based organizations and political organizations, refugees also exercise and negotiate authority. The camp space therefore is not an isolated zone of ‘exception’ but a pluralistic and networked web of legal and political relationships. As a result, it might be said that these refugee camps are paradigmatic of political power in contemporary society but not in the way that Agamben has imagined: not as legal and political voids but as legally and politically pluralistic. The governance of refugee camps is not a peripheral concern. The consistent use of camps as a response to forced migration began in the aftermath of the Second World War and in the intervening years a shift from ‘asylum to containment’ (Shacknove 1993) has become ever more apparent. Of course, not all camps are the same. At one extreme, the Dadaab and Kakuma camps in Kenya hold nearly half a million people in conﬁ ned camps with tightly restricted freedom of movement. Very different conditions prevail in neighbouring Uganda where refugee camps are loosely self-settled communities and refugees are free to work and travel. Yet while there are many differences between these contexts there are also many similarities: their origins in forced exile and divestment of property, dimensions of containment and segregation, vulnerability to national and international politics – and, above all, the experience of forging a new society in supremely challenging conditions. The need to recognize refugee camps as complex cultural and political worlds also suggests a productive analogy with prisons research. Until the mid-twentieth century, prisons were not considered sociologically interesting or signiﬁ cant but were merely holding pens for problem citizens. It was only with the pioneering work of Clemmer (1940), Sykes (1958), Mathiesen (1965) and others that prisons came to be recognized as sites where previously Downloaded by [University of Defence] at 21:28 09 May 2016 unrelated individuals joined in the creation of a shared social world:

[T]he prison, like other social groups, has a culture. (Clemmer 1940: 85)

[W]e must see prison life as something more than a matter of walls and bars, of cells and locks. We must see the prison as a society within a society. (Sykes 2007 [1958]: xxx) 4 Governing refugees

It does not take a great leap of imagination to see that a modern ‘Society of Captives’ is represented by the millions of refugees ‘warehoused’ in conditions which deny mobility and blunt entrepreneurship but which are also spaces for the creation of a unique culture and society. The parallel should not be stretched too far. Refugee camp residents have a much greater degree of shared past and future than a prison ‘inmate society’, as entire families and even villages often fl ee together.3 Refugees also have considerably greater freedom of movement than prisoners. Even in Thailand, where the camps are offi cially ‘closed’ and freedom of movement is restricted, refugees regularly travel between camps, back to Burma (perhaps to visit family, or to attend a wedding or funeral) or throughout Thailand (for employment, education or to attend meetings and training sessions); while Thai villagers travel into the camps to use the medical facilities, attend English classes or trade goods in the camp shops and markets. Through such journeys and transactions, networks are formed between refugees in camps and a range of other local, national and international actors. These connections are often narrow and partial but they connect refugees to a much wider world than the camp confi nes would suggest. Despite these important differences between a prison and a refugee camp, the classic works of penal sociology offer useful points of comparison on aspects of socialization and normalization within a camp environment. Insights from prisons research that are of particular interest for analysis of refugee camps include the identifi cation of key sites and processes of individual adaptation (including a distinctive prisoner ‘argot’ and a distinctive set of respected emotional traits, including dignity, stoicism and self-restraint); adaptation as a process shaped both by individual character and the environmental and situational factors of confi nement; and, crucially, the recognition that even in a total institution, ‘total power’ is impossible (Sykes 2007 [1958]; Irwin and Cressey 1962; McEvoy 2001: 28–31). In his classic study of an American penitentiary Sykes (2007 [1958]) found that prison guards needed not only the compliance of prisoners but their active participation in prison life, and that a peaceful, orderly institution was more likely where authority was devolved to inmates. Conversely, rebellions were most likely to occur when rules were enforced in a manner which reduced the leadership capacity of prisoner leaders, leading Sykes (ibid: 25) to conclude that ‘a sustainable Downloaded by [University of Defence] at 21:28 09 May 2016 social order cannot be coerced from above’ and ‘you cannot promote free will – acting with self-control and foresight – by extinguishing it.’ In responses to migration, as in responses to crime, ‘incarceration seems to be taking up the slack of governance’ (Simon 1998: 599, 603) and studies of refugee camps can benefi t from a similar analytical journey into the nature of power, the adaptive processes required to create and shape a distinctive ‘camp community’ and the role of a refugee population in both governing and consenting to be governed. These questions are revisited after a short introduction to the refugee camps on the Thai–Burma border. Governing refugees 5

Camps on the Thai–Burma border As noted above, people from Burma have sought refuge in Thailand for more than three decades. Most refugees in camps in Thailand have fl ed from the border regions of Karen State and Karenni State, in the Southeast of Burma. The Karen and Karenni ethnic groups have been fi ghting for self- determination effectively since the birth of Burmese independence and have experienced intense discrimination and persecution at the hands of the Burma Army. While most international attention has focused on Burma’s pro-democracy movement, the ethnic confl icts in Burma’s borderlands have been the site of uncountable human rights violations committed over several decades. Testimonies of these violations have been gathered by local organizations, international organizations and academics, who report extortion, forced labour, sexual violence, torture, village destructions and massacres (e.g. KHRG 2006, 2008, 2011; KWO 2004, 2007, 2010; Delang 2000; UN General Assembly 2011, 2010, 2009). One consequence of this protracted confl ict has been large-scale displacement, both internal and cross-border.4 For those who have fl ed to Thailand, the approach of the Royal Thai Government has changed over time from non-intervention to strict regulation and control, and settlements which began as many small, broadly village-like spaces have been consolidated into nine large, tightly regulated camps. The Royal Thai Government has not ratifi ed the 1951 Refugee Convention and avoids describing people from Burma as ‘refugees’. Instead, asylum is granted on a temporary basis to ‘displaced persons’ who are ‘fl eeing fi ghting’, provided they remain in the ‘temporary shelters’. Presently there are nine temporary shelters on the Thai–Burma border. Seven of these are ‘Karen’ camps, represented and managed by a (KRC) Karen Refugee Committee (Mae Ra Ma Luang, Mae La Oon, Mae La, Umpiem Mai, Nu Po, Ban Don Yang and Tham Hin), while two are managed by a Karenni Refugee Committee (Ban Mai Nai Soi and Ban Mae Surin, also known as Site 1 and Site 2).5 Each camp has a distinct character, shaped by factors such as proximity to towns or to the border, length of existence, ethnic composition, the personalities of key leaders and even the surrounding landscape. Mae La, the largest camp on the Thai–Burma border, is sited next to a main highway and acces-

Downloaded by [University of Defence] at 21:28 09 May 2016 sible by public transport, allowing any voyeuristic tourist to bypass the ofﬁ cial entry point (where a camp pass is required) and clamber over the simple wire fence that marks the camp boundaries. Mae La is bordered by the main road at its front edge and by a rocky mountain face at the back, creating a hemmed-in, claustrophobic atmosphere. Further south, Umpiem Mai is also close to a main road but is surrounded by an expansive landscape of farmed land and rolling hills. This, in turn, creates a very different environment from the dense forest and rugged mountains surrounding Mae La Oon and Mae Ra Ma Luang, the camps I visited most frequently. 6 Governing refugees

Mae La Oon and Mae Ra Ma Luang are very close to the Burmese border in the province of Mae Hong Son.6 Reaching them requires several hours’ drive along a winding road which, though little more than a dirt track, is heavily traffi cked by large fl atbed trucks with rations deliveries, sleek silver SUVs belonging to international organizations and ancient Honda motorbikes piled high with people and their belongings. After the long drive through almost entirely unpopulated land the bustling refugee camp is a shock. Bamboo huts are densely packed together and narrow dusty roads wend through a teeming mass of people and animals and vehicles. Small shops display snacks, drinks, packets of cheroots. Closer to the heart of the camp, painted signs and banners mark the offi ces of community organizations, which are decorated with faded posters from public education campaigns: ‘sexual and gender based violence is a crime’, ‘resettlement information is free’, the symptoms of H1N1 bird fl u. The simple bamboo huts, children playing and noise and clutter of everyday life can give the impression that these camps are simply a large-scale version of the poor rural villages that are found elsewhere in the Thai borderlands. A closer look shows otherwise. Shops sell Burmese brands of snacks and cigarettes, Burmese and Karen language DVDs and music, Burmese food – mohinga and kono noodles, tea-leaf salad, syrupy sweet drinks. Women and young children cover their skin with thanaka paste to protect them from the sun. Many people wear ‘Karen clothes’ of a woven sarong and tunic or t-shirts bearing the logo of international aid organizations. Even time runs differently here: some people choose to keep ‘Karen State’ time, which runs thirty minutes earlier than ‘Thai time’; with the result that many meetings and social events have an unintentionally staggered start. Most refugees in these camps were subsistence farmers prior to becoming refugees and are capable of self-suffi ciency with very basic agricultural resources but there is very little space to farm or grow food, and few opportunities for work. The vast majority of residents are dependent on external assistance for food, shelter materials, even soap and a toothbrush – yet donor fatigue, combined with rising rice prices, has consistently reduced the money available to assist them (Chusri et al. 2011: 53, 67–8). This problem has become even more pressing as political changes inside Burma have led to changing patterns of aid. Substantial resources have been allocated to new programmes ‘inside’ Downloaded by [University of Defence] at 21:28 09 May 2016 the country, while there have simultaneously been drastic cuts in services and supplies to refugees in camps. Refugees’ food rations are supplied by The Border Consortium (TBC), which also supplies shelter materials and other resources, such as charcoal for cooking fi res. In 2013 funding cuts forced TBC to reduce rations to the most basic ingredients (rice, yellow beans, fermented dried fi sh (a Karen food staple), cooking oil and salt)7 and to implement a graduated scale whereby refugees are categorized according to whether they are ‘Most Vulnerable’, ‘Vulnerable’, ‘Standard’ or ‘Self-Reliant’ and receive a different food ration accordingly, ranging from 8kg–12kg rice monthly. Governing refugees 7

As one of my interviewees said, ‘If we compare here [Mae La Oon] and there [Burma], here is better. But here is still bad.’ Life is somewhat easier for those who can ﬁ nd employment but Thai policy prohibits refugees leaving the camp and those who do are vulnerable to exploitation, arrest and deportation. International organizations have advocated for a comprehensive approach to sustainable livelihood and income generation capacity for several years but to date the Royal Thai Government has restricted this to vocational training in areas such as car mechanics, knitting, cooking, and child care (CCSDPT/UNHCR 2007). Despite the limited opportunities available in the camp, daily life is structured and industrious. As later chapters will explain in detail, much of the work of camp management is handled by refugees, as participants in camp committees, section committees and a range of community-based organizations. From sunrise to sunset, loudspeaker announcements are broadcast throughout the camp to inform residents about meetings and events occurring that day, such as registration checks by international agencies, school meetings, church events and other community activities. Beyond administrative activities, family, friends and faith are the foundations of life in the refugee camp, much as they are anywhere else, and the precariousness and anxiety of life as a refugee can be at least partially buried beneath the everyday concerns of raising children and taking care of a family. It is impossible not to admire the tenacity and courage that refugees display in coping with life in camp: ‘Like the tropical bamboo that is the main building material for their temporary shelters the refugee communities have proven over the decades to be resilient, ﬂ exible and tough’ (TBC 2010a: 2). Over the three decades of their existence these refugee camps have been affected by changing conditions inside Burma and by changing policies of both the Royal Thai Government and international donors. The most powerful driver of change since the camps were created has been an international resettlement programme which between 2005 and 2013 resettled more than 100,000 refugees from the Thai–Burma border to ‘third countries’ including America, Australia, England, Norway, Sweden and Finland (IOM 2012).8 Despite the scale of departures, the net population in the camps has barely decreased as tens of thousands of new arrivals have replaced those who left. The impact of this population shift has been profound. Some changes Downloaded by [University of Defence] at 21:28 09 May 2016 have been positive (such as a surge of international remittances to supplement the camp economy during a period of donor fatigue and withdrawal) but a massive ‘brain drain’ of experienced teachers, medics and community leaders has placed great strain on camp administration and capacity. Resettlement also introduced uncertainty about the camps’ future as refugees feared it was a precursor to camp closure and the forced return of refugees to Burma. In 2013, this uncertainty increased dramatically as political changes in Burma raised the prospect of repatriation; a prospect that at the time of writing is under discussion between the United Nations High 8 Governing refugees

Commissioner for Refugees (UNHCR), Royal Thai Government and the Government of Burma but which is viewed with great trepidation by the wider refugee population – who, after all, have little reason to feel conﬁ dent in the good faith of Burma’s political leaders (Burma Partnership 2012; Eh Na 2013). It is of course impossible to predict what the future might bring for Burma or for its refugees. However, in this time of political ﬂ ux, it is instructive to consider past experience. Despite the population shifts and political pressures, these camps have remained broadly stable and safe, and they have retained strong community structures and social capital. This book examines how and why that has been the case, arguing for recognition of these camps – and, by extension, other refugee camp contexts – as sites of growth as well as of suffering, where refugees themselves show considerable agency and creativity in coping with protracted encampment. The following sections outline three dimensions of the political and social order of the camps on the Thai– Burma border: refugee agency, culture and community; de-territorialized sovereignty; and legal pluralism and contested authority.

Agency, sovereignty and legal pluralism

Agency, culture and community If camps are typically presented as exceptional sites which are outside the realm of normal experience, those who live in camps are similarly pathologized, no longer recognized as individuals but as a collective ‘refugee’ (Peteet 2005: 70). In the same fashion as other humanitarian discourses which have come to be driven ‘from above’ by technical experts and external consultants, refugee voices have been largely absent from the strategic reviews, objectives and programme implementation of humanitarian agencies (Harrell-Bond 1986, 1999). Marginalization of refugee voices continues even where policies exist that are intended to guarantee participation (Rothkegel et al. 2008; Meyer 2006; Horn 2010a; Rajaram 2002; Olivius 2013). Marginalization of refugees can be linked to the power dynamics inherent in international aid and to the pervasive policy construction of refugees as vulnerable victims (see further Chapter 4). Contrary to these reductionist dis- Downloaded by [University of Defence] at 21:28 09 May 2016 courses, considerable evidence exists for the adaptive, agentic capacities of refugees and other displaced persons. This agency is expressed in numerous ways, from subversion of aid policies (Kibreab 1993, 2004) to outright protest (Moulin and Nyers 2007; Holzer 2012), to the ways in which ‘forced migrants creatively reconstitute their social identities in relation to displacement’ (Lubkemann 2008: 17). In Sudan, Kibreab (2005: 24) found that Eritrean refugees ‘responded to their losses and to the challenges of displacement by developing new and broader forms of social networks, occupations, skills, survival strategies, relationships, divisions of labour and social organization’. These studies resonate with the heuristic proposed by Essed et al. Governing refugees 9

(2005: 2–3) of four dimensions of agency among refugee population: the ways that bureaucracies engage with refugees, the ways that refugees assert and express identities, the ways that host societies develop and engage with policies for refugees and (crucially for the present discussion), ‘the fi nding [. . .] that refugee camp life, apart from causing losses and traumas, can also have some gains.’ Recognizing possible gains of encampment is essential if we are to move beyond the ‘pernicious premises’ of refugee camps as anomic spaces. In situations of political violence, confl ict can lead to forms of breakdown but it can also produce ‘sites of innovation and reordering resulting in the creation of new types of legitimacy and authority’ (Duffi eld 2001: 6). Similarly, the social upheaval that displacement causes can at times lead to constructive social change. An extreme example of the double-edged sword that this dynamic represents can be found in the increasing number of women village leaders in Karen State, appointed with the hope that Burma Army troops would treat them less brutally than male village leaders (KWO 2010). The strategy has hardly worked, with women reporting both witnessing and experiencing horrifi c atrocities. Yet, in the space of thirty years the appointment of female village leaders in this area has moved from ‘highly unusual’ to ‘commonplace’, changing villagers’ conceptions of leadership and political authority as a result (ibid: 9). Similarly, in refugee camp settings, traditional gender expectations may be challenged by a lack of employment opportunities, by changing avenues of economic support and control, or by alternative models of social ordering encountered in the host society or promoted by international agencies. While hardly an unqualifi ed good, protracted stays in refugee camps may open up access to education for women and girls, which in turn may provide new avenues for employment and positions of leadership. Refugee camps also have the potential to foster positive social change as a hub for civil society. Within the camps in Thailand, residents are not merely passively surviving but are actively engaged with life, through family, church, education and service to the community. To use the example of gender again, women have organized an impressive network of civil society organizations to represent their interests and to support others, though organizations including the Karen Women Organization, Karenni Women Organization, Muslim Women Association Downloaded by [University of Defence] at 21:28 09 May 2016 and Burmese Women Association. This community activism is an avenue for refugees to help each other, but it can also be seen as an important assertion of political identity by women who are otherwise systematically denied agency in their relationships with both Thai authorities and international organizations (O’Kane 2007). The point here is not merely that internal networks and memberships are important but they may even be the most important avenue for refugees in coping with exile (Sorensen and Vincent 2001: 273; Landau 2012). Of course, refugees’ space for autonomous action is limited by the circumstances of encampment and the restrictions under which they live. 10 Governing refugees

As Giddens (1984: 1–28, 25) has analyzed, agency is exercised in a dialectic relationship with the rules, forces and relationships that organize social life. Agency and structure exist as a duality, a twinned process of ‘structuration’. This is a helpful analytic tool for the political life of a refugee camp, where refugees’ agency is both constrained and enabled by a multitude of forces: laws and regulations of the host state, policies of international organizations, and less concretely deﬁ ned (but often powerfully normative) cultural, religious and attitudinal values.

The state: de-territorialized sovereignty The primary structural constraint on refugee agency is state sovereignty and this is where any analysis of power and authority in refugee camps must begin. The nation-state and the categories it establishes (citizen/non-citizen, state/non-state, legal/illegal) are the central organizing forces of refugee lives (Agamben 2005; Haddad 2008; Malkki 1995a). However, while the state is integral to establishing the juridical category of ‘refugee’, focusing only on the national level risks neglecting expressions and exercise of political authority and identity beyond the state. A territorial notion of sovereignty has for many theorists been superseded by a practice-based deﬁ nition, such as that proposed by Comaroff and Comaroff (2006: 35):

[T]he more or less effective claim on the part of any agent, community, cadre, or collectivity to exercise autonomous, exclusive control over the lives, deaths and conditions of existence of those who fall within a given purview, and to extend over them the jurisdiction of some kind of law.

Recognizing the practice of sovereign authority rather than its symbolic possibility repositions the state executive as one power-holder among many: a particularly important and necessary power-holder, certainly, but not an exclusive actor (Ferguson 2004; Hansen and Stepputat 2005). In itself this is hardly a new development. There have always been parallel authorities, communities that to a greater or lesser degree chose to exist outside or were pushed outside the dominant political order. Nevertheless, some aspects of Downloaded by [University of Defence] at 21:28 09 May 2016 the contemporary political order are genuinely new, including the sheer number of international and transnational actors involved in governance activities, the growing role of non-state actors in military and security capacities, the political clout and economic power of non-governmental agencies, and the effect which the interactions between all these actors has on the articulation and reproduction of norms (Benda-Beckmann et al. 2009: 3). This proliferation in new forms and institutions of governance has consequences in the creation of new forms of authority and new roles for both non-state actors and the nation-state. Governing refugees 11

Sovereignty, the state, and non-state governance are themes that have been perceptively analyzed by James C. Scott. In his most recent book, Scott (2012: xx) once again focuses on the spheres of unexamined political life, the realms ‘outside the visible spectrum of what usually passes for political activity’. His core insight here, as in much of his other writing, is to recognize that while the idea of the state is extremely powerful, much of the actual ‘work’ of politics and political relations takes place far outside the structures of a national executive. In short, the state as a unitary executive is central to our conceptual models of political power but it is often less central to people’s experience of political power. A refugee camp highlights precisely this paradox: the host State plays an essential role in deﬁ ning the conditions and circumstances of encampment at a political level but on a daily basis is largely invisible, even irrelevant, to the refugee population. Yet in policy and scholarly analysis it is more often refugees that are rendered invisible. Analyses of camp management assume that sovereign control is exercised by the host state or its ‘surrogate’, the UNHCR, and that a refugee population is the passive recipient of this governance (Slaughter and Crisp 2009). For example, Colson (2004: 108) describes displaced populations as subject to a ‘double chain of administration’ from host government and humanitarian agencies, while Wilde (1998: 121) suggests that, ‘The power relationship in many development camps is relatively straightforward, however. UNHCR runs the entire operation, with minimal involvement by the host state’. Voutira and Harrell-Bond have described the nature of authority in an ‘ideal-type refugee camp’ as:

Starting from the top down, a refugee camp is administered by ofﬁ cials assigned by the host government whose power is enforced by the presence of armed police or paramilitary personnel. [. . .] The management and distribution of material assistance is carried out by international nongovernmental (NGO) humanitarian agencies [. . .] On the receiving end are the refugees. (1995: 210)

This hierarchical model radically simpliﬁ es the nature and practices of camp governance, at least as it operates within the camps in Thailand. Three broad Downloaded by [University of Defence] at 21:28 09 May 2016 sets of governance actors operate within these camps: Thai authorities, international agencies and refugee-led organizations. These are diverse categories, containing multiple sub-groups. The ‘Thai authorities’ includes the Ministry of Interior, the military and policing structures, Border Patrol and Ranger forces, Forestry Department and other ofﬁ cials. More than twenty international agencies and organizations operate in the refugee camps, each of which is a complex institution in its own right with hundreds or even thousands of staff. Unusually, the lead agency in camp management in Thailand is not the UNHCR, but a consortium of donor organizations known 12 Governing refugees

as The Border Consortium (previously known as Thai–Burma Border Consortium and as Burmese Border Consortium). However, in addition to national and international authorities these camps also have a strong and coherent refugee-led administration, which in two camps is led by the Karenni Refugee Committee and in seven camps is led by the KRC. These Refugee Committees are the primary avenue of communication between the refugee camps, all relevant Thai authorities, and international organizations. Each camp has a ﬁ fteen-member camp committee which has primary responsibility for camp management and is the ﬁ rst point of contact for the local Thai Authorities, NGOs, surrounding Thai villages and the refugee population. The range of governing skills and processes used echo those described by Chatterjee (1993: 53–80) within settlements created by partition-era refugees in Calcutta, marginalized communities which nevertheless maintain complex relationships with surrounding citizens and with authorities at multiple levels. Thus, it might be said that refugees suffer from a simultaneous absence and surfeit of statehood: as refugees, they lack territorial citizenship but as camp residents they are subject to the exercise of sovereign authority by multiple actors, including the host state, international humanitarian agencies and refugee leaders. (These organizations, and the network of relationships, agendas and interests within which they operate, are the subject of Chapter 5.)

The law: justice, order and legal pluralism The fragmented nature of authority is particularly evident in relation to the administration of justice. There has been little policy attention to the issue of law in refugee camps but for camp residents it is a crucial concern which has a direct bearing on physical safety and well-being. In a camp housing tens or even hundreds of thousands of people a whole range of social problems may occur, including matters related to administration (such as registration of births, marriages or deaths), family disputes (inheritance, divorce), civil disputes (breach of contract, debts and loans) and criminal cases (which can run the gamut from petty theft to murder). The surrounding context of a refugee situation may also give rise to security threats, whether from armed Downloaded by [University of Defence] at 21:28 09 May 2016 groups in the state of origin, surrounding villages in the host state, or even from the authorities entrusted with refugees’ protection. In some protracted refugee situations camp populations live in conditions of intense insecurity with almost daily acts of violent crime (Crisp 2000: 601). The 1951 Refugee Convention states that refugees are entitled to access the courts of a host state on the same terms as a national of that country.9 In practice, refugees’ access to the justice system of a host state is extremely limited: in a study of thirteen refugee situations, including Burmese refugees in Thailand, da Costa (2006: 21–22) found that the overwhelming majority Governing refugees 13

of crimes that occurred were resolved by refugees themselves. The failure of a host state to extend policing functions to a refugee camp has led some ana- lysts to suggest that there is ‘no law’ in refugee camps or that camps are sites of unfettered power (Harrell-Bond 2004: 27; Verdirame and Harrell-Bond 2005: 334; Agier 2011: 82). In reality, there are likely to be numerous systems of order maintenance and community management. The systems used will vary, but might include written camp rules or by-laws, committee structures of dispute resolution and systems operated by religious or cultural leaders. The range of legal infl uences on decision-making is typically even broader, including the law of the host state, the law of the refugee’s home state, international human rights law, laws drafted by non-state armed groups, and religious codes or values. These systems may have signifi cant limitations but they are not without meaning, authority and normative effect. The refugee camps in Thailand show considerable regulatory complexity: a host state which has imposed many restrictions on the refugee population but has little interest in policing refugee disputes, a refugee-led camp committee which enforces a set of rules drawn up by a committee of elders and, increasingly, international organizations seeking to harmonize camp practice with both Thai law and international human rights standards. While the Royal Thai Government has legal jurisdiction over the camps, the vast majority of disputes and problems are resolved within the refugee community. Despite this limited access to Thai courts the camps maintain a high level of social order and political stability, and refugee-led committees have a high degree of popular legitimacy (see further Chapters 5 to 7). This situation points to an important truth of social ordering: that the alternative to order imposed from above is not necessarily chaos but can be the construction of social order from within. Similar conclusions have been reached by other authors who have shown that poverty and marginality do not equate to a lack of social order or social capital and, further, that the maintenance of social order in such circumstances is not spontaneous or accidental but is the product of sustained effort and specifi c practices (e.g. Burton 1978; Small 2004; Suttles 1968). Understanding how this occurs is essential to understanding the nature of refugee protection in camps. The role of customary or non-state justice systems is gaining increasing attention within development policy as an important component of realizing Downloaded by [University of Defence] at 21:28 09 May 2016 ‘access to justice’ (Harper 2011; Ubink and McInerney 2011; Pimentel 2010a, 2010b). In refugee settings, while there has been some recognition of the function and value of refugee-led justice systems, programme activity has largely continued to focus on increasing refugees’ access to host state policing and justice structures (UNHCR 2006b: 25–37). Hesitant engagement with refugee-led justice systems may be related to concerns that such systems violate human rights standards or due process, that they are a ‘parallel’ legal system in violation of state sovereignty or that they are inevitably second-best to formal courts. However, this is not necessarily the case. 14 Governing refugees

‘Internal’ refugee-led structures may attract high levels of popular support and legitimacy (as they do in Thailand), and may play a vital role in camp management. This is not a new debate. Dispute resolution was one of the earliest fi elds of anthropological inquiry, when it was established that an absence of ‘courts and constables’ is not necessarily an absence of legitimate authority (Malinowski 1926; Gluckman 1965). Analyses of the relationship between colonial and indigenous legal orders explicitly challenged ideas of ‘legal centralism’ (the perception that law is coterminous with the European legal model of a centralized, professionalized criminal justice system) with recognition of ‘legal pluralism’ and the diversity of legal orders (Griffi ths 1986). The central fi nding of this work was that in many countries, if not most, state law is not the most important regulatory force. Instead, localized practices typically operate as ‘juridical micro-climates, specifi c juridical practices which emerge from the social relations in these areas and which, although informal, have a high effi ciency’ (Santos 1984: 98). A second crucial insight was that the relationship between indigenous justice and colonial legal systems was not hierarchical but dialectic, and that legal subjects shift between these sites with fl uidity and agency (Nader 1969a; Moore 2000 [1978]; Merry 1988; Starr and Collier 1989). Pluralistic relationships can fundamentally alter the institutional and normative basis of local justice systems. Institutionally, the co-existence of multiple legal orders can encourage ‘forum-shopping’ by disputants, while normatively it can lead to the transformation of rules and principles. This latter point is captured by Moore’s (1973: 720) concept of a ‘semi- autonomous social fi eld’; a system which ‘can generate rules and customs and symbols internally, but [. . .] is also vulnerable to rules and decisions and other forces emanating from the larger world by which it is surrounded.’ This concept of a cross-fertilization of ideas and values is a valuable one for the environment of a refugee camp, where ‘internal’ policing and security systems are both infl uenced and constrained by the laws of host governments and the preferences of international organizations. Recently there has been a revived interest in legal pluralism, as a proliferation of transnational and supranational legal institutions have introduced new layers of relationships between legal authorities and suggested a need for Downloaded by [University of Defence] at 21:28 09 May 2016 deeper interrogation of the role of law in society, on global and local scales (Mundy 2002: xix; Merry 2006a; Rosen 2006; Santos 2006). Rule of law reform is increasingly at the forefront of development and humanitarian agendas, with ‘access to justice’ seen as a means to transform society, advance social justice and reduce inequality (Commission on Legal Empowerment of the Poor 2008: 1). Law is a bigger export than it has been at any time since colonialism and in a similar fashion to that era a ‘system of thought’ is being exported with the dominant legal models (Comaroff and Comaroff 2006: 1–57). Most infl uential in this regard is international human rights Governing refugees 15

law, the ‘lingua franca of global moral thought’ (McEvoy 2007: 418). Human rights promise universality, equality and social justice but the international human rights regime has also been criticized for neo-imperialistic ambition, replacing one set of principles for a world community (nationalism, empire and religion) with another (‘installing liberalism and the conditions of free trade’) (Brown 2004: 461; see also Chimni 2006; Kennedy 2002; Mutua 2002; Douzinas 2007). The alienating potential of this is obvious. As Speed and Collier write: ‘Only when the production of human rights becomes a project in which many voices participate will human rights cease to be an imposition of Western values and become [a] cooperative venture’ (2000: 882, emphasis in original). These changes in transnational legal practice have reverberated around the world, reaching even to the refugee camps on the remotest corners of the Thai–Burma border. After many years in which justice and security for these camps was effectively abandoned to the refugee population the past decade has seen numerous international programmes on these issues. In the chapters that follow, I examine the impact of these programmes on the existing refugee-led camp management systems. In doing so, I explore several of the themes discussed above – the production of order beyond the state, the institutional relationships through which law is created and enacted, and the inﬂ uence of international human rights discourse on local justice practice.

Methodology This research fi ts within the broad tradition of ‘law and society’ scholarship, with particular attention to an inter-disciplinary dialogue between anthropology and criminology. However, it is a socio-legal study carried out in a context where neither the ‘society’ nor the ‘law’ could be taken for granted. Establishing the contours of each required approaching the refugee camps holistically, with a view to understanding the various infl uences that have shaped their existence and development, recognizing the many stakeholders in camp management, and seeking to understand what this climate of inter- secting authority means for refugees. My goal was to contextualize refugees’ experiences of encampment (bearing in mind that to fully understand the social, historical, cultural and political infl uences on these camps – or on any Downloaded by [University of Defence] at 21:28 09 May 2016 society – is an impossible aim.) My research focused on the experiences of Karen refugees. The ‘Karen’ and ‘Karenni’ camps have different refugee-led administrations but all nine camps are administered similarly by TBC (though not necessarily by other NGOs, not all of which operate in all camps) and the two groups have a great deal of common experience. Ethnic plurality has increased in some camps since the beginning of the resettlement programme but the overwhelming majority of camp residents are still Karen. Furthermore, not all camps have experienced signifi cant change. In Mae La (the largest camp 16 Governing refugees

on the border) 16 per cent of the population is now non-Karen, and in Umpiem Mai this number rises to 25 per cent; but in Mae La Oon, Mae Ra Ma Luang, Ban Don Yang and Tham Hin camps, the non-Karen population is less than 5 per cent.10 I conducted fi fteen months’ fi eld research on the Thai–Burma border between 2007 and 2012, including around eight months living in a camp and working as an English teacher in the camp education system. This provided the opportunity to live within the refugee community, to observe daily life, to meet repeatedly with camp leaders, teachers, justice and security workers and many ordinary refugees, to observe the justice process, and to access the justice and security records that were kept by camp staff. I conducted fi eldwork interviews in Mae Sariang, Mae Sot, Mae Hong Son, Chiang Mai and Bangkok, and in four refugee camps: Mae La Oon, Mae Ra Mo Luang, Mae La and Umpiem Mai.11 The majority of my research was conducted in Mae La Oon. The fi rst phase of my research (September 2009–June 2010) focused on refugee-led governance and justice systems. Return visits (January–April 2010 and April 2012) focused primarily on international projects related to justice.12 Throughout the various fi eldtrips I conducted 170 interviews with members of camp administrations, community-based organizations, international non-governmental organizations and others, in addition to many more informal visits, conversations and meetings which I documented in daily fi eld notes. Though my work concentrated on the ‘Karen’ camps I also conducted a small number of interviews with Karenni refugee leaders and community-based organizations. The primary benefi t of extended fi eldwork is the additional rigour provided by the opportunity of testing data collected over an extended period and against multiple sources. Over repeated conversations, often with the same individual, I learned what questions to ask and how to ask them for the most expansive responses. Lubkemann (2008: 34) describes a similar experience in research with refugees in Mozambique and concludes: ‘It is this ultimately enhanced knowledge of “how to ask” that accrues through this process that I take to be the hallmark strength of anthropological history.’ At the outset of my work I was warned that refugees in camps would not speak freely about their experiences of governance, crime and justice. Mindful of Downloaded by [University of Defence] at 21:28 09 May 2016 such warnings, and recognizing that matters of governance and authority are inevitably sensitive, I allowed others to dictate the terms and fl ow of information. In this, I followed Malkki (1995b: 51) who suggests that in her experience,

‘the success of the ﬁ eldwork hinged not so much on a determination to ferret out “the facts” as on a willingness to leave some stones unturned, to listen to what my informants deemed important, and to demonstrate my trustworthiness by not prying where I was not wanted.’ Governing refugees 17

In this work I highlight many largely unreported aspects of life in a refugee camp, including the existence of a cohesive community and a vibrant social world. However, recognizing that displacement can create ‘new possibilities, new relationships and new forms of political organisation’ (Kibreab 2005: 24) is not intended to imply support for the use of refugee camps or complacency regarding their existence. While it is important to recognize that refugees in Thailand have responded to the challenges of encampment with remarkable resilience (and that in their experience there is much that deserves close examination), this does not alter the fundamental nature of protracted encampment, or the many deprivations and indignities it imposes on refugees.

Chapter overview The book falls loosely into two parts, with the fi rst three chapters addressing aspects of history, culture and community and later chapters analyzing the practice of camp management, administration of justice and the consequences of legal pluralism. Chapter 2, ‘The Karen in Burma: confl ict and displacement’, provides a political background to the creation of the refugee camps, with an overview of Burma’s ethnic politics and Karen experiences of colonization, nationalism, confl ict and displacement. Chapter 3, ‘The camp community’, explores the refugee camps as sites for the construction and practice of community identities. It has frequently been stated that a refugee camp cannot constitute a ‘community’ (Hyndman 2000: 140; Harrell-Bond 2004: 27; Colson 2003: 10). This chapter seeks to demonstrate that the refugee camps in Thailand are indeed dynamic societies with their own shared culture and social norms. Three overlapping ‘camp communities’ are identifi ed: a community of shared identity as ‘camp people’ or refugees, an imagined community of ‘the nation’ and a community of shared governance. The most potent signifi er of community is the imagined community of national identity. The second part of the chapter details some of the ways in which an ethno-national ‘Karen’ community is constructed and mobilized in the refugee camps, including through narrative and memorialization, language and clothing. Downloaded by [University of Defence] at 21:28 09 May 2016 Life in a camp is not a rupture from refugees’ past experience but is in many respects a continuation of it. Chapter 4, ‘The governance palimpsest: order maintenance in Southeast Burma’, considers the historical context of law and governance within Karen territories of Burma and examines how these practices from pre-camp life have been incorporated into camp governance. The Karen refugee situation illustrates the fragmented nature of sovereign authority and law, but it is not a manifestation of devolved governance in the ‘post-regulatory state’ (Crawford 2006). Instead, many techniques of camp governance are relics of a society that never fully came 18 Governing refugees

within the state, of a community justice process that was never entirely rationalized: not a post-regulatory state but a pre-regulatory state. This chapter traces historical eras and ideologies of order maintenance, and identifi es a variety of techniques for the production of order, including village-level sovereignty, supernatural and spiritual beliefs, the state- building project of the Karen National Union and the present context of confl ict and contested governance. The infl uence of these various processes on community governance is illustrated with discussion of responses to adultery, consistently responded to as a very serious transgression within Karen society. Chapter 5, ‘Sovereigns and denizens: camp governance and “the refugee”’, shifts attention from techniques used to produce order and considers the ideologies that stimulate governance and regulation. Karen systems of self-government have enabled autonomous camp management but refugees are not the only authorities with a role in governance. Refugee community leaders are confronted with other authorities such as the Royal Thai Government and the UNHCR. This chapter examines the governance approach of the three primary sets of actors in the camp (the Thai authorities, international humanitarian organizations and refugee leaders) and identifi es their main governance objectives. Broadly speaking, the Thai authorities seek the containment and control of refugees, while international organizations promote greater freedom and human rights protection. Refugee leaders and representatives are required to implement both sets of demands, while also pursuing their own goals of ensuring the continued existence of the refugee camps and preserving cultural and ethnic identity. This chapter also examines the role of the Karen National Union in governance, as part of a wider debate on refugee militarization and the political identity of refugees. Pluralistic relationships and ideologies are also apparent in relation to the administration of justice. After decades in which security and justice issues were wholly absent from international policy agendas, and in which refugee leaders were expected to cope independently, ‘access to justice’ has recently become a priority for the UNHCR and other international organizations (notably the International Rescue Committee [IRC]). Programmes have included the creation of Legal Assistance Centres, assistance in referring cases to Thai court and efforts to defi ne the jurisdiction and procedures to be used Downloaded by [University of Defence] at 21:28 09 May 2016 by camp justice. Chapter 6, ‘The struggle for ownership of justice’, examines the refugee-led camp justice systems, some key programmes by international agencies, and the consequence of the interactions between the two. Chapter 7, ‘Enacting interlegality: human rights and local justice’, analyzes the normative consequences of a pluralistic legal environment by tracing the local impact of international human rights discourse. The chapter compares local responses to two areas of human rights that are considered priorities for the UNHCR and other international organizations: women’s rights and children’s rights. Regarding women’s rights, refugee-led community Governing refugees 19

organizations have served as ‘knowledge brokers’ (Merry 2006b) to encourage acceptance of gender equality by camp leaders and residents. Regarding children and youth, however, human rights are largely seen as an imposition and a restriction on the camp’s authority, in particular on the ability of camp leaders to respond to youth delinquency. The chapter concludes with a discussion of the impact of a legally pluralistic environment and the opportunity for the manipulation of values in ‘forum shopping’ between legal orders. Chapter 8, ‘Beyond encampment’, recaps the main arguments of the book, outlines some possible reasons why these camps have remained stable when so many others do not, and suggests the implications of this for efforts to promote ‘refugee self-reliance’. The chapter concludes by considering the future for refugees in Thailand in light of political changes inside Burma and discussions regarding return and repatriation.

Notes 1 In 2011, the total population of concern to UNHCR was 35,444,128, of which 9,674,153 people were in camps and settlements (UNHCR Global Trends Data 2012). This number does not include Palestinian refugees from 1948 and 1967 (more than 1.5 million of whom live in camps in Lebanon, the West Bank and Gaza) nor refugees who are not registered with UNHCR. So, for example, in December 2011 the UNHCR population of concern in the nine Thai–Burma camps was 88,011 but TBC’s verifi ed caseload was 137,742 (TBC 2011). 2 It is diffi cult to offer precise evidence of the relative security in these camps as there are no reliable records. Indicative numbers can be found in a survey of more than 2,000 refugees in Ban Mai Nai Soi, Ban Mae Surin, Mae La, Umpiem Mae and Nu Po, where 367 people reported having experienced crime in the previous twenty-four months. Of these, 227 had experienced actual or attempted property offences (theft, burglary or criminal damage), while seventy-nine were victims of actual or attempted violent or sexual offences (IRC 2010: 46). By way of comparison, the rate of past-year police recorded crime in England and Wales (2012–2013) was sixty-six per thousand; a dramatic drop from 2002–2003 rates of 114 per thousand (Offi ce for National Statistics 2013). 3 It is possible that in situations of political imprisonment, an ‘inmate society’ may have stronger bonds of shared past and future (see e.g. Buntman 2003, McEvoy

Downloaded by [University of Defence] at 21:28 09 May 2016 2001, Peteet 2005). The comparison between prisons and refugee confi nement is also stronger in relation to immigration detention, where asylum claimants have no freedom of movement and are confi ned within a single institution (Simon 2000). Finally, in the event that refugees are able to return home they usually do not confront the same level of stigma that a prisoner faces on release (though refugee return of course presents its own social and political challenges). 4 Burma’s ethnic politics – and the resulting confl ict, persecution and displacement from Southeast Burma – are explained in detail in Chapter 2. 5 Of the two camps under the Karenni Refugee Committee, Ban Mae Surin also has a majority Karen population. 20 Governing refugees

6 Mae La Oon and Mae Ra Ma Luang are a ninety minute walk or fi fteen minute boat trip apart, with Mae La Oon situated closest to the Burmese border. Mae Ra Ma was created in 1994 and has experienced no further consolidations or relocation since that date. Both camps are situated along the banks of the Mae Sariang river, but where Mae La Oon is squeezed into a steep valley, Mae Ra Ma lies on a fertile plain. Houses in Mae Ra Ma are typically larger and often have a small patch of land alongside. The contrast with Mae La Oon is visible in vibrant vegetable gardens, fruit orchards and fl ower trellises. River water is harnessed by hydropower generators which provide electricity to many houses in the camp. 7 In 2011, the monthly food ration provided for an adult was 15kg rice, 1kg yellow beans, 750 grams fi shpaste, 1 litre soybean oil, 0.125kg chilli, 0.3–0.5kg salt and 0.25kg sugar. Due to funding cuts, rations were reduced in 2012 and again in 2013. In 2013, the ration pack no longer included chilli and sugar, and the fi shpaste and salt allowances were reduced to 0.5kg and 0.2kg respectively (S’Phan Shaung 2013). 8 The term ‘third country’ refl ects the assumption that refugees fl ee one country (Burma) to seek asylum in a second country (Thailand) and are resettled to a receiving third country. 9 See Article 16 (1): A refugee shall have free access to the courts of law on the territory of all Contracting States; and 16(2) A refugee shall enjoy [. . .] the same treatment as a national in matters pertaining to access to the Courts. [. . .] 10 See TBC (2013). As at March 2012, the population composition was: Mae La Oon 99.3% Karen, 0.57% Burman; Mae Ra Ma Luang 99.76% Karen, 0.14% Burman; Mae La 84% Karen, 2.87% Burman, 13.23% Other; Umpiem Mai 74.47% Karen, 14.07% Burman, 3.56% Mon, 7.9% Other; Nu Po 77.76% Karen, 9.17% Burman, 1.85% Mon, 11.22% Other; Ban Don Yang 95% Karen, 1.2% Burman, 3.08% Other; Tham Hin 98.5% Karen, 0.99% Burman, 0.51% Other. Population fi gures are from March 2012. 11 Every camp on the border is slightly different and it is not always appropriate to generalize from the experience of one camp to another. I conducted camp-based interviews in four camps (Mae La, Umpiem Mai, Mae La Oon and Mae Ra Ma Luang), but the majority of my camp-based fi eldwork was conducted in a single camp, Mae La Oon. 12 I was less successful in accessing members of the Thai authorities. When I contacted members of the police and court structures, I was told that all interview requests had to be fi ltered through a central authority and that personnel would be selected for me to talk to. I decided this would not be likely to lead to productive conversation and did not make further attempts to interview Downloaded by [University of Defence] at 21:28 09 May 2016 representatives of the Thai authorities. There is an important gap in research on the offi cial response to refuges and migrants from Burma in Thailand, but I was not well situated to conduct it. However, while the role of the Royal Thai Government is essential to understanding policy choices and possibilities in the realms of camp governance and justice, the government perspective was not essential to understand and analyze refugees’ responses to those policies. Chapter 2 The Karen in Burma: conflict and displacement

To understand the experiences of Karen refugees, it is necessary to understand their displacement in political and historical context. This chapter provides an introduction to Burma’s ethnic politics, and an overview of the Karen experience of missionization, colonization and nationalist revolution. The fi nal sections explain the context of confl ict and displacement in Southeast Burma, introducing the refugee camps on the Thai–Burma border. Recent political liberalization in Burma has given rise to hopes that the country will transition to democratic governance and that refugees may be able to return home in safety. This is a dramatic change after several decades of militarism. Burma gained independence from Britain in 1948 but liberal governance was short-lived: a military coup deposed the democratic government in 1962 and a series of military juntas controlled the nation for the next fi fty years, fi rst under the leadership of Ne Win (1962–1988), followed by the State Law and Order Restoration Council (1988–1997) and the State Peace and Development Council (SPDC) (1997–2010). Beneath the name changes, military rule continued unabated, as did repression, censorship, arbitrary detention and a litany of other human rights violations. Mass protests challenging military rule were brutally suppressed in 1988 and again in 2007. In 2003, the then-ruling SPDC announced that it was implementing a seven-stage ‘Roadmap to Democracy’, where the fi nal stage was to ‘build a modern, developed and democratic nation’.1 National and international scepticism only had cause to deepen as the various stages along the roadmap Downloaded by [University of Defence] at 21:28 09 May 2016 were passed: the creation of a National Convention to draft the Constitution (which was entirely dominated by military actors); the resulting Constitution (which guaranteed a continued role for the military in governance); a national referendum to approve that Constitution (which took place days after the devastating landfall of Cyclone Nargis, yet still contrived to secure a ‘yes’ vote from more than 92 per cent of Burmese citizens); and general elections which returned 129 of 168 available seats to the military-backed USDP (amid reports of fraud, ballot-stuffi ng and vote-rigging). 22 Governing refugees

The penultimate step on this roadmap was reached in 2010, when a national legislative assembly was convened and ostensibly replaced the military regime with a civilian government. At this stage there was little to indicate a genuine handover of power would occur – but then the process began to diverge from expectations. In November 2010 the leader of the National League for Democracy, Aung San Suu Kyi, was released from house arrest. In September 2011, construction was halted on the Chinese-funded Myitsone Dam, a ‘development’ project in northern Kachin State which would have displaced tens of thousands of people. In October 2011, more than a hundred political prisoners were released in the ﬁ rst of a series of amnesty processes. In April 2012, by-elections were held in which the National League for Democracy won forty-three of the forty-four seats it contested, and in May 2012 those representatives (including Daw Aung San Suu Kyi) took up their seats in Parliament. General elections have been scheduled for 2015, with a pledge that these will be free and fair. The international response was swift and decisive. The European Union lifted sanctions, vast sums of development aid were pledged and world leaders ﬂ ocked to visit Burma. However, while the political changes to date are undoubtedly important, their true reach remains unclear. Certainly they must be understood in the context of a nation with enormous social problems and deep divisions. Economic development, democratization and demilitarization will all be necessary to lasting political change but the most compelling challenge to the dominant narrative of liberalizing transition – and the most profound gap in much international analysis – is the position of Burma’s ethnic nationalities.

Burma’s ethnic politics Burma is the original ‘plural society’, a phrase coined by Furnivall (1939: 446) to describe its mosaic of ethnic diversity and the diffi culties of building a cohesive nation from such disparate fragments. Eight ethnic ‘nationalities’ are recognized (Burman, Arakanese, Chin, Shan, Kachin, Karenni, Karen, Mon) and many more sub-groups.2 Precise population fi gures are impossible as the last national census was conducted in 1983, but Downloaded by [University of Defence] at 21:28 09 May 2016 the ethnic Burman population is estimated to comprise approximately two-thirds of a total country population of approximately fi fty million (Steinberg 2010: xiv).3 Ethnic diversity has posed a governance challenge throughout Burma’s history. Superfi cial territorial cohesion was imposed by the British colonial occupation but collapsed even before independence. In 1947, an agreement was reached between the incoming government and representatives of the Chin, Kachin and Shan peoples to grant those nationalities administrative autonomy within an overarching Union of Burma. This, the Panglong The Karen in Burma 23

Agreement, was not refl ected in the 1948 Constitution, which provided only a limited right of secession to Shan and Karenni states and gave no recognition to ethnic nationalities which had not participated in the talks, including the Karen and the Mon. Later, the 1974 Constitution imposed even more direct authority over the ethnic territories, dividing Burma into fourteen administrative regions: seven divisions in the centre of the nation where the dominant ethnic group is Burman, surrounded by seven ethnic states where the majority population is a non-Burman ethnic nationality. The distinction between ‘division’ and ‘state’ has little practical signifi cance as political authority is claimed equally across all territory.4 In short, while Burma’s border territories and ethnic nationalities were brought within the Union of Burma at the onset of national independence, failure to grant political autonomy to ethnic groups was and remains the primary cause of confl ict between these groups and the Burmese ruling regime. Uprisings against the Burmese Government began before British withdrawal and have continued ever since. The fi rst ethnic group to take up arms was the Karen National Union (KNU), in 1949.5 Many more followed: Lintner (1999: 480–95) identifi es sixty-one non-Burman armed groups as operating at various times between 1948 and 1998. Some of these groups fell away in the natural attrition of changing political climates while others reached ceasefi re agreements with the ruling regime (see Kramer 2009). Since 2010, the Burmese Government has embarked on negotiations with the remaining non-ceasefi re groups, including the KNU/Karen National Liberation Army (KNLA), Shan State Army-South and the Chin National Front (though in the same period, the Kachin Independence Army returned to active confl ict after a sixteen-year ceasefi re which operated from 1994– 2010). A provisional ceasefi re was signed between the KNU and the Burmese Government in January 2012. At the time of writing (August 2013) negotiations for a lasting peace agreement were on-going.

Who are the Karen? The term ‘Karen’ suggests a discrete ethnic category but the several million Karen people in Burma and Thailand form a population with diverse linguistic, religious, political and territorial backgrounds (Hinton 1983; Downloaded by [University of Defence] at 21:28 09 May 2016 Harriden 2002; Delang 2003). It is thus more accurate to consider ‘Karen’ as an umbrella term for multiple sub-groups, including Sgaw, Pwo, Bwe and others (and, drawn at its broadest, potentially also including the Karenni and Pa-O, two groups which have their own ethnic self-determination movements). Estimates of the Karen population in Burma vary greatly, with Burmese military leaders asserting three million and Karen nationalists claiming a ﬁ gure more than twice as high. The two most widely spoken Karen languages are Sgaw (predominantly spoken by Christian and highland Karen) and Pwo (predominantly spoken by lowland/delta Buddhist Karen). 24 Governing refugees

Karen live in all parts of Burma, with sizeable centres in the Southeastern border areas of Karen State and Tenasserim division and in the lowland territories of the Irrawaddy Delta. There is also an indigenous Karen population in Thailand. Due to the difﬁ culties of researching inside Burma, most recent ethnographic work on Karen society has been conducted among Karen communities living in Thailand (Lehman 1979; Andersen 1979, 1981; Keyes 1979; Rajah 1986, 2002; Delang 2003; Hayami 2004), in refugee camps (Lang 2002; Tangseefa 2006, 2007; Horstmann 2011a, 2011b) or in insurgency areas (Cusano 2001; South et al. 2010; South 2012a). Much less is known about the ‘other’ Karen populations inside Burma, who in fact constitute the majority Karen population (Thawnghmung 2008, 2011b). Karen nationalism propounds a message of Karen unity and a single Karen family. It has been argued that this message shoe-horns untenable linguistic and religious complexity into a narrow ideal of ‘Karenness’ (Harriden 2002; Kuroiwa and Verkuyten 2008). Undoubtedly the constituency of Karen people is wider than the propaganda of political nationalism often acknowledges. However, despite the substantial differences, it is also the case that ‘very large numbers of people do subscribe to a distinct Karen identity’ (South 2008: 15). Recognizing that Karen identity has a real existence and meaning does not imply that it is a ﬁ xed or static category. Historically, and presently, Karen people have interacted with other ethnic groups through trade, employment and intermarriage. The depth of these interactions is apparent in the use of multiple identities. For example, Thai- speaking Karen may choose to ‘pass’ as Thai in many interactions, using a Thai name and speaking Thai language, yet will use a Karen name and Karen language when with other Karen people. Such processes of situational and strategic deployment of ethnic identity have a long history in this region (Moerman 1965). As a pre-literate society until the arrival of Christian missionaries in the early 1800s, there are few historical records of Karen societies prior to that time. Unlike other ethnic groups such as the Shan or Mon, the Karen historically held no great wealth or power and appear to have existed on the peripheries of other kingdoms rather than holding centres of power in their own right (Aung-Thwin and Aung-Thwin 2012: 48–9). Karen oral histories describe a nomadic past of persecution and displacement following a ‘river of Downloaded by [University of Defence] at 21:28 09 May 2016 running sand’. This was interpreted by Baptist missionaries in the nineteenth century as a reference to the Gobi desert, indicating a migration pathway through Mongolia and China to present-day Burma. A second theory that emerged during this period was that the Karen people were a ‘lost tribe of Israel’ and that their migration had begun in the Holy Lands, a version believed by many of the refugees in Thailand. In the pre-colonial era, non-Burman ethnic groups (including the Karen) had substantial autonomy but existed within tributary relationships of patronage and taxation, a model of rural village governance that is common The Karen in Burma 25

throughout Southeast Asia. Scott (2009) has argued that the Karen and other ethnic groups in the mountain massifs of this region made a deliberate choice to reject the central state (and its voracious demands for tributes, labour and other resources) in favour of autonomous self-government. While the facts interpreted by Scott in reaching this conclusion – that many Karen lived in remote highland regions, maintained autonomous village management systems and languages and largely avoided prolonged contact with agencies of the central state – could be interpreted as simple survival tactics rather than a ‘distinctive positioning’ (ibid: 8) or strategic adaptation, perhaps the real proximate cause is irrelevant. Whether the Karen actively rejected the nation-state or whether their isolation was a historical and territorial accident, it is apparent that they were successful in maintaining autonomous governance long beyond most other societies. As Scott (ibid: 26) recognizes, this may have important implications for resilience and coping strategies:

Noting how the Karen are spread throughout many ecological and political zones, Renard believes that the remarkable suppleness of their social structures, their oral histories, kinship patterns, subsistence techniques, cuisine and architecture is adapted for travel and change. If necessary, most Karen groups can turn on a dime. It is a quality that has great adaptive advantages and has served them well.

Scott’s analysis is historical, but in remote areas of Southeast Burma village level local autonomy survives to the present day. A Karen Human Rights Group (2008: 33) report states that ‘villagers generally see land rights as something to be determined and resolved locally rather than distributed or veriﬁ ed by some distant and foreign (including centralized state) power’ while Heppner (2006: 18) has described village level governance as intimately related to the success of coping and resistance strategies of displaced people in Karen State (see also South et al. 2010). It has also been extremely important for the refugee camps, as is examined further in Chapters 4 and 6.

Christian missions and colonization Christian missionaries began to arrive in Burma in the early 1800s, with the Downloaded by [University of Defence] at 21:28 09 May 2016 English Baptist mission beginning in 1807 and the ﬁ rst American Baptist missionaries arriving in 1813. Their efforts met with little enthusiasm among the predominantly Buddhist population in central Burma but the response was very different among non-Burman populations, particularly the Karen. After the ﬁ rst ‘Karen convert’, Ko Thah Byu, began to evangelize, mass baptisms were conducted of hundreds or even thousands of Karen at a time (Mason 1847; Smeaton 1887). The extraordinary reception for Christianity among the Karen is usually explained in relation to symmetry between the Christian message and a Karen myth which told of a ‘golden book’ of knowledge that had 26 Governing refugees

been lost and would be returned to the Karen by the ‘white elder brother’. When white missionaries arrived preaching the importance of the holy bible, this was received as the promised return of the golden book.6 Christianity never became the majority religion among the Karen in Burma but it has had a powerful and lasting infl uence nonetheless. As in many other contexts, missionaries enhanced literacy by developing a Karen script (adapted from the Burmese alphabet) and establishing schools. The missionary infl uence remains apparent today as although the majority of the Karen population are Buddhist or Animist, the political and military leadership of the Karen nationalist movement is overwhelmingly Christian and the majority of literate Karen are Christian. There is also a strong Christian infl uence in the refugee camps, where churches serve variously as a social hub and focal point, a source of education and entertainment and, at times, a mechanism of social control (Horstmann 2011b).7 These functions are broadly similar for Buddhist monasteries, Muslim mosques and Christian churches but the dominance of Christianity is clearly visible, leading to an environment in which as one of my interviewees commented, ‘many Buddhists are becoming Christian but no Christians are becoming Buddhist’.8 Among those I spoke with who had converted to Christianity, several had done so in order to marry a Christian spouse, others to conform to the practice of their fellow students or friends and neighbours, while one man said that he simply ‘liked their songs’. This may suggest that religious conversion owes less to theology than to the existence of a critical mass – the majority of the population are Christian, which encourages others to convert in order to share the same activities, entertainments (and, perhaps, opportunities). The tradition of mass baptisms also continues: in January 2010 at Mae Ra Ma Luang camp, 456 people were baptized in a single event. The Karen Baptist Committee apparently recognizes the importance of the refugee camps as a site of religious conversion, with the adoption of a chapter from Genesis as a motto: ‘You intended to harm us, but God intended it for good, to accomplish what is now being done, the saving of many lives.’9 Soon after the Baptist missions fi rst arrived in Burma colonization began. The process took place over a series of three Anglo-Burmese wars, from 1824–1926 (when the fi rst colonial administrators came to Burma and Tenasserim was annexed), 1852–1853 (when lower Burma was annexed) Downloaded by [University of Defence] at 21:28 09 May 2016 and 1885–1886 (when the last King of Burma, Thibaw Min, was exiled to India and Upper Burma was annexed (Myint-U 2007). Colonization had important consequences for the Karen and their relationships to government. The British colonial approach was to ensure the reach of the state throughout a national territory but there were manifest diffi culties in extending government to the malarial, mountainous eastern border region. Administrators were content to avoid the attempt, choosing instead to see the area as ‘perfectly worthless in itself [. . .] almost impracticable, even for an elephant’ (quoted in Ghosh 2005: 174).10 The Karen in Burma 27

Diarchic rule was implemented, distinguishing Burma proper (under direct rule, with local governance structures abolished) and the borderland frontier areas (under indirect rule, permitting village-level governance structures to remain in place). Indirect rule did not immunize the Karen from colonial inﬂ uence but it did permit continuity in local governance that was blocked in the heartland of the nation (Safman 2007: 55; Taylor 2009: 80–3). The Karen in Southeast Burma were less tightly controlled by colonial authority than the Burman majority but despite this (or perhaps because of it) they were also more supportive of colonial rule. Following the familiar strategy of ‘divide and rule’, Karen and other ethnic nationalities were favoured in the British administration and heavily promoted in the civil service, police and military structures. An educated Karen middle class emerged that was loyal to the British administration and played an essential role in suppressing Burman resistance to colonization, such as during the rebellion led by Saya San between 1930 and 1932, when up to 20,000 rebels were killed or captured (Lintner 1999: 32). These governing processes have had a lasting impact on the ‘Union of Burma’ and particularly on relationships between the Burman majority and minority ethnic groups.

The birth of Karen nationalism The new Karen middle class – centred in educated Christians from Rangoon and the Delta area – became the catalyst for a new Karen nationalist movement (Gravers 1999: 24–30; South 2008: 13–16). A Karen National Association (KNA) was formed in 1881 as an organization of political fellowship. Under its direction several nationalist symbols were adopted that remain in use today, including a Karen fl ag, a national anthem and ‘Karen New Year’ as a national day of celebration. The chairman of the KNA, San C. Po (2001 [1928]: 81), had romantic visions of Karen independence but at this stage it was conceived as a distant ambition to be achieved through political and non- violent means: ‘“Karen country”, how inspiring it sounds! What thoughts, what manly feeling, what wonderful visions of the future the words conjure forth in the mind of a Karen!’ It was only in the aftermath of the Second World War that the struggle for Downloaded by [University of Defence] at 21:28 09 May 2016 Karen independence began in earnest. During the war, the Karen fought on the side of the British in the belief that the British would later help secure an independent Karen nation. Terrible reprisals were infl icted on Karen villages by Japanese and Burmese forces but British withdrawal from Burma took place with no effort to secure special accommodation for the Karen. The KNU was formed in 1947 and took up arms in 1949, fi rst through the Karen National Defence Organization, later known as the Karen People’s Liberation Army and subsequently (and currently) as the Karen National Liberation Army. The demands of the Karen nationalist movement were defi ned in 1950 28 Governing refugees

by then-KNU leader Saw Ba U Gyi in four principles which remain the touchstone of Karen nationalism today:

1 For us, surrender is out of the question. 2 Recognition of the Karen State must be completed. 3 We shall retain our arms. 4 We shall determine our own destiny.

The KNU claim for self-determination rests on an argument of the sustained experience of ethnic discrimination and oppression, and a fundamental sense of ethnic difference:

Originally, Karen and Burman belong to two different racial origins (Mongolian and Tibeto-Burman), religions, philosophies (Animism and Buddhism), social customs and literatures. They belong to two different civilizations which are based mainly on conﬂ ict ideas and conceptions. Their aspect on life and of life are different. [. . .] To yoke together two such nations under a single state, one in numerical minority and the other as a majority, must lead to growing discontent and ﬁ nal destruction of any fabric that may be built up for the Government of such a state. (Po Chit 1947: 1)

It has been argued that this Karen sense of ethnic difference is a product of the colonial encounter (Renard 1990: 106; Cheesman 2002: 203). However, it is also apparent that ethnic identities were not entirely an invention of colonization. Perhaps it is more accurate to say that while colonial- era governance did not create a sense of difference and separation, it institutionalized it, and in doing so both hardened ethnic identities and made them more consequential. Unlike many postcolonial revolutionary movements, the KNU did not reject and oppose its former colonial oppressor. On the contrary, while Burmese nationalists clearly sought to break from the colonial order and defi ned themselves in distinction from it, Karen nationalists sought continuity with the colonial order and defi ned themselves in relation to it (Gravers 1999: 78–9).11 Downloaded by [University of Defence] at 21:28 09 May 2016 Insurgency Formed in 1947 and still in existence today, the KNU is sometimes described as the world’s longest running self-determination movement. In the early years of insurgency the KNU was strong in popular support, manpower and weaponry and its members were confi dent that their demands would be met within a few years at most (Lintner 1999: 96). However, for the next sixty years the revolution followed an almost consistently declining trajectory. The recent history of the KNU was dominated by the fi gure of General Bo Mya, president of the KNU from 1976 until 2000 (South 2008: 39–41; The Karen in Burma 29

Thawnghmung 2011b: 112). As a Seventh-day Adventist, fervent Karen nationalist and equally fervent anti-communist, Bo Mya’s personal convictions had a signifi cant impact on the direction of the KNU. His anti-communist stance aligned with the preferences of both Thailand and the United States and attracted considerable fi nancial support for the KNU (Smith 1991: 296–300). Bo Mya was equally infl uential for his opportunism and entre- preneurialism. During the 1980s and early 1990s, the KNU controlled substantial portions of the Thailand–Burma border and money fl ooded in from cross-border traffi c in timber and black market goods (Bryant 1997). In this period, though the possibility of defeating the Burmese army by military force was remote, the KNLA was a powerful guerrilla force and the two sides were locked in stalemate. The KNU also developed a political administration within Karen areas of Eastern Burma. At the outset of the revolution, the KNU named their proposed Karen State ‘Kawthoolei’ and divided the territory into seven administrative districts (each of which corresponds to a KNLA Brigade area) and further divided into townships. A civil service was established to administer the territory, with departments including Health, Education, Finance, Defence, Transport, Communications, Mining, Agriculture, Forestry, Information and Justice (Thawnghmung 2008: 26; Falla 1991). While never reaching a fully functioning parallel state, schools and clinics, courts and prisons were created within the ‘liberated areas’ which established the KNU as the de facto government.12 Community organizations were also formed to advance nationalist objectives and serve the Karen population, including the Karen Youth Organization, Karen Women’s Organization and the Karen Baptist Convention. This KNU state-building project was impeded by a growing imbalance of power between Karen insurgency forces and the Burmese Government. In 2007, the International Institute for Strategic Studies calculated that more than 40 per cent of the Burmese national budget was spent on defence and military expenditure (with a mere 0.4 per cent spent on health and 0.5 per cent on education). The number of troops in the national military doubled between 1988 and 2007, from 180,000 to 395,000. Against this vast force, KNU/KNLA troop size has been estimated at between 3,000 and 5,000 (Keenan 2012b: 1; South 2008: 56). KNLA troops were Downloaded by [University of Defence] at 21:28 09 May 2016 not only vastly out-numbered but were also woefully under-equipped: soldiers were furnished with WWII-era rifl es and insuffi cient ammunition and fought without pay, on inadequate food rations and in some cases even without shoes. Politically the KNU/KNLA was further weakened by counter-insurgency tactics which isolated the movement from other ethnic insurgencies and divided it internally. A series of damaging factional splits were likely fostered to some degree by the Burmese military regime but also refl ected genuine discontent with KNU/KNLA leadership failures and the escalating violence and suffering within Karen State (see South 2008: 57–68; Smith 1999). The 30 Governing refugees

most damaging of these was the formation of the Democratic Karen Buddhist Army (DKBA, later renamed the Democratic Karen Benevolent Army), which allied with the Burmese military to overthrow the ethnic insurgency headquarters at Manerplaw in February 1995. Manerplaw was considered a ‘liberated area’ and was home to a coalition of insurgency forces. Its overthrow sent thousands of people fl eeing to refugee camps in Thailand and was a critical blow for the KNU. The DKBA split was led by disaffected Buddhist Karen, ostensibly in protest at the dominance of Christianity within the KNU but also bound up in discontent with poor leadership, corruption and the affl uent, secure life- styles of KNU leadership (many of whom were living in large houses in Thailand, where their children attended school and university) in contrast to the poverty and insecurity of the military rank and fi le. The DKBA defection marked the beginning of a series of damaging defeats for the KNU. By 2012, the organization had lost control of almost all its territory inside Burma, almost all its sources of cross-border income and much of its leverage with the Burmese Government. Health care and education provision inside Burma was drastically cut back or led by internationally funded groups such as the Free Burma Rangers, the Backpack Health Worker Team and the Karen Teacher Working Group. A number of reasons can be posited for the decline of the KNU. A large measure of responsibility must be placed with poor leadership and a failure to seize political opportunities when they surfaced. Failure to manage internal dissent created fertile ground for splits and factionalism, while a failure to build strong alliances with other ethnic insurgencies allowed the SPDC to broker ceasefi re deals and consolidate its forces against a smaller number of opponents.13 Finally, the KNU suffered from changing political alliances as Thailand developed increasingly profi table political and economic relationships with the Burmese regime. The long history of confl ict has shaped political governance in Burma’s ethnic areas, with several authorities seeking political power and control over land, resources and populations and establishing an ‘emerging political complex’ involving ‘various government and nongovernment actors in ongoing contestation, negotiation, discord, cooperation and/or complicity over the nature and composition of political authority’ (Callahan 2007: 49; Downloaded by [University of Defence] at 21:28 09 May 2016 Oh 2013). Actors in this ‘emerging political complex’ include the Burmese regime and its army, the KNU and other non-state political and military factions, logging companies, the Thai Government, humanitarian organizations operating cross-border relief programmes and ‘indigenous’ civil society organizations. This environment of multi-layered political power is not unique to Karen areas but is also evident in other ethnic borderlands, including Kachin State, Karenni State and Mon State (Callahan 2007; Grundy- Warr and Dean 2003; Dudley 2003; South 2003).14 The extent of power each group possesses shifts according to variables such as the intensity and The Karen in Burma 31

location of confl ict, the character and political sympathies of key individuals, and the national politics of the Burmese and Thai Governments. This environment requires constant vigilance from all actors in reacting to the changing circumstances, and as such might be read as indicating political skills of analysis, accommodation and compromise as well as underlying instability (Callahan 2007: 49). This climate of competing sovereignties has prompted some observers to challenge representations of the KNU as the ‘sole legitimate representative’ of the Karen, noting that it is primarily associated with Southeast Burma (and therefore with a minority of the Karen population) and even in that region shares governing authority with a variety of other armed groups, civil society organizations, religious representatives and local village leaders, and evokes a range of responses from civilians, from loyal allegiance to active resistance (Thawnghmung 2011a; South et al 2010: 15–17). However, while the KNU is not the only non-state actor in Southeast Burma and is certainly not the voice of all Karen people, it continues to be the most signifi cant political and military power in Karen State. The continued importance of the KNU not only for Karen State but to Burma’s national politics is confi rmed by its on-going peace negotiations with the Burmese Government and its membership in the United Nationalities Federal Council (a coalition of political and military representatives from Burma’s various ethnic nationalities). Of course it is impossible to defi nitively assess the popular legitimacy of the KNU without a free and fair electoral process. However, surveys and reports from Eastern Burma suggest that it is considered to be less predatory than other armed non-state actors and certainly less so than the Burma Army (see also Callahan 2007: 36; Thawnghmung 2011a: 78–9; South et al. 2010: 16; South 2012a: 200–1). The relationship between an armed group and community may be one of broadly symbiotic partnership, as has been the case in some areas of Northern Ireland in terms of a relationship between a local population and the Irish Republican Army (e.g. McEvoy 2001; Sluka 1989). Alternatively, the relationship might be more clearly one of oppression, such as paramilitary policing operations in parts of rural Colombia (Taussig 2003). The relationship between the KNU/KNLA and the wider Karen population appears to sit somewhere between these two extremes, enjoying very high Downloaded by [University of Defence] at 21:28 09 May 2016 legitimacy in some geographical areas and among some constituencies but also at times committing acts of oppression, including extortion and forced military recruitment (Thawnghmung 2011b: 120–1).

Displacement In the sixty years of the Karen insurgency, there have been only fi ve rounds of peace negotiations (Keenan 2012a).15 A ceasefi re was signed in January 2012 but no fi nal agreement had been reached by mid-2013. It is not clear 32 Governing refugees

what would comprise an acceptable settlement for the KNU leadership. A proposal put forward in January 2012 focuses primarily on the protection and welfare of villagers in Karen State and does not mention demands for political autonomy or recognition.16 If the current negotiations do result in a formal settlement, it will mark the end of one of the world’s most protracted confl icts. As in all modern confl icts, the primary victims of this confl ict have been the civilian population. During the 1960s, 1970s and into the 1980s Burmese forces employed a counter-insurgency policy known as ‘Four Cuts’, so-named for its objectives of cutting links between insurgent organizations and sources of food, funds, intelligence and recruits (Smith 1991: 258–62). The Four Cuts policy was notorious for its cruelty but the violence infl icted by Burmese troops in Karen state continued to escalate from the mid-1990s. The catalogue of human rights violations has been extensive and extreme including forced labour, forced relocation, extortion, arbitrary detention, torture, killings and sexual violence (UN General Assembly 2011). Villagers have been forced to serve as human minesweepers or compelled to porter heavy loads until they collapse. Rape of women from ethnic nationalities has been widespread (KWO 2004, 2007; Woman’s League of Chinland 2007; Apple and Martin 2003). Nearly 4,000 villages were destroyed in Southeast Burma between 1996 and 2011 (TBC 2012: 2). Burma Army troops have been responsible for the greatest number of human rights violations against civilians but rural farmers and villagers are also called upon to give food and other support to KNLA soldiers and to breakaway groups such as the DKBA, Karen Peace Force and Tha Ka Sa Pa (Thawnghmung 2011a: 78).17 For rural villagers trapped between multiple warring forces the confl ict has affected every aspect of daily life, crippling education, social development, income generation and health. One consequence of the protracted violence and insecurity has been massive displacement. In 2012, The Border Consortium estimated that at least 400,000 people in Southeast Burma were internally displaced (2012: 2). For many, displacement has been a repeated, even regular, occurrence. In interviews with thirty-six internally displaced people (IDPs) in Karen State, South (2008: 85) found that they had collectively experienced more than 1,000 displacement incidents, with fi ve people reporting more than 100 acts of displacement each. Downloaded by [University of Defence] at 21:28 09 May 2016 Displacement conditions are extremely dangerous, with a high risk of sickness or death due to lack of access to health care, malaria from sleeping without shelter, or as a consequence of childbirth. Human rights organizations have documented the courage and resilience shown by rural villagers in developing coping strategies to manage displacement and resist leaving their land permanently (KHRG 2008; Malseed 2008; Heppner 2006; Burma Ethnic Research Group 1998). Village leaders and villagers have resisted the demands of the Burma Army and other forces with strategies including ‘negotiation, bribery, lying, outright refusal, confrontation, various forms of The Karen in Burma 33

discreet false compliance, jokes and counter-narratives and temporary evasion’ (KHRG 2008: 6). In some of these efforts, internally displaced populations were quasi-organized in their resistance and employed a range of tactics to evade and subvert control. However, ‘coping’ for many also entailed acquies- cing to the demands of the military, perhaps attempting to negotiate a more manageable situation but often simply bowing to demands for food, labour or other resources (Cusano 2001: 160; South 2012b: 188). Throughout the 1970s and early 1980s, the Burmese military in Karen State followed a pattern of dry-season offensives and wet-season retreats, and villagers under attack echoed this movement; crossing into Thailand to escape a military offensive, and returning when the troops departed. In 1984, for the fi rst time, Burma Army troops did not retreat when the rains came and the villagers were trapped in Thailand, leading to the creation of the fi rst ‘temporary shelters’. Over subsequent decades the refugee population grew, peaking at approximately 150,000 and remaining relatively constant at that number until the international resettlement programme began.18 A much larger population – upwards of one million people from Burma, of all ethnic groups – live in Thailand as undocumented migrants, though it has been estimated that up to half of these people would likely qualify for refugee status (Vungsiriphisal et al. 2011: 12). The jobs available to migrants are often unpleasant and dangerous, such as in factories, sweatshops, brothels, construction, illegal logging or intensive farming. They are also exploita- tively underpaid (with migrants paid less than a third of the salary a Thai casual worker could expect) and vulnerable to arrest at any point (Human Rights Watch 2012: 1–9). Registered refugees represent less than 10 per cent of Burmese nationals living in Thailand but they have received by far the largest proportion of aid and assistance. The camps also provide an environment of relative security which is absent for both IDPs in Southeast Burma and undocumented migrants in Thailand. Flight to a refugee camp is often the culmination of many years of intermittent displacement and may occur only after something happens to make previous coping strategies less viable, such as the death of a family member, crop failure or serious illness (Cusano 2001: 160). Becoming a refugee can therefore be an active choice (albeit one with few alternatives) for which there is a level of preparation, psychologically and sometimes Downloaded by [University of Defence] at 21:28 09 May 2016 economically. It is not uncommon that one or two members of a family will come to the camps while others remain behind in Burma. This embeds the refugee camps in continuing cross-border relationships, despite offi cial restrictions on freedom of movement. When the fi rst refugees began to arrive in the early 1980s, the Royal Thai Government perceived it as a temporary problem and a relatively minor one compared to the infl ux of Indochinese refugees that was occurring at the same time. There was even a political value in hosting Karen refugees, as a valuable ‘buffer’ between Thailand and its historical enemy, Burma. Thus, for the fi rst 34 Governing refugees

decade or more of the camps’ existence, the Royal Thai Government played a minimal role. Refugees from Burma were allowed to self-settle and to organize their own communities around existing village structures. There was a relatively high degree of autonomy and freedom, with opportunities to forage for food outside camp boundaries and to pursue (limited) income generation opportunities in neighbouring Thai villages (Bowles 1998; Lang 2002). During this period the Royal Thai Government denied access to the United Nations High Commissioner for Refugees, instead requesting a consortium of donors that had been serving Indochinese refugees to also provide services to refugees on the Thai–Burma border.19 This entity, now known as The Border Consortium, continues to play the primary role in service delivery to and advocacy for refugees. In the context of the buffer strategy, the Royal Thai Government provided direct and indirect support to the Karen insurgency, including permitting KNU leaders and their families to live in the refugee camps without interference (Lang 2002: 141–2). However, by 1995, declining KNU territory meant that the buffer was no longer effective – as was dramatically proven by a series of military attacks on the camps by Burma Army and DKBA forces between 1995 and 1998 (ibid: 155–8). In January 1997, three camps were attacked simultaneously and set on ﬁ re, destroying thousands of houses and leaving many homeless. The Royal Thai Government responded with camp consolidations, the stationing of Thai soldiers in every camp, prohibition on movement outside of the camp and tighter controls on new arrivals crossing the border (Bowles 1998). As economic ties continued to strengthen between Thailand and Burma, any political capital that the refugees had once represented was lost. By 2004, Thai policy towards refugees from Burma had deteriorated to such an extent that Human Rights Watch released a highly critical report, calling on the Thai Government to ratify the 1951 Refugee Convention and its associated Protocol, to stop deportation of refugees, to put in place proper screening and status determination processes for asylum seekers, to permit media access to the refugee camps and to provide adequate health care, shelter and educational opportunities for refugees. More recent reports remain critical of Thailand’s treatment of Burmese refugees and migrants, pointing to labour exploitation, refoulement and deportation, and proﬁ teering (Human Downloaded by [University of Defence] at 21:28 09 May 2016 Rights Documentation Unit 2008; Human Rights Watch 2012). Refugee registration is ostensibly regulated by ‘Provincial Admission Boards’. These ‘PAB’ have not operated since 2006 and the vast majority of people who have come to the refugee camps since that time have been unable to obtain formal registration.20 As only registered refugees are eligible for international resettlement the ratio of registered/unregistered refugees in the camps has been transformed in recent years). The UNHCR has been permitted to exercise a protection mandate in the refugee camps since 1998. One of its roles in this capacity (as in every The Karen in Burma 35

refugee situation) is to seek a durable solution for refugees. UNHCR recognizes three durable solutions for refugee situations: local integration, third-country resettlement and voluntary repatriation. In Thailand, local integration was rejected by the Royal Thai Government and repatriation was precluded by continuing confl ict and human rights violations inside Burma. Resettlement was initially resisted by the Thai Government as a probable ‘pull factor’, but was eventually agreed to. Between 2005 and 2012, more than 100,000 refugees were resettled from camps in Thailand to countries including the United States, Australia, England, Norway, Sweden and Finland (IOM 2012). Resettlement is increasingly perceived as a desirable durable solution in protracted refugee situations, offering an alternative to life in limbo in a camp and presenting potential opportunities for employment, education and mobility (UNHCR 2003b). However, resettlement also presents signifi cant challenges, for those who leave and for those who remain behind (see further Chapter 3). For those who leave, resettlement offers a route out of the camp – but at a cost which cannot be fully understood before departure.21 In 2008, the director of TBC found that refugees resettled to the United States faced many challenges of language and a lack of institutional literacy in accessing welfare and entitlements (TBC 2008a). A study several years later found that life was becoming easier for some but overall resettlement had proved a ‘mixed blessing’ (Kenny and Lockwood-Kenny 2011). Strong community organizations had developed to plug the gaps in service provision, with one resettlement agency quoted as saying that even shortly after arrival in the US, ‘the Karen had the best community support systems they had ever seen among refugee groups’ (TBC 2008a: 4). This suggests that the mechanisms of community organization which the Karen have developed over decades of confl ict, displacement and encampment are resilient enough to reform in industrialized society. It may also suggest something even more profound about the governance heritage of the Karen refugee societies: that a history of living without effective state service provision has enhanced their capacity to adapt to American (or Norwegian, or Australian) society, precisely because they are accustomed to self-suffi ciency and to working outside state structures.22 Downloaded by [University of Defence] at 21:28 09 May 2016 Notes 1 As announced by then-Prime Minister Khin Nyunt on 30 August 2003, the Roadmap to Democracy had seven stages: (1) to reconvene the National Convention to draft a national Constitution; (2) step-by-step implementation of the process necessary for the emergence of a genuine and disciplined democratic system; (3) draft a new Constitution based on the principles laid out by the National Convention; (4) adopt the Constitution via a national referendum; (5) hold free and fair elections to constitute a national legislative assembly based 36 Governing refugees

in accordance with the new Constitution; (6) convene the new national legislative assembly; and (7) build a modern, developed and democratic nation. 2 In 1989, when the military regime changed the country name from ‘Burma’ to ‘Myanmar’, it also changed the name of the Karen (to Kayin), Karenni (to Kayah) and Arakanese (to Rakhine). 3 Noting that the fi gures are unreliable, Steinberg (2010: xiv) cites 1983 census fi gures in establishing the Burman population as 69%, and a range of national population estimates from 47 to 58 million. 4 To use the example of the Karen, though an administrative district of ‘Karen State’ is recognized by the Burmese regime the ethnic insurgency movement has been fi ghting for something different, a politically autonomous ‘Karen State’. 5 At the outset of the revolution, the struggle was for a fully independent Karen State. This has more recently been replaced by a willingness to settle for political autonomy within a federal structure. 6 This story is still told among the Karen. Interestingly, it was told to me in several variations; sometimes with emphasis on the Karen brother’s foolishness in losing or failing to recognize the value of the book but on another telling it emphasized deceit by the White brother (i.e. that he ‘stole’ the book and did not tell his Karen brother). This echoes Hayami’s (2004) fi nding that Karen folktales are changed to accommodate the circumstances in which they are told, perhaps to chime with current themes or events, or to make a particular point to the listener. 7 In Mae La Oon, seven religious groups are members of the camp ‘Religious Union’: four Christian denominations (Baptist, Anglican, Roman Catholic, Seventh-day Adventist), Buddhist, Animist and ‘White Elephant’, a religious cult based around the worship of white elephants. In terms of population numbers, it is estimated that close to three-quarters of camp residents in Mae La Oon are Christian, while 24% are Buddhist and 3% are Animist (UNHCR/ILO 2007: 64). 8 Field notes Mae La Oon, February 2010. 9 As used in, e.g. the 2010 Karen Baptist Committee calendar. 10 Unfortunately for Burma’s ethnic nationalities these territories have recently come to be very valuable indeed, with ‘development’ projects including timber logging, hydropower damming and a massive gas pipeline. The latter two projects in particular have resulted in many human rights violations and extensive displacement (Earthrights International 2011). 11 Karen villages in Southeast Burma suffered deeply during the Second World War for their support of British forces but the British military is remembered today with respect, even with affection. I met two middle-aged Karen men called Downloaded by [University of Defence] at 21:28 09 May 2016 ‘Wingate’ (evidently named for Orde Wingate, commander of the Chindit forces in Burma during the latter half of the war) and was told the story of ‘Grandfather Longlegs’ on many occasions (‘Grandfather Longlegs’ being a nickname for a British army major who surrendered himself from hiding to stop reprisals against Karen villagers). 12 The effectiveness of this administration is disputed. Thawnghmung (2008: 56) quotes Alan Smith in describing the KNU government as ‘a warlord administration with an on-paper administrative structure’. 13 The KNU’s failings have been replicated by other revolutionary movements in Burma. The Chairman of the Shan Restoration Council, Lt-General Yawd Serk The Karen in Burma 37

(2010) has identifi ed six key failures of the Burmese ethnic insurgencies: failure to unite across ethnic nationalities; failure to agree on concrete policy objectives (e.g. federalism versus full independence); failure to focus on defeating the enemy and instead becoming distracted by self-interest and the management of internal splits; lack of strong military and political leadership within organizations; ceasefi re policies weakening the situation of all (those who agreed ceasefi re lost political ground and leverage, while those who did not were weakened by the greater concentration of Burmese forces); and each insurgency operating independently and relying on guerrilla tactics rather than combining forces with a central command capable of large offensives. Each of these points can also be applied to KNU policy over the last fi fty years. 14 This experience of shadow sovereignty is not unique to the Karen but has been repeated in other ethnic borderlands. Contrasting Kachin and Karenni experiences of governance, Grundy-Warr and Dean (2003: 94) note that there are some similarities (relative autonomy during colonial rule, the birth of an nationalist movement and resistance to rule from Rangoon) and important differences (e.g. the Kachin Independence Organization has considerably more authority and administrative control inside Burma than the Karenni National Progressive Party does). Nevertheless, they fi nd that both cases challenge conventional views of sovereignty, displaying a multi-layered, lapped, power environment where ‘political territoriality is greatly infl uenced by contingent, fl uid sets of relations, past and present political and social formations, and by regional geopolitics’. 15 Between the Karen National Defence Organization and the Anti-Fascist People’s Freedom League (AFPFL) Government in 1949; between the National Democratic United Front (NDUF) and Ne Win’s ‘caretaker government’ in 1959; between the NDUF and Ne Win’s military government in 1963; between the KNU and the military State Law and Order Restoration Council in 1995; between the KNU and the military SPDC in 2004–2005, and since 2011 between the KNU and the USDP Government. See Keenan 2012a. 16 The eleven point proposal includes the following points: (1) Establish a nationwide ceasefi re and immediately cease military operations in ethnic areas. (2) Guarantee the human rights and safety of all civilians. (3) Build trust among the people. (4) Support the basic needs of the people and ensure that development projects have the full participation and support of local villagers. (5) Allow national media outlets to participate in the peace process. (6) Immediately stop forced labour, arbitrary taxation and extortion of villagers. (7) Release all political prisoners and provide solutions to settle land rights issue. (8) Set out principles for all parties to ensure a genuine peace process. (9) Ensure the legitimacy of Downloaded by [University of Defence] at 21:28 09 May 2016 representatives involved in negotiations, provide adequate time for their consultation with respective constituencies and establish a clear role for third parties. (10) Initiate a plan for monitoring and ensuring the transparency of the peace process. (11) Establish a fl exible process that guarantees progress towards sustainable peace, and in which all parties speak straightforwardly and avoid using words that may be misinterpreted. (See Keenan 2012b: 7). A clearer demand for autonomy in ethnic areas has been made by the United Nationalities Federal Council (UNFC 2013), which issued a statement in August 2013 asserting its position that the ‘present Union of Burma/Myanmar’ should be formed ‘into a Federal Union of national states and nationalities states, having 38 Governing refugees

national equality and self-determination’ and ‘to practice federal democracy in this Federal Union’. 17 Reports based on testimony of Karen villagers indicate that the Burma Army is most feared, while there is a greater degree of support for and accommodation with KNU forces. Despite political support for the KNU, some villagers preferred to be under DKBA control as they believed this brought less likelihood of reprisals by Burma Army forces (South et al. 2010; Thawnghmung 2011a: 79). 18 Monthly population fi gures are reported on the website of The Border Consortium: www.borderconsortium.org. Three fi gures are given: TBC’s verifi ed caseload (which includes all persons verifi ed as living in the camps); TBC’s feeding fi gure (those who personally attend rations distributions); and the MOI/UNHCR population of registered refugees. 19 It is not entirely clear why the Royal Thai Government refused a role to UNHCR. It seems probable that a number of factors were relevant: a desire to avoid repeating the experience with Indochinese refugees where the availability of international support and resettlement were believed to have been ‘pull factors’ attracting refugees to Thailand; a desire to avoid granting the status of ‘refugee situation’; and a basic political miscalculation: ‘From interviews with RTG representatives, it seems clear also that the RTG did not expect the situation would last so long and become protracted.’ (Vungsiriphisal et al. 2011: 56). 20 This included, in 2009, the refusal of entry to more than 4,000 people who had been forced into Thailand by an SPDC–DKBA offensive. These people were denied entry to the refugee camps though they were permitted to remain on Thai soil and to receive aid and assistance for a short period of time. They were strongly pressured to return to Burma though their village area had been heavily mined. By September 2010 only a few of this group remained in Thailand (KHRG 2010). 21 People choose to resettle because they hope it will provide them with a better future. Berger (2010: 72) imagines a meeting between two migrant workers in western Europe, where one man, ‘full of the excitement of arrival’ says, ‘Here you can fi nd gold on the ground. I am going to start looking for it.’ His friend replies, ‘That is true. But the gold fell from very high in the sky, and so when it hit the earth, it went down very, very deep.’ Like Berger’s migrant workers, refugees resettled from Thailand often fail to fi nd the gold they are searching for, instead facing unemployment, stigma and the disillusionment of abandoned dreams. 22 As the political climate inside Burma has begun to alter, refugee repatriation is also being considered. This proposal is examined in Chapter 8, Beyond Encampment. Downloaded by [University of Defence] at 21:28 09 May 2016 Chapter 3 The camp community

In the dark times, will there also be singing? Yes, there will be singing. About the dark times. Brecht, Motto (1987: 320)

Unquestionably, life is hard for refugees in camps in Thailand. There is not enough food or money, freedom is curtailed and refugees have very limited control of their life choices. Houses are simple bamboo huts which provide little privacy and even less protection from the elements. An entire generation of young people has been raised entirely in camps, where they attend schools that are unaccredited to any formal education system and thus offer limited potential for accessing further education. Resettlement gave registered refugees a choice about the future but it was an unenviable one of leaving to an unknown destination and alien culture or staying in a camp under the ever-present fear of closure. The undesirability of living in a camp is eloquently apparent in the choice of refugee leaders and community organizations to locate their operations in Thai towns and villages, further emphasizing that the refugee camps are home for those who have no other options. Yet if there is any compensation for the many manifest difﬁ culties it is the thriving social capital and community which exists within the refugee camp – and furthermore, as this chapter will argue, which the camp environment produces. Downloaded by [University of Defence] at 21:28 09 May 2016

Camp communities Describing a refugee camp in terms of community has been rejected by some of the foremost scholars in refugee studies. For example, Colson (2003: 10) claims that refugees are too diverse to constitute a community and can have little in common but resistance to the camp administration. Harrell-Bond (2004: 27) notes that of the many refugee contexts she has witnessed: ‘None of them could unfortunately be described as “communities” They are better 40 Governing refugees

described as places where disparate populations have been lumped together and where at times every sort of anti-social behaviour is manifested.’ Hyndman (2000: 140) is blunter still: ‘citizens live in communities; refugees live in camps’. Certainly many camp situations appear to experience much higher rates of violence, corruption and insecurity than is present in Thailand, notably those in Kenya and Somalia (Human Rights Watch 2013b; Crisp and Jacobsen 1998; Crisp 2000). However, claiming that refugees lack points of common connection is not persuasive: all camp residents share the experience of political violence, loss of homeland and exile, as well as a shared living environment and future. These are much more meaningful points of connection than most communities might expect to share. In Thailand, the points of connection run even deeper, as the camp populations are broadly homogenous, with the majority of residents sharing an ethnic identity and often also a common religion.1 Entire villages fl ed to the refugee camps together, and took up residence in camp sections populated by families and neighbours, people who were embedded in long-standing and continuing relationships. It seems likely that resistance to recognizing refugee camps as communities is infl uenced at least in part by the particular connotations of the term. As Bauman (2001: 1) has suggested, community ‘feels good’ and we aspire to be part of a community. The same could hardly be said of a refugee camp. However refuting the existence of community by reference to unpleasantness of camp environs or lack of pre-existing cultural bonds suggests a somewhat essentialist notion of ‘community’, ignoring the well-established insights of Cohen (1985) and Anderson (2006 [1983]) that communities are not necessarily rooted in territory or tradition but are also constructed or imagined on grounds of social identity. Furthermore, while it is undoubtedly true that a refugee camp is not a pleasant or desirable place, it is not devoid of culture and community. Agier (2010b: 46) has described the social structures that exist within a refugee camp as ‘communities without identity’, but identities are profoundly important to refugees in Thailand, as they are for Saharawi refugees in Algeria, Palestinian refugees in the West Bank and Lebanon, and Bhutanese refugees in Nepal (Wilson 2010; Farah 2009; Rosenfeld 2004; Hanafi and Long 2010; Banki 2008). Despite the tendency to see camps as a rupture from past expe- Downloaded by [University of Defence] at 21:28 09 May 2016 rience, refugees do not lose pre-camp identities during encampment but draw upon these identities and reinterpret them in ways that offer meaning for life in camp. This is a dynamic process of identity construction that takes place through a host of practices, including story-telling, memorialization and material culture (Dudley 2010; Malkki 1995a). As a result, community identities may be salient to refugees in two respects: as the affi liations and memberships that they possess on arrival at a camp, but also as the specifi c ways in which those affi liations are practised or gain meaning within the camp environment. The camp community 41

‘Community’ is of course a ‘terminally vague’ term, which is notoriously contested and resistant to defi nition (Cotterrell 1997: 18; Matthews 1988: 10). In the discussion that follows, I have been most infl uenced by Cohen’s (1985) defi nition of community as a boundary-maintaining entity, defi ned simultaneously by the affi liation that unites its members and excludes non- members. This defi nition is particularly valuable for its recognition that ‘community’ membership is not discrete and exclusive but can exist in co-extensive overlapping scales. This is apt for Mae La Oon and the other refugee camps on the Thai–Burma border, where there is not a single camp community but rather multiple overlapping camp communities. Three of those are considered below: a situational community of those who live within the camp boundaries, an ethno-nationalist community of Karen refugees and a political community of shared governance.2

A situational community: ‘We are refugees’ The fi rst form of community within the refugee camps is situational and territorial, consisting simply of camp residents. In Sgaw Karen, two terms can be translated as ‘refugee’. The fi rst (bgha ba kaw ba ke) translates into English literally as ‘people who face diffi culties’ and is the term in translating terms such as ‘Refugee Convention’ or ‘World Refugee Day’. The second term (de ga wa po) translates as ‘camp people’ and is the term used consistently to discuss camp residents and life in camp, and particularly to explain the shared interests of camp residents in maintaining a peaceful and stable camp. The category of ‘camp people’ is established on the basis of the shared experience of encampment and of the deprivations, vulnerabilities and stigma that encampment imposes: from rations-cuts and aid reductions to the ever- present fear of attack by Burmese forces. Camp people also share the need for the continued existence of the camps. To ensure this, refugees believe they must ‘live quietly’, to avoid antagonizing Thai citizens and authorities and to avoid actions which may bring negative attention to the camps. As one pastor in the camp explained to me:

Our refugees and the villagers have understanding with each other. But there are some do not obey the law and if one, two, three are disobedient Downloaded by [University of Defence] at 21:28 09 May 2016 all the camp people are oppressed more and more. For example, some people cannot [afford to] give the villagers food to cut the bamboo. They only go secretly. And then the villagers say that they are thief, they steal their bamboo, their wood [. . .] If we can bribe them every time, okay. Because our refugees have to bribe. First we have to bribe the Thai authority. Two, we have to bribe the Thai villager. Three, we have to bribe the Thai security, the Or Sor. For example – we want to go here, we give them something. “Ok, you can go”. So it is not easy for refugee people. Some who are okay, have a ﬁ ne job, some who are working in the 42 Governing refugees

NGOs, they have pocket money and they can do this thing. But if we are simple refugees we cannot do anything. We have to stay. I don’t go anywhere. I don’t want to go about illegally. Also, what do we say, they said that if we go illegally, if they see, they will ﬁ re at us with a rusty musket [laughs]. If they will do or they will not do we don’t know. But they order, and we are afraid.3

‘Camp people’ not only fear the Thai authorities and Thai villagers, but also military attack from inside Burma. Between 1995 and 1998, the camps on the Thai–Burma border were repeatedly attacked by Burma Army and DKBA forces. These attacks impressed two messages on the refugee population: that cross-border attacks are possible and that if they occur, neither Thai security forces nor international NGOs will be able to protect the refugees. When I asked residents of Mae La Oon about their greatest fears and worries living in the refugee camp, the ﬁ rst answer was almost always a fear of attack by the Burma Army or DKBA forces.4 However, while the shared experience and shared risks of encampment establish a minimal level of cooperation it is little more than that. The banner of ‘camp people’ is applied to all within the camp bounds and thus subsumes other differences, such as ethnicity. Precisely because of this breadth of application, ‘camp people’ is not a deeply felt identity. Much more meaningful bonds are forged in shared ethnicity.

An ethno-national community: ‘We are Karen’ In the seven ‘Karen’ camps, the most powerful signiﬁ er of community is the ethno-national status of Karen. As has been mentioned, ‘Karen’ is a large umbrella, encompassing peoples from Burma and from Thailand, who speak different languages and practise different traditions. Being Karen has a particular status and meaning in the refugee camp, partly derived from representations of Karenness that are associated with the Karen nationalist movement (Sgaw-speaking, Christian) but also related to the experience of being a refugee. Neighbouring villagers who are ethnically Karen and speak Sgaw Karen language are seen as fundamentally Other to Karen refugees and described as ‘Thai villagers’: ‘They are not Karen any more, though they Downloaded by [University of Defence] at 21:28 09 May 2016 speak Karen. They say they are Thai, but they don’t speak Thai.’5 As this suggests, ethnic identities are not static. A Thai citizen who marries a Karen refugee and comes to live in the camp might be viewed as ‘becoming Karen’. Similarly, a Karen woman who marries a Burman man may ‘become Burmese’. A change in identity may be ascribed even if the move takes place within the camp boundaries. For example, in Mae La Oon, a Karen woman who married a Burman man and moved to Section 13 (the so-called ‘Burmese’ section of the camp) would be considered by other Karen camp residents to have ‘become Burmese’, not as an inevitable outcome of the marriage but the The camp community 43

result of adopting language and living environment. This fl exibility of identity is not specifi c to the camps; ethnic identity as practiced rather than simply inherited has been recognized as occurring in this region of Thailand for many decades, even centuries (Lieberman 1978). One young man told me that his goal in resettlement was ‘becoming gawlawah’ (a white foreigner), which he saw as achievable with ‘two things – a passport and a credit card. A passport is freedom. [. . .] I will not be a refugee anymore.’6 The position of ethnic nationalities inside Burma has for many years been one of subjugation within a strategic project of ‘Burmanization’ or ‘Myanmafi cation’ (Houtmann 1999; Gravers 1999: 113). In the camps these pressures no longer apply and ethnic identity can be freely expressed and mobilized. As will be explained in more detail below, the refugee camps are constructed as Karen in multiple dimensions, from the use of Sgaw Karen as the primary language of education and administration, to the calendar of days of celebration. Such activities assert the refugee camps as a space where the Burman–Karen relationship of dominance that applies inside Burma is reversed. For Karen refugees, this is liberating and affi rming – but for Burman refugees it is less so. This is increasingly important as a growing non-Karen population has come to live in the refugee camps, in large part a consequence of the international resettlement programme. The largest camp on the border, Mae La, now has approximately 16 per cent non-Karen residents (including a sizeable Muslim minority).7 In Mae La Oon, more than 98 per cent of the population is Karen (TBC 2013). The remaining 1 to 2 per cent is a Burman population of between 300 and 400 people, most of whom live in a designated ‘Burman’ section of the camp, Section 13. In addition to participating in the general camp management structures, Section 13 operates a parallel governance system that is led by the All-Burma Students Democratic Front (ABSDF).8 The ABSDF is a student army formed in the aftermath of the student protests in Rangoon in 1988. Forced to fl ee Rangoon, thousands of protestors made their way to the border regions where they formed the ABSDF and fought alongside armies of the ethnic nationalities, including the KNLA. In 1995, the headquarters of the Karen insurgency at Manerplaw was captured by the Burmese military forces and the Burman student revolutionaries became refugees along with the Karen in the area. Though there are no physical Downloaded by [University of Defence] at 21:28 09 May 2016 boundaries between the ABSDF section and the rest of the camp, the ethnic divide is continually reinforced. One ABSDF member in Mae La Oon had been a refugee for more than ten years but still felt excluded from the mainstream camp society:

Their view is that this is the Burmans. I understand. But we ﬂ ed together. We fought together. We stayed together. We are all refugees. We are not ‘the Burmans’, not ‘the Karen’ here in the refugee camp, but they cannot think about it like that. Their feeling controls their mind. Because when 44 Governing refugees

they were a child the military regime killed their father and mother, and sister, and destroyed their land. They recognize that this is a Burmese. They hate the Burmese.9

However, while ethnic difference has a direct impact on relationships of trust, the environment is not one of straightforward institutional discrimination. Burman residents of Section 13 in Mae La Oon told me they had experienced ‘hate speech’ and racist abuse from Karen refugees. They had reported this to the camp committee and claimed to be happy with the way that their complaint was responded to, saying there was ‘no problem’ with the camp committee but only with some other refugees.10 That they were willing to report this incident to the camp committee at all suggests that discrimination is isolated rather than institutionalized. A similar conclusion was reached in a recent evaluation of camp management across all nine border camps (Turcot 2012: xiii). In the camps that have a growing non-Karen population, there are several examples of accommodation and cooperation, notably in Mae La, where a Committee for Coordination of Ethnic Groups has been created, with representatives of Burmese, Shan, Lisu, Kachin and Karenni ethnic groups. The primary basis for division is arguably not ethnicity but rather length of stay in camp. Since the international resettlement programme began, each camp has experienced a large inﬂ ux of new arrivals. The label used to describe this population was not ‘camp people’ but bgha thaw (new people). One camp committee member in Mae La told me that he had received a complaint about this label, to which he had replied:

But you are new people, why do you feel bad about this? You are different from the ones with ID who arrived before 2005. ‘But why are we different?’ There are three kinds of people in the refugee camps. The ﬁ rst arrived with just their clothes and a bag and their shoes. They were prepared during the war. The second don’t have clothes, they don’t even have shoes. They ﬂ ed from the war. The third are not real refugees at all. They sold their house, arrived here with money, arrived here by car, even arrived by aeroplane. You are the third group. Downloaded by [University of Defence] at 21:28 09 May 2016 A community of shared governance A third community within these refugee camps is based on cooperation and shared governance. Research with displaced communities and migrants has consistently shown that family and community networks are central to coping with displacement. Among urban refugees in South Africa, Landau (2012) found that social networks and group membership were more determinative of protection than formal recognition as a refugee. Reviewing response strategies of displaced communities in ten countries, Sorensen and Vincent (2001: The camp community 45

273) suggest that ‘the successes of a displaced community appear to be closely linked to the availability of effective and accountable community organizations and leadership’. As described in the previous chapter, community organizations have had a prominent role in responding to both internal and cross-border displacement from Southeast Burma (Karen Human Rights Group 2008; Cusano 2001) and are central to the administrative management of encampment. Indeed, the level of internal refugee-led organization and administration is one of the most striking aspects of the Thai–Burma border camps:

Today the camps strike the visitor as busy, matter of fact, orderly places. They rely on outside relief agencies for material and technical support. But the daily graft of refugee administrators, teachers, religious leaders, medics, garbage disposal people, sanitation experts, and shopkeepers keeps them running. [. . .] They cope with the kind of social pressures and problems that occur anywhere and with traumas that are special to the camps, usually armed with little more than community spirit or neighbourly concern. (TBC 2010b: 2)

The structures of management and organization which were set up in the camps were essentially transplanted from Karen village and community life. That the Karen were able to set up life as refugees with their governance structures largely intact is an anomaly in refugee situations but one which proved successful (South 2003: 184). In 1998, Bowles (12) noted that organization within the KRC was ‘integral to refugee autonomy and self- suffi ciency’ and permitted the camps to maintain a ‘village atmosphere’. In the early years of the camps’ existence, their residents were almost exclusively ethnic Karen and their management structures refl ected this, from the KRC to the community organizations of the Karen Women Organization (KWO) and Karen Youth Organization (KYO). KWO and KYO are community- based organizations but they are also fi rmly part of the Karen political and military elite. More specifi cally, they are part of the KNU elite. KWO was established in 1949 with the explicit aim of supporting ‘the KNU revolutionary movement’ and raising the political consciousness of Karen Downloaded by [University of Defence] at 21:28 09 May 2016 women (Curwen 1989). KWO is now an internationally recognized women’s rights organization and though its links with the KNU remain strong (between 2008 and 2012 the General Secretary of the KNU was Zipporah Sein, a former Chairperson of KWO) this relationship is no longer emphasized.11 Instead, KWO represents itself as a social welfare organization which seeks to advance conditions for women through education and training programmes, research, social welfare (e.g. hygiene packs to new mothers, or assistance to those with disabilities), research and advocacy work, fund-raising and other activities. KWO and other community organizations perform vital 46 Governing refugees

work within the camp environment in terms of providing support to refugees, advocating for their needs with international agencies and donors, and at times in providing security to particularly vulnerable individuals. Gender equality in the refugee camps outstrips that in Karen state, a state of affairs which owes a great deal to KWO’s education and mobilization on women’s rights. Within both KWO and KYO the ideological connection with the KNU is important but it is not all-consuming. In relation to the Karen Youth Organization, Brown (2012: 169) found that KYO members represented their organization differently to different audiences, variously emphasizing its status as a community-based organization (to international NGOs), as a KNU organization (to KNU alliance groups and to Karen youth), and as a social work group (also to Karen youth). These are large organizations – KWO has more than 50,000 members in total – and not every member thinks alike. Those who work with community organizations may be motivated in part by politics (i.e. because they support the struggle or self- determination or because they are Karen) but they may also be inﬂ uenced by much more prosaic considerations: it is an avenue of work and service in a context where there are few such avenues, it provides some social capital and it brings a measure of social status.

Constructing a political community The existence of camp communities is signiﬁ cant not only in terms of the boundaries of membership but for the identities members possess. Community identities are of course not static but can be created and shaped by a variety of means, including symbols, stories, media and material traditions (Cohen 1985; Anderson 2006 [1983]; Hobsbawm 1983). For refugees in Thailand, appeals to tradition and culture are regularly used to inculcate and strengthen the ethno-national identity of ‘Karenness’. This is done in numerous ways, including through performative dance and ceremonies (MacLachlan 2006) and through music repeating core themes of ‘sacriﬁ ce, unity, bravery and loyalty’ (Cho 2011). The following sections examine three sites where the ‘invention of tradition’ is particularly strong: historical narratives, language and clothes. Downloaded by [University of Defence] at 21:28 09 May 2016

Narratives of identity and past Stories are a powerful means to both deﬁ ne and perform identity, making narrative a valuable tool in the creation and construction of nationalisms (Hobsbawm 1983; Anderson 2006 [1983: 12–36]; Gupta 2001: 191). Stories may serve a dual function of expressing one identity and reinforcing its difference from an Other and in doing so can establish identity and not merely reﬂ ect it, as Donnan (2005: 75) has concluded in relation to Protestants in the The camp community 47

borderlands of Northern Ireland: ‘It is relating the narrative that makes them “Protestant” and not because they are a “Protestant” that they relate the narrative.’ In refugee situations, the centrality of narrative in defi ning and enacting ethnic identity has been most fully developed by Malkki. Analyzing the experiences of Hutu refugees in Tanzania, Malkki (1995b: 52–104) found that refugee camps were powerful sites for the transformation of ways of thinking and for the nurturing of ‘an elaborate and self-conscious historicity’ among the refugee population. Stories of the past were used to solidify historical experience in a coherent cosmological ordering and ethnic essential- ism, creating a shared body of historical knowledge and interpretations: a ‘mythico-history’. Narratives, stories and jokes repeated key themes, including the historical persecution of Hutus, the boundaries and contrasts between Hutu and Tutsi, the opposition between good and evil, and a favourable interpretation of the colonial era. In effect, these stories consolidated a sense of Hutu identity and hardened a sense of ethnic difference from ethnic Tutsi. Importantly, Malkki (ibid: 197–8) found that the construction of a mythico- history which occurred among camp-based refugees did not occur among town refugees, who tended to be assimilated within the host society to a greater extent. A similar process of mythico-history production is evident in Mae La Oon with a startling similarity of core themes: the historical persecution of Karen, contrasts between Karen and Burman, and an interpretation of the colonial past as benefi cial to the Karen. There is a consistent emphasis on kinship and on the Karen family as a binding group identity, an identity which is fostered ‘at homes, in schools, at church, and through newspapers and magazines’ (Thawnghmung 2008: 22). Support for the KNU and KNLA is encouraged from childhood among refugees, and nationalist symbols (the fl ag, pictures of Karen ‘martyrs’ such as former KNU-leader, Saw Ba U Gyi) are prominent in many households. The school history curriculum, approved by the KNU, describes the Karen as a lost tribe of Israel, continually persecuted and forced to move on until eventually reaching Burma. As one student succinctly explained to me, ‘Our Karen people are not lucky. Throughout history we have been wanderers. First we came from Babylonia, then Mongolia, China, Burma, and now Third Country.’12 There is a direct Downloaded by [University of Defence] at 21:28 09 May 2016 echo of Zionist narratives of victimhood, exploitation and exile, narratives which help to bind ethnic identities and nationalist sympathies across the disparate linguistic and cultural constituencies that comprise ‘Karen people’.13 Karen identity is defi ned in terms which are sharply opposed to Burman identity. Cheesman (2002: 199) describe Karen identity as shaped by three primary discourses: that Karens are ‘oppressed, uneducated and virtuous’. These are often combined in a single narrative, as popular stories and jokes repeat tropes of the simple, honest Karen betrayed by cunning Burmans or cruel Thais.14 48 Governing refugees

Narratives which celebrate traditional Karen culture and glorify the revolution are consolidated by memorializing practices. Eight national days are recognized in the Karen refugee camps, almost all of which are linked to the struggle: Karen National Union Day, Karen National Liberation Army Day, Karen National Defence Organization Day, Karen Women Organization Day, Karen Youth Organization Day, Kawthoolei Day, Karen Resistance Day and Karen New Year.15 On these days, there is usually a ceremony of some sort, with speakers from the relevant organizations and some entertainment, such as a camp-wide sports tournament or a singing competition. These days are not celebrated with much intensity: in Mae La Oon in 2009, KNU Day was celebrated by inviting two KNU members to speak to students at the school I was teaching in. The men were elderly, in their seventies, with rheumy eyes. When one man spoke the other nodded off, as did many members of the audience. As a snapshot of the revolution it was poignantly symbolic – ageing leaders, tenacious but long past enthusiasm, divorced entirely from the younger population. Despite the lack of enthusiasm, such events were widely considered to be important and necessary, and social messages from schools, leaders and other organizations consistently emphasized the responsibility of young people to remember the conﬂ ict, and to ‘help their people’. Christian churches organized regular prayer vigils dedicated to those inside Burma. At one vigil I attended those present were instructed:16

The refugee camp is a place of refuge and safety but people should not forget what is happening in Karen State. They should not get soft or weak but should remain strong to do God’s work. Young people in particular have a responsibility not to desert the people in Karen State but to go there and help them after they get an education.

One high school principal told me that the biggest challenge presented by the camps was the risk that young people would forget about Karen State, would have no personal memories of living there and no sense of obligation to those who remained. He tried to ensure this did not happen by bringing leaders from Karen State to talk to the students and by requiring students to wear Karen ‘traditional clothes’ at least once a week. Another practice used by Downloaded by [University of Defence] at 21:28 09 May 2016 this and other schools was to take senior students to Karen State for special occasions, such as Karen New Year.17 As a school trip, this is a supremely diffi cult one, involving several days walk through jungle and mountains, with the threat of being found by Burma Army troops or of being prevented returning to the camps by Thai Border Patrol. Understandably, some students (and their parents) were reluctant to go. Others felt it was their moral duty; one student told me: ‘This is our country. If we do not care, then who will?’ Ambivalence around travel to Karen State refl ected the confl icting messages which young people in camps receive. On the one hand, they are The camp community 49

repeatedly told of the terrible conditions inside Karen State, the poverty, hunger, malaria, landmines, forced labour and military attacks. Yet they are also repeatedly told that these conditions mean that they have a duty to use their education and relative privilege to ‘help their people’.

Language Language is at the heart of individual and collective experience. In situations of ethnic confl ict, language can take on even greater signifi cance, refl ecting heightened emotional identifi cation and in some cases, a level of personal risk in being identifi ed as a member of a particular group (Eriksen 1993). For Karen refugees, the ability to use their ethnic languages is one of the few ‘gains’ of encampment. The offi cial language in Burma is Burmese, and this is the language of all schools and government services. In all the camps under the KRC, Sgaw Karen is the primary language used in schools, churches and public meetings. In Mae La Oon, only one school in the camp uses and teaches Burmese language. The production of Sgaw as ‘the’ Karen language is part of a broader simplifi cation of Karen identity, refl ecting the composition of the KNU’s senior members: Baptist and Seventh-day Adventist Christians, Sgaw- speaking, educated elites (South 2008: 37, 2007a; Oh and Stouwe 2008). However, Sgaw is not a common language of all Karen but is spoken primarily by Christianized Karen and those from highland Karen State. Oh and Stouwe (2008) have estimated that in the four Karen refugee camps in Tak province only 40–56 per cent of the refugee population were ‘native’ Sgaw speakers. In the three other camps, including Mae La Oon, they estimated 75–88 per cent of refugees were Sgaw-speaking. By reinforcing a pan-Karen identity and a Karen cultural supremacy, the preference for Sgaw Karen as the offi cial language of life in the refugee camps has obvious nationalistic benefi ts (South 2008; Oh and Stouwe 2008). As a practical matter, however, it has equally obvious disadvantages. In a country as diverse as Burma, a common language has utility beyond its status as the language of the oppressor and prioritizing Sgaw Karen risks isolating the Karen from other ethnic nationalities and further reducing opportunities for a cooperative ethnic opposition to the Burman majority. Downloaded by [University of Defence] at 21:28 09 May 2016 Sgaw as the primary language of education in the refugee camps is problematic for non-Sgaw students (a demographic which is growing in several of the camps) and even for some Sgaw Karen, as it is not uncommon that Karen who have been schooled in Burma can speak Karen language but are unable to read and write it. Nevertheless, ‘Sgawization’ is a sensitive issue to raise with refugee leaders, as mentioning the utility of learning Burmese language not only implies the possibility that the Karen revolution will fail but also echoes the Burmanization they are grateful to have escaped (Cusano 2001: 167). 50 Governing refugees

Interestingly, while it may seem that there would be considerable beneﬁ ts for refugees in speaking Thai language, there is much more interest in learning English. Many school books are produced in English, and some schools deliver almost the entire curriculum in English. A high value is placed on learning English for the perceived opportunities it may create, such as potential jobs with development organizations. There is some evidence of an emergent refugee camp contact language of a Sgaw Karen peppered with English language development lingo and acronyms such as ‘resettlement’, ‘third country’, ‘NGO’, ‘SGBV’ (sexual and gender based violence) and ‘TOT’ (training of trainers).

Clothes Material objects – particularly textiles and clothes – also help to create a sense of identity within the refugee camp (Dudley 2010). Clothes are a highly visible form of self-representation and can play an important role in performing political, ethnic or religious identities (Abu-Lughod 2002). As in many other nations, clothes became highly politicized in Burma during the colonial era as both a target of nationalism and a focal area for anti-colonial resistance. Edwards (2008: 11) describes a 1930s campaign by the Dobama Asiayone (‘We Burmans’) Association against Western clothing and for traditional national dress, and this theme of clothing as a means to defi ne political identity and express political resistance recurred in Burma throughout the twentieth century. For the Karen in Burma, the politicization of dress may be traced back to San C. Po, who proposed a Karen national dress together with the adoption of other nationalist symbols (the Karen fl ag, national anthem, and the creation of Karen New Year as a national day of celebration). San C. Po’s proposal was to establish an entirely new costume which combined elements of European and Burmese style. Illustrated in Burma and the Karens, his proposed national dress included an ‘evening wear’ option of trouser, jacket, shirt and bow-tie, an ‘ordinary costume’ of loose trousers, shirt and European jacket, and a ‘roughing it’ outfi t of trousers, shirt and felt-brimmed hat (2001 [1928]: 87–9). Nearly a century later national dress is profoundly important to Karen ethnic identity, albeit with no trace of San C. Po’s neo-colonial designs. Karen Downloaded by [University of Defence] at 21:28 09 May 2016 traditional dress is similar for both men and women, with gender distinctions established in colour and pattern rather than design. Women wear hand- woven sarongs in an array of colours, tied at the side to make a fl uid straight shape. Men wear a lighter cotton longyi, which is tied by twisting the fabric and knotting it in the front. The ‘Karen shirt’ is a square tunic top, cut to fall open at the neck. The Karen shirt is identical in design for both genders, but while men traditionally wear a red shirt, women’s shirts come in a variety of colours and are often decorated with embroidery and seeds. Traditionally, The camp community 51

girls and young women should wear a long white woven dress before marriage, with the sarong and tunic only to be worn after marriage. In 1862, Mason predicted that ‘the native Karen dress will, in a few years, become almost extinct’, replaced by foreign manufactured cloth and styles (1862: 219). A century and a half later, the same style and patterns she described are seen as intrinsically and authentically Karen and a crucial marker of political and ethnic identity. When I asked refugees, ‘What is Karen? What does it mean to be Karen?’, the ﬁ rst answer was always ‘Karen language’, closely followed by ‘Karen clothes’. Camp schools encourage students to wear Karen clothes, typically by designating one day of the week in which traditional dress should be worn. Public events often include a ‘traditional dress’ competition in which young people parade on a stage and give a short speech explaining what they are wearing and how it is made. On 22 May 2009 (Karen Youth Organization Day), the winners of the competition offered the following explanation:

Karen girls wear white because it is traditional. We carry a bag to carry betel nut and betel leaf. We weave the bag with bamboo. We wear a special necklace if we are single. Boys wear a traditional shirt, woven by women. We wear red colour because it makes us strong. We wear long sleeves to protect us from mosquitoes. We wear a cane and bamboo hat, woven by men. We carry a knife to cut trees and we carry it in a special holder woven of bamboo.18

Despite the emphasis on past and tradition, clothing trends in the camp are very different from those inside Burma. The regulations of colour and dress length for women have largely disappeared and it is very unusual for a young woman in the camps to wear the full length white tunic, perhaps doing so only for a special ceremony. Women now wear any colour of shirt at any age though it is still more common for the red-and-black combination to be worn by older, married women. Karen shirts are produced in a range of ‘modern’ colours – electric blue, vivid pink and purple. Perhaps the biggest shift in style is towards Western and Thai-style clothes. Dudley (1999) describes Karenni women as feeling stressed and traumatized when they had to discard traditional dress and begin to wear the clothes available in the refugee camp. Downloaded by [University of Defence] at 21:28 09 May 2016 In contrast, Karen women – particularly young women – seem to have embraced the opportunity for new fashions. Young women wear trousers and shorts, while young men have embraced Western-style clothing (jeans, shirts) and the opportunity to express individuality in appearance through growing long hair or ear-piercing. Though decidedly tame from a Western perspective, elder members of the community saw this as a worrying decline in moral standards: 52 Governing refugees

The young people here and the young people in Burma are very different. If the young people in Karen state ﬁ nd a teacher is teaching them, they listen carefully. They have short hair. No earrings. They give respect. [. . .]We respect the teacher as a god in Burma. In refugee camp, no. Some of the students have long hair, earrings, tattoo. When they see the teacher they don’t say hello, they don’t give respect [. . .] They get more and more rebellious. I don’t know. Even their dress, they dress so bad. Not deﬁ nite Western clothes but not Asian, but between Western and Asian. Between two cultures. Not really Western, not really Asian.19

A further sign of the camps’ position ‘between two cultures’ was the introduction of Thai school uniforms. In 2010, the higher level schools in the camps (post-tenth grade) adopted Thai uniform to be worn for one day or more each week. Girls disliked the uniform as the skirt was tighter and shorter than a Karen sarong and was less comfortable for both walking and sitting in. The introduction of Thai uniforms was also resented as it was perceived to send a message that refugees must become more ‘Thai’.20

Dynamic communities An emphasis on Karen traditions is not unique to the refugee camps. Cheesman (2002: 200) reports a speech by an elderly Karen man on Karen New Year of 1998 at a location near Rangoon: ‘Our people must wear their clothes, speak their language and study their texts every day, or all will be lost!’ What is unique to the camps is both a lack of alternative narratives to challenge the dominant representation of ethnic identity, and a site- specifi c interpretation of these traditions. As in the lines by Brecht quoted at the opening of this chapter, identities practised in the ‘dark times’ of displacement are reinterpreted for that context. The ethnic markers of stories, language and clothes all gain their legitimacy and power from the perception that they are ‘traditional’ but they are performed in a very specifi c manner in the camps, in a way that differs from ethnic expression inside Burma and among Karen populations in Thailand. The stories that are told in the camps are tailored to the experiences and environment of refugees, telling of the suffering and victimization of the Karen Downloaded by [University of Defence] at 21:28 09 May 2016 and emphasizing the capacity to overcome those experiences. Language itself is evolving – Thai is rarely used, and even Burmese is not common (at least in Mae La Oon and Mae Ra Ma Luang), while English words are scattered into conversation. ‘Karen clothes’ are popular but unusual designs and colour combinations are used in the more modern environment of the camp. Other clothes very clearly refl ect the conditions in which refugees live, such as t-shirts provided by NGOs. Refugee camps are frequently described as liminal spaces, existing between two (or more) societies, cultures and traditions (Turner 2005a: 99; Dudley The camp community 53

2010: 188). For Karen refugees, the importance of ‘Karenness’ is constant but the precise meaning of this identity is changing in (and as a result of) encampment. In the Karenni camps, Dudley (2010: 34) has similarly described the reshaping of cultural identity as ‘“traditional” became a dynamic, constructed and contested notion within the refugee community’. The centrality of a notion of tradition allows social change to be controlled in a measured process, ensuring that continuity with the past remains paramount. Thus, the balance that is struck between tradition and transformation is that of Lampedusa’s Sicilian nobles (2007: 19): ‘if we want things to stay as they are, things will have to change’.

Community interrupted: resettlement for those who remain Ensuring that ‘things stay as they are’ in the refugee camps was impossible after resettlement departures began. The capacity within camp committees and community organizations that had been built up over years was shattered as the most capable and knowledgeable staff departed for ‘third country’. The departure of teachers, medics and leaders had a direct impact on camp management capacity. In Mae La Oon, 40 per cent of teachers applied to resettle to America in 2009–2010 (KRC 2009). Those who remained were faced with the challenge of providing effective services with reduced capacity (Banki and Lang 2007; Thompson 2008). They were also faced with an infl ux of new arrivals, many of whom were perceived as ‘not real refugees’ but rather economic migrants seeking resettlement.21 Resettlement also indirectly reduced the cohesiveness of the ‘camp communities’ as those who intended to resettle disengaged from life in camp, perhaps leaving school or quitting their job. While they waited for their departure date – an unpredictable process, which might take several months or even years – they became bored and frustrated. Mentally they were already far away from the camp but physically they remained trapped in it. A rise in youth disobedience was dated to the beginning of the resettlement programme, as was a rise in ‘debt and loan’ cases. One judge consulted me to fi nd out what he could do to ensure that a person resettled in America repaid money borrowed from a refugee: did I think that the police in America would Downloaded by [University of Defence] at 21:28 09 May 2016 take action if he called them to report this? A consistent challenge in the resettlement programme, from the announce- ment of the programme through to the allocation of departure dates and destinations, was the absence of adequate information to refugees (Banki and Lang 2007; Shukla and Olsen 2007). This had potentially serious consequences for those who left but it also had a signifi cant impact within the camps, giving rise to an alternative information economy of rumours and gossip. Sunstein (2010: 6) has suggested that rumours spread because they refl ect deeper beliefs of the body politic: ‘they fi t with, and support, the prior 54 Governing refugees

convictions of those who accept them.’ Such prior convictions often relate to fears and perceived vulnerabilities, which in turn may be rooted in real dangers (Scheper-Hughes 2008). During my time in camp I was told new stories about resettlement on a near-daily basis: that after the resettlement process the refugee camps would be closed; that the UNCHR had bought land close to Chiang Mai and would relocate refugees there; that refugees would be forced to return to Burma; that refugees would be relocated to camps on the Burma-side of the border, under UNHCR control; that refugees who did not resettle would be given Thai papers and permission to work and assimilate in Thailand; that refugees would be resettled to Brazil; that refugees would be resettled to Japan.22 Each of these rumours was given equal weight and considera- tion and I was solemnly consulted about their implications: did I think Brazil would be a good country to live in? What about Japan? What language would refugees be required to learn? What religion did people follow? The most surprising, and also the most persistent rumour was that resettled refugees would be eaten by white foreigners (gawlawah aw pwakenyaw – ‘the white people eat Karen people’). This rumour was repeated endlessly, to the extent that it was publicly acknowledged and refuted by the resettlement agencies (IOM 2008: preface). It echoes a long-standing story told by Karen parents to young children to scare them into good behaviour (‘Behave well, or the white foreigner will come and eat you!’) but in the context of resettlement became a widespread and apparently genuine fear, with adherents even among the more educated members of the camp society. Describing a similar rumour mill within Kakuma camp, Feyissa and Horn (2008: 16) comment that:

People starved of vital information about their future will devour anything they can obtain. The news then becomes a hot point of discussion and speculation for about a month, after which it disappears to nowhere, as it came from nowhere. And hopelessness and emptiness return, with fear and uncertainty ﬁ lling the minds of those refugees who had such hope a few weeks earlier. Downloaded by [University of Defence] at 21:28 09 May 2016 There was also speculation about the motivations of the organizations and nations involved in resettlement. Some claimed refugees were being taken to America to become slaves, like people from Africa in the past; or to be conscripted into the army.23 Others believed that international organizations were trying to end the Karen insurgency inside Burma. This view gained more adherents after the US removed the KNU/KNLA from its list of banned ‘terrorist organizations’ and permitted KNLA combatants and former combatants to apply for resettlement:24 The camp community 55

Do they want to break our KNU? Because our people inside will be left without defenders. For individual families, maybe it is good for them, but for our people, their situation will become even worse.25

Resettlement also placed tremendous emotional strain on the wider refugee population. Behind each decision to go or stay there was a very painful reckoning of costs and benefi ts, and every departure was a wrenching separation from friends and family who would likely never be seen again. Again this had consequences for those left behind, adding another layer of sadness and separation to the losses that displacement had imposed. Fundamentally, resettlement was a process that happened to refugees – they did not invite the programme, had little understanding of how it operated and when it would end, and were expected to make life-defi ning choices on the basis of very little information. The incessant rumours indicated the deep vulnerability and insecurity that this engendered.26 The language used in talking about resettlement refl ected this powerlessness, with the most common phrase being ‘America is coming to take our Karen people’. As one young man (who later resettled himself) explained to me:

When resettlement ﬁ rst happened, our leaders were against it. They wrote letters saying, ‘Please don’t take our people’. But they said, ‘Can you pay for rice?’ And the leaders could not because we are weak. So now they come and take many people.27

Notes 1 Ethnic diversity has increased in some camps in recent years, but the majority population in each camp (by a substantial margin) continues to be Karen/Karenni. See endnote 9. 2 Similar categories are used by O’Neill (2010) in analyzing ‘Asylum, Migration and Community’. 3 Interview #54, Mae La Oon, 28 May 2009. 4 This may have been a particularly acute concern in Mae La Oon, which of all the Thai–Burma border camps is closest to the border (3.5km). 5 Field notes, May 2009.

Downloaded by [University of Defence] at 21:28 09 May 2016 6 Field notes, May 2009. 7 There have been Muslim refugees in the camps since the early 1990s but the population has recently grown. The term ‘Muslim’ is often used by refugees as though it is an ethnic rather than religious identity. TBC (2010b: 21) recognizes three ‘sides’ of the Muslim refugee community: (1) ‘Karen Muslims’ born and bred in Karen state who fl ed to the refugee camp to escape confl ict and human rights violations in their home area. Many members of this group arrived between the 1980s and early 2000s. (2) ‘Central Burma/Western Muslim caseload’ – these people fl ed to Thailand to escape conditions in Burma, which may have been human rights violations or economic. Members of this group typically arrived 56 Governing refugees

during the 1990s and early 2000s. They are primarily involved in trade and retail in the refugee camps. (3) ‘Thai-based Muslims’ – this group in general came to the refugee camps from other communities within Thailand in order to exploit economic opportunities. This group came to the refugee camps during the late 1990s and 2000s. 8 The origin of the parallel governance structure is not clear. Burman refugees informed me that it had been the decision of the KNU and KRC, while a KRC member insisted it had been the choice of the ABSDF. 9 Interview #49, Mae La Oon, 18 May 2009. 10 Interview #50, Mae La Oon, 23 May 2009. 11 Curwen (1989) quotes the then-KWO Constitution as listing four aims: (1) to free Karen women from oppression; (2) to raise the political consciousness of Karen women; (3) to be active in the KNU revolutionary movement, in accordance with KNU’s policies; and (4) to have equal rights with men. Currently the KWO website, www.karenwomen.org (accessed 7 January 2011) lists seven objectives of the organization. These focus on human rights, gender equality, freedom from oppression and protection of Karen culture. None mentions the KNU. 12 Field notes, 31 May 2009. 13 One of my interviewees criticized the parallel with Israel on the grounds that Karen people were not nationalistic enough: ‘So some pastors, they say, we are descendants of Israel – I told them, “Ah, Israel is not like this. Don’t be talking like that. Israel they are very faithful community and people. Not like Karen. Karen they run away here and there.”’ (Interview # 95, Mae Sariang, 12 March 2010.) 14 The KNU (2006: 1) lists ‘key characteristics’ of the Karens: ‘simple, quiet, unassuming and peace loving people, who uphold the high moral qualities of honesty, purity, brotherly love, cooperative living and loyalty, and are devout in their religious beliefs.’ An earlier KNU publication, Karens and Communism, identiﬁ es eight characteristics of Karen people: the knowledge that there is God, the divine being; high moral and ethical standards; honesty; simple, quiet and peaceful living; hospitality; language; national costumes; aptitude for music (Renard 1990: 106). 15 Kawthoolei is the Sgaw Karen name for their desired Karen nation. It has been variously translated in English as ‘green land’, ‘ﬂ ower land’, ‘the land burned black’ (Smith 1991: 141) and ‘land without evil’ (Rogers 2004). 16 Field notes, 24 Feb 2009. 17 Field notes, 20 Feb 2009. 18 Field notes, 22 May 2009. Downloaded by [University of Defence] at 21:28 09 May 2016 19 Interview #44, Mae La Oon, May 2009. 20 Field notes, 31 March 2010. Issues of youth disobedience and generational relations are returned to in Chapter 8. 21 An ambition that was disappointed, as resettlement was only available to refugees who had been registered with the UN and the Royal Thai Government before 2005. 22 The last proved to have a grain of truth, as Japan did accept a small number of refugees for resettlement in 2010; twenty-seven refugees in total, all residents of Mae La camp (Miyajima 2010). The camp community 57

23 Field notes, 11 May 2009. 24 The United States Department of Homeland Security excludes immigration opportunities to anyone who has provided ‘material support’ to a listed terrorist group. Until 2006 the KNU/KNLA was on this list. The ABSDF, which also has members in the refugee camps, was not similarly exempted until 2011. This led to much confusion among Burman refugees in Mae La Oon (many of whom were ABSDF members) as to why they were apparently not being given the same resettlement opportunities as Karen refugees. My perception of the situation – that it was an oversight – was not considered credible, and conspiracy theories circulated accordingly. Interview #50. Section 13 Committee, Mae La Oon. 23 May 2009. 25 Mae La Oon. Field notes (conversation with teacher, who subsequently resettled), April 2009. 26 Turner (2004: 245) has described rumours in Lukole refugee camp in Tanzania, where grand conspiracy theories circulated (for example, that the Rwandan government had sent beautiful Tutsi women to seduce Bill Clinton and ensure he was receptive to their policies). These rumours demonstrate a level of interest in, and familiarity with, global affairs that I rarely encountered in Mae La Oon. I was in the refugee camp during very signiﬁ cant global events (including the inauguration of United States President Obama in 2009 and the Haiti earthquake in 2010) and rarely encountered any awareness of their occurrence. Rumours were concerned with the world inside the camp boundaries, and particularly with the international resettlement programme. However, they echoed the stories related by Turner in refugees’ evident desire to make sense of their circumstances and to reposition their own experience at the centre of distant processes (for example, in claiming that the resettlement programme had been created to end the Karen insurgency). 27 Mae La Oon. Field notes, 11 May 2009. Downloaded by [University of Defence] at 21:28 09 May 2016 Chapter 4 The governance palimpsest: order maintenance in Southeast Burma

The force of habit, the awe of traditional command and a sentimental attachment to it, the desire to satisfy public opinion – all combine to make custom be obeyed for its own sake. In this the ‘savages’ do not differ from the members of any self-contained community with a limited horizon, whether this be an Eastern European ghetto, an Oxford college, or a Fundamentalist Middle West community. But love of tradition, conformism and the sway of custom account but to a very partial extent for obedience to rules among dons, savages, peasants, or Junkers. [. . .] in the main these rules are followed because their practical utility is recognized by reason and testiﬁ ed by experience. (Malinowski 1926: 52)

There is a perception, even an expectation, that refugee camps are spaces where community bonds break down and values disintegrate. Yet the refugee camps in Thailand have remained relatively stable and secure throughout tremendous political and demographic ﬂ ux. Violent crime occurs, but it is exceptional. From a social science perspective this presents an important question: if these camps are not anomic, then what are the processes of social ordering that are being marshalled? To answer this question it is helpful to look beyond the immediate situation of the refugee camps to investigate the governing processes that were used in the communities from which people ﬂ ed. Santos (2006: 47) has described Mozambique as a ‘palimpsest of political and legal cultures’ where ‘very different political legal cultures and very Downloaded by [University of Defence] at 21:28 09 May 2016 different historical durations are inextricably intertwined’. A similar palimpsest of legal cultures exists within Karen State, where village sovereignty, supernaturalism, missionisation, colonialism and political revolution have each left traces upon approaches to law and order – and which can in turn be traced in the practices of dispute resolution and order maintenance which are used in the refugee camps. These eras of governance have not evolved in linear fashion but have overlapped. Practices and beliefs that were followed before colonization (such as village dispute resolution) continue to serve as the cornerstone of political governance, and have been supplemented but never The governance palimpsest 59

fully replaced by newer additions, such as Christian values or the state- building project of the KNU. New layers of governance have changed attitudes towards some crimes and punishment but overall there is considerable continuity in social values and concerns. This is particularly evident in the policing of extra-marital sex, consistently regarded as an extremely serious transgression even as the rationale for its punishment has shifted from supernaturalism to Christian morality to legal (or quasi-legal) authority. Attempting to defi ne ‘law’ is challenging, as if the boundaries are narrowly drawn (for example, requiring formal codifi cation or professional practice) it excludes many norm-defi ning and enforcing authorities, while if the defi nition is over-broad it misses the particular cultural impact and infl uence that makes ‘law’ something different from convention or etiquette (Tamanaha 2008, 2001; Starr and Collier 1989; Nader 1969a). It is more helpful to think of governance practices in Southeast Burma in terms of order maintenance; i.e. the desired outcome of dispute resolution rather than the institutions or mechanisms used to resolve disputes. Following Benda-Beckman and Pirie (2007: 1), ‘Order, in this sense, connotes a prescription of “the way things need to be” (e.g. which activities and behaviours should be permitted, and which proscribed) in order to ensure a secure environment.’ The desire to ensure social order lies at the heart of security and governance practices. Benda-Beckmann and Pirie (2007: 1–2) identify some characteristics of the concept of order: order can be subjective or objective, order can exist even in apparently disordered situations (such as a confl ict zone or a refugee camp) and order does not exist spontaneously but must be generated and produced through specifi c practices. In this respect criminal law is just one tool among many methods of producing order, such as custom, tradition, religion and even violence (Johnston and Shearing 2003: 7). Therefore, it is not that ‘order maintenance’ describes non-state forms of policing while ‘law’ refers to the formal justice system. Rather, codifi ed and un-codifi ed systems co-exist and operate simultaneously as part of a network of order maintenance mechanisms. Similarly, the primary actor in securing order is not necessarily the nation-state. In Burma’s ethnic territories, multiple state and non-state actors seek political power and control over land, resources and populations. This governance complexity has implications for the administration of justice Downloaded by [University of Defence] at 21:28 09 May 2016 as it establishes a patchwork of power-holders, none possessing formal jurisdictional authority yet all engaged in policing or dispute resolution activities in various ways. The Burmese national court system operates at township and district level but this formal legal system is rarely used by ordinary villagers in Karen State who instead must rely on an assortment of informal governing authorities: village leaders, members of non-state armed groups, even soldiers of the Burma Army. Unsurprisingly, this leads to highly inconsistent and often unsatisfactory outcomes.1 There is an echo of the patrimonial governance climate described by Kelly (2006: 113) in the 60 Governing refugees

West Bank, where the legitimacy of the Palestinian Authority has withered as other agents have grabbed power and control. However, despite the absence of an effective, universal rule of law it is apparent that ordering processes do exist and function (albeit with varying degrees of effectiveness), and that these processes have normative force. This chapter examines the infl uence of village sovereignty, moral narratives, supernaturalism or ‘spirit-law’, Christianity, and ‘KNU law’ on the defi nition of harms and maintenance of order in Southeast Burma; and traces the extent to which these various sources of authority have been imported into the refugee camps. The fi nal sections consider the impact of changing frames of governance on perceptions of harm and wrongdoing, focusing on attitudes towards monogamy and a belief in ‘one husband, one wife’.

Order beyond the state

Village sovereignty As was described in Chapter 2, colonization transformed Karen nationalist politics but was much less infl uential at the level of local governance. Every village has a leader, chosen by the villagers. Disputes and problems within a village are resolved by this leader together with other elders. Describing dispute resolution in a Thai-Karen village, Andersen (1979: 320) found that a fi nal decision was agreed after ‘a debate which is governed partly by the existing rules, that are known to most villagers, and in particular by the social-historical knowledge of the Elders’. Village leaders acquire status but receive little material benefi t from their role. There is no automatic inheritance or transfer of leadership functions and no particular privileges attach to the role as a matter of right. Thus, the role of village leader refl ects a social value system that is both hierarchical and egalitarian. In a study of Karen society in the 1920s, Marshall (1997 [1922]: 129) found that a leader worked together with other villagers and did not levy any taxes or demand payment, though gifts would often be given to him in recognition of his work and popularity: ‘The Karen had no caste of chiefs, no royal family, or even a privileged social class. Every member of the com- Downloaded by [University of Defence] at 21:28 09 May 2016 munity shared alike in the ordinary tasks and the privations or prosperity of the seasons.’ Not only does the role of village leader lack attractive privileges, in recent years it has brought considerable danger as village leaders are frequently targeted by Burma Army troops to supply resources (including forced labour) or singled out for particularly brutal treatment. In the hope that soldiers would be merciful towards women, some villages appointed female leaders. The KWO (2010) found that female leaders continued to be targeted by Burma Army forces, and reported witnessing or experiencing horrendous The governance palimpsest 61

abuses, including execution, burning alive, gang rape and torture. Though rural Karen villages are conventionally patriarchal, a young female village leader who has been elected strategically will still command the obedience of all villagers (in part, it seems, because villagers recognize that she is assisted in the exercise of power by village elders) (Heppner 2006). The refugee camps were formed through the uprooting and displacement of entire villages, and the management structures developed in the refugee camps show a great deal of continuity with these practices of village-level governance. Institutionally, every camp is divided into administrative sections, each with a ﬁ ve-member section committee (comprised of a section leader, section social welfare representative, section security leader, KWO member and KYO member). Disputes are sent ﬁ rst to the relevant section leader, and will be referred upwards to the camp committee or justice team only if agreement cannot be reached. Refugees acknowledge the continued relevance of pre-camp governance models to their current way of life:

We live peacefully in the camp because we are Karen people. Before, we lived in the jungle and we lived quietly. For example, in a village all villagers will listen to the village head. If he decides something [everyone is] obedient and quiet. And when we moved to the camp there are lots of people so we have to divide it into sections and then peace and reconciliation will be the section leaders’ duty, and also the camp leader’s, and also everyone’s.2

In our culture we can live stable together. We are Karen. We don’t need law and justice. In the past, in our traditions we have no high-level law but only community management. But we can survive with our management system. We love peace and unity and can live peacefully without high level law.3

There is also continuity in the type of sanctions imposed. Criminal justice approaches in the heartland of Burma were transformed during the colonial era, when punishments focused on the offender’s body (such as ﬂ ogging, mutilation, execution or exile) were replaced with a prison system focused on containment. The simple bamboo huts which had served as detention Downloaded by [University of Defence] at 21:28 09 May 2016 units were replaced with secure concrete institutions, located outside urban areas and built in accordance with the latest Victorian prison architecture (Wintin and Brown 2005).4 This transition from personalized, public punishment to depersonalized, private punishment is a classic trajectory in the history of punishment practices (Pratt 2002; Garland 1993). However, it was not followed in rural highland Burma, which has largely retained the informal social practices of pre-colonial times. Simple bamboo huts continue to serve as detention centres, with additional sanctioning options including ﬁ nes, ritual ceremonies, violence and execution. 62 Governing refugees

A particularly serious punishment in Karen traditional justice was expulsion from the village, a sanction which might well lead to death if the offender was cast adrift from the support structures of the community. A person who repeatedly violates the rules of the village might be banished or may decide to leave voluntarily if s/he cannot accept a decision of the village leader. This is a common technique of dispute resolution in mediation-focused societies, where if a dispute proves long-standing or intractable, one party will simply move away. In the Karen context, Marshall (1997 [1922]: 288) described the practice of expelling wrongdoers from the village as ‘a primitive quarantine’ to protect the rest of the village from the repercussions, while Scott (2009: 264) presents it as an adaptive process for escaping from disputes and obligations and maintaining local harmony. Again, this practice continues in the camps, both in terms of voluntary relocation to avoid a dispute with family or neighbours and in the possibility of expulsion from the camp for those who repeatedly refuse to obey the camp leaders. (See further Chapter 6).

Moral narratives Tools for ‘order maintenance’ are not always immediately apparent in the absence of legal codes and institutions. One example of this is the role of moral narratives and didacticism. The earliest English language descriptions of Karen village life and governance are found in the writings of Christian missionaries to Burma. In an essay published in 1868, Mason records dicta and rules as narrated by elders of a Bghai (Bwe) Karen village.5 As with much of the writing of this era, Mason’s tone is jarringly patronizing and dismissive for modern readers but the core content depicts something surprisingly recognizable within modern Karen society: village-level organization was a broadly egalitarian social order led by a village leader elected by village residents. There was no written legal code or regulations. Instead, the elders of each village were repositories of knowledge and of ‘right’ behaviour, conveying moral, social and legal norms to the younger people and determining disputes in accordance with these norms. Mason reports the instructions of village elders as exhorting values of hard work and the avoidance of laziness, generosity to those in need, thrifty management of Downloaded by [University of Defence] at 21:28 09 May 2016 resources, humility, obedience, cooperation, honesty and harmony:

Do not hate each other [. . .] do not defraud, do not be dishonest in your transactions [. . .] do not do evil to each other [. . .] do not steal. (1868: 138).

What you do, do thoroughly and completely. We love happiness; and our greatest happiness is to clear our ﬁ elds and to build our houses. (1868: 133) The governance palimpsest 63

The elders’ teachings reinforce the positive consequences of correct behaviour and promise terrible repercussions for bad behaviour:

Children and grandchildren, love peace and you will live in peace and live to be old. He who loves peace, his house will be established and it will be permanent. [. . .] He who walks in peace, will enjoy peace. He will have associates, he will have friends [. . .] He will have no adversaries, he will have no enemies. (1868: 135)

He did evil, and his evil returns upon him; and he grieves, and weeps, and suffers anguish; and when he dies he has no one to bury him, and his body remains in the ﬁ eld, and the birds devour it. (1868: 138)

In essence, the philosophy of Karen village governance was that a well-ordered community is a peaceful one, and order was maintained by following good practices and avoiding bad ones. Many authors have recognized that Karen cosmology prizes obedience to tradition above development and change (Marshall 1997 [1922]: 37; Cheesman 2002) so it is perhaps not surprising that the moral didacticism recorded by Mason remains much in evidence among Karen communities today. Moral instruction is conveyed in a range of social forums, including religious ceremonies or public events, and in ‘lighter’ narrative forms such as songs, poems and jokes (Andersen 1979; Hayami 2004). Short poems – hta – historically served many functions in Karen society and continue to be performed and written at funerals, among other social occasions (Fink 2003; Mischung 2003). In my own fi eldwork, camp leaders and judges continued to emphasize moral values when resolving disputes, repeating much the same principles that Mason encountered: work hard, show humility, respect elders, don’t gossip or quarrel, avoid dishonesty, selfi shness and sexual contact outside marriage. A premium was placed on peaceful co-existence and the avoidance of quarrelling, gossiping or otherwise escalating disputes.6 In Mae La Oon, any event where many people were congregated served as an occasion for disseminating pro-social moral instruction. Often, this Downloaded by [University of Defence] at 21:28 09 May 2016 role was undertaken by spiritual leaders who intoned lessons and prayers for hours at a time, frequently leading audience members to fall asleep. In camp dispute resolution practices, particular emphasis continued to be placed on perceived cultural norms of ‘Karenness’ such as honesty, peacefulness and confl ict-avoidance:

We don’t have any problems with law. For our Karen people, the big problems we make small and the small problems we make go away. We don’t want problems. Our Karen people, they are honest, simple, pure. 64 Governing refugees

Here in the refugee camp there are more than 50,000 people but you never hear anything about big problems. Everything is very simple and our law is very simple. We solve problems in a simple way. Our Karen people are always simple. That’s how we became refugees [laughs].7

A Sgaw Karen language report on ‘Karen Traditional Justice’ describes Karen villages as peaceful societies which experienced few disputes because Karen people are quiet, honest and non-confrontational and where,

[the] old or senior people listened to both sides and if it is a big case they made it smaller and if it was a small case they made it disappear, and they settled the case to make both people understand and satisﬁ ed with each other. (Drum Publications 2006: 30)8

Likewise, at a ceremony conducted by Mae La Oon’s small animist community in January 2010, the prayer (as translated by my research assistant) emphasized peaceful living and repeated the idealized notion of Karen communities as harmonious:

In the old days our ancestors were like a dove, they lived peacefully and never quarrelled. So today in our time, when we drink this strong alcohol, we should not quarrel but we should thank God and he will send us good things for our children, look after us and protect us from bad things.9

The centrality of moral didacticism to social life suggests an intriguing possibility, that these moral narratives and messages establish a set of values and expectations which operate as a crucial mechanism for maintaining social order: in effect, a cultural code of citizenship, repeated at numerous sites and fully integrated into daily life. Geertz (1973: 448) is widely quoted for a defi nition of culture as ‘the stories we tell ourselves about ourselves’.10 This phrase subtly differs from the sentence Geertz actually wrote, but its reso- nance refl ects a wider truth about the importance of narrative in shaping identity, infl uencing thoughts and behaviour. Religion offers one narrative framework through which expectations are conveyed but narratives and Downloaded by [University of Defence] at 21:28 09 May 2016 stories may also be important at a communal or even individual level. In a rather different context, in interviews with ex-prisoners, Maruna (2001) found that those who had successfully desisted from crime explained their past experiences and future hopes in terms of redemption and ‘making good’ that were markedly different from the self-narratives of those who had returned to crime. Engel and Engel’s (2010) study of legal consciousness among ordinary people in Northern Thailand found that changing notions of legalism and legal culture were refl ected in ways of explaining the causes behind personal injury; these narratives had moved from emphasizing The governance palimpsest 65

spiritual retribution to (variously) Karmic justice or personal negligence, but rarely embraced the concept of external fault that is the basis of civil liga- tion in formal Thai law. Such studies suggest that Geertz’ language might be taken even further: that the ‘stories we tell ourselves about ourselves’ are not only a form of cultural deﬁ nition but have profound implications for individual behavior, and therefore for understanding patterns of crime and order- maintenance. It might be hazarded that the stories Karen refugees tell – with their emphasis on the importance of moral and ethical behaviour – play a vital role in establishing shared expectations of citizenship and in maintaining stability and order within these communities; inside villages in Burma and Thailand, and in the refugee camps.

Animism or ‘spirit-law’ The mention of Mae La Oon’s animist community points to another important infl uence on understandings of social order within the refugee camps, which is that of supernaturalism. It is diffi cult to generalize about ‘Karen’ animist beliefs as there are many differences between districts and even between villages. Karen villages appear to have maintained considerable autonomy at the spiritual level too, with a multiplicity of small cults, individual prophets and diverse animist practices (Hayami 2004: 177–200). Furthermore the nature of such belief systems is that they exist as a cosmology and therefore impact on multiple sectors – religion, story-telling, health and medicine, law and dispute resolution, rituals and ceremonies. It is impossible to fully understand their impact and meaning without deep and dedicated anthropological research. However, even in the absence of such a comprehensive analysis it is apparent that spiritual beliefs continue to exert infl uence on social practices both inside Burma and in the camps. The following sections sketch how and why this is the case, drawing on colonial– era publications (Cross 1854; Mason 1868), fi ndings of research conducted among Karen living in Thailand (Rajah 1986; Andersen 1979, 1981; Hayami 2003, 2004) and insights and comparisons from my own fi eldwork. Traditional knowledge among the Karen is infl uenced by perceptions of spiritual cause and effect which have implications both for conceptualiza- tions of and responses to wrongdoing. The basis of Karen animist belief is Downloaded by [University of Defence] at 21:28 09 May 2016 spirits – k’la – which are territorial in nature. Everything has a k’la – a human being or an animal, but also a tree, a plant, a knife, a river, a plank of wood or a stick of bamboo in a house frame. In addition to k’la belonging to a particular being or object, there is an overarching spirit, hti k’sa kaw k’sa, (the god of heaven and earth) as well as many lesser spirits which govern weather processes such as wind, sun and moon, lightning and thunder. The k’la is not identical with the ‘soul’, as it can travel apart from its host, but it is linked to vitality. If a living thing becomes weak – human, animal or plant – it is believed that the k’la has travelled away. The human k’la is 66 Governing refugees

believed to wander from the body during sleep or times of physical weakness, and can also be detained by spells or witchcraft. If the k’la becomes separated from its ‘host’, the host will die. To prevent this, offerings and ceremonies can be made to attract the k’la back to the body. K’la can be destructive forces, but a person will not suffer negative effects if his/her ‘tso’ is in place. The tso is located on the top of the head and so particular care is taken to protect the head, e.g. by avoiding walking underneath a house or low-hanging tree (which may possess a dangerous k’la). An annual ‘wrist-tying’ ceremony is conducted, where a thread is tied around a person’s wrist to keep the k’la attached to the body. The importance of the k’la is refl ected in dietary advice which seeks to maintain spiritual strength in times of weakness, such as pregnancy, childbirth and postnatal care. For example, a pregnant woman should avoid most ‘strong’ or ‘hot’ foods (a category that includes chilli and some meats and fi shes), while some other ‘strong’ foods are prescribed to help recovery after illness (e.g. dog meat is believed to help recovery after childbirth and from malaria, while papaya and banana are believed to make malaria symptoms worse). Karen animism recognizes a great variety of spirits which are dangerous to human k’la, and this has a direct link to community governance. Activities which are proscribed are often prohibited because they are believed to be offensive to the spirits rather than harmful per se, and leaders’ responses are intended to appease the spirits rather than to redress the victim. Falla (1991: 246) suggests that, ‘Almost any taboo, custom or prohibition in Karen life may be explained as placating the spirits. Offend them, and the crops will fail, the pigs and buffaloes will die and the people sicken.’ Hayami (2004: 142) lists examples of acts that might anger the spirits: ‘Extra-marital sexual liaisons, theft, fi ghting, confl ict, gambling and speaking bad words within the community boundaries.’ There are some slightly surprising exemptions to the list of prohibited acts – for example, as far as I could ascertain neither suicide nor abortion is considered to bring spiritual repercussions.11 In contrast, extra-marital sex is believed to be particularly offensive to the spirits, and retribution for this offence will be imposed on an entire village or community in the shape of bad weather and failing crops. Dispute resolution at village level was traditionally understood and is still Downloaded by [University of Defence] at 21:28 09 May 2016 sometimes explained in terms of ‘heat’ and ‘cooling’. Transgressions make the land ‘hot’ and unstable and anger the spirits, but spiritual displeasure can be averted by making the land ‘cool’ again. Managing this process requires an intermediary and in animist villages, in addition to a village chief there will be a designated spiritual leader (in Sgaw Karen, bu kho) who leads spiritual ceremonies and interventions. In Karen villages in Thailand, Andersen (1981) described the role of the bu kho as ‘priest-like’, noting that while in the past each village would have its own religious leaders, by the time of her research there were many less such leaders than in the past, with villagers travelling The governance palimpsest 67

instead to a neighbouring village for rituals or ceremonies. The process for cooling the land must be determined by the village bu kho in each case but will typically involve an animal sacrifi ce and ceremony. After the ceremony has been completed, the problem is considered solved. Sometimes, negative impacts will be apparent before there is any knowledge of wrongdoing; in such a case, the bu kho’s task is to divine the reason for spirit retribution (tools for this task include sticks or chicken blood) and make the appropriate appeasement ceremony. As the foregoing illustrates, spiritual beliefs, communal rites and sanctions (and in particular the need to keep the village ‘cool’) played an important part in maintaining social order in village societies. This ‘spirit-law’ was effective because it bound a village together in a shared destiny and politico-religious community. If the spirits were displeased, and not subsequently appeased, the whole community would suffer. Yet by participating in the designated rituals and ceremonies problems could be resolved and disaster averted. Village residents therefore had reason to use the existing structures as well as reason to believe in them. Territoriality was central to these beliefs and practices, and the community interest in policing declined if acts were committed beyond the territorial reach of the local spirits. Engel and Engel (2010: 46–76) describe a similar belief system operating in Northern Thai villages, centred around household and village spirits, powerful life forces whose power is territorially rooted and declines on moving away from the ‘sacred centre’. Ceremonies and offerings were made to the local spirits, with individuals appointed to manage the process and compel participation. Spiritual connections bound family members and villagers together in a spatially-defi ned social order and created a collective interest in redressing violations. They describe this belief system as ‘the remembered law of sacred centres’, a term which captures both the spiritual and territorial dimensions of this normative order. Territoriality is similarly central to Karen understandings of community policing. Hayami (2004: 141) describes one Karen animist village as recognizing the forest beyond the village as literally lawless, where the normal rules do not apply; so, for instance, gambling, which is prohibited in the village, is permitted in the forest. Similarly, when outsiders come to a Karen animist village they may be expected to adhere to local spirit-placating Downloaded by [University of Defence] at 21:28 09 May 2016 practices. Several foreign visitors to Karen animist villages have described having to submit to a marriage ceremony after they stayed in a house where an unmarried man was present. Hayami (2004: 158) describes one woman – a Thai teacher – protesting that she was already married. She was told this was irrelevant; her marriage ceremony had no meaning in the village, while the local ceremony was purely to appease the local spirits and would not need to have any binding value beyond the village. This fl exible understanding of the marriage contract (otherwise taken extremely seriously in Karen society, as will be explained below) is echoed in Marlowe’s (1979: 74) description of 68 Governing refugees

a Karen man who had been married several times but did not consider he had been ‘truly’ married because his wives were not Karen:

As long as I marry Thai girls and sleep with Thai girls it’s all right. I can divorce them or leave them, it doesn’t matter. Now one day when I am ready to settle down I will marry a Karen girl in church. But you must understand, when I do that, that’s it. I can’t ever get divorced, or leave her, or play around again. That is why I keep away from Karen girls.

This moral relativism is a pragmatic response to a very practical challenge of any community justice system, which is the ability to extend jurisdiction to those who do not choose to accept it. A similar pragmatism is apparent in spiritual practice in the refugee camps. The emphasis on ritual and sacriﬁ ce in animism is also difﬁ cult to maintain when people have no money to buy animals and extra food. If they are no longer able to observe the practices, for some people it is a relief to adopt a religion which does not require them, such as Christianity. The animist leader I met with in Mae La Oon acknowledged that he was not really able to observe the practices correctly in the camp because he was the true bu kho only of his home village and not of the spirits in the camp. Though he performed key ceremonial rituals of the annual calendar as far as he was able, he viewed his role as one of preserving cultural knowledge rather than truly communicating with the spirits of the area (which would be the responsibility of a local bu kho). Instead, he claimed that the Thai-Karen villages surrounding the camp performed the rituals on behalf of the camp population, including ceremonies of appeasement. So, for example, if the local villages heard of a case of adultery between refugees, they would perform a ceremony to placate the local spirits.12 A bu kho can serve as an intermediary with the spirits, but other efforts to harness or direct supernatural forces (such as by the use of plant poisons or enchanted objects) are viewed as intensely risky, potentially returning against the person who uses them and causing sickness or death. In the past, Mason (1868: 149) wrote that if a Karen person was found guilty of having used supernatural powers, ‘it is considered a meritorious deed to put him to death’. To this day the possession of ‘evil’ objects still carries the death penalty inside Burma – a penalty which is still exacted on occasion, as in a case reported by Downloaded by [University of Defence] at 21:28 09 May 2016 the Mon Forum (2008), where a husband and wife were executed by DKBA forces for suspected witchcraft, though their neighbours described the man as a traditional healer and his wife as ‘just a normal person’. In the refugee camps, witchcraft is recognized as a crime but the likelihood of prosecution is minimal as an allegation must be supported with actual evidence of possession and use of supernatural objects.13 The possession of supernatural objects (examples of which include an ‘evil hand’ or an ‘evil stone’) is not permitted and witchcraft will be ‘proven’ as the cause of death if an unusual object is present in the body, such as a coconut in the stomach. The governance palimpsest 69

This had never happened in the knowledge of my sources, though allegations of witchcraft do occasionally occur. One middle-aged woman explained to me that she had been accused by her daughter-in-law of implanting a coconut in her stomach to kill her. When the daughter-in-law visited camp medical staff, she was told that her stomach pain was caused by a gastric ulcer but persisted in the allegation. According to the mother-in-law, members of the camp social welfare department met with her family and advised them that it was not good for a family to quarrel, that they should live together peacefully. Instead, the woman’s son and daughter-in-law opted to move to a different section of the camp, far away from the woman’s house; this, as noted above, is a classically ‘Karen’ response to an intractable dispute. The role of spirit-based belief systems in the camp has to be interpreted in light of a further layer in the governance palimpsest, that of Christianity. Most community leaders and camp committee members are Christians who are aware of the Western scorn for supernatural subjects and are embarrassed to discuss the topic. One camp leader that I asked about witchcraft dismissed the topic entirely:

In the Karen tradition, in the past we have a special law on spells. Like, if you have enough evidence, a speciﬁ c item like a giant’s palm or an evil stone, if you have some evidence you can be punished by special law on spells. Here, it is rarely punished. In Karen state, it is up to death penalty, it’s really harsh. But here, there is not enough evidence. It is difﬁ cult to understand which item can be used for spells. It is not practical, not logical. So now, we don’t even talk about that kind of law. You can’t be punished for it. It’s not logical. It can’t happen.14

Few other Karen refugees, even other Christians, expressed such strongly rational views. In Mae La Oon, the population is overwhelmingly Christian yet the ‘wrist-tying’ ceremony remains one of the most widely celebrated ceremonies in the annual calendar.15 Many Christian Karen profess not to believe in spirits, but there is an almost universal fear of ghosts. When the camp leader claimed that spells ‘can’t happen’ my research assistant was visibly shocked. After the interview I asked whether he agreed that witchcraft was not logical. His response was a pragmatic one: ‘If I say Downloaded by [University of Defence] at 21:28 09 May 2016 I believe in spells, foreigners like you will think I’m crazy. If I say I don’t believe, maybe the spirits will be angry with me. So I say – maybe. Maybe yes, maybe no.’16

Christianity As was detailed in Chapter 2, the elite of the Karen nationalist movement are predominantly Christian and this dynamic has been incorporated into the camp environment. Christianity is a minor religion for the Karen population 70 Governing refugees

as a whole but it is the dominant religion in the refugee camps, particularly in Mae La Oon. Some reasons for this were discussed earlier (see Chapter 2). Another reason for conversion can be the practical difﬁ culties of practicing traditional animism in the refugee camps. As noted above, spirits are believed to be located in particular sites and objects. When people travel away from their home area, the power – protective and destructive – of these spirits declines. Furthermore, ceremonies of tribute or appeasement have a cost outlay that many refugees simply cannot afford. Such challenges lead some refugees to convert to Christianity, in a decision that is both logical and strategic. An example of this occurred among my own acquaintances, when a student at the school that I taught in was found to have been in a sexual relationship with a girl from a neighbouring Thai-Karen village. The girl became pregnant and the (animist) village leader demanded that the boy provide a buffalo for a sacriﬁ ce. As he could not afford a buffalo, he instead converted to Christianity and married the young woman in a Christian ceremony, thus performing the only other ritual that was capable of cleansing a pre-marital sexual relationship of its negative repercussions. Christian missions transformed local conceptions of law and order in many of the sites where they operated, with control of sex and sexuality seen as a fundamental component of the mission civilisatrice (Chanock 1985; Merry 2000a). Monogamous marriage was a particular concern: in Hawai’i, missionaries campaigned against promiscuity and adultery (Merry 2000a: 19, 221–257), while in many African nations opposition was directed at polygamy (Comaroff and Roberts 1977; Hunt 1991). The impact of Christian missions on Karen sexual morality is less clear-cut. In part, this is because only a minority of Karen people were converted to Christianity. More importantly, sexual conservatism pre-dated the arrival of Christian missionaries. Indeed, it is likely that the shared importance placed on monogamy and the nuclear family is one of the reasons for the Karen receptiveness to Christianity. As was discussed above, Karen ethnic identity is tightly bound up with strict behavioural expectations and a strong prohibition against extra-marital sex. The belief that Karen people are of particularly strict moral virtue is shared by other ethnic groups, including Burmans and Thai (Cheesman 2002; Hayami 2003). Baptist missionaries among the Karen applauded the emphasis on monogamy and interpreted it as another indication that Karen Downloaded by [University of Defence] at 21:28 09 May 2016 people were unusually close to Christianity and a ‘lost tribe of Israel’ (Mason 1862: 126). However, while Karen Christians are censorious of adultery, Karen animists are equally or even more so. As was explained above, animist ‘spirit- law’ sees adultery not as a crime between individuals or a sin in the eyes of God but as an act that will bring bad fortune to the entire village. This creates an incentive for policing which extends beyond the notion of individual punishment and into community self-protection. Among Karen villagers in Thailand, Hayami (2003: 120) found that animist villagers The governance palimpsest 71

viewed their Christian neighbours as morally lax (and were particularly appalled that pre-marital sex was not considered problematic if the couple subsequently married) and believed that abandonment of animist beliefs had lowered moral standards:

In some ways, rather than Christianity bringing sexual moral[s] and monogamous family to a heathen amoral society, one might say that Christianity brought in effect a deterioration of sexual sanctions based on rituals on both the communal and family levels. Ironically for missionaries, Christianity was seen from the non-Christian Karen to be the source of moral degradation.

The Karen National Union and Kawthoolei Law Since the 1950s, the KNU has operated as the de facto state authority within ‘liberated areas’ of Karen State. Though the territory entailed by this has shrunk drastically in recent decades, some remnants of its ambition for a civil service remain. In the KNU governing agenda, health, education and welfare were early priorities, with law coming much later.17 The ‘Kawthoolei Law’ is modelled on the Burmese Penal Code (itself a product of the British colonial system) with courts at three levels: township, district and headquarters. Judges are appointed by the KNU and no formal training in law is required or provided, though judges I spoke with had attended a short course (between one and three weeks) of training by the Burma Lawyers’ Council. In deciding a case, judges hear evidence from witnesses and reach a decision and judgment. No lawyers are involved and a ﬁ nal decision is reached by the judges alone on the basis of the written KNU law. Kawthoolei Law thus represents an attempt to create a state legal system in accordance with the colonial model, including penal sanctions and a punitive rationale for policing. The elements of consensus and traditional knowledge which characterize village council decision-making are apparently absent,18 and though there is a Justice Minister, in practice higher judicial decisions are often decided by the KNU Executive Committee. The narrow ideal of Karenness represented by the KNU elite – Sgaw- speaking, Christian – is also apparent in the administration of justice. Downloaded by [University of Defence] at 21:28 09 May 2016 Christianity has been a potent motivation for senior members of the KNU, particularly General Bo Mya (KNU president from 1976 to 2000), whose Seventh-day Adventist beliefs led to an increasingly hardline approach to the prosecution and punishment of extra-marital sex (Smith 1991: 393). Even at the height of KNU authority only the most serious cases would be referred to KNU courts. Now, with very little territory and resources, KNU law has limited reach. When I interviewed a judge from KNU headquarters, he said that only two cases had been decided by that court in 2009.19 The capacity to impose penal sentences has also decreased with losses of 72 Governing refugees

territory. Instead, an offender may be sentenced to work for the KNU or KNLA troops for a ﬁ xed time of months or years, where they may have to assist with cooking, portering and other ‘support’ activities. In particularly serious cases, execution may be ordered. I was told of this occurring in a case where a young girl had been badly beaten and raped but it is also a possible penalty in cases of murder and of spying or informing. Given the limited reach of KNU law, village leaders remain the ﬁ rst point of action for any dispute with a possibility of referral of more serious cases to the KNU. Both systems are susceptible to corruption, as one person from Mu Traw district explained:

I think KNU justice is better than village. Because the village leaders, they are not educated. They give a punishment, if it is their relative, they do not punish. But also, KNU are the same. [laughs] If someone does a mistake and they belong to their organization, they do not punish. The same! It’s not fair.20

Elites in power tend to construct legal systems that protect their own interests (Starr and Collier 1989; Chanock 1985; Moore 1989) and in this the KNU is no different. In 1992, New Yorker journalist Jon Lee Anderson spent time with KNU members and KNLA troops inside Karen State. It was already apparent that the KNU’s territory and authority was slipping and that Karen society was entering into a time of ﬂ ux, when increased contact with new authorities and external forces would change the parameters of the society. Pointing to practices of recruiting prisoners into the KNLA and the execution of adulterers Anderson (1992: 194) concluded that, ‘The KNU’s administration has become less equitable the more desper- ate the situation has become’. A more recent example was given to me by an interviewee, who described an incident where two villagers were arguing over the ownership of a honey tree. On appealing to the local KNU commander to resolve the problem, he promptly declared the tree to be the property of the KNU.21 The role of the KNU in the refugee camps is discussed in Chapter 5. Refugee leaders assured me that the KNU has no jurisdiction over cases between refugees but can only deal with cases involving KNU members Downloaded by [University of Defence] at 21:28 09 May 2016 or people living in KNU controlled areas. One KNU judge explained that ‘Our leaders say, do not disturb the camp. They do not belong to us, they belong to the NGOs [. . .] the refugee camps have KRC law, not KNU law’.22 However, until 2011, the KRG camp rules and regulations clearly stipulated that certain offences (including murder, weapons possession and espionage) should be sent to ‘the mother organization’. In addition, camp leaders still recognize KNU jurisdiction if the offences were committed inside Burma; so, for example, during my ﬁ eldwork a woman from Mae La Oon was summoned to a KNU court inside Burma to give The governance palimpsest 73

evidence against her adulterous husband. He was sentenced to ﬁ ve years in prison.

‘One husband, one wife’: perceptions of harm in policing adultery The previous sections have outlined several frames of governance that have inﬂ uenced Karen refugees. However, while rationales for policing and governance have repeatedly shifted from pre-colonial times through to the patrimonial present, it is less clear how much change has been effected on underlying beliefs and perceptions of crime and harm. As mentioned above, the conservative values of Karen society have been commented upon since the ﬁ rst missionary contact (Marshall (1997 [1922]: 192, 287; Hayami 2003: 112). Sexual activity before marriage is utterly prohibited and the boundaries of forbidden activity are drawn very broadly to include essentially any private contact between a single girl and an unrelated man. In the past, this could lead to banishment from the village or community and even today in many Karen villages a young couple who are known to have been alone together will be expected to marry, subject to the decision of their parents and the village leaders. As one person told me, ‘it’s probably impossible to calculate how many young people have been forced into marriage for being seen alone together.’23 The phrase ‘one husband, one wife’ was repeatedly used to describe Karen beliefs and the prohibition against adultery. This was also described as including bigamy:

Then we have our traditional law. Old law, law that came from our parents’ parents’ parents. For example, that one husband can have only one wife. He cannot take a second wife when his ﬁ rst wife is still alive. This is our tradition, it came from our old parents.24

However, Mason (1868) had described the practice of ‘concubine wives’ (a second wife married while the ﬁ rst wife is still alive, with all spouses living together) as permissible where both women agreed, the village leader granted permission and a ceremony was conducted to appease the spirits. Downloaded by [University of Defence] at 21:28 09 May 2016 When I asked people in the refugee camps whether they had ever heard of or encountered a case where a man had two wives simultaneously, several people described cases from their villages where this had occurred and even claimed that the practice was only stopped after the introduction of KNU law. One Karen leader agreed that in the past village leaders had been able to sanction ‘concubine wives’ but said this had happened only in some areas and only in a few cases. The KNU ended the practice by strong enforcement mechanisms. Between 1981 and 1985, he said, ‘they had to shoot a lot of people’.25 74 Governing refugees

Whether or not many people were in truth executed for the offence, it seems that in relation to bigamy, the language of tradition was manipulated to engi- neer desired social reforms (a familiar pattern in legal anthropology, e.g. Starr and Collier 1989). The converse is true in relation to the policing of adultery, where the language used to justify its prosecution has changed repeatedly but the underlying perception of harm has remained constant. Adultery is the most important policing concern of Karen leaders and the KNU, and is arguably the area which leads to the most unjust and disproportionate penalties. This is not a recent concern: as was explained above, adultery has been taken very seriously by Karen leaders since pre-colonial times. For Karen animists, the prohibition of adultery was inspired by the belief that extra-marital sex was offensive to local spirits, who would punish the community by bad weather, crop failure and poor harvests. Thus, though adultery was a private offence it had collective repercussions and the community had a shared interest in policing it. Wrongdoing may be inferred from the consequences; so, if crops fail it is apparent that the land has become ‘hot’ and the spirits are displeased. Then, the spiritual leader must take steps to determine what has occurred and to cool the land. The village leader and elders would determine the punishment for adultery, and so had some latitude in sentencing, but penalties were often very harsh. Descriptions of the responses to adultery include banishment, death by poisoning, public humiliation (e.g. to be made to eat from a dish on the fl oor like a dog, or to be stripped naked and chased around the village), compensation in the payment of money or food; and ritual (in the sacrifi ce of a chicken or buffalo and performance of prayers) (Marshall 1997 [1922]; Drum Publications 2006; Falla 1991: 260). In the Kawthoolei Law drafted by the KNU, adultery was a capital offence. Anderson (1992: 175) has suggested that the KNU rarely enforced this penalty and only did so during the 1980s as a pragmatic response to poor morale and defection from the front lines following a spate of spousal betrayals: in the words of the KNU Chief Justice of the time, ‘Excessive mis- tresses is death. Just one mistress is usually ten years’. My own interviewees told me that execution for adultery continued well into the 1990s. Even where the death penalty was not imposed, adultery was still a very serious crime which could lead to a lengthy prison sentence, exile from a community Downloaded by [University of Defence] at 21:28 09 May 2016 or expulsion from the KNU/KNLA (though in recent years the punishment was more likely to be conscription into the KNU/KNLA, in an effort to provide resources for the struggling army). A Karen village prison visited by Falla (1991: 260) held twenty male prisoners of whom three were jailed for murder and seventeen for adultery. Of four female prisoners, one had been jailed for pre-marital sex and three jailed for adultery – with the latter group all receiving sentences of between ten and fi fteen years. Of course Karen society is not unique or even unusual in the seriousness with which fi delity in marriage is approached. Sex is a crucial area of regulation The governance palimpsest 75

in almost all societies and adultery is frequently one of the most serious sexual offences (Donnan and Magowan 2010: 7; Berkowitz 2012). A review of prohibitions of adultery in ancient legal orders (Murray 1961) found that common rationales for policing the offence were as a violation of the husband’s property right in his wife, or as a possible corruption of a family bloodline (which again shows a clear link to property and inheritance). From the earliest appearances in legal codes, male and female offenders were treated differently. As Berkowitz (2012: 20) writes, ‘Virtually nothing consumed ancient lawmakers more than female inﬁ delity, and very few crimes were so severely punished’. Interestingly, neither a property rationale nor gendered enforcement is apparent in Karen society. Historically and presently both partners are punished equally (though as there is no detention site for women in the refugee camp the man will often receive a sentence of detention while the woman may be required to spend an equal amount of time working for the KWO). The core value at stake seems to be protection of the family unit rather than inheritance, property or lineage.26 When I asked Karen refugees why adultery should be punished, some mentioned the traditional belief that adultery displeased the spirits and would cause the land to become dangerously ‘hot’, when it must be ‘cooled’ by special prayers and ceremonies. More frequently my respondents mentioned emotional harm and disruption to the family, including to children of the marriage(s). This emotional harm was recognized as real and at least as serious as the harm suffered by a victim of theft or assault. Adultery was also seen as a root cause of other social harms, including domestic violence, divorce and potential alcohol abuse or addiction. Finally, and most compellingly for most of my interviewees, was an argument from custom. It was important to punish adultery because ‘we are Karen’ and because the Karen tradition is to have one spouse for life, ‘one husband, one wife’.27 Respondents occasionally mentioned a Christian rationale for punishment, such as one member of the KRC who suggested that the camp justice workers were proxies for Divine retribution:28

In our Karen culture we cannot give money [as redress for adultery]. They have been married in a church with a Pastor. They have agreed to the system of one husband, one wife. They pray to God and so maybe Downloaded by [University of Defence] at 21:28 09 May 2016 if they make a problem God will punish them. But God cannot punish them directly – so the people in the camps, the camp leader and camp committee should make punishment for them.29

In the refugee camps, adultery is still treated as a serious transgression but in comparison with the executions and twenty-year prison sentences imposed inside Burma the sanctions are relatively lenient. In 2009, camp justice in Mae La Oon dealt with seven cases of adultery and one of sexual ‘misbehaviour’ of ‘hugging and kissing’.30 The standard punishment for cases of adultery was 76 Governing refugees

a sentence of three months detention for the man, two months detention for the woman and a fi ne of between 500 and 1000 baht each. The term of detention could be reduced or avoided by paying an additional fi ne of 500 baht for every month sentenced. In principle, adultery is not actively policed but will only be dealt with if there is a complaint, which is usually made by a wronged spouse. Nevertheless, adultery was one of the most commonly policed cases in the camp. The centrality of adultery to camp policing practices has been challenged by international agencies, including UNCHR and the IRC. The position of these agencies has been that the rules of the refugee camp must conform with Thai law, and as adultery is not a crime in Thai law it should not be treated as a crime in the refugee camps either.31 After some years of negotiation (see Chapter 6), this position has been defi ned in a set of ‘Mediation and Dispute Resolution Guidelines’ for the refugee camps:

Under the Thai law, adultery is not a crime therefore a person should never be held in detention if s/he has committed adultery or has been accused of adultery. Based on shared community values, camp justice shall encourage and, if possible, restore good relations between the spouses involved in an adultery dispute.32

A mediation or group conferencing approach is envisaged, with possible outcomes including a promissory note, counselling, compensation, community service, religious teaching or moral and ethical instruction from the camp leaders, or divorce. This de-criminalization of adultery has not yet been applied in all nine border camps, but its prospect provoked strong reactions from community organizations and refugees as a direct challenge to culture, tradition and the prevailing moral order:33

If we look at Thai law, sometimes it’s not the same as our culture. It seems like they don’t value our culture. If we let this happen, more adultery will happen. Even though we stay in Thailand, every ethnic group has their own culture, values and expectations. For Karen, we practice only one husband, one wife.34 Downloaded by [University of Defence] at 21:28 09 May 2016

In our Karen tradition it is one husband, one wife. So our tradition is becoming lost.35

Notes 1 Two cases illustrate the extreme variation in responses to social problems. One interviewee told me of an incident in her home village near Pa’an (the capital of The governance palimpsest 77

Karen State) where a Burma Army soldier offered a bag of rice in compensation for raping an elderly woman (and this was considered an exceptional concession of responsibility). Another described a case in Karen State where villagers lynched and murdered someone for ‘anti-social and loutish behaviour’. 2 Interview #74, Mae La Oon, 20 February 2010. 3 Interview #81, Mae La, 6 March 2010. 4 Wintin and Brown (2005) note that Rangoon Central Jail (no longer in operation) and Insein Prison (still the most feared prison in Burma) were both constructed according to the ‘Pentonville model’ of wings radiating from a central hub, permitting the categorisation and separation of prisoners. However, some continuity was preserved between precolonial and colonial disciplinary practice, including the use of convict labour and reliance upon convicts as jailers. These elements permitted some undermining of the prison’s hegemonic force as jailors were susceptible to bribery and manipulation, thus allowing prisoners some leverage to improve their conditions of confi nement. 5 Bghai or Bwe is a sub-group of Karen ethnicity, with its own distinct language but recognized as part of the larger Karen ‘family’. 6 These values are all apparent in proverbs emphasizing respect for parents (‘if your children and spouse die, you can fi nd new ones; if your parents die, they are gone forever); social hierarchy (‘an arrow is good to shoot; the young are good for asking to do things’) and peaceful co-existence (‘silent fl atulence spreads widely, whispering creates traps’). Proverbs available in both Sgaw Karen and English translations at www.drumpublications.org (accessed 10 August 2013). 7 Interview #82, Mae La, 6 March 2010. 8 Translated by Zion Lay. The page reference is to the Sgaw Karen original. This description of the ‘Karen’ philosophy of justice was echoed by several of my interviewees. See further Chapter 6. 9 Field notes, 30 January 2010. 10 The original phrase occurs in Geertz ‘Notes on the Balinese Cockfi ght’, summarizing the centrality of the cockfi ght to Balinese culture: ‘Its function, if you want to call it that, is interpretive: it is a Balinese reading of Balinese experience, a story they tell themselves about themselves.’ Thus, the ‘story’ he has in mind is performative rather than narrative. 11 Abortion is certainly perceived as ‘bad’ – but as a social or legal rather than spiritual problem. Karen refugees perceived suicide as a risk in situations which appeared trivial to me, such as rejection of a courtship advance or public embarrassment. It is perhaps relevant that a study comparing twenty-one of the world’s richest societies found suicide to be the only social ill which occurs more Downloaded by [University of Defence] at 21:28 09 May 2016 frequently in egalitarian societies than in non-egalitarian societies (Wilkinson and Pickett 2009). 12 Interview #68, Mae La Oon, 30 January 2011. 13 Witchcraft accusations and their implications for refugee protection are considered in Schnoebelen (2009). 14 Interview #88, 7 March 2010. 15 A good indication of how this is rationalized by Christians is provided in a short video produced by Karen News (2013a) which shows the wrist-tying ceremony taking place in Mae La camp. Though this is a core animist ritual, it is described 78 Governing refugees

by refugees in the video as a cultural event, passed down from Karen ancestors, celebrating the unity of the Karen people. In this way, the event is stripped of its religious meaning and replaced with an important political message. 16 Field notes, 7 March 2010. 17 The fi rst codifi cation of a Kawthoolei Law took place in the mid-1970s, in a document prepared by leading members of the KNU, including Htoo Htoo Lay, a Rangoon-educated lawyer who still holds a senior role in the KNU Executive today. 18 As my research was based entirely on the Thailand side of the border, I was not able to witness any KNU court practice, and this summary is based on interviews with judges and with villagers from KNU areas. 19 Interview #83, 6 March 2010. 20 Interview #6, 14 January 2009. 21 Interview #105, notes from phone interview, 1 September 2010. 22 Interview #82, 5 March 2010. 23 Email, 2 November 2010. This strict approach could be interpreted as protecting women (by ensuring that a man who has sexual contact with her will marry and provide for her and any children they have) but the social taboo of pre-marital contact is also susceptible to manipulation, as one interviewee explained: ‘I think that if you ask [our] students, especially the ones from Brigade 1, they see this a lot. They say – “Oh, when we stay in our house we have to shut our door and be very careful, because if that man likes you he will come and attempt to rape you at night and then your mother will ask you to marry that man straight away.”’ Interview #4, 11 January 2009. 24 Interview #41, 1 May 2009. 25 Field notes, 12 March 2010. I was told of one relatively recent case of a man inside Karen State who had been arrested for having two wives. He pleaded that he had married his second wife before the law was written and so was allowed to keep both wives – but the village was warned that any subsequent cases would be charged with adultery, as a crime carrying the death penalty. In the same village, a man who was married to two sisters was punished by being exiled to another part of Karen State, though not formally prosecuted. Interview #105, 1 September 2010. 26 In other respects too, Karen conventions do not emphasize property rights in spouses or children. For example, lineage patterns are unusually short: there is no ‘family name’ shared between spouses or passed on to children and therefore no way of connecting family members other than personal knowledge of a blood relationship. However, blood relationships are meaningful for inheritance. Downloaded by [University of Defence] at 21:28 09 May 2016 According to Karen traditions of inheritance, if one parent dies, their children will inherit his/her property. Even if the surviving spouse later remarries and has further children, the children of the new marriage have no claim on the property of the fi rst spouse (Drum Publications 2006). 27 Even if a spouse dies, it was necessary at one time to conduct a ceremony before the surviving spouse was permitted to remarry. This is no longer done in every case, but the ritual is still recognized. In 2009, a couple were arrested by the Camp Security staff in Mae La Oon for remarrying after a divorce (the same spouses were involved in both marriages) without requesting administrative permission for the remarriage. They were detained for one month before being The governance palimpsest 79

permitted to remarry. When I returned to Mae La Oon in 2012, this case was referred to by several interviewees as a reason why divorces should not be granted. 28 Churches also play a role in opposing adultery, through a managed process of confession, repentance and forgiveness: ‘If you belong to a church and if you commit adultery, you have to go [. . .] Normally you have to write a letter, and the Pastor will read it out in the church, in the whole church during that service, and you are asking to leave. And in another week or month, that depends on you, if you want to come back and be in the church members, then you ask for acceptance and forgiveness and again they will read out the letter from you, and then you have to come up yourself in front of the church and then they will shake hands with you and welcome you back into the church.’ 29 Interview #93, 12 March 2010. 30 The breakdown of cases is as follows: 21 alcohol; 1 drugs; 7 adultery; 1 sexual misconduct; 1 disobeying camp rules; 3 stealing; 1 stealing teak/wood; 1 stealing rations. In the same year, camp security recorded 89 incidents involving 246 individuals. Incidents not referred to the camp justice team included cases of drunkenness, fi ghting, breach of curfew, disputes with security staff and wearing military uniform. These case records should not be understood as a defi nitive picture of dispute resolution in Mae La Oonn: many more cases are solved at section level than at camp level, and some cases are also resolved by other departments of the camp committee, such as social welfare. 31 Though the UNHCR/LAC approach to policing adultery is consistent with Thai law there is some precedent for recognizing adultery as a crime in refugee camp policing. In Sierra Leone, UNHCR developed camp by-laws in conjunction with the refugee population. UNHCR initially refused the community insistence that witchcraft and adultery be punishable crimes, but conceded when it was apparent that the entire process would be rejected otherwise ‘and the potential rejection by the refugees of the by-laws on these issues would push witchcraft and adultery outside the agreed system and impair UNHCR’s role to monitor and infl uence the proceedings’ (UNHCR 2006c: A4). The primary distinction between this example and the situation in Thailand is that Sierra Leonean law recognized witchcraft and adultery as criminal, and thus there was no contradiction between the camp law and the national law. The primary argument against recognizing adultery as a crime in the camps in Thailand would therefore seem to be one of protecting state sovereignty rather than protecting human rights standards. 32 Guidance Note 10, Mediation and Dispute Resolution Guidelines (copy on fi le with author). 33 Interview #139, Mae La, 24 April 2012. Downloaded by [University of Defence] at 21:28 09 May 2016 34 Interview #153, Umpiem Mai, 26 April 2012. 35 Interview #156, Umpiem Mai, 27 April 2012. Chapter 5 Sovereigns and denizens: camp governance and ‘the refugee’

Refugee camps are often described as a ‘state within a state’ (Suleiman 1999; Griek 2006a: 1; Ramadan 2009: 158; Wilde 1998). However, this suggestion of contained, nested power does not accurately capture governance in the camps in Thailand, where three sets of actors exercise authority in camp management (the Thai authorities, international agencies and refugee leaders)1 and power ebbs and fl ows between them, with considerable autonomy devolved to refugee leaders. In this environment, ‘sovereignty’ is not wholly exercised by the Royal Thai Government but nor is it wholly devolved to the United Nations High Commissioner for Refugees or any other international agency; instead the climate is one of diffuse and negotiated authority between multiple sovereign or quasi-sovereign actors. In this, these camps exemplify a wider trend in global governance, whereby sovereign power – ‘the will to rule’ – is not limited to national governments but is shared among many bodies, including transnational agencies, corporations, communities, ‘big men’ and leaders (Hansen and Stepputat 2005: 4; Appadurai 2003). Just as multiple actors exercise sovereign powers, refugees may defi ne their primary political subjecthood in relation to entities other than a nation-state. Despite lacking the formal status of citizenship, refugees are regulated by multiple authorities. As denizens rather than citizens, they may hold multiple political affi liations simultaneously, ‘depending on the number of domains of governance through which their lives are regulated’ (Shearing and Wood 2003: 408). Accordingly, the state–society relationship that is applicable in a refugee camp – certainly within the camps in Thailand – is not Downloaded by [University of Defence] at 21:28 09 May 2016 the singular sovereign-and-citizen but a pluralistic sovereigns and denizens. The ‘real’ sovereign power in this context is of course the Royal Thai Government. Thai ministries and government departments with a role in refugee policy-making include the Ministry of Interior and its Offi ce for the Coordination of Displaced Persons, the Ministry of Defence, the Ministry of Foreign Affairs, and the National Security Council. Local administration is conducted via a District Chief/Provincial Governor (in Thai language Nai Amphur) and a deputy District Offi cer/Camp Commander (Palat). One Camp Commander is appointed for each camp and this is the Thai offi cial with Sovereigns and denizens 81

primary responsibility for the day-to-day running of the camps, in conjunction with the refugee-led camp committee. The army manages camp security with the assistance of Border Patrol Police outside the camps and paramilitary rangers (Or Sor) stationed inside each camp. In general, Thai authorities are likely to become involved where problems affect Thai citizens, property or policies but internal affairs are largely left to the camp administration. For many years, the preference of the Royal Thai Government was that internal camp management was handled by refugees themselves. To borrow Buur’s (2005: 192) term, the Thai authorities effectively ‘outsourced’ sovereignty for the internal affairs of the refugee camps to refugee leaders. Camp administration in all seven Karen camps is overseen by the KRC, which maintains regular communication with the Royal Thai Government and all relevant Thai authorities, with international agencies and non- governmental organizations, and with individual camp committees. Each camp is managed by its own camp committee which is responsible for camp- level policy and implementation and is the fi rst point of contact for the local Thai authorities, NGOs, surrounding Thai villages and the refugee population. Camp management is further divided into section committees and ‘ten-household’ leaders. Camp committees’ work in basic assistance and service delivery is supplemented by the activities of refugee-led community- based organizations. In most refugee situations, the UNHCR takes on a key role in camp management and the provision of services to the refugee population. In the Thai–Burma border camps, UNHCR was not allowed access until 1998. Instead, key service provision has been undertaken by TBC. TBC’s approach to refugee assistance has been one of promoting self-reliance. This has allowed a great deal of latitude for self-governance and as such has been an important infl uence in the development of these camps. All non-governmental organizations working in the refugee camps participate in a coordination network, the Committee for the Coordination of Services to Displaced Persons in Thailand (CCSDPT), which meets on a bi-monthly basis to coordinate action for refugees.2 The signifi cance of this climate of contested sovereigns is not simply the existence of numerous authorities but that these authorities have different motivations and interests in governing. Language can create ways of seeing Downloaded by [University of Defence] at 21:28 09 May 2016 and acting; not simply reporting knowledge but also producing it and it is now well understood that the language used to describe refugees has practical consequences for both refugee identities and humanitarian interventions (Zetter 1991, 2007; Malkki 1995b: 8–17; Rajaram 2002). This chapter explores the relationship between the labels applied to refugees and the governance regimes that are created to manage them. Globally, two primary discursive threads exist in refugee policy: host governments see refugees as threatening and seek containment, while humanitarian agencies approach refugees as victims in need of protection. These 82 Governing refugees

labels and goals seem oppositional but they share a denial of refugees’ agency and individual capacities. In practice, this denial of agency takes different forms. For example, in Thailand, national laws and regulations impose strict regulation on refugees but at the local level many of the most restrictive aspects are often negotiable. Furthermore, the effective devolution of governing authority to refugees themselves has helped ensure strong and cohesive internal management structures within the camps. In contrast, international agencies’ policies are ostensibly centred on the needs of refugees but may offer limited room for dissent. The two approaches also establish very different expectations of refugees. The demands of the Thai authorities are relatively clear, if authoritarian: ‘If possible, please don’t make a problem. You are not our people’.3 In contrast, a protection approach creates more complex expectations of a refugee population. The nexus between protection and victimization can make it diffi cult for refugees to be heard and acknowledged when they are not victims. When refugee leaders in Thailand have resisted projects proposed by international agencies, there have been two repeated avenues of criticism: that refugee leaders and representatives are not legitimate community representatives (but are acting from self-interest) and that refugee leaders and representatives are not legitimate governance actors (but are proxies for a non-state armed group, the KNLA). This dynamic is analyzed in detail below. The fi nal sections of this chapter examine the potential for non-state armed groups to be partners in refugee protection, suggesting that opposition to such groups may be infl uenced by conceptions of ‘sovereigns and denizens’ rather than by the empirical threat they present; i.e. the continuing belief that sovereign functions can only be executed by a state, and that refugees are subjects of sovereign authority rather than legitimate agents of it.

National policies and the ‘discourse of danger’ Theoretical recognition of de-territorialized sovereignty captures an important dimension of the changing character of state power. However, while the state may not be the only actor carrying out governance functions it continues to play an important, typically pre-eminent, role. Certainly in the management of refugees and asylum claimants there is little to indicate a decline of Downloaded by [University of Defence] at 21:28 09 May 2016 the nation-state. The ‘national order of things’ exerts a powerful force; so powerful that those who are uprooted from it are seen as dirty, polluting, ‘matter out of place’ (Malkki 1992: 34). Haddad (2007: 127) describes the ways in which this underpins discursive construction of refugees as threatening and dangerous:

Within this discourse of danger, several myths are perpetuated that construct the refugee as a potential source of pollution: the refugee causes unemployment and so pollutes the economy; the presence of the refugee Sovereigns and denizens 83

indicates a loss of control over sovereign borders and so pollutes authority; the refugee weakens the national identity of the host society and so pollutes social cohesion; and the refugee brings disease and so pollutes the well-being of citizens.

This discourse of refugee-as-threat has been deployed by many governments to justify aggressive anti-immigration policies aimed at closing porous borders and stemming refugee ﬂ ows. The fear it engenders justiﬁ es remarkable actions, such as the legal excision of national territory in Australia so that ‘boat people’ cannot claim asylum on arrival (Rajaram 2007; Soguk 1999: 211). In Thailand, recognition of refugees as a threat to national security is apparent in the placing of responsibility for refugee policy with the Ministry of Interior, Ministry of Defence and National Security Council. However, it was only after refugee camps in Thailand were attacked by Burma Army and DKBA forces (1995–1998) that they were consolidated and more directly controlled (Bowles 1998). Thus, it might be said that containment of camp- based refugees has been consonant with the extent to which they are perceived to be a threat. More recently, representation of refugees as threats has been apparent in Thai political discourse at the highest levels. Former Thai Premier Thaksin Shinawatra is quoted as saying:

They [migrants from Burma] must stay in their places. They must be controlled. [. . .] They live here and give birth to a lot of children. They shot our students. They bring diseases long gone from our country back to us, including tuberculosis and elephantiasis. They sell drugs and rob and kill our people. (Human Rights Watch 2004: 20)

In 2008, another Thai Prime Minister expressed similar sentiments. Discussing proposals to relocate Burmese Rohingya migrants to a deserted island, Samak Sundaravej said refugee camps were not deterring migrants from entering Thailand:

To stop the inﬂ ux, we have to keep them in a tough place. Those who are Downloaded by [University of Defence] at 21:28 09 May 2016 about to follow will have to know life here will be difﬁ cult in order that they won’t sneak in. (quoted in Nanuam 2008)

Refugees are seen as threatening to national security, to the natural environment and to public health. Regulations for the camps reﬂ ect this, with prohibitions on leaving the camp boundaries, on cutting bamboo, timber or vegetables, on working outside of the camp, on income generation within the camp, on education beyond basic literacy level and restrictions on 84 Governing refugees

ownership of property such as motorcycles.4 These laws are enforced by police outside the camp, by the Thai Palat with responsibility for the camp and by Or Sor stationed inside each camp. Refugees arrested outside the camp risk being deported, in breach of the international law obligation for non- refoulement of refugees.5 Inside camps, the Or Sor role is primarily to contain the refugee population rather than to manage internal camp affairs, and the primary responsibility for ensuring that refugees obey regulations is placed with camp committees. As a consequence, even the most controlling policies of the Royal Thai Government depend for their implementation on the cooperation of the refugees themselves. Camp committees are responsible for ensuring the smooth functioning of the camps, with tasks ranging from guarding and distributing rations, arranging and facilitating training sessions, workshops and other programmes led by international organizations, completing reports and disseminating information to the camp population. A central part of their responsibility is to ensure that Thai laws and policies are obeyed, as a member of the camp administration explained to me: ‘If the camp people live quietly, the Thai authorities are happy.’6 During my fi eldwork, announcements were repeatedly broadcast by loudspeaker across the camp to warn residents that, ‘No camp person is to leave the camp. The Thai authorities say that if a camp person picks even one bamboo shoot he will have to pay a fi ne of 500 baht.’7 The importance of this message was underscored by its broadcast in both Sgaw Karen and Burmese languages, where routine announcements of camp meetings and events were broadcast in Sgaw Karen only. Other announcements warned camp residents when Thai police or other authorities were expected to visit the camp, so that shops might be closed and other signs of prohibited income generation could be hidden, and so refugees could avoid being caught outside the camp boundaries. For the most part refugees tolerate the restrictions and direct protest or confrontation is exceptional. Instead, policies are subverted and circumvented by a range of approaches including bribery, exploitation of loopholes and negotiation with sympathetic offi cials. With negotiation and a little money permission can be granted to do many things, including to leave the camp to work, to cut bamboo, to travel outside the refugee camp (perhaps back to Karen State or to travel within Thailand), to attend training programmes and Downloaded by [University of Defence] at 21:28 09 May 2016 to own offi cially banned goods such as an unlicensed motorcycle or mobile phone.8 Maintaining such fl exibility requires money, language skills, confi - dence and connections, and negotiation with authorities is primarily handled by appointed refugee leaders and representatives. One study found that the majority of refugees who leave the camp ask permission before doing so, usually from their section leader but also at times from the camp committee or directly from the Thai Or Sor (UNHCR/ILO 2007: 21). When permission is given to leave the camp, the refugee is issued with a slip of paper to show to the Or Sor when leaving and re-entering the camp. If a permit holder fails Sovereigns and denizens 85

to return before their pass expires or if a refugee leaves the camp without permission their food rations may be stopped until they return. Refugees also negotiate relationships with citizens and authorities beyond the camp bounds. Again, the KRC and individual camp committees take the lead in this process, though the Thai camp commander may also be involved. The relationship with host communities is often diffi cult in refugee situations, especially where camps are located in areas where the local population is itself poor and under-resourced. Mae La Oon and Mae Ra Ma Luang are in an isolated area several hours’ drive from a sizable town. In some respects villages near to the camps benefi t from the refugee presence – by a ready market for any goods they have to sell, by accessing schools and medical facilities in the camps – but villagers complain that the camps have reduced their agricultural land, that refugees cut bamboo and timber illegally, that they pick vegetables and plants from the forest, form relationships with local women, steal things and cause problems when drunk. Camp committees work to smooth over this relationship, both by providing assistance to villages (typically bags of rice) and in meeting with village leaders and individuals to respond to specifi c disputes.9 Again, refugees have little negotiating power in this relationship. Refugees have even been killed outside camps, and camp leaders can do little to investigate or to ensure redress to relatives (see e.g. Poe Kwa Yay 2012). In one such incident, in 2009, a refugee was shot and killed outside Mae Ra Ma Luang camp. The person responsible said that it was an accident; it was dark and he had thought he was shooting at an animal. The camp committee asked for a mere 2,000 baht (US$60) as compensation for the victim’s family.10 Refugees’ efforts to negotiate and navigate the restrictions under which they live could be seen as ‘peasant resistance’ or ‘weapons of the weak’ (Scott 1985), small acts by comparatively powerless individuals which seek to rebalance the scales of political and economic power a little more favourably. Alternatively, it may be closer to ‘corruption as governance’ (Nelken 2009), an apt term for a context where the pliability of individual offi cials and institutional structures has a dramatic impact on daily life. As one of my interviewees said, ‘We can say that they [Or Sor] understand our situation. Sometimes if we give them a little money it helps them to understand.’11 The motivation by local Thai offi cials is not always and entirely economic. Some Downloaded by [University of Defence] at 21:28 09 May 2016 individuals have genuine empathy and concern for refugees and some Or Sor and other Thai offi cials are themselves ethnic Karen, speak Karen languages and have friends and relatives within the refugee community. It should also be noted that ‘corruption’ is an integral part of many transactions with offi cials in Thailand and in Burma – and indeed, across much of Southeast Asia (e.g. James 2010) – and is neither unique to the refugee camps nor especially aberrant behaviour. The key point is that at the micro-level of individual relationships macro-policies are transformed. Dudley (2010: 36) provides an example of this from Ban Mae Surin in the late 1990s, when 86 Governing refugees

relations between the refugees and local Thai offi cials were generally good but the local authorities were being pressured by their superiors in Bangkok to return refugees to Burma. The response was to stage a forced repatriation, a videotape of which was sent to Bangkok. The tape shows Thai soldiers with rifl es watching refugees laden with packs walk back towards Burma – but the packs were empty and the refugees were not even walking towards Burma but deeper into Thailand. Negotiating fl exibility in the implementation of rules makes camp life more bearable but it is highly unstable and unpredictable. Some camp schools offer post-tenth grade education (a higher level than is offi cially permitted) but they do so with the threat of being closed down at any time. Community organizations such as the KWO and the KYO maintain offi ces outside the refugee camps but do so with the constant fear of these offi ces being raided and staff arrested. Refugees travel between camps and outside of camps, but every journey is fraught with danger and the possibility of arrest. Fundamentally, the relationship between refugees from Burma and the Thai authorities is characterized by an overwhelming imbalance of power:

I myself was arrested by Thai police. And what really hurt me was that you cannot say anything, you have no voice. They give you a paper with lots written on it and say, ‘Is this what happened?’ And you have no choice. You cannot say no. You must say yes to whatever they say. As an illegal person from Burma you are a very small person. Even a regular Thai person is a very small person to the police but an illegal from Burma is nothing.12

As this speaker indicates, the savvy response by a refugee in an offi cial encounter is not to ‘resist’ (at least, not in the way that term is commonly understood) but to conciliate. Refugees have limited integration into Thai society and on very unequal terms. They are extremely vulnerable to exploitation, and if it occurs they have few avenues for redress (Human Rights Watch 2012). Refugee leaders have slightly more power than ordinary refugees, but their freedom is also narrowly circumscribed and they are very careful in how far they attempt to extend it. Securing refugees’ continued presence in Thailand may be seen as a constant process of negotiation and conciliation, a Downloaded by [University of Defence] at 21:28 09 May 2016 process for which the KRC has primary responsibility. On the rare occasions where camp-based refugees have made direct protest about mistreatment, the response by the KRC has been to appease the authorities. For example, in Mae La camp some years ago, students were celebrating at the end of the school year and two disturbances broke out at different ends of the camp. At one site, an Or Sor fi red into the ground and the bullet ricocheted away, grazing a refugee but not causing serious injury. The young people rioted in protest, accumulating from across the camp and setting fi re to cars belonging to the Or Sor. The response of the KRC was to accept responsibility and pay for the Sovereigns and denizens 87

damage, angering members of the ‘community’ inside and outside of the camp, who felt that the KRC should have asserted the refugees’ rights against a reckless show of force by Thai ofﬁ cials.13

Spatial ordering and ownership Refugee ‘threats’ are controlled by rules and regulations, but they are also controlled by the spatial environment of the camp itself. Refugee camps have been aptly described by Agier (2010b: 62) as a form of ‘quarantine’ to separate polluting refugees from a host population (yet, it is worth noting, which can simultaneously be considered a humanitarian space to protect refugees in their victimhood). Inside this quarantined space the design and organization of camps has important consequences for refugees. Hyndman (2000: 95) has analyzed the spatial organization of refugee camps in Kenya. Noting the extent to which they are ordered to suit the needs of administrators rather than residents she asks, ‘whose geography is this?’ A similar design hierarchy is apparent in the Thai–Burma border camps. Most obviously, the camps have been built to constraints imposed by the Royal Thai Government. The construction of permanent structures is prohibited in the ‘temporary shelters’ and so all buildings are made with leaf roofs and bamboo walls. Fire is an ever-present danger, with particularly serious fi res in 2012 (in Umpiem Mai, when more than 1,000 houses were destroyed) and again in Ban Mae Surin in 2013 (when thirty-seven people were killed and more than 2,300 left homeless). Furthermore, bamboo deteriorates rapidly and a house must be rebuilt after two or three years. The diffi culties of refugee life are considerably enhanced by the fact that the camps are in protected forest reserves and cutting bamboo is illegal. Though Mae La Oon is surrounded by forest where bamboo grows freely, refugees must depend on building materials provided by TBC. These materials are only suffi cient to build a small house, requiring refugees to supplement their allocation from other sources. In addition to the material and labour cost, refugees risk arrest both for being outside the camp and for cutting prohibited forest reserves. The combined effect of the prohibition on construction of permanent structures and on cutting bamboo is a cycle of risk and vulnerability simply to maintain basic shelter. Downloaded by [University of Defence] at 21:28 09 May 2016 As with the camps described by Hyndman, camp design refl ects the needs and preferences of international agencies rather than refugees. The administrative centre of Mae La Oon, the Camp Offi ce, is located at the top of a steep hill where it is easily accessible for trucks and cars but is less convenient for the majority of the camp population. When resettlement began, a processing offi ce was built just outside the main entrance of the camp, guaranteeing that the majority of camp residents would have a lengthy walk up a very steep hill to reach it – grandparents and small children among them. No provision was made to assist refugees in reaching the resettlement offi ce.14 88 Governing refugees

Through such design choices the camp space not only contains refugees but establishes a hierarchy in which camp residents’ needs and preferences are subordinate to those of Thai authorities, Thai citizens and international organizations. Hyndman’s question as to ‘whose geography is this?’ can be taken a step further to ask ‘whose camp is this?’ In Thailand, this question was answered deﬁ nitively during resettlement when departing refugees were forbidden to sell their house. Overcrowding combined with the difﬁ culty of securing bamboo means that good houses – large, spacious, relatively new, in a convenient location – are in short supply. Those who were leaving for resettlement would sell their house to someone else in the camp, providing the departing family with a small sum of cash (typically around 5,000 baht/$170) and allowing a newly arrived or newly created family to set up in their own home. This practice was prevented by the Thai authorities, on the basis that refugees could not have any property right in the refugee camps, despite the fact that the labour costs (and often much of the material costs) of house construction were borne by the refugees.

International protection and the refugee as victim Together with the Thai authorities and refugee leaders, the third group of governance actors in these camps are international agencies. Host governments often establish camps precisely because they intend international agencies and donors to provide for a refugee population. International agencies provide assistance in all aspects of camp management and service provision, from water and sanitation to food and shelter to health, education and transportation. This role has been explicitly recognized as quasi-sovereign, as ‘rivalling that of the state and local authorities’ (Colson 2004: 111; Wilde 1998) and a form of ‘humanitarian government’ (Agier 2010a). UNHCR is the global agency mandated to ‘provide international protection’ for refugees (UNHCR 1950: 1), a function that it carries out in more than 100 countries. In many of its ﬁ eld operations, UNHCR has the primary role in camp management, and an involvement in refugees’ lives that is considerably greater than that of the national government. Slaughter and Crisp (2009:8) have described UNHCR as a ‘surrogate state’, complete with its own territory (refugee camps), citizens (refugees), public services Downloaded by [University of Defence] at 21:28 09 May 2016 (education, health care, water, sanitation etc.) and even ideology (community participation, gender equality). In contrast to the ‘discourse of danger’ through which national governments justify hardline immigration policies, international agencies conceive of their role in terms of a discourse of protection (Rajaram 2002: 253; Nyers 2006; Agier 2010b: 63; Malkki 1995b: 9). Policies of protection are more discursively sympathetic than policies of containment but the foundational premise – that refugees are victims who must be protected – has important consequences for refugees. Indeed, ‘humanitarian government’ (Agier 2010b) Sovereigns and denizens 89

can be even more intrusive than the containment pursued by host states, as has been reported from camps in Tanzania, Kenya and Lebanon (Turner 2010; Verdirame and Harrell-Bond 2005; Hyndman 2000; Peteet 2005; Hanafi and Long 2010). Several of these authors single out the ‘biopolitical’ nature of international assistance, where refugee populations are managed and monitored through intrusive head counts, health checks and registration processes (Hyndman 2000; Hanafi and Long 2010; Peteet 2005; Ramadan 2013). Population registration in particular is a frequent battle-ground between international agencies and the refugee population, with NGOs attempting to prevent fraud by increasingly aggressive methods: marking refugees’ hands with indelible ink, herding them into pens to be counted and registered, conducting dawn raids on houses to check the number of residents (Verdirame and Harrell-Bond 2005; Hyndman 2000: 127). Many of the negative dimensions of NGO–refugee relationships have been traced to the conceptualization of refugees as victims, which establishes a relationship of charity rather than rights and therefore a disjuncture between refugees’ rights entitlements and their ability to realize those rights (Harrell-Bond 1986, 1999; Hanafi 2008: 2). UNHCR has explicitly tried to address these problems with policies that advocate ‘refugee self-reliance’, ‘community-based participation’ and a ‘human rights based approach’ (UNHCR 2006b, 2006c, 2008b). These goals have proven diffi cult to actualize, with evaluations of fi eld operations showing that refugee participation continues to be somewhat thin and superfi cial (Meyer 2006; Horn 2010a, 2010b; Rothkegel et al. 2008). It is extremely diffi cult for refugees to demand participation. Where there are tens of thousands of refugees in a single camp, it can be diffi cult for any individual to be heard at all. Feyissa and Horn (2008: 20–1) have described the extreme measures taken by refugees in Kakuma camp – home to more than a hundred thousand people – to get the attention of UNHCR staff:

Refugees in Kakuma know very well that anyone with a security problem in the camp can appeal for protection. It is also known that resettlement is more likely to be offered to those with security problems in the camp. Those without severe and obvious security problems in the camp feel they have no opportunity to make their case and explain their situation Downloaded by [University of Defence] at 21:28 09 May 2016 to UNHCR and therefore they come to believe that the only way they will be heard is if they ‘create’ an emergency situation.

The authors go on to describe ‘staged dramas’, strategic inter-ethnic marriages (where a protection risk might be ascribed), attempted suicides and even the harrowing story of a man who created a protection incident by dousing his wife with kerosene. It seems that in Kakuma, refugee ‘protection’ demanded refugee victimization – and refugees felt they would only be attended to if they provided it. 90 Governing refugees

In these examples, the disempowering effects of international aid are essentially the result of concerned paternalism, a by-product of methods of management and communication rather than an intentional effect. Other dimensions of refugee protection are more explicitly disempowering. UNHCR’s role in international refugee protection is deﬁ ned as ‘entirely non- political; it shall be humanitarian and social’ and the primary tools for its activities are deﬁ ned as international law and the ‘diplomatic’ function of negotiating with national governments (UNHCR 1950: paras 2 and 8). These provisions echo the humanitarian ideal of politically neutral, impartial aid – but they also establish a philosophy of refugee protection which is oriented towards national governments and international law and which ignores or erodes local organizing capacity (Chimni 1998: 354). The statist orientation of the United Nations regime, including the UNHCR, is undeniable: it is devised by states, funded by states and views the world in terms of relationships with states (Hathaway 1991: 114). The construction of refugee situations as a state problem and a state concern represents refugees as subjects of the refugee regime, and generates mistrust of refugees in any other role, including as political actors. When a refugee community acts politically, organizes in a quasi-statelike fashion and exer- cises governing functions, this contravenes the ‘national order of things’, in a blurring of categories that engenders distrust and suspicion (Malkki 1995a). In the operational bias for state over community, the refugee protection regime can be compared with the city and village planners described by Scott (1998) in Seeing Like a State: systems that operate with an imperialist, high- modernist model of statehood, where local knowledge is invisible and irrelevant. As examples of such systems Scott suggests the planned city of Brasilia and Soviet collectivist farms. Less concretely, but displaying many of the same ‘high-modernist’ principles of centralized bureaucracy, lies the international system of refugee protection. The requirement that UNHCR’s operations should be ‘entirely non- political’ has been extended both to the space of refugee camps and to the bodies and minds of refugees. For example, in Lukole refugee camp in Tanzania, political organization within the refugee population was explicitly prohibited. Turner (2005b: 314) has described the power struggle between the contested sovereignties of state, NGOs and refugees, whereby Tanzanian Downloaded by [University of Defence] at 21:28 09 May 2016 authorities had ultimate power over refugees’ presence in the country and their freedom of movement but international agencies managed the daily administration of the camp and ‘refugees themselves seek to [. . .] create pockets of sovereign power outside the reach of either the camp comman- dant’s restrictions or UNHCR’s benevolent control’. This ‘benevolent control’ had internal contradictions, which Turner suggests are more widely encountered in the practice of humanitarian management: relief agencies removed choices from the refugee population yet simultaneously expressed concern at the possibility of dependency, while refugee participation was an ostensible Sovereigns and denizens 91

goal yet autonomous organization was considered threatening politicization. Banning political organizations did not eradicate refugees’ political agency but rather drove it underground:

Because of its secrecy and clandestine nature, the politics is all the more mystical and powerful. Thus, in a paradoxical manner, the attempts by UNHCR and MHA [Tanzanian Ministry of Home Affairs] to void the camp of politics have created a sense of politics that is much more powerful. (2005b: 330)

Similar dynamics are described by Peteet (2005: 52) from Palestinian refugee camps in Lebanon, where there has been a consistent decline in the authority of community structures from the 1970s and 1980s (when the ‘resistance movement wielded considerable power and was fairly autonomous’ and built networks of community support and service delivery in areas including health, education, training and leisure) to the more recent past, where community structures have fractured into political factionalism. Hanaﬁ and Long (2010: 141) attribute this outcome in large part to international organizations which were ‘complicit in the dismantling of Palestinian governance structures, and supportive of their replacement with the bureaucracies of humanitarian and aid agencies’.15 Such analyses suggest that a breakdown of community structures during protracted displacement may not be entirely due to the indirect consequences of aid dependence or societal trauma but could also be the direct outcome of humanitarian policies which undermine existing community structures and supplant those structures with those of international agencies. They also indicate that contrary to the perception of a supportive, benevolent ‘surrogate state’ there is often an environment of mutual distrust between refugee populations and international agencies.

Promoting self-reliance The trope of refugee as victim has been less infl uential in Thailand than in many other contexts. As has been mentioned, the primary agency in camp Downloaded by [University of Defence] at 21:28 09 May 2016 management in Thailand is not UNHCR but TBC. TBC was established in 1984 as a network of concerned donor organizations to respond to what was then perceived to be a short-term emergency situation of a refugee fl ow from Burma to Thailand. Thirty years later it remains the primary source of assistance and relief aid to refugees in this area, providing food rations, shelter materials and capacity building assistance. From the outset of its work with refugees from Burma, TBC promoted refugee empowerment, refl ecting the cohesion and value of existing community governance structures: 92 Governing refugees

We found something on the border that was quite different to other refugee situations we had seen [. . .] Although the Karen had already been struggling for a long time, their communities had remained intact, and their own social and governing structures were still in place. It made sense to support and recognize the Karen Refugee Committee. (TBC 2004:19)

Refugees have therefore been afforded a relatively high degree of latitude in self-governance and the extent of community governance in Thailand is repeatedly highlighted as something perhaps uniquely impressive about these camps:

It seemed the Karen and others were very wary of handing over their future and losing their ability to make decisions for themselves. I was impressed by the level of organization they had in the village-like settings. It was pleasing to see the people genuinely working hard to do the best they could as a community. (Phillip Passmore, TBC 2010b:36)

The key thing was the camps were managed by the refugees so they had the sense of responsibility and ownership that came out of that. (Ray Hasan, Christian Aid, TBC 2010b:111)

Through the various community structures, the refugee population clearly asserts sovereign capacity and a ‘will to rule’. As Kelly (2006: 130) recognizes, people who are excluded from formal judicial structures may adopt ‘processes, concepts and rituals from the formal legal system and apply them in alternative contexts, as they try to bring some semblance of order to their lives’. Within the refugee camps in Thailand, rules and rituals reproduce the classic systems and ceremonies of national government, leveraging these classic attributes of state authority to reinforce their authority – and their nationalist claims. The camp committee operates as a central authority and camp executive, with departments for health, welfare, food distribution, security, education and other functions. Camp committee members are Downloaded by [University of Defence] at 21:28 09 May 2016 elected every three years in a camp-wide process.16 Much as political prisoners may experience prison as a site for the development of political skills and organization (McEvoy 2001; Buntman 2003; Nashif 2008) these refugee camps have been a site for developing capacities of community governance. For some in the camp administration this was seen as preparation for an autonomous or independent Karen State:

The ﬁ rst thing is that we do not have independence yet, so when we are refugees it is a good time for us to learn how to manage our people, and Sovereigns and denizens 93

also to learn from the foreigners and the NGOs, and then one day we will get independence and we will be able to go back and to manage ourselves.17

Others described the camp system itself in terms appropriate to national governance:

There is no speciﬁ c guidance and direction from UNHCR and KRC related to community management. So it depends. We need to plan and we need to look at our administration and management. Because we would like to lead our community, manage our community as the community of a democratic community.18

For more than two decades, the camp administrations managed internal affairs without any formal external support. In 2005, TBC began a camp management programme to provide stipends to camp staff, support community leaders and strengthen camp governance. The average monthly stipend in 2010 was 900–1000 baht, or approximately $1 per day (TBC 2010c). Since 2005, there have also been many changes to the camp administrations as many experienced leaders and community workers departed and an equivalent number of ‘new arrivals’ came to the camps. In some camps, a growing non-Karen population has presented practical challenges (such as the provision of halal rations to Muslim refugees) and political ones, such as whether these structures that have developed so directly from ethnic and ethno-nationalistic conceptions of Karenness are also fl exible enough to accommodate other populations. A recent evaluation of the camp management structures across all nine camps suggests that they are (Turcot 2012). Based on responses from more than 1,000 informants, the basic conclusion of this evaluation was a positive one: for refugees from Burma in Thailand, encouraging self-reliance has worked. Camp management systems were viewed positively but realistically by refugees. There was clear understanding of their limitations but ‘refugees do not call into question the refugee camp management model per se’. There was no evidence of systemic discrimination on ethnic grounds. The electoral process was taken seriously as an avenue to remove bad leaders from offi ce. Desirable leadership characteristics were per- Downloaded by [University of Defence] at 21:28 09 May 2016 ceived to include a level of education but also high standards of personal morality and behaviour and a willingness to sacrifi ce for the community. Section leaders and camp leaders were rated with high levels of satisfaction: 82.6 per cent of participants rated their section leader as ‘good’ or ‘very good’, and 85.8 per cent rated their camp committee as good or very good. Very few participants considered the camp leaders to be ‘bad’: 3 per cent for section leaders, and 1 per cent for camp committees (Turcot 2012: 43). This evaluation reached three primary conclusions: that these systems have great value for the present environment of the camps; that they also 94 Governing refugees

have value for refugees’ future return to Burma; and that these systems have developed and survived because of shared values and mutual trust between the various actors and authorities comprising the camp governance ‘ecosystem’ (i.e. Thai authorities, international agencies and donors and the refugees themselves). Trust was at the foundation of TBC’s relationship with refuges from the outset. In recent years this mutual trust has been undermined, in part by a proliferation of different organizations and management styles, and in part by allegations that the camps have been ‘militarized’ by a relationship with the KNU. These issues are explored in detail below.

De-legitimizing local sovereignty The KRC’s monthly report always includes a message of gratitude to the Royal Thai Government and to all NGOs working in the refugee camps.19 Yet the relationship between international agencies and the refugee population is not as straightforward as this might suggest. In my conversations with ‘ordinary’ refugees (i.e. those without a role in the camp administration), they most frequently talked in terms of generic ‘NGOs’ rather than naming speciﬁ c agencies and often had a rather vague understanding of which organizations provided which services – and, indeed, why they were present at all. One man speculated that it was because British people felt guilty about betraying the Karen during the Second World War, while another suggested that NGO staff were motivated by greed, and were working to keep refugees in camps so that they would continue to draw their high salaries. Most common was a slightly bemused gratitude, whereby refugees were apprecia- tive of the support provided but unclear as to quite why and by whom it was being provided. TBC in particular was rarely discussed by name, which seemed curious given its long-standing and central role in these camps. In contrast, UNHCR was quite often directly mentioned by name and singled out for negative comment. This may have some relation to the fact that TBC carried out its work for many years by actively avoiding ‘branding’ (i.e. promoting its name on vehicles, staff uniforms, t-shirts or other products distributed to refugees), while the United Nations is a globally famous agency. It seems more likely that it is related to the nature of programmes and their implementation. For Downloaded by [University of Defence] at 21:28 09 May 2016 example, while TBC is certainly considered by refugees to be an ‘NGO’, it is primarily involved in direct service delivery and, as noted above, has consistently supported refugee self-governance. UNHCR’s protection mandate for the Thai–Burma border camps gives it a much less central role there than in, for example, Kakuma or Dadaab camps in Kenya. Nevertheless, it has had a leading role in the registration of refugees, in the international resettlement programme, and presently in discussions regarding repatriation. These are controversial and highly intrusive programmes which have had a far-reaching impact on camp life. Sovereigns and denizens 95

A perception of UNHCR as a powerful force was at times a source of fear. Refugees in Mae La Oon talked of ‘UNHCR law’ and believed that UNCHR had a police force that might arrest them for any wrongdoing. There were also more familiar problems of poor communication between international agencies and refugees. Much as the Thai authorities depend on the camp administration to enforce their policies, international agencies depend on camp committees to arrange meetings, conduct translation and facilitate programme implementation. Yet refugees felt that UNHCR and other organizations did not understand or respect the community structures. The head of the social welfare department in one camp said that UNHCR staff had never met with him or acknowledged his work. He laughed, suggesting that they apparently did not realize that there was a social welfare department in the camp. In another camp, the camp leader described his relationship with international agencies as increasingly antagonistic:

The camp committee is like the community that doesn’t have any life. It is in the hands of NGOs. When someone asks us to do something, we must do it [. . .] At the moment, the camp committee needs cooperation. We don’t want to be the leader, or to be under the leader, we just want to be friends [with NGOs]. We don’t want to be the camp committee any more if the UN and NGOs don’t need us. We are tired of being embarrassed in front of our people. We just want to be simple residents.

Others, from both Karen and Karenni community organizations, expressed similar sentiments:

Whenever I talk about UNHCR I feel so tired. For me, we worked with our community about twenty years already. Many times we felt that the NGOs could not recognize our experience or our skills or our knowledge. But then, even though, we can work with them and we can share experience. But they do not respect any more. But I want to say, they don’t know the reality of the needs of the community. We are a community and we work for the community. They come from outside. How can they understand the community?20

Downloaded by [University of Defence] at 21:28 09 May 2016 NGOs need to respect CBOs and leaders. Sometimes they think the community doesn’t know anything but we have been here a long time.21

Poor communication has been particularly problematic in relation to administration of justice programming (see Chapter 6), in gender programming (see Chapter 7), and in the resettlement programme. Camp leaders were concerned that refugees were not being provided with enough information to make a reasoned decision about the costs and beneﬁ ts of resettlement and that some of those who were being resettled were poorly equipped to 96 Governing refugees

cope: elderly people, people with mental illnesses, people with no English language skills and no intention of acquiring them. One social welfare member talked to me about his frustration with NGOs apparent lack of sense of responsibility for those who were leaving and suggested that I write to UNHCR and IOM to request that they provide more, and better, information about life after resettlement. He thought my request may be listened to where his had been ignored: ‘they tell us they want suggestions but they do not do what we suggest’.22 In Mae Ra Ma Luang, the camp leader tried to counter the unambiguously positive representation of resettlement by displaying pictures and documents highlighting the potential diffi culties: unemployment, poverty, the risk of homelessness. This earned him the reputation with the resettlement agencies of being ‘diffi cult’.23 He and other community leaders who spoke out against resettlement were believed to be acting with an agenda to ‘keep refugees in camps’. This is one example of a more widely experienced phenomenon, which is that while international agencies ostensibly promote participation and consultation, if refugee leaders oppose a policy it is claimed that they are simply protecting their own interests. In effect, to reject the message, international organizations seek to reject the messenger. There are two primary ways in which this occurs in Thailand: by claiming that refugee leaders are not legitimate community representatives (but are oppressing the refugee population) and by claiming that they are not legitimate governance actors (but are proxies for a non-state armed group). As noted above, opposition to resettlement was considered an attempt to control the refugee population, blocking their opportunities in order to maintain the camp hierarchy and constituency. Similar claims have been made in relation to justice and dispute resolution, where UNHCR has repeatedly stated that ‘reluctance on the part of refugee leaders who did not support the greater involvement of Thai authorities [was] because they did not want to relinquish their control over the refugees.’ (2006b: A8; Costa 2006: 58). In relation both to resettlement and the administration of justice, the assumption that refugee leaders’ opposition was motivated solely by self-interest allowed agencies to ignore their concerns. However, there are also credible reasons why opposition may refl ect real concern for the safety and well-being of other refugees – and real fl aws in the policies of international organizations. Downloaded by [University of Defence] at 21:28 09 May 2016 Resettlement is diffi cult for many people, and community leaders could legiti- mately fear what the future might hold for those who were ‘applying with their eyes closed’ (Banki and Lang 2007: 22). There were also reasonable concerns about the consequences for the camp administration of a massive reduction in skills and capacity. Such feelings do not necessarily show that community leaders don’t have refugees’ interests at heart but could mean precisely the opposite; resettlement made life harder for those left behind and it didn’t necessarily make life better for those who left. Similarly, in relation to the administration of justice camp leaders have been encouraged for many Sovereigns and denizens 97

years to ‘live quietly’, resolve problems internally, and avoid making demands of the Thai police and authorities; it is not unreasonable that they are concerned about the possible consequences of doing otherwise. It is of course possible that some camp staff over the years have proﬁ ted from their positions, but there are many more rewarding avenues of proﬁ teering available in the Thai–Burma borderlands. For more than twenty years, all camp staff worked on a voluntary basis. Now, they receive an ‘incentive payment’ of approximately $1 per day. In exchange for this, they are expected to be available to any camp resident twenty-four hours a day, seven days a week. And, as the evaluation cited above indicates, they are overwhelmingly supported by camp residents as doing a good and necessary job (Turcot 2012: 43). The second way in which refugees’ voices are rejected is by claiming that they are not legitimate governance actors but are proxies of a non-state armed group – in the Karen camps, the KNU, and in the Karenni camps, the Karenni National Progressive Party. This claim is examined in detail below.

Refugee governance: the role of ‘non-state actors’ TBC’s approach to camp management allowed refugees to retain their governance systems and political identities. This was empowering for refugees, for all the reasons outlined above. More recently it has been por- trayed as naïve, even negligent, for enabling refugee militarization (South 2007a: 63; Callahan 2007: 37). At the outset of this refugee situation, aid was directed to organizations close to the KNU (and, in the Karenni camps, the Karenni National Progressive Party). This was in part due to the Royal Thai Government’s preferred approach to refugee assistance – for international agencies to have a minimal presence – but it was also because these organizations were considered trustworthy representatives of the refugee population. This was not an unreasonable assessment, with precedent in, for example, the role of Polisario in relation to Saharawi refugees in Algeria, the Central Tibetan Administration in India and the Palestinian Liberation Organization in Lebanon (Farah 2009, Frechette 2002). The role of the KNU in Thailand is less clear-cut than in these examples, but it was certainly infl uential. In the fi rst instance, its members were in many respects among the best equipped for management. KNU leaders took key roles in the KRC because they had Downloaded by [University of Defence] at 21:28 09 May 2016 education, administrative experience and pre-existing relationships with Thai offi cials, NGOs and international donors (Thawnghmung 2008: 22). Existing links with Thai intelligence, with transnational church networks and with Western donors also opened up avenues of support and protection for the refugee camps. Precisely because they were the political elite, KNU leaders were the natural leaders in the fi rst response to the refugee situation. More recently, the involvement of KNU members in camp governance has been criticized as turning the camps into an extension of revolutionary politics and making the refugee population hostages to the insurgency. Many 98 Governing refugees

allegations have been made about the role of the KNU in these camps: that it taxed and extorted rations from refugees to feed soldiers inside Burma, recruited soldiers from the camps (including children), diverted humanitarian funds for military activities and otherwise dominated the refugee population (Thawnghmung 2008: 22). In 2007, the UN Special Rapporteur on the Situation of Human Rights in Myanmar reported allegations ‘that the militarization of refugee camps at the border has put the safety of civilians at risk.’ (United Nations Human Rights Council 2007: para. 52). International organizations were blamed for creating this situation. South (2007a: 63) has suggested that the willingness of international organizations and humanitarian agencies to accept the KNU and the KRC as legitimate representatives of the refugee population helped consolidate the power of a KNU elite and even contributed to prolonging the confl ict. Callahan (2007: 37) concludes that international aid agencies ‘struggle to maintain politically neutral assistance programmes in the border refugee camps, most of which are in fact manipulated by Thai and KNU agencies and personnel’. Ensuring ‘the civilian character of refugee camps’ is a global policy concern (UNHCR 2002, 2006e), and these allegations of militarization in Thailand prompted donors to demand more transparency and oversight of camp management. Refugee committees and camp staff responded by re-structuring the organizational hierarchy of the KRC to exclude any relationship with the KNU, separating the camp education structures from the KNU Education Department and re-writing the camp rules. As individual memberships and allegiances remain the same it is questionable whether the intended ‘purge’ was fully achieved. However, I would not question this end-point, but rather whether an alternative approach may have been more effective: to recognize a relationship between the KRC and the KNU and work within this relationship to promote transparent governance. UNHCR is particularly concerned by the role of ‘non-state actors’ in these camps and has repeatedly asserted that the camps are ‘administered by camp committees that are themselves politicized and directly linked with the political/military ethnic factions that exercise social and political control over the camps’ (Costa 2006: 58), that ‘constant care is required to maintain the civilian nature of the camps’ (UNHCR Thailand 2006b: 24) and that ‘as part of [UNHCR’s] protection mandate, there is a continued need for vigilance Downloaded by [University of Defence] at 21:28 09 May 2016 about the potential interference of non-state actors’ (Turcot 2012: xiii, 65). One report explains that ‘refugees have reported many incidences of abuse and even death at the hands of the KNU and KNPP [. . .] Forced labour, including carrying supplies in mine laden border areas, and summary executions are among the incidents reported’ (UNHCR Thailand 2006b: 24). No further information is given to explain the number of reported incidents, where they occurred and whether they have been investigated and verifi ed. In a more recent evaluation UNHCR representatives clarifi ed that they do not suggest the camp administrations are directly supporting KNU activities and Sovereigns and denizens 99

‘acknowledged that there is no evidence that the camps are being used for combatants to be treated and rest before returning to combat’ (Turcot 2012: 65, 71). The primary concern appears to be the potential for abuse of power, forced recruitment or military domination rather than its actual occurrence. Though UNHCR representatives in Thailand have taken a strong stance against engaging with ‘non-state actors’ a recent report indicates that while there is no agency policy on this issue, in practice UNHCR fi eld operations frequently do engage with armed groups. The report ‘employs the point of view that as a matter of principle UNHCR should engage all non-state armed actors where necessary and if possible’, with the understanding that the landscape of non-state armed groups has changed and so has UNHCR’s engagement with them (Keogh and Ruijters 2012: 1). It is diffi cult, if not impossible, to untangle either the precise role of the KNU in camp governance or the extent to which coercive power has been involved. The KNU is associated with the Karen and to some degree with a narrow sub-section of the Karen people: Sgaw-speaking, Christian, nationalist. Refugees in camps are to all extents and purposes a captive constituency and thus are uniquely susceptible to ‘soft’ coercive mechanisms (gossip, shunning, perceived sense of obligation) as well as to the ‘hard’ threat of violence. However, as I have argued previously, its relationship with the wider refugee population is more complicated than one of simple domination (McConnachie 2012). Weapons and military uniforms are banned in the camps (and the latter is actively policed, even extending to seizing camoufl age jackets or t-shirts from young men) and there is no overt evidence of militarization. The allegations that are made in relation to KNU involvement are for the most part linked to the provision of rice to KNU soldiers or the fact that the families of KNU soldiers are refugees living in the camps. More serious allegations – which are indeed serious and include the execution of suspected informers – appear to be isolated and exceptional cases. The camp structures have many limitations but they also have very important assets such as popular recognition and support, understanding of the local context, and proven commitment to the community. It seems probable that the involvement of the KNU in the earliest stages of this refugee situation facilitated many of these assets. Of course, acknowledging that KNU involvement in the refugee camps may have had benefi cial effects does Downloaded by [University of Defence] at 21:28 09 May 2016 not preclude recognizing that there may also have been corruption and abuse of power. Nor does acknowledging the achievements of the camp management systems entail ignoring refugees’ entitlement to representative and responsive political leaders. Ultimately, the most compelling argument is a pragmatic one: the potential for developing transparent governance is greatest when there is transparent engagement with governance actors. As Turner experienced in Lukole camp, banning political organizations will not stop political identities from being meaningful to refugees, but instead will likely drive political organization underground, potentially making those 100 Governing refugees

identities more meaningful (and perhaps even more appealing) while simultaneously removing the possibility of a constructive infl uence from international actors. Rather than rejecting the simple possibility that ‘non-state actors’ can also be legitimate governance actors, a more productive approach would be to openly acknowledge their role and use this as a baseline to negotiate terms and techniques of governance. The KNU has proven its willingness to commit to international standards by signing formal Deeds of Commitment on the protection of children from the effects of armed confl ict and on the protection of women and girls during armed confl ict (Geneva Call 2013, 2011, 2010). Several years prior to signing this international commitment, in 2008 the KNU made a commitment not to recruit child soldiers from the refugee camps. This shows openness to dialogue about international standards and a willingness to formally commit to them, which are precisely the characteristics that would assist in the development of ‘good governance’. In the Thai–Burma border camps, this may be a moot point: the KNU role is now at most a peripheral one and it is questionable whether the camps themselves will exist for much longer. As a point of wider relevance however, the question of whether ‘non-state actors’ can be legitimate governance actors is an important one. As this chapter has shown, the refugee camps are sites of multiple sovereignties and of multiple denizenships. For the international refugee regime, a ‘legitimate sovereign actor’ may be synonymous with a national government, but for refugees from Karen State - and across Burma - the state (whether ‘home’ or ‘host’) is more likely to be viewed as a predatory rather than protective force. In practical terms, refugee camps are sites of de-territorialized sovereign authority. Refugee protection programming will be strengthened by rejecting the familiar approach of ‘seeing like a state’ in favour of understanding where power is exercised and by whom. The extent to which non-state actors should be able to exercise governance functions is also relevant to the administration of justice, which is the subject of the next chapter.

Notes 1 Each of these groups constitutes a broad category with multiple members. Downloaded by [University of Defence] at 21:28 09 May 2016 Certainly not every member of the Thai police, military, Border Patrol and civil service thinks or acts alike, and nor does every staff member of every international organization, but these groupings do share (at least some) policy goals and practices. 2 Currently, eighteen NGOs and agencies are members of CCSDPT, involved in supporting health care, training medics and equipping hospitals, supporting schools and education supplies, ensuring water and sanitation, children’s nurseries, malaria control and treatment: Adventist Development and Relief Agency; ARC International; Catholic Ofﬁ ce for Emergency Relief and Refugees; DARE Network; Finnish Refugee Council; Handicap International; International Sovereigns and denizens 101

Rescue Committee; Jesuit Refugee Service (Asia Pacifi c); Malteser International; Right to Play; Solidarité; Shanti Volunteer Association; Taipei Overseas Peace Service; The Border Consortium; World Education; Women’s Education for Advancement and Empowerment; the United Nations High Commissioner for Refugees and ZOA Refugee Care Netherlands. See www.ccsdpt.org (accessed 10 August 2013). 3 Interview #170, 5 May 2012. 4 The Royal Thai Government has issued Guidelines for Government and NGO Offi cials including Relevant Agencies on Providing Services to Displaced Persons from Burma in the Temporary Shelters for Displaced Persons in Kanchanaburi, Tak, Mae Hong Son and Ratchaburi (Ministry of Interior, RTG, January 2011) (see Turcot 2012). Other regulations are imposed by Thai immigration laws including Immigration Act 1950 (amended 1979, 1992); Alien Registration Acts 1950, 1952, 1954; Nationality Act 1965 (amended 1992); Alien Employment Act 1978; Law on the Categories of Work Not Permitted for Aliens 1979; Laws on Form of Employment and Activities in Which Aliens Can Engage (1979 and 1985) (Baek and Subramanium 2008). 5 A study conducted by the IRC in Karenni Sites 1 and 2 and Mae La camp found that 27.1% of respondents had been arrested by Thai authorities and 16% had been forcibly returned to Burma at least once in the previous fi ve years. The fi gures were substantially higher in Mae La than in the other two camps, with one in four respondents having experienced refoulement in the previous fi ve years. This likely refl ects the fact that Mae La is close to the town of Mae Sot and many camp-based refugees leave the camp illegally to work (IRC 2007: 40). 6 Interview #17, Mae La Oon, 15 Feb 2009. 7 Field notes, January 2010. 8 The Royal Thai Government Guidelines for the Temporary Shelters permit refugees to leave the camps for reasons of occupation, education or resettlement. 9 Camp committees also mediate and resolve problems between refugees, processing administrative matters of divorce or adoption, and dealing with acts of physical or sexual violence or any other disputes that may arise. See further Chapters 6 and 7. 10 Interview #57, NGO staff member, Mae Sariang, 1 June 2009 11 Field notes, 20 February 2009. 12 Interview #1, Mae Sariang, December 2008. 13 Field notes, March 2010. More serious still was an incident in December 2008, in Karenni camp Ban Nai Soi, when a young male refugee was shot and killed. The shooting occurred when a crowd of refugees had gathered at the Or Sor station Downloaded by [University of Defence] at 21:28 09 May 2016 to protest mistreatment of refugees in detention. Shots were fi red, supposedly into the air to force refugees to disperse, but hitting a seventeen-year old boy and killing him. Refugees rioted, burning the Or Sor station and several vehicles. The Thai Palat responsible for the camp was transferred to a different post, but the camp committee was ordered to pay for the damage caused. 14 Not all demands are acceded to. When refugees were relocated to the site that is currently Mae La Oon, the Thai authorities instructed that the camp must be built in a strict plan, with all houses built to the same dimensions and laid out in rows. Camp leaders baulked at this regimentation and allowed refugees to build their houses as they pleased. It was impossible for the Thai authorities to enforce 102 Governing refugees

their plans on several thousand house-builders simultaneously and the most important priority was achieving the camp relocation, so – on this occasion – the refugees succeeded in their resistance. 15 A study by Hyndman (1997: 19) suggests that this undermining of local capacity may be intentional. UNHCR staff in Kenya considered a proposed refugee self- management project to be ‘dangerous’ as it risked ‘reviving traditional power’ and ‘elders’ enclaves of autocratic authority’. The UNHCR Head of Sub-Offi ce felt that UNHCR was the pre-eminent governing authority for Dadaab and should remain so as, ‘We have succeeded in breaking up the traditional structures of power (in the camps)’. 16 The electoral process is rather complicated. In the 2010 camp elections in Mae La Oon (which I witnessed), each administrative section in the camp with the exception of Section 13, the ‘Burmese’ section, selected fi ve representatives to vote, providing a pool of seventy voters. An additional three people per hundred were chosen by the section leaders. A list of nominees for the camp committee was circulated in advance and on the appointed day the selected pool voted for fi fteen camp committee members. Votes were cast and counted publicly: each voter placed their voting sheet into a sealed cardboard ballot box. When all votes had been cast, the box was opened and the votes were counted and tallied in front of all present. There was no opportunity for fraud in this part of the process. The fi nal count resulted in a tie between the fi fteenth and sixteenth candidates. All voters had to recast a vote to choose which candidate would be selected. The new camp committee then voted internally to allocate positions, beginning with the ‘executive committee’ of camp leader and camp secretaries. With less than 10% of the adult camp population participating in the elections the system is far from full suffrage. However, in the 2010 elections, there was some turnover in camp positions, indicating that the voting system could produce change. It was also clear that there was genuine support for the incumbent camp leader, who was re-elected with 247 of 285 votes. 17 Interview # 21, member of camp committee, Mae La Oon, 26 February 2009. 18 Interview #17, member of camp committee, Mae La Oon, 15 February 2009. 19 E.g. from August 2009: ‘We shall always be grateful to the Thai people and Government for their sympathy and understanding in allowing us safe refuge. We are also indebted to the NGOs for their Supports and Services’ (KRC 2009: 6). 20 Interview #169, Notes from phone interview, 4 May 2012. 21 Interview #167, 3 May 2012. 22 Field notes, 11 May 2009. Downloaded by [University of Defence] at 21:28 09 May 2016 23 Interview #59, NGO staff member, Mae Sariang, 1 June 2009. Chapter 6 The struggle for ownership of justice

We lawyers just cannot help being Darwinian. We simply cannot shake off our assumption that some legal cultures are more developed than others. We prefer written law to oral law; we are happier with professional judges than with people’s rough justice; and – need I say? – we just love cultures that have their own lawyers. (Huxley 2001)

Dealing with crime and disputes is a key challenge of protracted encampment that has been almost entirely ignored in international policies for camp management. This lack of attention to justice and security has been identiﬁ ed as a contributing factor in the high levels of violence and insecurity encountered in some refugee settings; including, most recently, for refugees from Syria living in camps in Jordan (Refugees International 2012; Verdirame and Harrell-Bond 2005; Crisp 2000: 610–2; Griek 2006a). In Thailand, as the previous chapter explained, trusted community-led systems have taken the primary role in maintaining stability and security. In these camps serious violent crime is relatively rare but many other crimes and social problems occur, from theft, assault and domestic violence, to gambling, drug and alcohol abuse, to the more culturally speciﬁ c concerns of adultery and pre-marital sex. As with camp management generally, the Royal Thai Government has historically done little to police internal camp affairs and the vast majority of justice issues that occur in the refugee camps are resolved by

Downloaded by [University of Defence] at 21:28 09 May 2016 refugees (as is typically the case in refugee camp situations; see Costa 2006). Refugee leaders who have sought to engage the Thai justice system describe being rebuffed:

They say, ‘This is your community, this is your case.’1

They told me, ‘Please educate your people. You, yourself must take care of your people, not to be creating so many problems in our soil. Otherwise we also can get headache. And then our parliament will complain because the refugees behave like this.’2 104 Governing refugees

Sometimes we ask, and they say, ‘Do it yourself. This is your people, not our people.’3

The devolution of justice functions to refugees was done partly on ideological grounds – ‘these are not our people’ – but also with some practical justifi cation. Most of the refugee camps are remote and not easily accessible for investigations or policing activity, few refugees speak Thai language, local Thai police are themselves over-stretched and under-resourced, and refugees themselves largely prefer to resolve cases internally. The state absenting itself in this fashion is not unique to refugee situations. One report has estimated that informal justice systems deal with eighty–ninety per cent of all disputes in Malawi and sixty–seventy per cent in Bangladesh, are applicable to ninety per cent of land transactions in Mozambique and Ghana and are relied upon by eighty-fi ve per cent of the population in Sierra Leone (Wojkowska 2006: 12). Contrary to the Weberian notion of a national executive with sole authority, in many parts of the world legal systems beyond the state are the legal system. The infi nite variety in the type of systems used means there is limited value in thinking only in terms of ‘state’ and ‘non-state’ justice. There are many possible ways in which justice systems beyond the state could be differentiated, including by the nature of process (restorative, adversarial), the nature of settlement (ritual, compensatory, violent), or by the underpinning belief systems (religion, Marxism, supernaturalism). In terms of the primary focus of this book – the relationship between community, state and supra-state orders – models of informal justice range from completely autonomous community-based structures to state-controlled processes with minimal community autonomy. Discussion of justice systems beyond the state rarely attempts to disaggregate these different models, instead relying on the simple binaries of state/non-state, formal/informal. As Norrie (1999) recognizes, this ‘antinomialism’ radically simplifi es the universe of justice systems and re- inforces hierarchical thinking, whereby national programmes are considered implicitly superior and non-state programmes are evaluated in terms of what they lack rather than what they contribute. On the contrary, non-state justice systems are not merely an inferior version of state courts but can be an indispensable mechanism for resolving disputes that otherwise might not be addressed. Furthermore, the value of local dispute Downloaded by [University of Defence] at 21:28 09 May 2016 resolution is not restricted to an immediate case but includes the benefi ts gained from community participation in establishing shared values through rule defi nition and enforcement. As was discussed in Chapter 4, a capacity to establish and reinforce moral values, taking into account all relevant factors, is something that is deeply embedded with Karen village dispute resolution practices. This value is lost – or ‘stolen’ (Christie 1977) – when legal processes become professionalized and the knowledge that is prized is not the socio- cultural-historical knowledge of community elders and leaders but a technical understanding of law. The struggle for ownership of justice 105

The systems used by refugees in Thailand are described in some detail below. These camps are climates of legal pluralism, where refugee-led dispute resolution systems are also affected by other sources of law, including Thai law and international human rights law. However, the term ‘legal pluralism’ suggests equity between normative systems that does not exist in practice. Law is embedded with assumptions, power and class relations, and where asymmetric power relations exist a struggle for ideological dominance often ensues (Starr and Collier 1989: 7; Burman and Harrell-Bond 1979). Historically this has tended to play out with a nation-state incorporating alternative justice structures into the national regime or reshaping the local justice practice into a more ‘state-like’ form. This is generally conceived at policy level as an act of generosity, of law as ‘the gift we gave them’, but it has real and practical consequences (Fitzpatrick 1989). A formal relationship with the state transforms local justice practice, making it indistinguishable from ofﬁ cial dispute resolution structures by removing political power from local actors or diverting disputes into the state system (Cain 1985; Matthews 1988). Though total cooption is rare, once a community justice process is rationalized in this way it is profoundly altered. As McEvoy and Eriksson (2008: 188) recognize:

[T]he state’s hegemonic power and ‘will to dominate’ can ‘swallow up’ community-based programmes, professionalize them and reconﬁ gure them into the image of the state to such an extent that they ultimately lose their speciﬁ c community focus and indeed legitimacy.

Attempts to incorporate informal legal processes into a rational bureaucratic ideal type were a familiar theme during the colonial era. Today, rather than a colonial power or national government coopting local justice systems, legal change is more likely to be instigated by international organizations in accordance with international human rights law and neo-liberal principles of development (Comaroff and Comaroff 2006: 25). For some years, international rule of law reform was characterized by a ‘rule of law orthodoxy’ that was implemented by foreign experts, focused on state- level relationships and was concerned with the creation or reconstruction of state-level legal institutions (Golub 2003). More recently, there has been a Downloaded by [University of Defence] at 21:28 09 May 2016 shift towards ‘legal empowerment’ and engagement with non-state and customary justice systems (Pimentel 2010a, 2010b; Harper 2011; Ubink and McInerney 2011). Recognition of the role and value of non-state justice is important and long overdue, but engaging with customary justice systems presents practical and operational challenges that should not be under- estimated (Faundez 2010). The lack of formal institutions and codiﬁ cation can make it difﬁ cult to untangle their operational logic without deep local knowledge and understanding, while their sui generis nature prevents the easy transfer of project approaches from one environment to another. If misjudged, 106 Governing refugees

programmes intended to strengthen non-state justice systems risk instead undermining them (ibid). When the option of community- based dispute resolution is removed, it is not at all clear that its constituency take their problems to the formal legal system. Regarding personal injury cases in Thailand, Engel and Engel (2010) found that the declining use of traditional dispute mechanisms (in this case, a form of ‘spirit-law’ that was very similar to Karen animism) did not lead to more use of national courts and the pursuit of formal justice but rather to more avoidance: more frequently than in the past, people who experienced litigable personal injuries did nothing at all. This chapter explores such dynamics for the refugee camps in Thailand. It opens with an overview of the structures used to resolve cases by refugees, before going on to examine how these systems are viewed by camp residents and by international organizations. Programmes operated by UNHCR and IRC have sought to enhance refugees’ access to Thai justice by supporting referral of criminal cases to Thai courts, codifying camp rules and procedures and creating Legal Assistance Centres to advise refugees of their legal options. This work aimed to strengthen refugees’ access to Thai courts and to build capacity at camp level but in practice the latter goal has proven difﬁ cult to achieve. The ﬁ nal sections examine the consequences of this for the functioning of refugee-led dispute resolution.

Camp justice For many years no coordinated attempt was made to establish rules or justice procedures for the refugee camps in Thailand. In the late 1990s, a set of camp rules and regulations was drafted by members of the Karen Elders Advisory Board and KRC, to apply to all seven Karen refugee camps. These rules were revised in 2011 and remain the only codiﬁ cation of ‘law’ in the refugee camps.4 Their importance should not be overstated: few camp residents I talked with were aware of the existence of these rules and even fewer had any familiarity with their content. Nevertheless, their content gives a fairly good indication of the type of problems which refugee leaders anticipate dealing with in the camps, including prohibitions on criminal activities such as murder, physical violence, theft and sexual assault, rules pertaining to admin- Downloaded by [University of Defence] at 21:28 09 May 2016 istration (e.g. any overnight guests must be reported; every house must display a household number); hygiene (e.g. every person must obey rules of water supply and sanitation); public participation (e.g. camp residents must attend meetings if they are informed of them); and morality (pornography is prohibited, as is engaging in sex work, ‘illegal sexual engagement’ and adultery). Gambling, drugs and the production or sale of alcohol are prohibited, and there is a curfew on all non-essential movement around the camp between 9pm and 5am. Punishments include education, warning, cash ﬁ ne, public service, detention and ‘turn over to Thai authority’. The struggle for ownership of justice 107

Beyond the KRC Rules, no consistent model of dispute resolution applies in all seven Karen refugee camps, nor do the camps have identical governance climates. I conducted fi eldwork interviews with refugee leaders and community organizations from fi ve camps but the majority of my camp- based research and all of my case observations were conducted in Mae La Oon. Mae La Oon and its neighbor Mae Ra Ma Luang are generally considered to be among the most stable camps on the border, with the fewest serious crimes and the most effective justice systems. They have also experienced less external intervention on justice issues than some of the other ‘Karen’ camps, as will be discussed below. In Mae La, a single judge resolves cases and disputes, in conjunction with other camp leaders as he chooses. In Mae Ra Ma Luang, the camp leader appoints a panel of camp committee members to decide each case as it arises, usually including fi ve to seven members of the camp committee, elders or other infl uential camp residents. Mae La Oon has a more institutionalized system, with a dedicated justice team comprised of three judges. The judges are chosen by the camp leader and camp committee, and are supported by Security members who patrol the camp, arrest and detain offenders and conduct investigations (though the term ‘investigations’ is used loosely here, as transcripts of interviews by Security members suggest that only a stock set of very simple questions is asked: (1) Did you do it? (2) How many times? (3) Do you agree to accept punishment?) All camps do share a structural similarity of hierarchical governance. Veroff (2010) has described dispute resolution in Meheba refugee settlement in Zambia as a diffuse environment where refugees select from a range of possible authorities to decide their case. In the Karen refugee camps in Thailand, the maintenance of order is secured in a much more coherent and cohesive fashion. Most disputes and problems that occur in the camp are addressed at section level. If this is not possible, the case will be referred upwards to the camp committee or the Camp Justice team. The referral process is hierarchical, but it is also networked as a variety of organizations and individuals may become involved in a case, including social welfare, KWO, KYO or refugee staff of internationally-funded initiatives such as the Child Protection Committee or the Sexual and Gender-Based Violence Committee. Thus, ‘judges’ do not take decisions unilaterally but in consultation with Downloaded by [University of Defence] at 21:28 09 May 2016 several other individuals and organizations. The types of cases that arise can include theft, assault, disorder, fi ghting and other aspects of personal security and criminal law. Relationship disputes are a frequent occurrence, including cases of domestic violence but also situations of abandonment (such as where a man applies for resettlement claiming that he is single but in reality has a wife and children or a pregnant girlfriend) and a myriad of other potential problems between couples and families. Problems can also arise from the challenges of living in confi ned spaces without basic civic regulation. The houses in the refugee camps are 108 Governing refugees

packed very tightly together and a fractious baby or noisy teenagers can disturb literally hundreds of people. Furthermore, disputes are not conﬁ ned within the camp boundaries, but may involve Thai civilians (e.g. labour disputes, alleged assault or rape, or other problems relating to a relationship between a refugee and a Thai citizen), or non-refugees from Burma who are visiting the camp (perhaps a theft, or a broken promise of marriage, or the use of counterfeit money), or a former refugee who has been resettled to another country (who promised to repay a debt, or to sell goods on behalf of a refugee). Such cases – where one of the parties is not a refugee – are particularly difﬁ cult for refugee leaders to resolve.

Process: the pursuit of order The camp justice system shares many characteristics with mediation in ‘nonindustrial’ societies: there is very limited legal codifi cation, the process is led by community leaders of general authority rather than highly educated specialists, and decision-making is guided by values of consensus and harmony-seeking (Merry 1982; Abel 1982). It also shares many characteristics with the dispute resolution practices used in Southeast Burma, as outlined in Chapter 4, including a focus on the maintenance of order (understood as an absence of confl ict and acceptance of authority); reliance on a narrow set of shared values (reinforced by moral messages circulated publicly and privately; and a premium placed on the community good rather than individual demands. In some societies, acts that transgress the social order are tolerated or even respected as defi ance (Benda-Beckmann and Pirie 2007: 6). In the Kakuma and Dadaab camps in Kenya, Crisp suggests that the ‘social organization and culture’ of the refugee communities enhances a sense of impunity whereby ‘camp residents have a tendency to feel that they are above the law’ (2000: 619). This is not the case among refugees in Mae La Oon (or the other border camps) where deviation from the camp rules was generally met with intense disapproval and the importance of obeying rules was universally recognized. Moral guidance is central to dispute resolution processes within the camps (as was discussed in Chapter 4). In cases I attended, judges offered moral instruction in a variety of terms, telling an adulterous couple that Downloaded by [University of Defence] at 21:28 09 May 2016 they should respect their marriage vows made before God, a thief that the traditional way set down by the Karen ancestors is to be honest in all things, and a young person who had been fi ghting that he should obey his parents and teachers and live quietly in the camp. These ‘soft’ forms of social control provide clear narratives of morality and appropriate behaviour that are at least as important as the institutionalized aspects of dispute resolution. While the form of authority appealed to varied (from God to Karen ancestors to parents and teachers) it was clear that the camp rules were not the relevant reference. Rather, it was a shared ‘common sense’ of morality and appropriate The struggle for ownership of justice 109

behaviour, reﬂ ecting the standards of honesty, truth, duty and responsibility which were perceived as authentically ‘Karen’ (Cheesman 2002: 214). A desire for peace and harmony was presented as a quintessentially Karen trait, encapsulated in a phrase that I was repeatedly told explained the ‘Karen’ philosophy of dispute resolution: ‘if it is a big problem we make it small, and if it is a small problem we make it disappear’. This is a classic ‘harmony ideology’ and as such can mean that ignoring or minimizing conﬂ ict (‘lumping it’) is integral to the process of dispute resolution, as the primacy placed on social order subordinates individual preferences to the perceived communal good (Nader 1990). Ultimately, however, it is a different philosophy of justice and dispute resolution, one to which the adversarial formalism of the common law tradition is equally alien and undesirable and is likely to be seen as escalating rather than resolving disputes. Riles (2006: 61) has suggested that, ‘To be a professional lawyer [. . .] is to be agnostic about the ends and to be far more interested in the means’. This approach is antithetical to the Karen philosophy of dispute resolution, which is entirely concerned with the end goal of making a problem ‘disappear’ and rather less concerned with the means of achieving that.

Sanctions The process of dispute resolution and decision-making is participatory and consensus-based. However, its primary goal is not the healing or restoration of relationships between individuals (as in Native American dispute resolution practice [Zion 1999; Pimentel 2010b]) but the maintenance of social order (as has been described in Zincanteco courts [Speed and Collier 2000: 895] and in village governance in Ladakh [Pirie 2006]). In a further distinction from restorative justice systems, the outcome is not a reparation settlement or reconciliation ritual but a punitive disposal such as detention or a fi ne. Serious punishments are diffi cult to impose in the camps as few people have money to pay a fi ne and there are no secure detention facilities. In most camps, the ‘jail’ is a simple bamboo hut and wooden leg-stocks are used to shackle and restrain (and, as these stocks are painful and uncomfortable, they also add an additional punitive measure). The offender is not shackled publicly but detained in a bamboo hut.5 Family members are allowed to visit but otherwise Downloaded by [University of Defence] at 21:28 09 May 2016 an offender is kept with other detainees and guards. Some offenders are permitted to move around during the day provided they remain close to the security offi ce. After the detention period is complete the offender signs a note pledging to obey the rules in the future and is released. Proceedings throughout are low-key with little evidence of the theatre or drama so often considered an integral component of legal sanction, such as the ‘circus’ of community dispute resolution described by Christie (1977) in Tanzania, the drama of trial by ordeal described by Tonkin (2000) as occurring throughout Africa, or the wigs, gowns and other archaisms of British court room practice. 110 Governing refugees

Violence is not central to the justice process, in the sense that it is never offi cially imposed as a punishment, but I was told that offenders were often beaten by Security members in arrest and detention. The most serious punishment is exclusion or banishment from the camp, which may occur if agreement cannot be reached between disputing parties or if a person refuses to accept and abide by the authority of the camp leaders. As noted in Chapter 4, the threat of banishment or expulsion is a powerful sanction in small-scale cooperative societies and can operate as either a punishment or a means of confl ict avoidance. In the past, short-term banishment was used to punish young men who repeatedly broke the camp rules, perhaps by fi ghting, stealing or other such petty disturbances. If he (as it invariably was) refused to abide by the camp rules, he might be sent to join the KNU/KNLA for a period of some months and could return to the refugee camp once he proved willingness to obey the camp leaders. This sanction would be carried out in camps or supporting roles away from active combat. Nevertheless, such referrals have become rare – and perhaps stopped altogether – after vehement criticism by international agencies and actors that this practice constitutes forced recruitment of child soldiers.6 Despite the evidently punitive dimension of such processes as detention, banishment and referral to the KNU, these disposals were described by justice workers as rehabilitation and viewed as part of a process intended to transform offenders into good citizens who can contribute to their community. Even sending an offender back to Burma, some people said, was an opportunity for the person to learn a ‘good way’ of discipline, hard work and subordination. Describing such a case, one young person said:

[L]ater the camp leader and leaders here sent him to the KNU so that he will stop the ﬁ ghting here, he will be a good person. But when he [goes] back to the KNU, the leaders are not allowed to punish him, only to teach him so that he will know, ‘Oh I did a bad thing in the past, so right now I know I have to do the good thing’. But they give him enough food. They teach him.7

Though it seems improbable that sending young men to a conﬂ ict zone can be seen as an acceptable punishment, this rationale is not dissimilar to Downloaded by [University of Defence] at 21:28 09 May 2016 perceptions of military service in Western societies, which represent the army as a place for wild young men to learn discipline and self-control and to become patriotic citizens (Bouffard and Laub 2004). There is no evidence that the process works, as a teacher in Mae La Oon acknowledged:

[I]f young people go back to Karen State, in that area, they will know the situation in Karen State, how difﬁ cult it is to survive, and they will compare their life in the refugee camp and in Karen State and think about how to make their life better. Some people when they come back The struggle for ownership of justice 111

to the refugee camp, their life is getting better, but not all. It is difﬁ cult to treat, I think.8

In the emphasis on reintegration and rehabilitation as opposed to punishment, there is a suggestion of ‘reintegrative shaming’, sanctions imposed in a way that reintegrates offenders into the community rather than stigmatising them (Braithwaite 1989; Braithwaite and Mugford 1994). For example, an offender’s family members are always included in the process of dispute resolution and the rhetoric is of reform and improvement rather than humiliation. However, there is a point at which efforts to rehabilitate are sharply cut off and the offender may be sent out of the family by his parents or sent out of the camp by leaders. Reintegration could perhaps best be described as ‘all or nothing’. After a term of detention an offender apparently re-enters the community with no lasting stigma, providing s/he agrees to accept the authority of the camp leaders. However, if a person continues to re-offend or if there is a continuing problem, s/he may be utterly rejected. This limit to tolerance again indicates that the primary objective of the justice system is not to heal relationships between offender and victim but to heal the breach in the social order and ensure that the authority of the camp leadership is respected.

Legitimacy Wilson (2000: 87) has suggested that the extent of plurality within a legal system depends on one’s position. To the South African Justice Minister the national criminal justice landscape may seem to be populated by hundreds of co-existing justice processes, but to a petty criminal the justice system will look ‘relatively unifi ed and integrated’. Similarly, while I saw the camp as a jurisdictionally complex site, refugees in Mae La Oon saw a single hierarchy: cases should be reported ‘step by step’, fi rst to the section leader, then to camp-level leaders and only if the case could not be solved in the camp (e.g. if the crime was a very serious one or if the disputants could not agree) should it be referred outside to the UNHCR or the Thai authorities. This structure was understood by everyone within the camp and recognized as an appropriate pathway, ensuring that those best able to respond to a problem Downloaded by [University of Defence] at 21:28 09 May 2016 would be alerted immediately.9 Support for the camp governance structures was not uncritical. Practical limitations of the system were fully understood, including the diffi culty in responding appropriately to very serious crimes and the inability to assert jurisdiction over non-refugees (and thus to protect refugees against exploitation or violence that occurs outside of camps). Women’s organizations have campaigned for years to try to ensure that cases of domestic violence and sexual violence are recognized and responded to appropriately (see further Chapter 7). There was also criticism of individual decisions and of 112 Governing refugees

decision-makers, for example, that refugees with money could bribe their way out of punishment and that leaders or prominent people within the camp were given preferential treatment (so ‘ordinary’ refugees were punished for making and selling alcohol while section and camp committee members were not). Refugees also complained that those in leadership positions were sometimes uneducated and unqualifi ed. This was perceived as a growing problem due to resettlement departures. It was most frequently levelled at Security members, who were seen as ‘low status’, uneducated and prone to use violence during and after arrest. This may not be unrelated to the fact that Security often appears to be the poorest sector of the camp administration, which perhaps attracts less educated staff than other sectors of leadership and has certainly in the past received less external support.10 Nevertheless, with such caveats noted and within the (admittedly narrow) parameters within which they operate, the camp governance structures are fairly effective and even well-adapted for their purpose. The hierarchical, fl exible, order-focused systems allow for minor disputes to be dealt with at an early stage and in doing so they play an essential role in maintaining peaceful co-existence among camp residents. They also command high levels of legitimacy within the camp population and are vastly preferred to engagement with either the Thai authorities or international organizations. It is valuable here to distinguish between the legitimacy of institutions versus the legitimacy of process and personalities. The justice structures were criticized with regard to specifi c practices and decisions but this did not negate a preference for community decision-making and a perception that this was a legitimate and appropriate form of governance for the camp. This perception is important. In determining ‘why people obey the law’ Tyler (1990) found that the primary reason for obedience was not fear of punishment but a perception of fairness and a belief that the system is legitimate. Yet whether something is ‘fair’ or ‘legitimate’ depends on the measure used; it is in essence a subjective position that refl ects personal or communal values. Beetham (1991) identifi es three perspectives on legitimacy: philosophical, legal and social scientifi c. The fi rst is concerned primarily with the moral justifi ability of power, the second is concerned predominantly with the nature of rules, and the third is concerned with the belief of legitimacy within the society. Though there is no reason for these perspectives to be polarized – Downloaded by [University of Defence] at 21:28 09 May 2016 each, after all, can illuminate elements of the relationship between law and society – they are often pitted in opposition. In Thailand, divisions between legal and social scientifi c perspectives in particular have shaped evaluations of the camp governance systems on the Thai–Burma border. Camp committees, community-based organizations and refugees demonstrate a high level of support for and trust in the structures in general despite dissatisfaction with individual cases. International organizations, in contrast, place less weight on popular feeling than on adherence to internationally recognized procedural standards. The struggle for ownership of justice 113

‘We must go step by step’: sociological legitimacy The most obvious barometer of support for dispute resolution structures is people’s willingness to use them. A survey by the IRC asked more than 2,000 refugees in Mae La camp and in the two Karenni refugee camps in Mae Hong Son province, ‘If you were a victim of crime, where would you go fi rst?’ Sixty- seven per cent of respondents said they would go to their section leader or to camp justice; a further 13 per cent said either camp committee or camp security; 0.5 per cent said they would go to UNHCR, while only 1 per cent in total would go to the Thai camp commander, Or Sor, Thai police or Thai courts (IRC 2007: 53). A follow up survey three years later, with 2,138 refugees in fi ve camps, asked those who had experienced crime where they had sought advice. Camp structures were still greatly preferred to Thai structures, with 61 per cent of respondents using camp structures, 5 per cent going to UNHCR or NGOs and 4 per cent going to Thai authorities (IRC 2010: 46). These statistics could be interpreted in a number of ways: as support of the camp structures, fear of other structures, or as coercion by camp leaders to keep cases inside the camp. I found the fi rst two reasons to be much more prominent. In particular, refugees considered involvement with the Thai authorities to be dangerous, with the potential for a perpetrator to take repercussions against the individual victim or collectively against the camp. I was told of several occasions where threats had been made against victims of crime, their advocates and even an entire camp. This included a case in Mae La Oon where a woman reported to an international agency that she had been raped by a man from a village close to the camp. The case was referred to the Thai justice system, after which threats of physical violence were made against the victim, her family and the international agency staff member who had referred the case.11 This experience and others like it have instilled deep fear of the consequences of referring cases to Thai court:

If we go to the Thai police, maybe the Thai authorities will create a problem for the camp.12

We should go step by step. If someone went directly to the Thai police 13 Downloaded by [University of Defence] at 21:28 09 May 2016 maybe they will be hurt or even killed.

Thai laws can’t help refugees because they are only used to punish refugees.14

In my interviews, refugees also expressed genuine support for the camp structures as the appropriate site for dispute resolution:

We are Karen, we have our own leaders. They should solve our problems.15 114 Governing refugees

If we bring problems to them [Thai authorities] we can’t speak Thai, we don’t understand the legal system, we don’t have the same background. People should have their own law.16

In contrast to the belief in legitimacy expressed by camp residents (that these systems are appropriate, culturally relevant and safe) international agencies are concerned with its legal legitimacy: the practice of dispute resolution and its adherence to human rights standards. Reports by UNHCR describe the camp systems as failing to adhere to international standards on gender equality and due process and as lacking ‘a clear legal basis, transparency, fair procedures, and trained personnel’ (e.g. UNHCR 2006b: A8; UNHCR Thailand 2006b: 5). Popular support from within the refugee population – its social scientiﬁ c legitimacy – is recognized to exist but is considered misguided, the result of oppression or of not knowing better, and therefore invalid:

Refugees often prefer for the crime to be ‘solved’ under traditional justice systems that do not meet basic standards of due process and do not apply sanctions in line with international human rights standards or Thai law. Refugees’ preference for using traditional systems of justice stem from its familiarity, pressure from refugee leaders, and a desire to keep refugee problems hidden from host country authorities (for fear of eroding Thailand’s receptiveness to hosting the refugee problem). (UNHCR 2006b: A8)

The ‘due process critique’: human rights as trumps As discussed, the systems of dispute resolution and order maintenance used in the refugee camps are diffuse, ﬂ exible and largely un-codiﬁ ed. Assessing their practice through a lens of procedural legitimacy inevitably skews the analysis towards the areas where they are most different from formal court structures and away from recognition of their overall contribution to camp stability and refugee welfare. This is explicit in the introduction to a Burma Lawyers’ Council report on ‘The Situation of the Refugee Camps from the Rule of Downloaded by [University of Defence] at 21:28 09 May 2016 Law Aspect’:

Many administrative bodies do a good job contributing to stability of the camps and promoting the welfare of the refugee people. However, as the objective of this research is to study the ﬂ aws or failings of rule of law institutions within the camps, only those areas of concern will be highlighted. (2007: 4) The struggle for ownership of justice 115

Meyerstein (2007) has described a similar dynamic in the reaction of international organizations to the creation of gacaca courts in Rwanda to process crimes committed in the genocide of 1994. Gacaca incorporated a traditional community dispute resolution process into the Rwandan criminal justice system and as such presented a unique response to an unprecedented situation. This approach attracted extensive criticism from international human rights organizations, most frequently on procedural grounds. Meyerstein notes that this ‘due process’ critique established a juridical ideal of technical and procedural standards that the gacaca courts could never attain; and, indeed, which only a Western court could attain (see also Clark 2010: 132–67). A bias towards ‘formal’ and state courts can be obscured by the apparently reasonable formula of a ‘human rights-based approach’. According to this approach – often advocated by those who are generally sympathetic to informal justice systems, such as Golub (2003) and Pimentel (2010a, 2010b) – informal justice practices are acceptable, provided they do not contravene human rights standards. A pertinent example of this is the UNHCR Code of Conduct, which is also the agency’s guidance for engagement with refugee-led justice systems:

Respect the cultures, customs and traditions of all people. Strive to avoid behaving in ways not acceptable in a particular cultural context. However, if considered as directly contrary to an international human rights instrument or standard, the later applicable instruments should guide us. (UNHCR 2004b: i, 2006b: 30):

This seems like an entirely reasonable compromise between recognition of local justice systems and protection of human rights but it has consequences that are not immediately apparent. Normatively, the human rights-based approach establishes international human rights standards as a ‘trump’ card overruling all other values (McEvoy 2007; Meyerstein 2007: 480). The extent to which this is problematic depends on which right is at issue. Ignatieff (2001: 173) has argued that human rights activism should recognize ‘the elemental priority [. . .] to stop torture, beatings, killings, rape, and assault and to improve, as best we can, the security of ordinary people’. Brown (2004) Downloaded by [University of Defence] at 21:28 09 May 2016 develops this point further in arguing that to see the prevention of cruelty as ‘the most we can hope for’ is not defeatism but rather the fundamental goal of human rights advocacy. Translated to non-state justice systems, this approach would dictate that the primary concern should be with the core ‘prevention of cruelty’ rights (such as the right to life and prohibition against torture) rather than with narrowly deﬁ ned due process norms such as the right to legal counsel. Similarly, gender and ethnic representation are important long-term goals but failure to meet these goals should not delegitimize a non-state justice system, not least because equal representation is rarely seen 116 Governing refugees

in any formal justice system (including the UK, where of the eleven serving Supreme Court judges at the time of writing, ten are men, one is a woman and all are Caucasian).17 Jurisdictionally, the human rights-based approach grants external actors (such as UNHCR staff) the authority to decide when a community justice system is exceeding its functions, thus removing any certainty or self- determination from those local systems. In this, there is an echo of the ‘repugnancy clause’ which was a staple of colonial-era legal reform, whereby the practice of customary courts was acceptable provided it was not ‘repugnant to’ any provisions of written law or ‘contrary to morality, humanity and natural justice’ (Otlhogile 1993: 526).18 Describing a similar provision ordering the relationship between indigenous justice systems and the Mexican state, Speed and Collier (2000: 901) call it ‘a cruel hoax’ which gives the appearance of protecting indigenous authority while actually removing its autonomy. This not only reduces the capacity of local actors to conduct dispute resolution but by establishing human rights in opposition to community values also risks setting up resistance to human rights values as a foreign imposition. Neither of these outcomes helps to advance the ultimate goal of human rights protection. The conclusion to be reached from this is not that serious human rights violations by local justice systems should be tolerated but rather that the potential for human rights discourse to be used to reduce the power of local authorities must be recognized (Speed and Collier 2000). This is also not to say that human rights are inappropriate, inapplicable or unimportant to non-Western ‘cultures’ (see further Chapter 7 for a discussion of cultural relativist critiques). It is precisely because human rights values are important that they should be communicated in a way that local actors understand and as part of a conversation to which they can contribute. Drawing on many years of practical experience in working with community-based and indigenous justice systems, Braithwaite (2002) has outlined a process for achieving this in establishing community restorative justice programmes. The fi rst stage is to assemble key stakeholders to refl ect on a core set of principles, standards or rights. These might be grounded in the values of international or regional human rights instruments but should be expressed in terms which all participants Downloaded by [University of Defence] at 21:28 09 May 2016 can understand. Through a process of deliberation and agreement, local stakeholders should commit to these shared standards. Contested standards may be resolved by ‘refl exive practice’, a process of negotiation that refers back to values and principles that have already been agreed. Throughout, didactic training on law or human rights should be avoided in favour of training sessions which are a refl exive process of discussion between all participants. Over time, discussions should gradually expand outwards to include increasing numbers of the local population, before culminating in community-wide adoption of the standards. Through participation and The struggle for ownership of justice 117

consensus-building, human rights values and principles can be embedded into ways of seeing and resolving problems. With its recognition and understanding of the unique and lasting value to a community in participating in norm deﬁ nition and enforcement, this approach could be of particular value in a refugee camp context.

NGOs and the rule of law orthodoxy Until recently, international humanitarian agencies paid little attention to justice and security in refugee camps. As rule of law activities gained prominence within the UN system as a whole, they began to also be incorporated into refugee operations (Costa 2006; UNHCR 2006b). In 2004, the UNHCR in Thailand launched an ‘administration of justice’ programme. By 2013, access to legal assistance was central to its protection programming, and to the work of IRC. This work began in an environment where there was an evident need for external support. It was extremely diffi cult for refugees to access the Thai criminal justice system, or to receive any redress if the perpetrator was a Thai citizen. Women’s organizations had been working for years to determine the safest way for migrant and refugee women to seek justice for sexual and domestic violence, whether in camps or through Thai courts. Resettlement had placed the camp-led dispute resolution systems under increasing strain as experienced staff and leaders left, and as new types of crime became more common (including debt and loan cases, and youth delinquency; on the latter, see Chapter 7). Overall, however, the systems were ‘reformable’: there was no institutionalized violence or corruption, there was considerable respect for the existing structures, and many (though by no means all) refugee leaders were amenable to change. The ideal role for international organizations was therefore to retain the strengths of the existing system (popular legitimacy, simplicity, effi ciency) while addressing such problems as lack of resources, staff turnover and refugees’ power inequalities and vulnerabilities vis-à-vis non-refugees and the Thai criminal justice system. Of course, achieving this is rather more diffi cult than it sounds, to which a tranche of past and present law reform processes are testimony (Ubink and McInerney 2011; Harper 2011; Faundez 2010; Downloaded by [University of Defence] at 21:28 09 May 2016 Pimentel 2010a; Carothers 2003). The fi rst programmes on the administration of justice in the camps were led by UNHCR and broadly followed the ‘rule of law orthodoxy’ of state- focused, state-centric initiatives, focusing on consultation with Thai police and other offi cials, and on encouraging reporting of serious cases to Thai courts. A more ‘legal empowerment’ approach was represented by a project to create ‘Legal Assistance Centres’ (LACs) which serve as a conduit between refugees, camp justice structures and Thai authorities. 118 Governing refugees

Legal Assistance Centres Implemented by IRC with partial UNHCR funding, three Legal Assistance Centres have been operational since 2007 (in the two Karenni camps of Ban Mai Nai Soi and Ban Mae Surin and in the Karen camp of Mae La) and two more opened in 2012 (in Umpiem Mai and Nu Po). In the camps where it is present, LAC responds to refugees’ requests for legal advice and provides support in referring cases to the Thai legal system. Activities include training for camp residents on Thai law, basic legal concepts and human rights awareness; training for camp staff (including mediation and restorative justice techniques); material and technical support to camp staff; and a ‘law reform’ process of consultations to harmonize camp justice practice with Thai law and international human rights law (Hutchinson 2013; Harding et al. 2008). LAC has no investigative capacity or authority to resolve cases but provides education and assistance. In this role it also serves as a bridge or buffer between the camps and local Thai authorities. This function has been developed by, for example, bringing Thai court staff and police to visit Mae La camp in order that they can better understand the situation of refugees (Harding and Varadan 2010). LAC has also been able to secure refugees’ access to entitlements of the Thai legal system; such as victim compensation for the family of a refugee killed by a passing car. An evaluation of more than 2,000 camp residents in the sites where LAC operates found considerable success across a range of indicators, including increased awareness of Thai law, improved conﬁ dence in camp justice and perceptions of overall safety and security (IRC 2010: 3–5). Yet of the 2,138 respondents, only six had sought advice from LAC and ten had reported a dispute to LAC. Other refugees were indisputably the preferred option for seeking advice and for reporting disputes. In this the hierarchy of ‘step by step’ remained central, with section leaders as the clear ﬁ rst preference for both seeking advice (29 per cent) and reporting a problem (32 per cent) (ibid: 20, 43).19 LAC was created with the dual goal of enhancing refugees’ access to Thai law and strengthening the operation of camp justice structures. An early evaluation of the programme stressed the importance of the second aspect, advising against ‘a legalistic approach’ and advocating simple systems that

Downloaded by [University of Defence] at 21:28 09 May 2016 were based ‘on culturally acceptable practices, rather than trying to undermine or replace these’ (Aiken 2009: 42). Efforts were made to strengthen the capacity of camp administrations through training and resource provision, recruitment of refugees as paralegals and consultation with refugee leaders about the progress of cases. As the programme has developed, this work has continued but the legal empowerment component of LAC’s work has become less prominent than its work in strengthening access to Thai law. An institutional bias towards Thai law is also clear in stafﬁ ng, with Thai lawyers and law students taking a leading role in project implementation. While their commitment and professionalism is not in dispute, people with formal legal The struggle for ownership of justice 119

training are not necessarily the most appropriate people to work with community justice systems; indeed, they might even be the least appropriate people, as their measure of a successful system will typically be the legal system in which they have been trained. Power dynamics between ‘local’ and ‘international’ staff are a perennial challenge in international development (Pouligny 2006: 87–5). These dynamics take on an additional dimension in a refugee situation, where the relationships to be navigated are not only between international/national staff but international/national/refugee staff. The relationship between refugees and citizens of the host society is particularly sensitive. When this is combined with the charged dynamics of justice reform (which inevitably challenges deep-seated norms and cultural values), there is a risk that justice reform will be viewed as a Trojan horse to reshape refugees’ identities and encourage them to (in this case) ‘become Thai’. As one of my interviewees commented, when asked how he felt about LAC, ‘In fi ve years, the camps will be under the control of the Thai authorities. There will be no need for camp committees.’20 For its fi rst fi ve years of operation LAC was present in only three of the nine border camps, and only one of these (Mae La) was a ‘Karen’ camp. This has had obvious implications for equitable service delivery, capacity-building and participation. It has also had implications for consistent policy advice to refugees. In the camps where LAC is present, IRC is the primary avenue of external contact and advice on justice matters. Where LAC is not present, UNHCR is the lead agency on these issues. This is important because staff of the two organizations at times had different approaches to camp justice, including regarding the most important questions of all: which problems and crimes can be solved in camp and how they should be dealt with. This difference – and the tension between the ‘rule of law orthodoxy’ and the goal of legal empowerment – was particularly evident in efforts to defi ne camp rules and procedures.

Harmonizing camp rules and Thai law One of LAC’s primary objectives for strengthening camp justice was a ‘law reform’ process to clarify camp rules and harmonize camp practice with Thai law. Two concurrent law reform processes were instigated, one with members Downloaded by [University of Defence] at 21:28 09 May 2016 of the Karenni Refugee Committee, camp committees and community representatives of the camps of Ban Mai Nai Soi and Ban Mae Surin, and one with the KRC and members of the camp committee and community organizations from the camp of Mae La. The drafting process for the ‘Karenni camps’ included representatives of both camps. In Mae La, members of the camp committee, representatives of camp-based community organizations and members of the KRC were formed into an ‘Approval Committee’. The Mae La process has been described by a former LAC staff member as ‘an instructive example of norm transference in the context of development and 120 Governing refugees

humanitarian interventions’ (Hutchinson 2013) but its restriction to only one of the seven Karen refugee camps meant that it was not representative of the wider refugee population.21 The capacity-building impact of the law reform process was further weakened – perhaps fatally so – by its failure to produce a concrete outcome; a failure that was not due to resistance from refugees but to a lack of clarity and consensus among international agency staff about what the process was intended to achieve. There is relatively little normative clash between the crimes punished in the camps and under Thai law, with the most direct inconsistency being the camp practice of punishing adultery and pre-marital sex (which are not crimes under Thai law). Nevertheless, after four years, the IRC-LAC law reform process had not publicly circulated even a draft document. In February 2011, the KRC issued a revised set of camp rules and regulations that showed little evidence of an enhanced knowledge base from the years of law reform discussion. The revisions meet some very broad objectives (e.g. clarifying that camp rules do not confl ict with Thai law, avoiding defi ning adultery as a crime and removing any role for the KNU in cases involving refugees) but they are very unclear. There are four categories of offence: ‘rules’, ‘civil law suits’, ‘crimes’ and ‘civil law suit – crime’. Non-discrimination is categorized as a rule punishable by education, warning or referral to Thai authority. Showing or selling ‘x-rated movies or pictures’ is also a rule, punishable by a cash fi ne and one month confi nement. Offences identifi ed as civil law suits include the violation of ‘social and cultural norms’ of adultery and prostitution, which should be dealt with by ‘warning, compromising or action taken similar to that of action taken by related society’. Crimes include forestry offences, gambling, drugs, stealing, fraud, murder, human traffi cking, violent and sexual offences and some vague prohibitions such as ‘the illegal exploitation of children for self-gain which creates mischief in the child’, ‘any action that negatively affects the stability and peacefulness of the camp’ and ‘abusing the personal rights of others’. Each of these vague offences has only one permitted response: to refer to the Thai authority. Participants in the law reform process attributed the delay to a combination of staff turnover within IRC-LAC and to UNHCR opposition. Some UNHCR representatives in Thailand had been opposed to the idea of establishing a legal code for the camps from the outset. As one staff member Downloaded by [University of Defence] at 21:28 09 May 2016 said to me: ‘Law reform. It’s not law. [. . .] Why do they need law for the camps when there is a Thai law covering the whole of Thailand?’22 This representative also considered that the camp systems were a ‘parallel’ system in violation of Thai sovereignty and that any arrest, detention and punishment carried out by camp leaders was illegitimate. In response to such complaints, the Law Reform process was re-envisaged and re-named as Developing Mediation and Legal Standards. In October 2011, the process was changed again, dropping all reference to ‘law’ and becoming Mediation and Dispute Resolution Guidelines. Refugee leaders and participants from The struggle for ownership of justice 121

all sectors (including IRC-LAC staff) were deeply disillusioned with the repeated delays and lack of progress:

It’s been delayed again and again. The person responsible for the Law Reform from LAC has changed again and again. At the beginning we put so much effort to make a traditional law that is not against the Thai law. For example, before, adultery was a crime in our community, it can be sentenced to up to seven years in prison. We adapted quickly, we challenged the old community leaders and eventually we made agreement on that, that it’s not a crime. So many issues. And then it changed again. So eventually I decided not to join them. I quit from the Approval Committee.23

According to UNHCR, refugees cannot arrest refugees [. . .] Refugees cannot use law. Refugees cannot write their own law. They cannot create an article. They cannot punish a person. So what can we do? It is a big problem.24

In 2013, the Mediation and Dispute Resolution Guidelines were piloted in the camps where LAC operates (it is not clear if or when they will be extended to ‘non-LAC’ camps). The Guidelines – approximately eighty pages long in English – consist of twelve guidance notes, each addressing a broad area rather than a single offence. The issues covered fall into three types: core principles of dispute resolution (including guidance notes on general principles, absolute jurisdiction offences and confl ict of interest); methods of dispute resolution (including notes on alternative dispute resolution, mediation and group conferencing and arbitration) and specifi c case types (involving youth, rape, domestic violence, adultery, marriage and divorce, and debt and loan cases). The crux of the Guidelines is that serious offences must be sent to Thai courts but that other offences may be solved in camps where that is the preference of the parties. Cases that must be referred to Thai law include murder, rape, assault causing grievous bodily harm, narcotics offences, human traffi cking, offences against children, possession of fi rearms and forestry offences. Camp-based dispute resolution is underwritten with the principle Downloaded by [University of Defence] at 21:28 09 May 2016 that ‘all disputes that arise in the camps can be referred to the Thai justice system at any time’.25 The Guidelines do not resolve the central problems of defi ning camp rules and jurisdictions, but instead focus on establishing a process of dispute resolution. Camp justice is to be ‘alternative dispute resolution’ wherein detention is only to be used in exceptional cases or where there is an immediate risk of violence or fl ight from the camp. Other disposals include a warning, promissory note, fi ne, damages and community service order (CSO). Three forms of alternative dispute resolution are identifi ed: mediation, 122 Governing refugees

arbitration or group conferencing. The option closest to that already used by refugee leaders is group conferencing, whereby all interested and affected parties can participate (mediation will involve the disputants and a mediator, and arbitration will involve the parties to a dispute and any witnesses). It is helpful that the Guidelines clarify a role for camp leaders in resolving disputes in camp and that they offer a fl exible rather than legalistic guide to decision-making (in contrast to a 2010 draft of the ‘law reform’, which was a thirty-one-page document of dense legal defi nitions extracted from the Thai Penal Code). Formalizing the camp dispute resolution system as a restorative justice approach may be helpful, provided suffi cient investment and attention is given to supporting the implementation. Restorative justice has many advocates in Anglo-European jurisdictions, where it is seen as a valuable alternative to the ineffi ciency and inhumanity of the prison ‘punishment society’. A variety of studies have found that restorative justice approaches can increase victim satisfaction with the justice system, reduce re-offending and improve community safety. While it is naturally diffi cult to provide a defi nitive answer on whether restorative justice ‘works’, the evidence base has been assessed as ‘far more extensive, and positive, than for many other policies [in the UK criminal justice system]’ (Sherman and Strang 2007: 4). However, dispute resolution does not become ‘restorative justice’ simply by removing the possibility of punitive sanctions. The approach used by refugee leaders at present has many elements that are close to a restorative justice model (such as an emphasis on consent, participation and a negotiated solution) but it also relies on the sanction of detention. In my interviews in 2012, camp leaders found it diffi cult to understand how a justice process would be effective without this as ‘mediation is not good for cases where there is a clear right and wrong’.26 One LAC staff member acknowledged that encouraging restorative justice approaches had been ‘one of the areas of greatest resistance in what we’re trying to do’.27 Community-based organizations were not receptive to a restorative justice approach and perceived CSOs as giving them responsibility for the rehabilitation of ‘youth gangsters’, a task that they were not trained or resourced to carry out.28 This was linked to more deep-seated – and wholly reasonable – concerns about the capacity to effectively implement an unfamiliar justice model: Downloaded by [University of Defence] at 21:28 09 May 2016

The mediation process is part of camp justice, so if we are going to do mediation in camp, the people doing it really need to understand it, they need to have the necessary skills to do it. Right now, they don’t. If we don’t have camp justice any more, we really have to make clear which cases will go to Thai court. For absolute jurisdiction cases, it’s clear – what’s not clear is what happens with everything else. With camp justice, if there is to be a mediation system in camp, they need to choose the best and most skilled people to do it. Section leaders at the moment are not educated.29 The struggle for ownership of justice 123

Referral to Thai courts The Mediation and Dispute Resolution Guidelines are the culmination of several years’ effort to clarify the role of camp justice and encourage referral of serious crimes to the Thai criminal justice system. Between 2003 and 2006, UNHCR supported the referral and processing of only seven serious crimes involving refugees to the Thai criminal justice system (IRC 2009: 2). Later, IRC reached an informal agreement with the Thai Ministry of Interior regarding offences which should be dealt with in Thai court (the same ‘absolute jurisdiction’ offences that are identiﬁ ed in the Mediation Guidelines).30 Since LAC was created, many more cases have been referred from the refugee camps to the Thai criminal justice system, with 120 cases referred in 2011 alone (UNHCR Thailand 2011). However, not all of these cases are serious crimes. In fact, it seems that the majority of case referrals relate to immigration (‘illegal presence’) and forestry offences. Statistics provided to me by LAC show that sixty cases were referred from Mae La to Thai justice in 2009. Of these cases, 18 per cent were forestry offences and 11 per cent were illegal presence. Sexual violence cases (rape and attempted rape) and cases of physical harm (actual and attempted murder, manslaughter, assault and domestic violence) were a total of 37 per cent of referrals, i.e. twenty-two cases. During the same period, in Ban Mae Surin and Ban Nai Mai Soi, sixty-seven cases were referred to Thai court, of which cases of physical harm and sexual violence constituted only 12 per cent; cases of ‘illegal presence’ accounted for 75 per cent of all referrals. Similarly, in 2010, of seventy-seven cases referred to Thai justice from Mae La, 44 per cent were cases of illegal presence and forestry offences. Cases of physical violence and sexual violence accounted for only 27 per cent of the total.31 Of referrals from the Karenni camps, 66 per cent were cases of illegal presence. When serious crimes occur it is critical that refugees should have access to Thai courts. However, while the principle of refugees’ access to Thai courts has been afﬁ rmed, the reality is rather different. In the past referrals were resisted by both Thai police and refugees and to some extent this resistance continues. Refugees may not wish their case to be processed in Thai court for fear of discrimination as well as practical challenges such as language,

Downloaded by [University of Defence] at 21:28 09 May 2016 transport and lack of legal knowledge. For many years, camp leaders have been encouraged to resolve cases internally and there is a continuing fear that referring cases to Thai courts will lead to repercussions by perpetrators of crime, or that the Thai authorities will tire of the increasing demands that are being placed on them. International support is considered essential to overcoming the power imbalance that refugees face in engaging with Thai citizens or authorities:

LAC help us access Thai justice and this is very good for us. By ourselves we can’t do this. If a Thai citizen harms a refugee, they discriminate against us. But if LAC speaks for us, they can’t do this.32 124 Governing refugees

In addition, while access to Thai law is vitally important for responding to serious cases occurring in the refugee camps, it is less appropriate for the (much more common) low-level disputes that arise. International agencies may be driving the reform agenda but national offi cials have to implement it and the Thai criminal justice system is simply not equipped to provide the constant, continual low-level dispute resolution that section leaders currently conduct. Furthermore, camp staff and community organizations have found that the Thai police continue to prefer that they manage problems internally and at times refuse to accept cases, instead returning offenders to the camp without punishment.33 Confl icting advice from international agencies and from Thai authorities places camp leaders in a diffi cult position:

Ten days ago, I caught ﬁ ve people using drugs. I sent them to the Thai authorities and the Thai authorities sent them back yesterday. When I asked, ‘What did the Thai authorities do?’, they said, ‘They just took a picture and released us.’ This is a big problem. We send these young men who are drug addicts to the Thai authorities and then the Thai authorities send them back. This is very bad.34

When the [Thai] camp commander comes to the camp committee, he says one thing to the camp committee and another to the NGOs. For example, when the camp commander comes to the camp, they say to the camp committee that you can arrest this person. Then UNHCR come and say, ‘Why did you arrest this person?’35

‘One nation, one law’ While efforts to encourage and support referrals to Thai court have been valuable for particularly serious criminal cases, they have also led to confusion among camp security and justice staff about how to respond when referrals are rejected or are not possible. My interviewees suggested that the messages conveyed in NGO-led training programmes were at times misunderstood and misinterpreted:

Downloaded by [University of Defence] at 21:28 09 May 2016 In the past, if the neighbours see the husband beat his wife several times, they would report to the leaders and they would take action. Now, they understand that if the victim doesn’t tell the leaders then nobody can do anything. [. . .] They are told that if they do this it’s wrong, and if they do that it’s wrong. They have less conﬁ dence. They don’t want to take any responsibility or any action because different organizations have different ideas about what they should do.36 The struggle for ownership of justice 125

Sometimes, there is domestic violence in my section, but this is not against Thai law so I don’t dare to go and detain.37

Effective communication is always a challenge in international development programming, but misunderstanding has potentially serious consequences here. The comments above suggest a move away from an ‘order-focused’ approach (whereby section leaders would intervene where they perceived a problem to exist) to a ‘victim-focused’ approach (whereby section leaders believe they should wait for a victim to report a problem before taking action). It is signiﬁ cant that these speakers describe confusion in responding to domestic violence cases, as this is the single case type that has had by far the most external attention to encourage effective responses. The comments above also suggest that anxiety about the appropriate response may discourage camp staff from taking any response at all; an outcome that is likely to increase impunity and decrease stability rather than to strengthen ‘access to justice’. Much of the blame for this confusion, and for the lengthy wrangling around the law reform, was placed with the hardline stance of UNHCR staff in advocating ‘one nation, one law’:

Basically, I think that the UN are very strong about this – the only law is Thai law. And I think that that has sort of been undermining the traditional justice system. Not that there weren’t major problems with the traditional system, but instead of trying to improve and work on the traditional system their mantra is ‘Thai law is the only law’. [. . .] And it’s not like the Thai justice system and the Thai police are readily jumping up to ﬁ ll the justice roles, their justice roles in the camp. So it’s kind of like, they’re directing everyone to go through this justice system that isn’t really there.38

In some cases, we feel we could solve in the camp but UNHCR say no.39

UNHCR ask KRC to be careful with Thai law, it is ‘one nation, one law’.40

This highlights an important point, which is that the non-governmental Downloaded by [University of Defence] at 21:28 09 May 2016 sector is no more monolithic than the state. Between and within agencies there can be considerable policy differences which may be enshrined in agendas deﬁ ned at headquarters or may be related to decentralisation at the ﬁ eld level. In the refugee camps in Thailand, opposition to refugee-led ‘parallel justice systems’ was led by UNHCR representatives rather than the Royal Thai Government, which while presumably in principle opposed to a justice system that opposes its state sovereignty has taken no action to suggest that it considers the refugee-led structures to be such a system.41 I was repeatedly told that IRC-LAC staff were more sympathetic to community 126 Governing refugees

governance than counterparts in UNHCR. However, not all UNHCR representatives adhered to the stance of ‘one nation, one law’, while LAC staff did not promote legal empowerment with sufﬁ cient conviction to overcome UNHCR opposition. These institutional variations and power dynamics have had important consequences for justice programming in the refugee camps, as evidenced in the fundamental lack of clarity about the purpose and objectives of the law reform process. This lack of clarity not only affected the direction of international programming but placed refugee leaders in the challenging position of navigating the politics of international organizations as well as those of internal camp affairs:

In the ﬁ rst time, when we started law reform process, a lawyer from England she worked with LAC and she was very enthusiastic to create a law for the refugee camps. After more than one year, she left and another one came. Not too strong. […] He or she? Maybe a man. He listened to us, but not too much technical support from him. […] And then another came, weaker, another came, weaker. The last one, ‘I am not sure that we can get approval from MOI’. At the same time, UNHCR – UNHCR also have turnover and the last one very strong, very strong guy. A very strong UNHCR and a very weak LAC. So LAC can do nothing.42

LAC has to obey UN. UN is more important than LAC so whatever UN say, LAC have to follow.43

A tentative counter-critique Non-state justice systems operate all over the world, with varying degrees of effi cacy and legitimacy. In refugee situations, the extent to which refugee- led dispute resolutions systems are an asset to camp management will obviously depend on the specifi c circumstances of each context. Neverthe- less, it is highly likely that most disputes and social problems that arise in refugee camps will be solved at community level and thus that some form of local dispute resolution will be an essential component of camp management. Downloaded by [University of Defence] at 21:28 09 May 2016 As has happened with other rule of law projects, international agencies in Thailand have been critical of the legitimacy of camp justice while apparently overlooking the procedural failings of the formal system (Carothers 2003: 11). An assumption that Thai law is the appropriate destination for cases from the refugee camps overlooks the considerable fl aws in the Thai national justice system, which is chronically overstretched with delays in case processing, prison overcrowding and serious problems of corruption and abuse of human rights throughout the system. Ethnic minorities, refugees The struggle for ownership of justice 127

and migrants are at particular risk of mistreatment (Amnesty International 2002; Human Rights Watch 2012). In my own interviews with refugees who had been imprisoned in Thai jails, one man who had been imprisoned for theft described the conditions as so unbearable that he wanted to commit suicide,44 while another man, jailed purely for immigration offences, had been beaten so badly in prison that he still experienced pain from his injuries many years later.45 Yet when I raised the matter of prison conditions with a staff member of UNHCR he said, ‘it depends where you compare it to. Compared with my country, Thai prisons are pretty good’.46 In its very ﬂ ippancy this comment indicates a relative lack of concern with human rights violations when they occur in a national criminal justice system. This blindness to bureaucratized violence resonates with Cover’s (1986) work on the ‘violence of law’. Cover argued that when the state is recognized as the sole legitimate holder of authority we become inured to the ‘violence of legal acts’ by familiarity with its roles and institutions. This does not imply that non-state legal processes are not violent – on the contrary, he recognizes all law as backed with violence – but in state legal processes, the sanitized theatre of court process prevents appreciation of its real meaning. He wrote, ‘I think it is unquestionably the case in the United States that most prisoners walk into prison because they know they will be dragged or beaten into prison if they do not walk’ (ibid: 1607). Applied to the position for refugees in Thailand, this blindness to the violence of law permits the cruelty of Thai prison to be routinized (and as such deemed acceptable), while violence beyond the state (such as the referral of offenders to the KNU in Burma for punishment) is recognized for what it is and deemed unacceptable. In Rwanda, Meyerstein (2007: 492) describes an energetic ‘countercritique’ from members of the Rwandan Government and judiciary to criticisms of the gacaca courts, in a dialogue that he links to postcolonialism and the desire to assert an independent, authentically ‘local’ legality. In contrast, Karen refugees are for the most part deferential to Western knowledge, which tends to be associated with positive conceptions of political power and social development. In my interviewees with camp judges, section leaders and security staff, they often expressed embarrassment at what they viewed as a lack of sophistication in their practice. There was a perception that the camp Downloaded by [University of Defence] at 21:28 09 May 2016 processes of dispute resolution were vastly inferior to Western law and an assumption that I, as a Western-trained lawyer, would ‘look down’ on the processes used in camp. This attitude was less prevalent among those who were more educated and who had more experience in dealing with foreigners. These people were better positioned to compare the different legal traditions and recognized that legal formalism also had disadvantages, and that a notion of law as malleable and instrumental may not be appropriate for their community after all. One Karen leader who had participated with LAC in 128 Governing refugees

the law reform process described long lawyerly debates about words and deﬁ nitions of crimes, saying:

This is why I hate the law. In law there are two channels [holding hands out from his side, weighing them like a set of scales] but if you are clever in talking, if you are clever with words, then you will win the case.47

There was also some irritation with the human rights idealism of NGOs, which ignored the difﬁ culties faced by refugee camp justice and security staff:

Sometimes the community get annoyed. Sometimes they appreciate it if they get more knowledge. And they think, ‘Ok, so before we didn’t know but now we know so it’s better.’ But sometimes they feel like, ‘Hey, we’re not in our own country, we’re in the refugee camp, this is the best we can do.’ And normally they think that their decision is better than what the NGOs think, because it’s the community making the decision, judging other people who commit a crime in the community. So they feel that it’s more fair, and they look at it every way.48

There was also some recognition that this environment of institutional pluralism would have lasting consequences for community management practices, both for the refugee camps and potentially inside Burma:

Another concern I have is the long-term effect of this process, you know because it’s so much out of the hands of the community but it is given a certain status, part of that, uh, ‘mysterious knowledge’ of lawyers. These documents will have power in the future and they will have authority and so, you know, when refugees go back to Karen State, are these laws going to go with them? So how far reaching will it be, the effect on this culture or this community?49

There is something both familiar and ironic about this process – familiar, in that it repeats a process of rationalization of customary justice practices within the state which was more common in colonial times, yet ironic in that community-level governance practices among the Karen survived Downloaded by [University of Defence] at 21:28 09 May 2016 colonialism but are threatened by ‘state capture’ at a time of statelessness, as refugees. There is a further irony in the fact that elsewhere there has been a turn away from the state as the exclusive justice authority. In Europe and in America, criminologists and sociologists have increasingly noted a loss of faith in the penal system and punishment society (Braithwaite and Mugford 1994; Garland 2002, 1993) and emphasized the value of ﬂ exible and decentralized governance practices, and of more constructive relationships between community and state, such as community policing partnerships or community restorative justice initiatives (Crawford 1997, 2006; Johnston The struggle for ownership of justice 129

and Shearing 2003; Loader and Walker 2007). These proposals all suggest fruitful new directions in security knowledge which recognize the possibilities in governance beyond the state and are rooted in recognition of the limitations of national policing and prosecution structures, and of the value of community dispute resolution in its own right. Of course, an analysis focused on state capture risks over-simplifying the relationship between ‘community’ justice and other legal orders. This is particularly problematic in relation to human rights – which, after all, seek to emancipate and not oppress. Human rights standards may be used by international actors to challenge the autonomy of local actors but they can also be a tool for local actors to confront oppression and inequality. This chapter has identiﬁ ed a struggle to own and to deﬁ ne justice at the institutional level. The following chapter shifts the focus from institutions to ideas to examine the effect of human rights values on camp justice practice.

Notes 1 Interview #60, KWO member, 6 June 2009. 2 Interview #95, former camp leader, 12 March 2010. 3 Interview #59, KRC member, 4 June 2009. 4 The most recent version of the camp rules was revised and amended in February 2011. UNHCR and IRC prepared a set of Mediation and Dispute Resolution Guidelines that was circulated to community organizations in August 2013. These Guidelines are discussed further below. 5 In three camps, IRC-LAC has funded the construction of secure detention units, and trained security staff in appropriate detention standards. As a result, the use of stocks has been abandoned in these camps. 6 An earlier version of the camp rules was explicit about the role of the KNU in camp justice, stipulating ‘forward to mother organisation’ as the punishment for murder, death [sic], possession of ﬁ rearms and suspected informers (Burma Lawyers Council 2007, Annex 37). There is no mention of a role for the KNU in the updated version of the KRC camp rules (KRC 2011). The subject of sanctions for youth delinquency is returned to in Chapter 7. 7 Interview #25, Mae La Oon, 4 March 2009. 8 Interview #48, Mae La Oon, 15 May 2009. 9 This cohesion may well be signiﬁ cant for the success of these systems as while

Downloaded by [University of Defence] at 21:28 09 May 2016 there are in theory multiple possible legal orders, most refugees only recognize one, thus reducing the use of ‘forum-shopping’ (see further Chapter 7). 10 This is another area where there is wide variation between camps. Where IRC- LAC operates, security receives considerable technical and material support. Where IRC-LAC is not present (such as Mae La Oon and Mae Ra Ma Luang) Security members have received little or no dedicated training and assistance. 11 Another example occurred in 2003 in a Karenni camp, where a young girl and a woman were raped by two Thai Or Sor. The camp committee reported this case and it was prosecuted in Thai military court with the perpetrators sentenced to nine years in military jail. Throughout the process, the victims, their families and 130 Governing refugees

the camp committee were repeatedly threatened and urged to drop the prosecution. 12 Interview #24, Mae La Oon, 4 March 2009. 13 Interview with camp resident, Mae La Oon, recorded in fi eld notes, 2 Feb 2010. 14 Interview with former section leader, Mae La Oon, recorded in fi eld notes, 14 Feb 2010. 15 Interview #76, camp resident, Mae La Oon, 15 Feb 2010. 16 Interview #59, KRC member, Mae Sot, 4 June 2009. 17 See www.judiciary.gov.uk (accessed 2 August 2013). 18 This provision became the basis for challenging polygamy across much of sub- Saharan Africa, and it is an interesting twist that while the human rights-based approach also challenges local conception of sex and sexuality, it is no longer perceived promiscuity that is a problem but rather sexual conservatism, i.e. the policing of adultery and pre-marital sex. 19 Where respondents seek advice: section leader (29%); family members (21%); camp security (12%); camp management (5%); CBO (4%); camp justice (3%); zone leader (2%); religious leader (2%); Thai authority (3%), NGO (2%), UNHCR (1%); LAC (1%); Thai citizen (1%); other (2%). Where respondents report a dispute: section leader (32%); other family members (15%); camp security (14%); other camp residents (7%); camp management (6%); camp justice (6%); CBO (4%); zone leader 3%; religious leaders (2%); Thai authorities (4%); LAC (2%); NGO (1%); Thai citizen (1%); other (1%) 20 Interview #96, KYO member, Mae Sot, 14 March 2010. 21 Earlier, I outlined Braithwaite’s proposal for a participatory, consensus-based, values-focused approach to establishing bases for community-based restorative justice programmes. This approach would have been much more valuable for refugees in Thailand than the ‘Approval Committee’ process followed by IRC- LAC and UNHCR, where the process throughout was conducted in terms that participants were unfamiliar with: fi rst, the dense legalism of the Thai penal code, then the complex vocabulary of ‘Alternative Dispute Resolution’. Furthermore, the Approval Committee allowed only a small number of refugees to participate in the drafting and consultation process, in a role that was (as the name implies) primarily one of agreeing to a process designed and devised by others. It was clear that the KRC and camp committees were willing to cooperate with a law reform process and to revise the camp justice procedures; this goodwill could have been harnessed to hold a conversation about human rights values and standards in terms which refugee participants could understand. Allowing the wider refugee population to participate in the process of debating and negotiating Downloaded by [University of Defence] at 21:28 09 May 2016 standards would have strengthened understanding of the values and rationale underpinning the reform of justice and dispute resolution, thus increasing programme sustainability as well as local ownership. 22 Notes from interview #138, Mae Sot, 23 April 2012. 23 Interview #139, Mae La, 24 April 2012. 24 Interview #134, Mae Hong Son, 20 April 2012. 25 Mediation and Dispute Resolution Guidelines, Guidance Note #1. 26 Interview #139, Mae La, 24 April 2012. 27 Interview #65, Bangkok, 6 January 2010. 28 Interview #167, Mae Hong Son, 3 May 2012. The struggle for ownership of justice 131

29 Interview #167, Mae Hong Son, 3 May 2012. 30 These offences are also identifi ed in the Mediation and Dispute Resolution Guidelines as those that cannot be solved in camps and that must be referred to Thai courts. 31 Interview #170, Mae Sot, 5 May 2012. 32 Between January and December 2010, seventy-seven cases were referred to Thai justice from Mae La: illegal presence 33%; forestry offence 11%; drug use/ possession 1%; drug traffi cking 1%; child abuse 2%; attempted murder 1%; accidental death 4%; vehicle accident 5%; theft 4%; suicide 1%; rape of a minor 16%; rape 1%; physical assault 1%; physical harm 3%; other 14%; murder 2%. In Jan–Dec 2009, sixty cases were referred to Thai justice from Mae La: forestry offence 18%; illegal presence 11%; manslaughter 1%; murder 6%; other 9%; physical harm 7%; rape 2%; rape of a minor 12%; theft 2%; threat of murder 1%; human traffi cking 1%; accidental death 4%; attempted murder 5%; attempted rape of a minor 1%; child abuse 1%; child support 1%; divorce 1%; domestic violence 2%; drug traffi cking 4%; drug use 9%. Statistics as provided by LAC and on fi le with author. These statistics were provided as a pie-chart which gave the total number of cases reported to LAC and referred to Thai justice. For each individual case-type a percentage was given rather than a number of cases. 33 A DFID document cites LAC database data in claiming that in 2011, 93% of all appropriate serious crimes in the fi ve camps where LAC operates were successfully referred to the Thai justice system. My interviews with leaders and community organizations from four of those fi ve camps suggested that this is not the case. It is possible that camp staff do not tell NGOs if Thai police reject a case for fear of damaging their relationships with local authorities. See DFID Logical Framework [203400], Output indicator 4.1. Available online at http://projects.dfi d.gov.uk/ iati/Document//4082541 (accessed 1 September 2013). 34 Interview #142, Mae La, 24 April 2012. 35 Interview #159, Umpiem Mai, 28 April 2012. 36 Interview #152, Umpiem Mai, 26 April 2012. 37 Interview #157, Umpiem Mai, 27 April 2012. 38 Interview #165, Mae Sot (phone interview), 3 May 2012. 39 Interview #139, Mae La, 24 April 2012. 40 Interview #170, Mae Sot, 5 May 2012. 41 A UNHCR representative acknowledged to me that ‘Government offi cials can’t admit openly that there is a parallel legal process in the refugee camps, even if in practice they might prefer it’ and said that in conversations with several Ministry Downloaded by [University of Defence] at 21:28 09 May 2016 of Justice prosecutors, all had denied the existence of a parallel justice system in the refugee camps. (Notes from interview #138, Mae Sot, 23 April 2012 . 42 Interview #134, Mae Hong Son, 20 April 2012. 43 Interview #155, Mae Hong Son, 27 April 2012. 44 Interview #77, Mae La Oon, 16 Feb 2010. 45 Interview #54, Mae La Oon, 28 May 2010. 46 Field notes, 11 March 2010. 47 Interview #99, KRC member, Mae Sot, 16 March 2010. 48 Interview #4, KWO member, Mae Sariang, 11 January 2009. 49 Interview #91, Mae Sariang, 12 March 2010. Chapter 7 Enacting interlegality: human rights and local justice

The concept of human rights requires very careful, critical scrutiny, precisely because it is something around which we are organizing our political energies and where we’re focusing our hopes and ambitions for the future. (James Ferguson, in Schouten 2009: 5)

Earlier chapters have illustrated the ways in which these refugee camps are sites of institutional pluralism and contestation. This chapter moves the analysis beyond the institutional level to examine the normative impact of these multiple layers of law, borrowing Santos’ notion of the interaction between porous legal orders as ‘interlegality’ (1987: 297) to analyze the inﬂ uence of international human rights norms on camp justice practice. It examines the rights that are promoted by international organizations and how these assertions are received by the refugee population: how rights are enacted, both in the sense of the creation of law, and in the sense of local efforts to incorporate rights standards in dialogues for change.

Culture and rights As legal discourse and as a source of praxis for activists, human rights aspire to transformation and emancipation. Yet these aspirations may fail if rights are seen as imposed by external forces (Banda 2006: 19; Speed and Collier

Downloaded by [University of Defence] at 21:28 09 May 2016 2000; Douzinas 2007). Critical perspectives on human rights have highlighted the risk of human rights imperialism, of a top-down system of codiﬁ - cation and institutional creation that is ethnocentric, positivist (e.g. Kennedy 2002; McEvoy 2007; Chimni 2006) and makes inﬂ ated ‘claims to ethical purity’ (Riles 2006: 55). The most pervasive anti-rights critique has been framed in terms of cultural relativism, which in effect claims that there is no universal normative consensus and that different societies have different social values and should be permitted to practice them accordingly (Donnelly 1984; Freeman 2002: 101–30). Scepticism regarding the universality of human Enacting interlegality 133

rights is often directed at their origins, as the product of a highly specifi c (European) philosophical tradition and thus unsuited to collectivist or conservative ‘Asian values’ or ‘African values’ (Mutua 2001; Ibhawoh 2003). Cultural relativism has dominated rights scholarship for decades and in its milder expressions serves as an important warning against hubris and cultural imperialism. However, rigid assertions of cultural relativism potentially legitimate any activity, however cruel or unjust, on the basis of indigeneity. This ignores several of anthropology’s key insights: that culture is a fl uid concept; that traditional practices are often relatively recent inventions (e.g. Moore 1989; Burns 1989); and that when claims to a particular course of action are vigorously asserted it is usually wise to determine who is making those claims and what their interests in a particular outcome might be (Chanock 2002: 41; Brown 2008). This fi nal point is of particular relevance in human rights debates, where rejection of rights on grounds of ‘culture’ are often made by those who are most threatened by change to an existing social order (Ignatieff 2003). Invoking ‘cultural relativism’ then, can be a means to stall political dialogue and inhibit transformation when in practice ‘rights- talk’ is not always imposed from above and resisted from below but may be embraced locally as a means to catalyze social change. However, perhaps the most damaging limitation of cultural relativist critiques is simply irrelevance: focusing on whether rights are truly universal freezes analysis of the relationship between global rights discourse and local justice practices at a stage which rights implementation has already far surpassed. The question is no longer whether rights are imposed upon, or irrelevant to, non-Western cultures, but what rights mean in these contexts (Speed and Collier 2000; Wilson 1997; Merry 2006a; Cowan et al. 2001). It is apparent that there is no consistent clash between local culture and global rights but a more fl uid transfer of knowledge: some norms are accepted while others are resisted, in a series of ‘shifting patterns of dominance, resistance, and acquiescence’ (Wilson 2000: 86). It is also evident that for human rights to effect social transformation they cannot be imposed by elites but must be taken up locally. Understanding how, where and when this happens is essential to ‘shift the focus from law’s violence to law’s possibilities’ (Margulies 1996: 193–4) and realize the potential of rights as a vehicle of transformation and even emancipation. Downloaded by [University of Defence] at 21:28 09 May 2016 In addition to the institutional reforms and codifi cation processes mentioned in the previous chapter, international organizations in Thailand have made considerable efforts to educate camp leaders and community organizations about human rights tenets. Activities have included workshops and training sessions, the production of Codes of Conduct for camp management staff, translation of international conventions such as the Convention on the Rights of the Child into Sgaw Karen, and the distribution of t-shirts emblazoned with excerpts from human rights conventions and treaties. Human rights talk has fi ltered into the community, bringing a new 134 Governing refugees

conceptual vocabulary and, for some, a sense of possibilities. In somewhat superfi cial ways human rights are embraced, such as the celebration of International Refugee Day, or International Elimination of Violence against Women Day (for which, in November 2008, I attended a ceremony organized by the KWO and received a t-shirt – produced by an Australian organization – expressing the text of UN Security Council Resolution 1820 in Sgaw Karen: ‘violence against women is a crime’). Yet at a deeper level, there is competition between the different orders. Caldeira (2006) suggests that it is normal for national governments to adopt rights discourses in different spheres, as in Brazil where human rights and democracy had a profound impact on political structures but barely touched ‘law and order’ institutions such as police and prisons. Similarly, the KRG has become adept at leveraging ‘rights-talk’ to describe the actions of the Burma Army (e.g. see Brooten 2004: 176) but there is rather less traction for the application of human rights principles to intra-community affairs. In disputes arising within the refugee camp, refugee leaders are considerably more likely to refer to ‘tradition’ and ‘culture’ than human rights principles. The following sections analyze the relationship between local and global norms in two key spheres – gender and youth. Gender and youth are crucial status hierarchies in Karen society, as well as prominent policy areas for international relief organizations. The policy focus on women and children has benefi ts for these sectors in resource allocation, but from an ideological standpoint it refl ects a status of perceived vulnerability and victimhood – of ‘womenandchildren’, in Enloe’s memorable phrase (1999). The special status of women and children in human rights law is further emphasized by the fact that the two most widely ratifi ed human rights treaties are the Convention on the Elimination of Discrimination Against Women (CEDAW) and the Convention on the Rights of the Child (CRC). Of course, governments’ public allegiance to human rights texts is often not refl ected in compliance with their terms (Hafner-Burton and Tsutsui 2005; Banda 2006). Nevertheless, the fact that women’s rights and children’s rights attract such overwhelming state support is at least indicative of a value of being perceived to adhere to standards in these areas, even if this adherence is largely illusory. Local reception among the refugee population in Thailand has been very Downloaded by [University of Defence] at 21:28 09 May 2016 different in relation to ‘women’s rights’ (where there is strong local activism and a reasonable degree of local support for changing practices) and ‘children’s rights’ (where rights are often seen as an impediment to community control of delinquents). The following sections examine how global rights standards differ from traditional understandings of gender and generational norms, and outline the role of local actors in translating international rights standards to the local population. The chapter concludes by examining human rights ‘claims-making’ in a case before the Justice Committee in Mae La Oon. Enacting interlegality 135

Women’s rights as human rights Though few would now question gender rights as a legitimate component of human rights, the centrality of women’s rights to human rights is surprisingly recent (dating from the early 1990s), and the result of a deliberate political strategy to reframe feminist claims for gender equality as human rights (Bunch 1990; Friedman 1995: 19; Charlesworth 2005). Engaging issues of spousal abuse and sexual violence in terms of human rights law required a repositioning of human rights law from the ‘classic’ focus of protection of citizens from abuse of power by state offi cials and into the private sphere of intra-community and inter-familial relations (see e.g. Friedman 1995; Peters and Wolper 1995). Defi ning women’s rights as human rights also required glossing over diversity to represent women as united across cultures by the shared experience of gendered practices, including, for example, domestic violence, rape, control over reproductive capacity, discrimination and cultural expectations of femininity (Peters and Wolper 1995: 6). However, assumptions of universality are particularly tenuous here. In most societies, sex and gender lie at the heart of societal hierarchies and strategies of community ‘boundary- maintenance’ (Donnan and Magowan 2010: 3; Berkowitz 2012). Beyond this very general statement, the nature of sexual taboos and prohibitions varies greatly and even the most apparently straightforward claim for universalism – such as a prohibition against sexual violence – becomes more complex when measured against the range of existing cultural perspectives. Furthermore, the construction of ‘women’ as a universal category underplays other identities that may be as or more important than gender (such as nationality or ethnicity) in terms of both personal identifi cation and vulnerability to violence (Hyndman and de Alwis 2009: 88). For example, a female Karen refugee would very probably identify more closely with a Karen man than with a Western woman, and – given the discrimination and persecution that many Karen have faced inside Burma – may also consider that her ethnicity has a greater infl uence on her risk of attack than her gender.

Sexual violence in refugee situations

Downloaded by [University of Defence] at 21:28 09 May 2016 The categorization of women as vulnerable is even more marked for refugee women. Protection of refugee women and refugee children is an ‘organizational priority’ for the UNHCR (UNHCR 2006c: para 68) and a commitment to tackling violence against refugee women has been codiﬁ ed in numerous UNHCR documents (1991, 1995, 2001, 2004a, 2008c). There is a concentration on sexual violence in refugee situations, which has been described as a ‘pandemic [. . .] a systemic form of human rights abuse on the scale of a global epidemic’ (UNHCR 2001: foreword). In Thailand, protection of refugee women has similarly been a central area of intervention for international organizations. Principal concerns include 136 Governing refugees

early marriage, forced marriage, rape, domestic violence, human traffi cking and a lack of access to justice, with Sexual and Gender Based Violence (SGBV) considered ‘a persistent and far too prevalent problem’ which the camp justice structures lack capacity to respond to effectively (UNHCR 2006b: 19, 22). Others have referred to ‘alarmingly high levels’ of SGBV (Kenny and Lockwood-Kenny 2011: 223). In truth, it is very diffi cult to determine the extent of domestic violence and sexual violence in the refugee camps. Numerous agencies are involved in recording incidents but each agency’s records are confi dential and are not pooled. In one survey of more than 2,000 refugees from fi ve camps, twenty-one reported experiencing rape or attempted rape (IRC 2010: 46). Another survey of 861 women from three camps found that sixty-eight had experienced intimate partner violence, with twenty-one experiencing physical (rather than threatened) harm (Falb et al. 2013). These rates are considerably lower than those reported to occur in other refugee camp settings. Hathaway (2005: 444) writes that:

Somali refugee women in search of ﬁ rewood near three UN refugee camps in Kenya were also routinely raped, resulting in a sexual assault rate seventy-ﬁ ve times higher than would be expected in a community of comparable population. [. . .] In Tanzania, for example, 95 percent of the cases of refugees seeking protection inside Ngara camp related to domestic violence.

It is important to be aware of ascribing characteristics to a refugee situation simply on the basis that it is a refugee situation (Ibhawoh 2003) and to remember, as one of my interviewees pointed out, ‘Don’t say violence is part of Karen culture. Some husbands love and take care of their wives.’ Nevertheless, it is reasonable to assume that many more SGBV incidents are committed in the Thai–Burma border camps than are reported, as is typically the case in relation to such crimes. Furthermore, women’s organizations believe that SGBV offences are increasing both in number and in seriousness, describing an increase in cases such as the rape of elderly women, of young children, of children with mental disabilities, and of rape using weapons. Interviewees in Mae La claimed that the worst such offences were committed by men who came from outside the camp with the express intention of Downloaded by [University of Defence] at 21:28 09 May 2016 committing sexual violence against refugee women.1 UNHCR Protection Programming for the camps includes SGBV awareness and prevention as a core element (UNHCR Thailand 2006b). The projects included under this ambit include training and education, funding and support to ‘SGBV committees’ in all seven Karen camps, the drafting of a Code of Conduct to be signed by all security workers (including the Thai Or Sor), the funding of a safe house and provision of psychological care for victims of SGBV, a ‘community theatre group to combat SGBV stigma’ and general ‘support’ for women’s and children’s work conducted by community-based Enacting interlegality 137

organizations (UNHCR Thailand 2007: 3). In 2013, UNHCR-Thailand’s budget for gender-based violence and protection of children combined was US$2.2 million, with an additional US$3.5 million for ‘law and policy’ and ‘access to legal assistance and remedies’ (UNHCR 2013: 234). SGBV lacks a concise and consistent defi nition but generally refers to acts such as domestic violence, verbal abuse, physical and sexual abuse of children, forced marriage, marital rape, rape, sexual harassment, forced prostitution and traffi cking (e.g. UNHCR 1991, 2001). The notion of ‘SGBV’ is diffi cult to convey to Karen and Burmese refugees, for both linguistic and conceptual reasons. There is no Karen term equivalent to the English ‘gender’, so the most straightforward translation of SGBV is ‘violence-against-men-or- women’. This translation removes the gendered nature of the violence and thus the distinction between acts of SGBV and any other act of violence.2 Furthermore, the content of ‘SGBV’ does not always accord with refugees’ conceptions of sexual wrongdoing. The following sections examine the disjuncture between international norms and local realities regarding gender- based violence and consider the role of local interlocutors in educating the wider refugee population about women’s rights.

Defining sexual harm: adultery, rape and domestic violence As was explained in Chapter 4, some of the most serious taboos in Karen society are related to extra-marital and pre-marital sex. Adultery in particular is not considered a victimless crime but the cause of serious emotional distress and the root cause of other harmful acts, such as physical violence or the abandonment of dependents. The Mediation and Dispute Resolution Guidelines have ‘de-criminalized’ adultery but these guidelines have not yet been implemented in all camps. In Mae La Oon, adultery was still considered a very serious crime and, for those who were familiar with the terminology, it was also considered to be a form of ‘sexual-and-gender-based violence’. This was the case even among persons who understood that technically adultery was not SGBV, such as members of the NGO-trained SGBV committee:

If adultery is a crime or not, it depends on the culture. For ARC rules, it Downloaded by [University of Defence] at 21:28 09 May 2016 might not be a crime, but for the Karen community it is one husband, one wife. Not one husband and two or three wives.3

Rape is recognized as a crime in terms of the KRC camp rules but I found that the term ‘rape’ was used more readily when the perpetrator was a non- Karen, whether it was an SPDC/Burma Army soldier, a Burman refugee, Thai citizen or Thai ofﬁ cial. Incidents of pre-marital or extra-marital sex between a Karen man and woman were described in terms such as ‘man– woman problem’, which implied a mutual transgression.4 This was the case 138 Governing refugees

even where there were indications that the woman had not consented to intercourse. Several cases were described as ‘adultery’ that seemed to me more aptly characterized as rape. In one such case, a young woman with learning disabilities became pregnant after repeated intercourse with a married man. Though the girl denied that she had consented, a judge told me that there had been repeated sexual acts, and he suspected she had received payment (there was no evidence of this) and thus she was also at fault. The girl’s parents refused to accept payment from the man and elected to raise the girl’s baby themselves. Another case that was punished as adultery but seemed inappropriately characterized involved a man ‘giving’ his wife for sex to another man to whom he owed a debt. In this case, all three were punished: the wife and the creditor because they had committed adultery; the husband because he had permitted his wife to commit adultery. There was no inquiry into whether the woman consented to this arrangement and the possibility that it may have been a case of rape did not appear to have been considered. In these cases, sanctions indicated where blame was considered to lie but the language of ‘rape’ was never used. So, in the fi rst case described, the young woman was not punished in any way but the man was sentenced to four months detention and a payment of 4,000 baht to the girl for costs of childbirth (as her parents had rejected fi nancial payment). In the second case, the woman was ordered to pay a fi ne of 1,000 baht, the man with whom she committed adultery was to pay a fi ne of 2,000 baht, and her husband was sentenced to four months detention and a 2,000 baht fi ne. During my fi eldwork, only one case of rape was recognized as occurring in Mae La Oon. This was a case of exceptional seriousness, of the repeated rape of multiple young women in a dormitory by a man who held several positions of responsibility within the camp, including as the manager of this dormitory. It was referred to the Thai authorities, and the perpetrator eventually received a twenty-year prison sentence. This last case suggested that when a case is understood and recognized as rape, camp leaders would willingly refer it to the Thai criminal justice system. One diffi culty is, as the other examples indicate, that the issue of consent (which is so crucial to defi nitions of sexual violence in Western legal systems and in international human rights law) is not the core concern. For example, Downloaded by [University of Defence] at 21:28 09 May 2016 if sex occurs repeatedly, consent is presumed. Even very experienced women’s activists were not entirely sure whether this constituted rape:

Maybe it happened as a rape from the beginning and then later. [. . .] because I think for the Karen people, the ﬁ rst time when the man rapes them, they don’t feel like it’s a rape, or maybe they do feel like it’s a rape, but they don’t report it, and if it happens a few times then it’s more like consensual.5 Enacting interlegality 139

This was apparent in the transcript of an interview conducted by security staff in Mae La Oon regarding a case of adultery. The transcript recorded the following exchange:

Security: Tell me about the mistakes you made. Woman: I broke the law. Security: Did he force you or did you agree? Woman: The ﬁ rst time I didn’t agree but after several times we agreed. Security: You made a mistake. Will you accept punishment? Woman: Yes.6

It is clear that there are different understandings of harm in Karen custom and Western law, and this is one reason why incidents of sexual violence are often not reported to camp leaders – because they are not universally understood as harms. In a recent study involving close to 400 young people in Mae La Oon and Mae Ra Mo Luang refugee camps, one third of young women and 30 per cent of young men surveyed agreed that, ‘It is ok if sometimes a boy forces a girl to have sex if he loves her’ (Benner et al. 2010: 5). Similarly, domestic violence may not be considered as a crime, but as legitimate ‘discipline’ of an unruly wife. Researching perspectives on gender violence among male Karenni refugees living in Thailand, Brezovich found that consensual sexual acts such as adultery were described as violence, while a man hitting his wife was not considered to be violence because ‘violence is seen as different from “teaching a lesson”’ (2005: 30).

Reporting and accountability However, weak accountability for sexual violence is not solely due to lack of understanding of what sexual violence is. Barriers to effective responses include an inability for refugee leaders to resolve cases (e.g. if the perpetrator is a Thai citizen) or unwillingness to get involved in domestic disputes:

Sometimes a case seems like rape at ﬁ rst but after examination it is more like a relationship.7

Downloaded by [University of Defence] at 21:28 09 May 2016 Domestic violence happens again and again and the section leaders say ‘Oh you are always having problems’ and try to ignore it.8

Some Security members and section leaders denied that domestic violence occurs at all within Karen society. Others recognized it was a problem but found it impossible to police, citing cases in the past where they had intervened but the woman had pleaded to release the man without punishment.9 There was an echo of the discourses of ‘good victims’ and ‘bad victims’, described by Merry (2003) in relation to domestic violence 140 Governing refugees

prosecutions in Hawai’i, and women who withdrew cases from the system fell into the latter category. As a short-term remedy, victims of domestic violence can seek shelter in a camp ‘safe house’ that is funded by UNHCR and operated by refugee women’s organizations. There are no good long-term solutions. Divorce is possible in principle but in practice pressure may be placed on the injured party to remain within the marriage.10 Resettlement is also sometimes used as a protection response for victims of domestic or sexual violence, though this presents its own challenges for a woman (and often children) who have experienced serious victimization. It also does nothing to punish the perpetrator or protect others in the camps. Cases of rape can be resolved at camp level if that is the preference of the victim, and provided there are no aggravating circumstances. The Mediation and Dispute Resolution Guidelines recently circulated by UNHCR and IRC stipulate that rape cases must be referred to Thai courts if the victim is aged under fi fteen or if there were multiple perpetrators, the assault caused grievous injury, weapons were used, or if the perpetrator holds a position of authority in the refugee camp. Some of the diffi culties of referring cases to Thai courts were discussed in the previous chapter.11 To encourage reporting of sexual violence, UNHCR created ‘SGBV committees’ in each of the seven ‘Karen’ camps. First established in 2003 as a community awareness raising process, the SGBV committees were subsequently developed to take a role in recording, responding to and referring cases of sexual violence and were ‘provided with the knowledge and resources so that they can take the lead in responding to SGBV’ (UNHCR-Thailand 2008). In undertaking this role, the SGBV committees were to operate independently of the camp administrations in providing a confi dential avenue for reporting SGBV offences directly to UNHCR and its implementing partners. Expecting SGBV committees to ‘take the lead’ in SGBV case management has been extremely contentious, in large part because it placed the SGBV committees in opposition to existing services that were operated by the KWO (KWO 2011a, 2011b). In one camp, Umpiem Mai, the relationship between the SGBV committee and the camp committee deteriorated to such an extent that in 2010 the SGBV committee was expelled from the camp. A series of emergency Downloaded by [University of Defence] at 21:28 09 May 2016 meetings were held between Umpiem camp committee, Karen community organizations and international NGOs to try and resolve the issue. A decision was taken at a meeting in June 2010 that the SGBV committees should be disbanded in their current form, renamed as ‘community peace teams’ to conduct outreach and education (but not case management) and fully incorporated into the camp management hierarchy. This decision was later rejected by UNHCR on the grounds that their representative at the meeting did not have the authority to agree to such a decision. Three years later, the matter remained offi cially unresolved. Instead, each camp had reached its own accommodation with SGBV committees. In Mae La Oon and Mae Ra Ma Enacting interlegality 141

Luang, the decision to rename SGBV committees as ‘Community Peace Teams’ had been implemented, while in other camps they continued to operate as SGBV committees and continued to undertake case management activities. There is a perception among international NGO staff that ‘the camp leaders may put pressure on victims of SGBV not to report cases and not to seek remedy through the Thai legal system.’ (IOM, quoted in Huguet and Punpuing 2005: 19; UNHCR 2006b: A8).12 However, low-reporting of sexual violence is not unique to refugee situations or to the Karen community. Some aspects of camp life – lack of privacy, inability to leave or move away, language barriers – make reporting sexual violence particularly challenging for refugee women (Lattu 2008). Yet even in the UK, despite efforts to ‘take sexual offences seriously’, ‘sexual violence against women in general and rape in particular are among the least reported and most under-enforced crimes’ (Lacey et al. 2003: 478; Home Ofﬁ ce 2002).

Translating ‘SGBV’ Merry (2006b) refers to the appropriation of rights standards into local practice and dialogue as ‘vernacularisation’. Though it is not a consistent or predictable process, for human rights vernacularisation to occur at least some individuals and organizations must see the value in human rights for their society and translate those rights for others. The ‘translation’ that is required is not only of words but also of the shaping concepts and ideas. It is precisely the absence of such cultural translation that often limits the impact of international human rights education (Englund 2006). For refugees from Burma living in Thailand, vernacularisation of human rights has primarily been undertaken by women’s organizations. In the Karen camps, the largest organization is the KWO. The goals of KWO and international organizations are broadly aligned; to raise awareness of sexual violence within the refugee population, to encourage reporting of sexual violence or domestic abuse, to ensure that sexual violence against refugee women is taken seriously by camp justice systems and to support access to Thai courts where this is the choice of the victim/survivor. There are differences in how these goals are pursued. Community-based organizations are typically Downloaded by [University of Defence] at 21:28 09 May 2016 pragmatic, concerned with developing strategies for practical action rather than with absorbing the ﬁ ne detail of the philosophical or legal frameworks. This has been documented particularly with Muslim women’s activists who advise other women on approaches which work within Shari’a law, rather than challenging that law on human rights grounds (Merry 2006b; Snajder 2005). In effect, their work seeks to achieve the best form of justice available within the system as it is rather than as they would like it to be. A similar approach is apparent in a document produced by a coalition of Burmese women’s organizations, an ‘Automatic Response Mechanism’ for processing reports of sexual violence against refugee and migrant women 142 Governing refugees

(Women Exchange 2003). This text was produced after four years of consultation with more than 600 women from fourteen ethnic groups, and outlines a ten-step process from the fi rst contact with a victim to a fi nal case decision. The document recognizes that above all ‘the victim has the right to take the case to court’ and that her decisions at all points must be respected and supported. However, it also acknowledges that migrant and refugee women face unique vulnerabilities that may undermine their safety or well- being in seeking justice for sexual violence: the offi ces of a refugee community organization may be raided or closed down; confi dentiality is hard to achieve in refugee camps; international agencies may not be able to ensure protection in areas where they do not have a permanent presence; ‘if the perpetrator was a Thai soldier, police, person of authority, there may be threats to the women’s organization, to the camp committee, to the refugees’ (2003: 19). The document simultaneously explains legal entitlements, practical challenges and basic principles in responding to victims of sexual violence, in an approach that is rooted in human rights law but which is also fully aware of the gap between rights ideals and their realization. Of course, the international human rights framework is much more than human rights law but is rather an entire system with its own organizations, norms and ways of thinking (Merry 2000b: 90). Local actors must learn to communicate in the terms understood by the international donors and agencies that control the fl ow of resources and ideas. This leads ‘knowledge brokers’ to occupy a somewhat ambiguous position, caught between the local population and the international power-holders of the human rights system. It is extremely diffi cult to get the balance right, as Merry cautions: ‘inter- mediaries are always suspect because they are not fully in one world or the other’ (2006b: 48). Those who learn to navigate the international system can attract resources, status and prestige, but in doing so they risk losing their local authority and grassroots mandate: in effect, they become more like the international elite and less like the community they seek to represent. Conversely, to maximize their local impact, ‘knowledge brokers’ may avoid ‘rights-talk’ in favour of a more culturally moderated approach:

Women’s rights. Trying to change behavior it is taking a long time. It’s very sensitive to raise this within the community, really sensitive for Downloaded by [University of Defence] at 21:28 09 May 2016 staff. Some men are really against this. Mostly leaders understand about women’s rights but it takes a long time. You have to have a plan to work. You can’t challenge the leaders directly but have to work with them civilly, talk to them in a nice way so that they will understand. Go slowly and speak nicely in a soft way. If you just shouted ‘women’s rights’ you would never get there.13

As was discussed in Chapter 4, the KWO is part of the KNU ‘family’, politically (in the sense that it belongs within the KNU organizational Enacting interlegality 143

structure) and often literally (many members of KWO have husbands and sons within the KNU/KNLA). This relationship with the KNU is no longer a core element of KWO’s public face as a women’s rights organization but it continues to have a bearing on its local reception. Senior staff members of KWO do not live in the refugee camps but are based in Mae Sot or Mae Sariang. Thus, the views of policymakers within KWO will not necessarily be refl ective of the Karen ‘grassroots’, whether refugees, internally displaced or those living in Karen State. To some extent, this is unavoidable. The truly subaltern of Karen society are illiterate, impoverished, coping with protracted displacement and hardly in a position to mediate between international organizations and the wider community, while any organization with tens of thousands of members is liable to have a gulf between executive staff and mainstream membership. Nevertheless, it does impact somewhat on the representativeness of the organization and highlights the diffi culties of determining who is entitled to speak ‘for’ a population. However, KWO’s human rights advocacy has not brought it closer to international organizations at the expense of its local legitimacy. Rather, it has prioritized local legitimacy at the expense of its relationships with international actors, repeatedly challenging NGOs about policies and programmes it feels do not appropriately serve refugee women. In 2011, KWO circulated a document to international organizations working in the camps outlining its concerns: that international SGBV programming had dupli- cated existing services at community level, leading to confusion about where to report cases and how to manage victim support; that victims were being given insuffi cient information to make informed choices and insuffi cient support in proceeding with cases in the Thai justice system; that the camp- based structures were not being strengthened and that UNHCR and NGOs ‘were ignoring and undermining existing community systems and structures’ (KWO 2011a, 2011b). In return, Olivius (2013: 13–14) found that staff of international NGOs fail to acknowledge the depth and range of KWO’s role and perceived the organization as ‘political, illegitimate and less able to adhere to “international standards”’. However, at camp level KWO commands clear recognition as the primary actor in community social welfare. This is the case even among non-Karen women, such as a Burman woman I interviewed (who was also Downloaded by [University of Defence] at 21:28 09 May 2016 a member of her camp’s Burmese Women Organization and had worked for two years with the camp SGBV committee). This woman described her experience of domestic abuse, which her section leader had repeatedly failed to address appropriately. She said that perhaps she would turn to KWO to see if they could help her:

KWO can help a lot of women. Even when they face difﬁ culties, they try to support and help the women. Everywhere, KWO stand bravely and ﬁ rmly for the women. They go into the justice system with the women. 144 Governing refugees

Because of their work, many women trust them and rely on them. Many KWO work in progress because they work for the community and the women can trust and depend on them.14

Youth delinquency and intergenerational relations As with refugee women, refugee children are recognized as in need of special protection (UNHCR 1987, 1994, 2007). However, within the refugee community in Thailand there is no mobilization on children’s rights to match that on women’s rights. Though there is a Karen Youth Organization (KYO), its activities are primarily focused on youth fellowship and cultural celebration of ‘Karenness’, thus with upholding the traditional order rather than challenging it:

The ﬁ rst responsibility of KYO is to organize young people not to make problems in the community [. . .] for special Karen days, KYO takes responsibility to arrange and perform the Karen traditions. We can’t preserve all because we don’t have all the things for Karen culture here. But we try to do what we can to make sure that young people don’t forget their Karen traditions.15

Keck and Sikkink (1998: 196) have suggested that one reason for the success of gender rights as a global campaign is the existence of ‘transcultural consensus’ on the core issues of ‘preservation of dignity [. . .] protection from physical abuse [and] concern with bodily harm’. The international approach to youth rights, in contrast, refl ects a specifi c construct of childhood as a protected, age-bound space. The key text here is the Convention on the Rights of the Child which is, as was mentioned in the introduction to this chapter, the single most ratifi ed human rights treaty. This is somewhat anomalous, as it is also recognized to be a particularly problematic text, ‘which has at its core a universalized and essentialized view of the child based on Western assumptions about children’s best interests and a single standard of age (18 and under)’ (Bluebond-Langner and Korbin 2007: 244). Public discourse and social policy present two dominant representations of children, which are both polarized and contradictory: on the one hand, Downloaded by [University of Defence] at 21:28 09 May 2016 children are viewed as innocent and vulnerable, yet where they contravene these expectations they are often demonized as irredeemable delinquents (Montgomery 2001; Muncie 2004). These confl icted attitudes are apparent in domestic policy and in international development and humanitarian assistance. In Britain, the ‘young offender’ gained particular notoriety as a special policing category during the eighteenth century and perceptions of youth as a high-risk policing category remain prominent in criminal justice policy today (Ariès 1973; Muncie 2004). The picture is similar in societies as diverse as Rwanda (Sommers 2006) and Canada (where Ericson Enacting interlegality 145

and Haggerty [1999] describe the myriad ways in which young people are policed, monitored and controlled, including a police presence in schools, intelligence-gathering on young people, and the use of closed-circuit TV cameras in schools and housing areas). In international humanitarian work, the emphasis is ﬁ rmly on the vulnerability of children. In the sphere of refugee action, this is quite notable, with documents focusing on victimization and the need for international organizations to ‘take charge’ of children’s interests (e.g. UNHCR 1994; Save the Children 2007).16 In the emphasis on protecting children from adults, there is little recognition of the possibility of children as offenders or of how to respond when children are not vulnerable and weak but are assertive or aggressive. The following section examines the clash between the contradictory approaches to children as vulnerable/threatening in Mae La Oon, where youth delinquency is interpreted by refugee leaders as a source of disorder and a challenge to camp management but is minimized by international staff as ‘acting out’ the frustrations of refugee life.

Delinquents or victims? Age is a crucial status indicator within Karen society, as was explained in Chapter 4. Power hierarchies placing old men fi rmly at the head of the social order have been shaken by prolonged encampment, where the skills that are most valuable for survival are not folklore knowledge or farming techniques but English or Thai language, confi dence in negotiating with outsiders and other such non-traditional techniques. Young people brought up in the refugee camp have better access to education than their parents and while there is still a strong sense of respect for elders (and for most, a continued commitment to the Karen revolution), there is also a degree of empowerment among young people that threatens the prior hierarchy. In what is arguably an expression of this new generational confi dence, all of the Thai–Burma border camps have recently begun to experience problems of youth delinquency, with young men (and sometimes young women) organizing into territorial gangs, drinking alcohol and using drugs, fi ghting with each other and harassing other camp residents. The problem manifested fi rst and has been most serious in Mae La, the largest refugee camp on the Downloaded by [University of Defence] at 21:28 09 May 2016 border and the closest to the Thai town of Mae Sot. Camp security in Mae La described youth crime as happening ‘constantly’, with ‘too many cases to count’.17 In Mae La Oon, the organization of youth gangs has been less serious. Nevertheless, at any evening event (such as a school fundraising concert or church singing competition) camp security would be detailed to guard the event and prevent any trouble from young people. In 2010, a fi ght between two gangs resulted in serious injury to two young men who had to go to hospital in Chiang Mai for treatment. The offenders – more than twenty of them – were arrested by camp security and detained. Other camp residents 146 Governing refugees

had little sympathy for the offenders but were angry at their behaviour and afraid that it would have repercussions for the camp, as the Thai authorities had been informed of the incident to request permission to transport the injured boys to hospital. Camp leaders initially responded to youth delinquency through the normal justice sanctions of ‘instruction’ or guidance in the fi rst instance, followed by detention in the event of recidivism. When this failed to eradicate the problem, efforts were made to identify root causes of the behaviour. The sale of alcohol was banned in the refugee camps, and in Mae La Oon a ban was also proposed on the circulation of foreign movies (later reversed after an outcry from camp residents who were horrifi ed at the notion that the primary avenue of entertainment in the camp might be removed).18 These efforts proved unsuccessful in changing the behaviour of young people, and harsher sanctions were attempted, with recidivists referred to the KNU for punishment. As was detailed in Chapter 6, the punishment was usually to assist KNLA troops by cooking or carrying materials. Offenders were permitted to return to the refugee camp after they had proved that they were willing to abide by the camp rules. Staff members of international organizations view youth delinquency much more leniently than the refugee leaders, claiming that serious violence is rare and that young people may fi nd themselves in trouble for simple self- expression – for example, a teenage boy growing his hair long, or wearing Western clothes such as tight jeans or studded belts. Foreign staff also place greater weight on the challenges of encampment, particularly the confi nement and lack of opportunities that young people are faced with. The focus of NGO youth programming is not with the activities of youth gangs but with the needs of orphaned children, physical or sexual abuse against children, early marriage, and the risk of recruitment of child soldiers (UNHCR Thailand 2006b: 22). Refugee leaders’ desire for a more punitive approach to youth delinquency confl icts with these agendas for human rights and youth protection. Detention of young people has been resisted by international agencies and requests for international support to build ‘rehabilitation centres’ for young offenders were rejected. The practice of referring young offenders to the KNU for punishment was considered to be forced recruitment of child soldiers and Downloaded by [University of Defence] at 21:28 09 May 2016 thus a violation of international law:

We used to send them [young offenders] to the front lines, they would go for maybe one year and they did not like it, they would become afraid and want to come back and change their behaviour. But UNHCR do not want us to do this, because they say it is child soldiers. In UNHCR, if they are 16, 17, they call it child soldiers and say these young people cannot ﬁ ght. So now what can we do? We can do nothing. KWO do a lot, but it is for women, only for women. For the boys there is nothing. What can we do with them?19 Enacting interlegality 147

The thing is, here, sometimes, UNHCR get involved, so people under 18, when they are detained UNHCR say, ‘Oh, you can’t detain a child’. They always try to hold them to the highest standards.20

We don’t have a place to detain children, to give them education. We want a place where they can be held at night but during the day they can go to school and continue their education. But we don’t have anywhere like that in the camp.21

If I catch one of them stealing I will beat them myself before I hand them over to Security. Security can’t do it because they are afraid of children’s rights.22

As these speakers indicate, criticism of existing responses to youth delinquency has not been matched with the development of effective alternatives. The Mediation and Dispute Resolution Guidelines provide for youth crime to be resolved through group conferencing or arbitration, with sanctions to include a warning, promissory note or CSO. Detention is permitted only in exceptional circumstances to prevent escape or further violence.23

Why now? Moral panic and an ‘othering’ of deviance Muncie (2004: 52) notes that young people have been blamed for a perceived decline in social standards for centuries and that in every generation the ‘moral decay’ is encountered as unprecedented: ‘“Juvenile delinquency” is as much an index of adult perceptions as a description of youth behaviour [. . .] What actually constitutes “young offending” is in a constant process of (re) invention and (re)defi nition.’ Youth offending in the US and UK is at a point much further along the delinquency spectrum than in these camps, with youth gangs involved in organized crime, drug traffi cking and serious violent offences. Though the level of delinquency in Mae La Oon and the other border camps remains low in comparison with these national jurisdictions,24 youth gangs are a very real problem. Young people who drink alcohol, fi ght and cause problems are a small minority within the camp overall, but their activities have a signifi cant negative impact on camp life, and in doing so, Downloaded by [University of Defence] at 21:28 09 May 2016 make an already challenging environment even harder. Camp leaders and residents largely see youth misbehaviour as a challenge to authority that should be met with a tough ‘law and order’ response. The topic of youth delinquency has been the subject of meetings, discussions and consultations over a period of years and is consistently identifi ed as the primary policing problem in the camps. The level of concern has similarities with Cohen’s (2002 [1972]) notion of a ‘moral panic’, the crucial feature of which is not a deviant act per se but the response it provokes in scapegoating a category of perceived troublemakers and in the disproportionate response 148 Governing refugees

to the threat. The disproportionate nature of the response does not mean that there is no objective threat – on the contrary, the exaggerated response is possible precisely because there is a real vulnerability that is susceptible to exaggeration and a real perception that important values are threatened. The task for researchers is to ‘look beyond the tangible, the immediate, and the material, and try to understand, symbolically and culturally, what the panic represents to the participants involved’ (Goode and Ben-Yehuda 2009: 31). When I asked refugees why they thought young people were organizing in gangs and using violence, a variety of suggestions were offered, including the inﬂ uence of foreign movies and other foreign inﬂ uences and increased contact with Thai society:

Is it because the management and camp committee is not good or is it that the young people want to follow their needs, their own way? Is it because here we have a huge population in the same area? In our country we live in a village, only a few people, less people than here. Or is it because they see the movies, and they see the ﬁ ghting movies and they want to copy? Also, the other thing is that nowadays there are many human rights for the young people and everyone, so they use their rights in the wrong way. There are many things that we wonder about.25

This man was unusual in his suggestion that the living conditions of encampment may have contributed to youth delinquency. More often, refugees linked youth violence to increased Thai and international infl uences, and in particular to the beginning of the international resettlement programme. It is not uncommon that refugee situations alter intergenerational relationships (Hampshire et al. 2008: 31; Turner 2010), and in this case it might be that young people who plan to leave to another country no longer feel obliged to live by the camp rules or community expectations. Since the resettlement programme began, there have been many new arrivals to the camps, and these new arrivals are also blamed for many of the problems.26 The ‘real’ proximate cause is arguably irrelevant. More important is the local perception that blame rests with the external infl uences of foreign movies, Thai society, international resettlement and human rights education. This attribution of blame suggests that the real threat that is perceived by Downloaded by [University of Defence] at 21:28 09 May 2016 refugee leaders is of outside forces, of Thai society and of international resettlement. Reporting a visit to the Karen insurgency headquarters at Manerplaw in the early 1990s, Anderson (1992: 175–6) described KNU leaders as feeling their society to be under pressure from increased contact with outside infl uences including Thai society, and the Burmese student revolutionaries that had come to the area. In the refugee camps, this contact with ‘external’ forces has increased exponentially. Some of the camps are located in close proximity to Thai towns and have considerable contact with Thai people through work, trade and travel. Even the more isolated camps, Enacting interlegality 149

such as Mae La Oon, have some contact with Thai society through fi lms and movies, as well as the more threatening spectre of police, military and immigration offi cials. International infl uences are even more prominent, with foreign staff of international organizations and foreign volunteers coming to visit and teach in the refugee camps. This cosmopolitanism is frequently cited by refugees as an asset to their experience in camp, bringing the opportunity to learn English and to meet English speakers. However, international volunteers often espouse very different cultural values from the Karen ideal: sex before marriage, parenthood without marriage and homosexuality are issues that foreign volunteers discuss freely but which are considered ‘very dangerous and frightening’ by many Karen refugees.27 Chanock (2002: 44) remarks that during the colonial era, Western publics perceived colonial territories as spaces of sexual licentiousness which had to be controlled and ‘civilized’. Now, the position is reversed, with former colonial sites perceiving Western society as dangerously, contagiously promiscuous. This has an impact on the reception of human rights, which are associated with foreigners and thus are resisted not only because they are perceived as culturally inappropriate but because they are considered a vector for moral deterioration. Oh and Stouwe (2008: 602) found that that the principal and vice-principal of the only school in Ban Don Yang camp ‘attributed a rise in teenage sexual activity to the proliferation of human rights education by international NGOs’, while a teacher in Mae La Oon told me that when he proposed teaching a course on human rights to high school students, he was reprimanded by a member of the camp committee:

Why do you want to teach that? Human rights is a foreigner thing. If you have a machete, and you sharpen it, it can be dangerous. If children don’t know to use it, they can cause a lot of damage. If you teach this subject, you will damage our community.28

The impact of rights: claims-making and forum- shopping Human rights aspire to transform power relations and as such some collision Downloaded by [University of Defence] at 21:28 09 May 2016 with existing power structures is inevitable; the question is to what degree this collision is supported by segments of the local population and viewed as relevant to their needs and desires (e.g. for equality, for recognition). One of the means by which social problems or deviant behaviours come to be deﬁ ned as problematic is through ‘claims-making’, deﬁ ned by Spector and Kitsuse (2001 [1977]: 78) as ‘a demand made by one party to another that something be done about some putative condition’. The objective truth of a claim is irrelevant; more important is the desired goal. 150 Governing refugees

Claims-making is a key tactic in human rights practice, as Ferrante (2008: 184–6) neatly highlights by contrasting the language of a US Department of State report on human rights practice in China with that of a Chinese Government human rights report on the United States. Speed and Collier (2000: 80) suggest that claims-making is also an important indicator of the acceptance of human rights values, i.e. when local actors assert their claims in rights language it demonstrates their support for human rights values. It seems likely that this underplays the potential for exploitation of rights language. People engaged in claims-making are, by defi nition, in pursuit of a particular outcome, and there is a strong possibility for strategic deployment of rights language that (deliberately or accidentally) misrepresents real beliefs. Rights claims-making may also be an attempt to exploit the interstices between legal orders. Legally pluralistic societies afford opportunities for strategic exploitation, of ‘forum-shopping’ for the best probable outcome (Tamanaha 2008; Benda-Beckmann 1981). In the refugee camps in Thailand, for the most part, the cohesive nature of the camp governance hierarchy and the general acceptance by refugees that they should take a case ‘step by step’ seemed to work against forum-shopping. However, as the previous chapter explained, the cohesion of this hierarchy may be weakening. Perhaps because of these changes, I did fi nd some indication that the refugee population is now prepared to exploit the tensions between different legal orders and ‘forum shop’ for preferred outcomes. Unfortunately, this was not always done with the motivations human rights activists might hope for. This dynamic is well illustrated by a case that occurred between a twenty- one-year old Burman woman (‘Moe Kyi’) and a seventeen-year old Karen man (‘Saw Htoo’).29 Moe Kyi and Saw Htoo had been observed spending time alone in an empty house by a neighbour, who reported the case to the section security team. The couple were arrested and the section leader decided that they should be married immediately. Saw Htoo was tied up so that he could not run away, but his father heard of the arrest and came to rescue his son. Subsequently, Moe Kyi and her parents requested the camp leaders to order a marriage between the couple, on the grounds that the couple had had sex, and that Moe Kyi had consented with the belief that they would marry. When the dispute came to the Justice Committee it proved very diffi cult Downloaded by [University of Defence] at 21:28 09 May 2016 to resolve. Moe Kyi and her family insisted that marriage was the only solution and refused money or any other form of compensation. It was agreed by those present that if the couple had had sex, the appropriate response in Karen culture and in Burmese culture was that they should marry. Saw Htoo’s parents insisted that nothing had occurred between the couple. Besides, they said, Saw Htoo was under 18 and thus ‘a child’ too young to consent to a marriage. Various sectors of the camp structures became involved: the Child Protection Committee (an NGO-supported, refugee-staffed initiative) supported the boy and his family, and were concerned that his detention and Enacting interlegality 151

the proposed marriage were contrary to the Convention on the Rights of the Child. The SGBV committee (also NGO-supported, refugee staffed) supported the girl and said that the refusal to marry was an SGBV incident: ‘it is not SGBV but it is GBV [gender-based violence], because the boy will not marry the girl. In Karen culture, if they sleep together and do not marry it is a crime.’ The question became whether the camp leaders should order a marriage even against the wishes of Saw Htoo and his parents. After several hours of discussion the decision taken in the court room was that the couple should be married. However for this decision to be binding, all present had to agree. Saw Htoo’s parents protested that they would not accept such a decision and insisted that they wanted the case to be referred to ‘UNHCR law’. When the UNHCR fi eld offi cer was consulted some days later, it was advised that the couple could not be forced to marry. Instead, it was agreed that a cultural cleansing ceremony would be conducted, and UNHCR would expedite an application for Moe Kyi and her family to resettle to the United States.30 Ostensibly a case about pre-marital sex, this case had several dimensions that were not apparent from the conversation in the court room, including the infl uence of the international resettlement programme on the camp community, resistance to inter-ethnic marriage and anti-Burman discrimination. Though Saw Htoo’s parents insisted publicly that the couple had not had any sexual contact, it was not clear that they really believed this. Saw Htoo’s father told me that Moe Kyi was being unreasonable in refusing fi nancial compensation; he said that he had ‘made a mistake’ himself as a young man and the woman had accepted money without making a fuss. The reasons given by the parents for opposing the marriage (that Saw Htoo was legally a child and that the couple had not had intercourse) obscured a different set of reasons that were almost certainly more central to their opposition. One of these was their participation in the international resettlement programme. The family had been accepted to resettle to Australia and was waiting only for a departure date. If Saw Htoo were to get married, Moe Kyi would not be permitted to travel with them and Saw Htoo might have to stay in the refugee camp. Another reason – which seemed to rank highly in my conversation with the family – was the fear of inter-ethnic marriage. Many Karen refugees believe that inter-ethnic marriage inside Downloaded by [University of Defence] at 21:28 09 May 2016 Burma is a deliberate tactic by the military to ‘dilute’ Karen ethnicity. It was also resisted in this case for intensely personal reasons, as Saw Htoo’s father told me, ‘I lost everything because of the Burmese, and now my son wants to marry one?’31 On Saw Htoo’s ‘side’, then, the human rights claim of a violation of the Convention on the Rights of the Child (forced marriage of a minor) was used to bolster ethnic discrimination and avoid the cultural responsibility to marry a woman after having sexual contact with her. These ulterior motives were tacitly recognized by the judges, who had written a message in Burmese on a 152 Governing refugees

whiteboard at the front of the room: ‘What is law? Everybody is equal before the law, regardless of religion or nationality. If there is discrimination, it will be a crime.’32 On Moe Kyi’s ‘side’, the claim of an incident of SGBV was made to encourage a forced marriage which was seen as the best (and indeed only) way to protect her interests.33 Though both parties were engaged in human rights claims-making, they were not adopting the internationally mandated standards ‘correctly’ nor were they amending their underlying beliefs or attempting to change prevailing social norms: instead they were reinforcing them. Similarly, Saw Htoo’s family’s refusal to accept that the decision of the camp leaders in preference for consultation with UNHCR was not a sign of increasing legitimacy of UNHCR as a dispute resolution forum but was rather an exploitation of the legal plurality that UNHCR presented and the opportunity to ‘forum-shop’ for a more favourable outcome. My impression was that the misuse of human rights values here was not a deliberate attempt to undermine or ‘resist’ international impositions of law but the result of a lack of comprehension. It was an attempt to make sense of rights in an entirely rational way: by ﬁ tting them within pre-existing world views, and using rights language to bolster these claims. This is also entirely consistent with generations of past practice among the Karen in Southeast Burma, as was outlined in Chapter 4. In effect, this case suggested that human rights has become another layer of the ‘governance palimpsest’, whereby a new language is used to describe harms and their rationale for punishment but the underlying social and moral values remain largely unchanged.

Notes 1 Interview #146, 25 April 2012; interview #148, 25 April 2012. 2 In fact, a series of acronyms have been used to develop programmes on the same core issue, including gender-based violence (GBV), violence against women (VAW) and prevention of sexual abuse and exploitation (PSAE). This inconsistency has led to confusion among refugee community groups and the wider population as to the scope and breadth of the concept. 3 SGBV committee member, Mae La Oon, 27 January 2010. ARC (the American

Downloaded by [University of Defence] at 21:28 09 May 2016 Refugee Committee) is the implementing partner on the UNHCR-funded project to create and support SGBV committees in each refugee camp. 4 In general, the language used to describe sexual violence was often euphemistic. For example, a rape of two women by a Thai soldier was described to me by one person as ‘he had two girlfriends’ and by another as ‘he was in love with two girls’. Hyndman mentions a similar challenge in research with refugees in Kenya, where a rape was translated as ‘spousal assault’ (2000: 92). 5 Interview #4, 11 Jan 2009, Mae Sariang. 6 Field notes, Mae La Oon camp security (excerpt translated from written case records), 22 May 2009. Enacting interlegality 153

7 Interview #111, 12 April 2012. 8 Interview #148, 25 April 2012. 9 Interview #72, 15 February 2010. 10 If there is no violence or abuse within a relationship, divorce is even more diffi cult to obtain. In a case that I observed, a divorce was denied to a couple where the man had repeatedly committed adultery and requested a divorce; his wife was willing to remain married provided he reformed his behaviour. My translator was fully in agreement with the denial of a divorce, considering that the adulterer/ transgressor had no right to request a divorce if the ‘wronged’ spouse did not. One judge explained to me that couples will usually change their mind about divorce if they are given enough time to reconcile. He described one couple who had requested a divorce. He refused and told them to wait for one year, at which point he claimed that they were very happy together again. Whenever he met the couple he would laugh and say, ‘Do you still want that divorce?’ This was just one example of many similar outcomes, he assured me. On another occasion, he asked me what the process was for seeking a divorce in the UK. I explained that if there is adultery or unreasonable behaviour then the couple can be divorced immediately but if there is no ‘fault’ then the couple must live apart for some time (two years in England and Wales, one year in Scotland) before a divorce will be granted. He was triumphant, both that he followed the same practice (if there is fault, a couple will be divorced immediately but if there is no fault, they must wait) and that the waiting period he imposed was only six months. 11 Interview #146, 25 April 2012. 12 This view is also apparent in one of the core principles of the Mediation and Dispute Resolution Guidelines, that camp leaders ‘should not discourage nor obstruct’ camp residents from referring a dispute to Thai court. 13 Interview #167, 4 May 2012. 14 In fact, this woman’s praise was so effusive that I felt obliged to clarify that I did not work with KWO (as on this fi eldtrip I was staying in a camp boarding-house operated by KWO, though it is unlikely this interviewee knew that). She seemed confused by this and said, ‘I just want to tell you about the job that KWO do.’ Interview #125, Mae La Oon, 14 April 2012. 15 Interview #126, KYO Mae La Oon, 16 April 2012. 16 As an indication of changing thinking on this issue, in 2010 the UNHCR webpage on children was titled ‘Young and Innocent’. In 2013, it was titled ‘Protection and Building Resilience’ – see www.unhcr.org/pages/49c3646c1e8. html (accessed 10 August 2013. 17 Interview #144, 24 April 2012. Downloaded by [University of Defence] at 21:28 09 May 2016 18 Interview #44, 12 May 2009. 19 Interview #59, KRC member, Mae Sot, 4 June 2009. 20 Interview #4, KWO member, Mae Sariang, 11 January 2009. 21 Interview #111, 12 April 2012. 22 Field notes, Mae Ra Ma Luang refugee camp, April 2012. 23 The Guidance Note lists offences that should be send to Thai courts, including murder, non-compoundable rape, sexual assault, drugs, traffi cking, grievous bodily harm, or cases where the young person poses an immediate physical danger. As explained in the previous chapter, while the option of referral to Thai court is welcomed in very serious cases, for minor cases (such as smoking 154 Governing refugees

marijuana) there is a high probability that Thai police will simply return an offender to the camp without punishment. 24 The Mae La Oon security offi cer with responsibility for record-keeping informed me that the number of incidents involving young people rose from 200 incidents in 2007 to 300 cases in 2008. I could not verify these fi gures but believe they offer an indicative ‘ballpark’ fi gure of the number of security incidents involving young people in the camp. Interview #20, Mae La Oon, 22 February 2009. 25 Interview #14, former judge, Mae La Oon refugee camp, 11 February 2009. 26 There are some practical considerations that may make it more likely for new arrivals to come into confl ict with camp rules. For example, new arrivals will not receive food and other rations until they are registered with TBC, which may take weeks or months. In the meantime they are more likely to have to leave the camp in search of work. It is also possible that they are more likely to drink alcohol or contravene other camp rules. 27 Interview #109, social welfare, 11 April 2012, responding to the idea that having children out of wedlock might be socially acceptable (as I had explained it was in the UK) and going on to comment that ‘Young people have been living together [euphemism for pre-marital sex] because they mimic the NGOs. But NGOs can handle this. They use their rights in the right way. But the young people here are not able to control themselves.’ As an indication that this may be changing, seven young people in Mae La camp organized a Lesbian Gay Bisexual Transgender Intersex (LGBTI) community group. This group seems to have organically disbanded as its members left the camp for various reasons but its creation and existence is an important indication that a more open discussion about sexuality may be beginning to take place (Moses 2013). 28 Field notes, 11 April 2009. 29 These names have been changed. 30 ‘Cleansing ceremonies’ are used in the Karenni camps in responding to and resolving disputes but this was the only case in which I heard of one being used in Mae La Oon (and in this case the cleansing ceremony was decided in discussion with UNHCR rather than as a disposal from camp justice or the camp committee). 31 Interview #45, Mae La Oon, 13 May 2009. 32 In an example of the vagaries of memory, I had recalled that this message was written in English and that it had therefore been directed at me, as the only native English speaker in attendance, as in itself a piece of ‘claims-making’ on the part of the judges. On returning to my original notes from the event I found that the message had in fact been written in Burmese, and therefore was directed to Moe Kyi and her family as a message of support and solidarity. Downloaded by [University of Defence] at 21:28 09 May 2016 33 In the refugee camps, ‘forced marriage’ is increasingly requested in precisely these circumstances, where a man refuses to marry a woman who he has had sexual intercourse with (and who may be pregnant), often because he has registered with a resettlement programme. Interview#4, Mae Sariang, 11 January 2009. Chapter 8 Beyond encampment

Freedom is what you do with what’s been done to you. (Sartre, quoted in Gerassi 1989: 47)

The experience of Karen refugees in Thailand shows that camps can be functional societies even after decades of encampment. That is not to say that these camps are easy or desirable places to live; they are not, and those who live in them face many challenges and deprivations beyond the fundamental ‘unhealable rift’ of exile (Said 1984: 159). Nevertheless, protracted encampment has not destroyed community bonds, and this social capital has been a vital coping mechanism in managing displacement. We are accustomed to thinking of refugee camps as total institutions, where refugees are subject to the arbitrary exercise of authority and where opportunities for resistance are limited. In the terms of Sartre’s comment (above), the focus has been rather more on ‘what’s been done to’ refugees than on how they have responded to that. And yet, with all that has been done to refugees from Burma – forced from their land, confi ned in camps, subject to the policy whims of governmental and non-governmental agencies alike – they have somehow contrived to fi nd and exercise a form of freedom. This raises an obvious question: why here? Why have these camps remained safe when so many others do not? There is no single or simple answer to this question. As this book has emphasized, governance and authority in refugee camps must be understood holistically, as an inter-related normative universe Downloaded by [University of Defence] at 21:28 09 May 2016 rather than the result of individual forces or policies. The normative universe of the Thai–Burma border camps is a product of the infl uence of Thai policies, international agencies’ policies, the KNU, camp committees and the preferences of refugees themselves. If any of these strands of authority had been different, the refugee camps would have developed differently. Of course, this is not to suggest that all policies and decisions are equally important. Almost certainly, the nature of the refugee situation had an impact. In the mid-1980s, thousands of refugees from Vietnam, Cambodia and Laos were fl eeing to Thailand every day (Robinson 2000: 46). In contrast, 156 Governing refugees

refugees from Burma arrived relatively slowly, with the population expanding over decades rather than months (and with most being absorbed in Thailand as undocumented workers rather than as camp-based refugees). There were periods where the infl ux was much more intense, notably in the mid-1990s, but as an overall trend the population escalated relatively gradually. The camps evolved similarly, shifting over time from village-like settlements to closely regulated camps. As a result, governance structures were able to grow and develop alongside the development of the camps. This is very different from refugee situations where camps are created to immediately house hundreds of thousands of people. In this type of situation – such as is currently experienced by refugees from Syria in Turkey and Jordan – it is harder to imagine that strong, stable, trusted governance structures could be established entirely from within the refugee population. Demographic characteristics were also important. For the fi rst two decades of the camps’ existence the ‘Karen camps’ were almost entirely populated by ethnic Karen. In 2013, six of the nine border camps had a Karen population of 95 per cent or more and even the most ethnically diverse camp (Umpiem Mai) was 75 per cent Karen. This has allowed for consensus around governance approaches and provided precedent for self-governance. As Chapter 2 explained, the Karen territories of Southeast Burma were never fully ‘rationalized’ within a state structure. Though always maintaining relationships with outside authorities (including other ethnic groups, religious communities and the Burmese Government) governance was essentially conducted at village level. This remained the case through colonization, the state-building project of the KNU, protracted armed confl ict and the present climate of patrimonialism and an ‘emerging political complex’ (Callahan 2007: 49). As a result, while refugees from these areas have a long history of coping with confl ict and displacement they also have a long heritage of self- protection and self-reliance strategies. This has fostered skills of resilience, autonomy and self-suffi ciency – the same attributes that have been at the heart of coping strategies for IDP populations inside Burma, for refugees in camps and among refugees who have been resettled. This raises an intriguing question: is the strength of community organization and social cohesion among Karen refugees not despite but because of the absence of state protection? Is it possible that familiarity with village sover- Downloaded by [University of Defence] at 21:28 09 May 2016 eignty provided an adaptive advantage for these refugees? Scott (2012: xxii) poses a similar question in raising ‘two cheers for anarchy’, asking, ‘to what extent has the hegemony of the state and of formal, hierarchical organizations undermined the capacity for and the practice of mutuality and cooperation that have historically created order without the state?’ It certainly seems plausible that the political and legal heritage of autonomous governance has had a determinative impact on the Karen refugee experience as, accustomed to self-reliance, Karen refugees have been better able to form networks of support and service delivery in encampment. Beyond encampment 157

A shared Karen identity has also been important for the beliefs and practices that are considered (or constructed as) intrinsic to that identity, and that have lent themselves to cooperative co-existence in camps. Political governance in Southeast Burma has been shaped by several eras which have in turn shaped different frames of governing authority, including village level sovereignty, supernaturalism or ‘spirit-law’ and the pseudo-state structures of the KNU. These layers of belief and authority have formed a ‘governance palimpsest’ of spiritual beliefs, traditions and customs, moral didacticism and codifi ed law. This palimpsest also establishes the governance model used in the refugee-led camp systems, in terms of institutional approaches (e.g. hierarchical, egalitarian, participatory decision-making), specifi c practices (e.g. moral didacticism, consensus, participation, banishment) and consensus on values (e.g. respect for hierarchy, respect for order, communitarianism, monogamy). These ‘internal’ characteristics of the Karen refugee population ensured a capacity for autonomous governance and a template for implementing it; characteristics that will not be present in every refugee situation. However, it was equally crucial that national and international policies allowed this capacity to be exercised. In recent years the Royal Thai Government has tightened border controls and placed increasing restrictions on refugees’ freedom of movement and the political activities of Karen exiles in Thailand, but at the outset of this situation refugees were permitted broad self- settlement in communities where they had relative freedom to farm and to travel beyond the camp boundaries (Lang 2002; Bowles 1998). Above all, they were expected to manage themselves. That the Thai Government ‘outsourced’ its sovereignty to refugee leaders might be seen as an abdication of its responsibility to provide justice and security to refugees – but it may also have had a silver lining in allowing refugees to establish their own communities, according to their own rules and values. If large, closed camps had been insisted upon from the outset the resulting social pressures may have made it more diffi cult for the camps to sustain stable management. Related to this, it seems probable that another signifi cant policy decision by the Royal Thai Government was the refusal of access to UNHCR for the fi rst fi fteen years of this refugee situation. TBC provided material support and resources for camps but its pragmatic, partnership-focused approach also per- Downloaded by [University of Defence] at 21:28 09 May 2016 mitted a high level of freedom for refugee leaders, including tacit acceptance of a role for members of the KNU in camp management. A link between refugee populations in camps and political organizations in their state of origin is often seen as dangerous and threatening, based on experiences of refugee militarization in Central America, Indochina and the Great Lakes region (Lischer 2005; Lawyers Committee for Human Rights 2002; Stedman and Tanner 2005). For example, in refugee camps in Goma, Hutu militarized groups appropriate millions of dollars of aid for direct military purposes, and killed thousands of residents inside the camps (Terry 2002: 175). However, in 158 Governing refugees

Thailand, the role of the KNU in the refugee camps has not been so straightforwardly one of domination, and arguably had constructive dimensions, such as establishing management structures, political relationships and access to resources to support refugees (see further McConnachie 2012). TBC’s approach appears to have played a key role in allowing this refugee population to develop such a high level of autonomous self-management. If UNHCR had played a dominant role in these camps from the outset it is likely that the management approach would have been rather different. UNHCR’s fi eld operations are shaped by many variables, from the preferences of a host state to the personalities of senior fi eld staff, but the approach taken by UNHCR in countries where it has a leading role in camp management – including Tanzania, Kenya, and Uganda – suggest that efforts would have been made to establish an entirely non-political camp management governing structure (Turner 2010, 2005b; Hyndman 1997; Verdirame and Harrell- Bond 2005). This would have required dismantling the systems that were established organically through the KNU and KRC. It is ironic that the Thai authorities’ refusal to recognize refugees from Burma as ‘refugees’ – and thus their refusal to allow UNHCR more than a minimal role – quite probably enhanced the refugee population’s space for agency and autonomy in managing encampment. It seems that the policies of both the Royal Thai Government and TBC were signifi cant less for what they did than for what they didn’t do. In effect, both the Royal Thai Government and TBC allowed refugees to get on with the work of camp management largely without oversight; hardly the orthodox approach to refugee protection. TBC encouraged refugee-led management but it did not have a formal programme to support camp management until 2005. A programme of this nature might have been valuable at an earlier stage – but equally, the example of international intervention in the justice sector shows how diffi cult it is to intervene to improve local governance without undermining it. It is certainly possible that the decades in which camps were permitted to develop their own structures largely without interference were central to the stability that they developed. This apparently paradoxical outcome may be explained by returning to Giddens’ theory of structuration, discussed in Chapter 1. As Giddens (1984: 1–28) explained, agency and structure (rules, norms, constraints) are not Downloaded by [University of Defence] at 21:28 09 May 2016 separate entities but are entwined and interdependent. For Karen refugees in Thailand the relative absence of ‘external’ structural constraints at the national and international level allowed for greater expression of agency, while the ‘internal’ characteristics of the refugee population shaped the direction it took. Of course, the approach of outsourced sovereignty by the Royal Thai Government and non-intervention by TBC lies fi rmly in the past. Now, a considerably higher degree of control is imposed on refugees by both Thai authorities and by international agencies. The dynamics here are more Beyond encampment 159

complex than might fi rst be assumed. Chapter 5 explained the different governing interests and orientations of the two sets of actors whereby, very broadly, the Royal Thai Government sees refugees as a threat to be contained while international agencies see refugees as victims to be protected. The approach of the Thai authorities is not necessarily the more damaging of the two. As with other regimes of international aid, refugee assistance can have paradoxical effects. There are eighteen NGO members of the Committee for Coordination of Services to Displaced Persons, not all of which share TBC’s philosophy of supporting refugee self-governance. Where power and authority is situated with elite actors and state structures are relied upon to create change, international agencies risk stripping power and autonomy away from the very populations they seek to protect. Dispute resolution and order maintenance in the refugee camps manifest the same power dynamics and contestation that exist in camp management more generally, as refugee leaders struggle to resolve the contradictions between the preferences of the Royal Thai Government, international agencies and camp residents. The camp structures arise from and protect a tightly controlled community hierarchy while international organizations are guided by liberalism and the supremacy of individual rights and freedoms. To an extent, tensions between these approaches are inevitable (Krieken 2001). Resonating with the experience of other confl ict-affected communities, contested notions of justice are apparent in different views on institutional and procedural legitimacy, and in different constructions of harm (McEvoy and Mika 2002; Santos 1977). Dispute resolution in the refugee camps continues to refl ect the order- focused, hierarchical, communitarian principles of village sovereignty in Karen State but the layers of the governance palimpsest are not recognized by many external actors, who see the camps instead as tabula rasa – ‘the tablet wiped clean’ – a blank slate on which to inscribe their own governing ideologies. Chapter 6 analyzed the struggle to defi ne institutional structures for the administration of justice within the Thai border refugee camps, examining the functioning of the camp justice system as a model of community justice with high internal legitimacy but low ‘external’ legitimacy as perceived by international humanitarian agencies (particularly UNHCR), whose staff repeatedly express an institutional preference for Thai courts and a normative Downloaded by [University of Defence] at 21:28 09 May 2016 preference for international law. This bias in favour of state institutions paral- lels historical elite-subaltern relations during colonization and missionisation (Starr and Collier 1989; Moore 2000 [1978]; Merry 2000a), while the power dynamics in the refugee–NGO relationship echo that in other spheres of development and humanitarian action. Chapter 7 examined the normative consequences of this clash of justice ideals by tracing local reception of international human rights standards, particularly women’s rights and children’s rights. In relation to women’s rights, the level of local ‘vernacularisation’ by refugee activists has been high, leading 160 Governing refugees

to some changes in gender relations and growing acceptance of the need to recognize and prevent domestic violence and sexual violence within Karen society. Change is slow but there is genuine energy from within the refugee population to pursue it and to connect with national and transnational advocacy networks in doing so. Children’s rights, in contrast, have not found such willing local ‘knowledge brokers’ and are perceived by camp leaders and residents as a bar to effective camp management and a threat to intergenerational relations. Where responses to youth delinquency have altered in response to rights discourses it has been the result of external pressure rather than a genuine social movement within the refugee community. It is not clear what effect the prevailing climate of legal pluralism has had on the camps and their residents. Human rights discourses have been at least partially accepted in some spheres (gender equality, non-discrimination) and have been rejected in others (child protection). There is some evidence of human rights language being adopted for claims-making by refugees, but the use of ‘rights-talk’ did not necessarily refl ect a transformation of underlying beliefs and attitudes. In considering the infl uence of human rights on refugees’ beliefs and practices, it may also be worth refl ecting on the role of human rights more broadly as an avenue for refugee protection. Human rights education has been a key agenda of international humanitarian agencies working in the camp but the most pressing vulnerability for refugees in Thailand is not internal camp security but powerlessness in their relationships with other actors: the Burmese regime, Thai offi cials and much of wider Thai society. Discussing human rights discourses regarding Muslim women, Abu- Lughod (2002: 788) questions whether activists should focus less on saving ‘vulnerable’ groups and more on changing our own behaviour and the global inequalities that give rise to their situation: ‘Could we not leave veils and vocations of saving others behind and instead train our sights on ways to make the world a more just place?’ Similarly, it could be argued that the refugees’ needs might be better served if human rights advocacy focused less on transforming the internal dynamics of the camp society and more on securing a genuinely sustainable future for refugees. This is considered in more detail below. Downloaded by [University of Defence] at 21:28 09 May 2016 Overcoming ambivalence and the limits of self-reliance The foregoing analysis suggests that the strength of self-governance within these camps may owe more to historical and political accident than to a triumph of specifi c policies. This does not mean that the experiences of refugees in Thailand are irrelevant to other refugee situations. Perhaps the most important fi nding from this context is that the destruction of community networks is not an inevitable outcome of protracted encampment and that Beyond encampment 161

refugees can play a central, essential role in governing themselves: in short, that self-governance can work. This outcome is predicted in a series of UNHCR policies promoting refugee participation, a community-based approach and self-reliance (UNHCR 2006d, UNHCR 2008b). These policies are soundly justifi ed on both principled and practical grounds but in fi eld operations they have often been implemented in ways that fail or disappoint (Meyer 2006; Rothkegel et al. 2008; Hunter 2009). A consistent thread is an apparent ambivalence towards community representatives, whereby policies promote participation and consultation but in practice community structures are seen as threatening or dangerous to refugee protection (Horn 2010b; Olivius 2013; Turner 2010; Palmary 2008). This ambivalence has also been apparent in Thailand (though, importantly, not in the policy approach of the lead agency in camp management, TBC) and has been most clearly expressed in the concerns of UNHCR (and others) that camp committees and community-based organizations are proxies for non-state actors and that reluctance to refer criminal cases to Thai courts is a means for camp leaders to consolidate their power. This ambivalence establishes a layer of mistrust in engaging with refugees that undermines the prospect of good working relationships. It is also potentially counter-productive, as it risks alienating respected leaders and concentrating engagement with people who have less genuine infl uence and authority. Overcoming ambivalence towards community representatives does not have to entail naivety. First, it is essential to establish whether a ‘community’ exists to engage with at all. Among Karen and Karenni refugees this is unmistakable but many other refugee situations have a much higher degree of ethnic diversity and/or factionalism. If there is no entity with broad legitimacy (whether this is organized around ethnicity, nationality, faith, politics or on any other basis), the benefi ts of community governance will be much harder to achieve. However, where community leaders do exist, they should be included in any consultation and participation processes. The importance of remaining open-minded about community leadership was reiterated in a recent report clarifying that UNHCR fi eld staff should be willing to engage with armed groups where it serves the ultimate objective of refugee protection (Keogh and Ruijters 2012). Downloaded by [University of Defence] at 21:28 09 May 2016 However, the limits of self-reliance approaches are not solely related to operational challenges of inadequate consultation and ambivalence about the nature of community organization, but can also be about the goal that is being pursued. ‘Self-reliance’ approaches aim to enable independence but for refugees in camps independence will always be a relative concept. Camps are typically situated in rural, isolated areas where there are limited employment opportunities. Where jobs are available, refugees are unlikely to be able to compete on equal terms to host citizens, whether as the result of outright discrimination, of language barriers, of restrictions on their freedom of 162 Governing refugees

movement and right to work, as a consequence of their legal status, or because the sheer number of refugees looking for work has driven down rates. All of these factors also expose refugees to greater risk of exploitation and mistreatment, and make it harder to do anything about mistreatment if they experience it. Similarly, at the national level, refugees are dependent on the continued tolerance of a host government. Throughout Southeast Asia there have been many examples of ‘pushbacks’, forced repatriation and refoulement to demonstrate the limits of such tolerance. As a result, even the most successful ‘self-reliance’ approach is unlikely to remove the need for international support, because this support will always be essential not only for ﬁ nancial assistance but also for political advocacy. In Uganda, a ‘Self-Reliance Strategy’ for refugees was linked to UNHCR withdrawal and handover to the Ministry of Local Government. Kaiser (2002: 219) has eloquently described refugees’ disbelief that UNHCR would actually leave the settlements to the management of local government and their ‘fear and trepidation’ as the reality dawned. The success of the approach was further limited by a thin understanding of self-reliance as ‘the capacity to generate enough food to live on’ (ibid: 223) which failed to take into account the many practical and political challenges that refugees faced in becoming ‘self-reliant’ (Meyer 2006: 63–4). In Thailand, refugees exercise agency in numerous ways but ultimately within a limited sphere. The central decisions of their lives (the creation and closure of camps, the location of camps, the availability of ‘durable solutions’) are determined by national governments and international agencies. On a day-to-day basis they are at the mercy of local ofﬁ cials, employers and citizens. Resilience alone cannot compensate for this lack of power. As a result, the most valuable role for international agencies is that of helping refugees to negotiate their relative powerlessness by serving as a buffer between refugees and the host society. The Karen and other refugees from Burma are skilled political actors, but they lack the political leverage to pursue their own claims. This speaks to the general power imbalance in refugee lives and thus to the area where international organizations could arguably have the greatest impact on refugees’ protection and quality of life: by rebalancing asymmetries of power, advocating on refugees’ behalf and enabling them to achieve self- reliance on their own terms. Downloaded by [University of Defence] at 21:28 09 May 2016

Nothing about refugees, without refugees Several of the themes discussed throughout this book – the capacity within this refugee population, their political powerlessness, and the often challenged communication between refugees and international agencies – are at the forefront of the latest challenge to confront refugees: the possibility of return to Burma. Beyond encampment 163

The inauguration of a ‘civilian’ government in Burma in March 2011 presaged reforms including the release of hundreds of political prisoners, relaxation of censorship, and peace negotiations with several ethnic insurgencies, including the KNU. While predicting Burma’s political future is always a risky endeavor, it seems plausible that these changes will continue as Burma holds the chairmanship of ASEAN in 2014 and national general elections are scheduled for 2015. As the political environment inside Burma has begun to change, repatriation has become a very real possibility. However, changes in the heartland of Burma have not been matched in the ethnic borderlands. A provisional ceasefi re was signed between the KNU and the Government in January 2012 but Burma Army troops continued to solicit forced labour from villagers in Southeast Burma (KHRG 2013). In Kachin State, after the collapse of a sixteen-year ceasefi re, all-out confl ict has resumed and more than 100,000 people have been displaced. Persecution and massacres of Rohingya in Arakan state and generalized anti-Muslim violence in central Burma has been described as ‘ethnic cleansing’ and has also created new fl ows of internal and cross-border displacement (Human Rights Watch 2013a). In short, human rights violations are still occurring on an extreme scale, the Burma Army is still present in ethnic regions and there is not yet even a concrete peace agreement between the KNU and the Burmese Government. The extent of change remains far short of transformative. This may be the best opportunity yet for a peaceful resolution to ethnic confl icts in Burma (see South 2012a) but it is also a time of great uncertainty. In 2013, the offi cial positions of the Thai government and of the UNHCR were consistent: that conditions in Burma are not yet appropriate for an organized return. Nevertheless, as early as 2011, senior Thai offi cials made public statements that the political situation in Burma had changed and refugees should soon be repatriated (Chitradon 2011). International donors are equally keen to see a resolution to this long-standing refugee situation and withdrawal of funding to the refugee camps has forced TBC to reduce already meagre food rations even further (S’Phan Shaung 2013). A political conversation about repatriation has been underway for some time between the Royal Thai Government, the Burmese Government and the UNHCR (Burma Partnership 2012). These developments are perceived by refugees as pointing towards imminent camp closure and return to Burma, a terrifying prospect Downloaded by [University of Defence] at 21:28 09 May 2016 for many: what will happen to them? Will they be forced to leave Thailand? Where will they go? How will they survive? Many issues must be addressed before safe return of refugees will be possible, including the withdrawal of Burma Army troops, landmine removal, property restitution or land allocation, and provision for reintegration support and assistance. Above all, there must be confi dence in a stable and secure peace – and currently this is a long way from being established. In their response to the prospect of repatriation, community organizations are once again showing impressive leadership. The KRC has issued a statement 164 Governing refugees

emphasizing its hope that refugees may be able to return to their homeland but stating ten points that must be met to ensure that repatriation is safe. These points can be read as a comprehensive list of the challenges facing a repatriation process, including the need for a durable peace to be established; for land and livelihood opportunities for returnees; recognition for education and healthcare received in the camps; removal of landmines; and the necessity that any decision to return is truly voluntary and not coerced. In a KRC- conducted survey of 6,195 refugees in Tham Hin camp, 46 per cent of respondents wanted to resettle to a third country, 27 per cent wanted to remain in Thailand and 27 per cent wanted to return to Burma when the political situation is stabilized (Eh Na 2013). A core problem for the refugee population is lack of information. Repatriation is typically conducted as a tripartite process between a sending state, host state and the UNHCR, and there is no established mechanism for including the refugee population or its representatives as equal participants in dialogue. In Thailand, UNHCR and other members of the Committee for the Coordination of Services to Displaced Persons have issued a short statement of basic principles for repatriation, emphasizing that repatriation ‘needs to be voluntary’, ‘will take place under conditions of safety and dignity’, ‘may be “promoted” or “facilitated”’ and will include Governments, UNHCR, NGOs and refugees (CCSDPT 2012). The principles do not explain how ‘voluntariness’ will be guaranteed, what sort of provision will be made for those who return or what will happen to those who choose not to return: and it is these fundamental questions that most exercise refugees. News reports reﬂ ect refugees’ anxiety and sense of exclusion from the political discussion, with titles such as ‘Talk to us and listen to our voices’ (Karen News 2013b) and ‘Still not safe to send back refugees’ (Lah Khin Moo 2013). As with resettlement, this is a potential ‘durable solution’ for tens of thousands of people that is being proposed and planned with minimal consultation with the refugee population and community-based organizations. The result, as with resettlement, has been to give rise to a ﬂ ow of rumours and gossip that further increase anxiety and worry (Burma Partnership 2012). Whatever happens, the years ahead will hold new challenges for refugees in Thailand, whether of embracing change and return to Burma or of reconciling to dashed hopes and the continued uncertainty of life in camps. While it is Downloaded by [University of Defence] at 21:28 09 May 2016 impossible to predict what the future holds, one important lesson from past experience is that the Karen and other refugees from Burma are likely to rise to the challenges they face with dignity and resilience. This is a refugee population that has repeatedly proven its capacity to organize and manage its own affairs and that must surely be taken seriously as an equal partner in planning discussions and in any implementation. It seems appropriate to conclude by echoing their plea for political inclusion: ‘Nothing about refugees, without refugees’. References

Abel, R. (ed.) (1982) The Politics of Informal Justice: Volume 2, New York and London: Academic Press. Abu-Lughod, L. (2002) ‘Do Muslim Women Really Need Saving? Anthropological Reﬂ ections on Cultural Relativism and its Others’, American Anthropologist, 104(3): 783–90. Agamben, G. (2005) State of Exception, trans. K. Attell, Chicago and London: University of Chicago Press. ____ (1995) Homo Sacer: Sovereign Power and Bare Life, Stanford: Stanford University Press. Agier, M. (2011) Managing the Undesirables, Cambridge: Polity. ____ (2010a) ‘Humanity as an Identity and its Political Effects (A Note on Camps and Humanitarian Government)’, Humanity, 1(1): 29–45. ____ (2010b) On the Margins of the World: The Refugee Experience Today, trans. D. Fernbach, Cambridge: Polity Press. Aiken, M. (2009) Final Evaluation Report: The International Rescue Committee Thailand Legal Assistance Centre Project, July 2009, copy on ﬁ le with author. Amnesty International (2008) Myanmar: Crimes against Humanity in Eastern Myanmar, online at: http://www.amnesty.org/en/library/info/ASA16/011/2008 (accessed 10 August 2013). ____ (2002) Thailand: Widespread Abuses in the Administration of Justice, online at: http://www.amnesty.org/en/library/info/ASA39/003/2002/en (accessed 10 August 2013). Andersen, K.E. (1981) ‘Two Indigenous Karen Religious Denominations’, Folk, 23: 251–62. ____ (1979) ‘Deference for the Elders and Control over the Younger among the Downloaded by [University of Defence] at 21:28 09 May 2016 Karen in Thailand’, Folk, 21&22: 313–25. Anderson, B. (2006 [1983]) Imagined Communities, New York and London: Verso. Anderson, J.L. (1992) Guerrillas, New York: Random House. Appadurai, A, (2003) ‘Sovereignty without Territoriality: Notes for a Postnational Geography’, in S.M. Low and D. Lawrence-Zúñiga (eds) The Anthropology of Space and Place: Locating Culture, Oxford: Blackwell Publishing. Apple, B. and Martin, V. (2003) No Safe Place: Burma’s Army and the Rape of Ethnic Women, Washington DC: Refugees International. Ariès, P. (1973) Centuries of Childhood: A Social History of Family Life, trans. R. Baldick, London: Jonathan Cape. 166 References

Aung-Thwin, Michael and Aung-Thwin, Maiitri (2012) History of Myanmar since Ancient Times: Traditions and Transformations, London: Reaktion Books. Baek, B.S. and Subramanium, G. (2008) ‘Myanmarese Refugees in Thailand: the Need for Effective Protection’, Cornell Law School J.S.D, Doctoral Student Papers, Paper 1. Banda, F. (2006) ‘Women, Law and Human Rights in Southern Africa’, Journal of Southern African Studies, 32(1): 13–27. Banki, S. (2008) Bhutanese Refugees in Nepal: Anticipating the Impact of Resettlement, Camperdown, NSW: Austcare. Banki, S. and Lang, H. (2008) ‘Protracted Displacement on the Thai-Burmese Border: The Interrelated Search for Durable Solutions’ in H. Adelman (ed.) Protracted Displacement in Asia: No Place to Call Home, Aldershot: Ashgate. ____ (2007) Planning for the Future: The Impact of Resettlement on the Remaining Camp Population, Report Commissioned by the Committee for Coordination of Services to Displaced Persons in Thailand (CCSDPT), online at: http://www. tbbc.org/resources/2007-07-ccsdpt-resettlement-impact-study.pdf (accessed 1 February 2011). Bartolomei, L., Pittaway, E. and Pittaway, E.E. (2003) ‘Who Am I? Identity and Citizenship in Kakuma Refugee Camp in Northern Kenya’, Development, 46(3): 87–93. Bauman, Z. (2001) Community: Seeking Safety in an Insecure World, Cambridge: Polity. Beetham, D. (1991) The Legitimation of Power, Basingstoke: Macmillan. von Benda-Beckmann, F., von Benda-Beckmann, K. and Eckert, J. (2009) ‘Rules of Law and Laws of Ruling: Law and Governance between Past and Future’, in F. von Benda-Beckmann, K. von Benda-Beckmann and J. Eckert (eds) Rules of Law and Laws of Ruling: On the Governance of Law, Farnham: Ashgate Publishing. von Benda-Beckmann, K. (1981) ‘Forum Shopping and Shopping Forums: Dispute Processing in a Minangkabau Village in West Sumatra’, Journal of Legal Pluralism, 19: 117–59. von Benda-Beckmann, K. and Pirie, F. (2007) ‘Introduction’, in K. von Benda- Beckmann and F. Pirie (eds) Order and Disorder, New York and Oxford: Berghahn Books. Benner, M.T., Townsend, J., Kaloi, W., Htwe, K., Naranichakul, N., Hunnangkul, S., Carrara, V.I. and Sondorp, E. (2010) ‘Reproductive Health and Quality of Life of Young Burmese Refugees in Thailand’, Conﬂ ict and Health, 4:5. Berger, J. (2010 [1975]) A Seventh Man, New York and London: Verso. Berkowitz, E. (2012). Sex and Punishment: Four Thousand Years of Judging Desire, London: Westbourne Press. Downloaded by [University of Defence] at 21:28 09 May 2016 Bluebond-Langner, M. and Korbin, J. (2007) ‘Challenges and Opportunities in the Anthropology of Childhoods: An Introduction to “Children, Childhoods, and Childhood Studies”’, American Anthropologist, 109(2): 241–6. Bouffard, L.A. and Laub, J.H. (2004). ‘Jail or the Army: Does Military Service Facilitate Desistance from Crime?’ in S. Maruna and R. Immarigeon (eds) After Crime and Punishment: Pathways to Offender Reintegration, Devon: Willan. Bowles, E. (1998) ‘From Village to Camp: Refugee Camp Life in Transition on the Thai-Burma Border’, Forced Migration Review, 2: 11–14. Braithwaite, J. (2002) Restorative Justice and Responsive Regulation, Oxford: Oxford University Press. References 167

____ (1989) Crime, Shame and Reintegration, Cambridge: Cambridge University Press. Braithwaite, J. and Mugford, S. (1994) ‘Conditions of Successful Reintegration Ceremonies: Dealing with Juvenile Offenders’, British Journal of Criminology, 34(2): 139–71. Brecht, B. (1987) Bertolt Brecht: Poems 1913–1956, trans. John Willett, London: Routledge, Chapman and Hall. Brezovich, E.J. (2005) ‘Towards Men’s Involvement: Masculinities, Change and the International Rescue Committee’s Gender-Based Violence Program in Karenni Site 1, Thailand’, M.A. thesis, copy on ﬁ le with author. Brooten, L. (2004) ‘Human Rights Discourse and the Development of Democracy in a Multi-ethnic State’, Asian Journal of Communication, 14(2): 174–91. Brown, M.F. (2008) ‘Cultural Relativism 2.0.’, Current Anthropology, 49(3): 363 –83. Brown, S. (2012) ‘Youth Identity and Non-military Action in an Armed Opposition Group on the Burmese-Thai Border’, unpublished Ph.D. thesis submitted to School of Oriental and African Studies, University of London. Brown, W. (2004) ‘“The Most We Can Hope For …”: Human Rights and the Politics of Fatalism’, South Atlantic Quarterly 103 (2–3): 451–63. Bryant, R. (1997) ‘Kawthoolei and Teak: Karen Forest Management on the Thai- Burma Border’, Watershed, 3(1): 23–46. Bunch, C. (1990) ‘Women’s Rights as Human Rights: Towards a Re-Vision of Human Rights’, Human Rights Quarterly, 12(4): 486–98. Buntman, F. (2003) Robben Island and Prisoner Resistance to Apartheid, Cambridge: Cambridge University Press. Burma Ethnic Research Group (1998) Forgotten Victims of a Hidden War: Internally Displaced Karen in Burma, Chiang Mai: Burma Ethnic Research Group and Friedrich Naumann Foundation. Burma Lawyers’ Council (2007) Analysis on the Situation of the Refugee Camps from the Rule of Law Aspect, Mae Sot: Burma Lawyers’ Council. Burma Partnership (2012) ‘The Situation of Refugees on the Thai-Burma Border’, 11 December 2012, online at: http://www.burmapartnership.org/2012/12/the- situation-of-refugees-on-the-thai-burma-border (accessed 2 March 2013). Burman, S. and Harrell-Bond, B.E. (eds) (1979) The Imposition of Law, London: Academic Press. Burns, P. (1989) ‘The Myth of Adat’, Journal of Legal Pluralism, 28: 1–127. Burton, F. (1978) The Politics of Legitimacy: Struggles in a Belfast Community, London: Routledge and Kegan Paul. Buur, L. (2005) ‘The Sovereign Outsourced: Local Justice and Violence in Port Downloaded by [University of Defence] at 21:28 09 May 2016 Elizabeth’, in T.B. Hansen and F. Stepputat (eds) Sovereign Bodies: Citizens, Migrants and States in the Postcolonial World, Princeton and Oxford: Princeton University Press. Cain, M. (1985) ‘Beyond Informal Justice’, Crime, Law and Social Change, 9(4): 335–73. Caldeira, T.P.R. (2006) ‘“I Came to Sabotage Your Reasoning!” Violence and Resigniﬁ cations of Justice in Brazil’, in J. Comaroff and J.L. Comaroff (eds) Law and Disorder in the Postcolony, London and Chicago: University of Chicago Press. Callahan, M. (2007) Political Authority in Burma’s Ethnic Minority States: Devolution, Occupation and Coexistence, Washington: East-West Centre Washington. 168 References

Carothers, T. (2003) Promoting the Rule of Law Abroad: The Problem of Knowledge, Working Paper, Rule of Law Series, Democracy and Rule of Law Project, no. 34, Washington: Carnegie Endowment for International Peace. Cathcart, R., Decker, C., Ellenson, M., Schurmann, A., Schwarz, M., and Singh, N. (2007) People from Burma Living in Chapel Hill and Carrboro: An Action-Oriented Community Diagnosis: Findings and Next Steps of Action, University of North Carolina: Department of Health Behaviour and Health Education, School of Public Health. Chanock, M. (2002) ‘Human Rights and Cultural Branding: Who Speaks and How’, in A. An-Na’im (ed.) Cultural Transformation and Human Rights in Africa, London and New York: Zed Books. ____ (1985) Law, Custom and Social Order, Cambridge: Cambridge University Press. Charlesworth, H. (2005) ‘Not Waving but Drowning: Gender Mainstreaming and Human Rights in the United Nations’, Harvard Human Rights Journal, 18: 1–18. Chatterjee, P. (1993) The Nation and its Fragments, Princeton: Princeton University Press. Cheesman, N. (2002) ‘Seeing “Karen” in the Union of Myanmar’, Asian Ethnicity, 3(2): 199–220. Chimni, B.S. (2006) ‘Third World Approaches to International Law: A Manifesto’, International Community Law Review, 8: 3–27. ____ (1998) ‘The Geopolitics of Refugee Studies: A View from the South’, Journal of Refugee Studies, 11(4): 350–74. Chitradon, B. (2011) ‘Thailand Wants to Close Myanmar Refugee Camps’, Agence France Presse, 11 April 2011. Cho, V. (2011) ‘Rocking and Rapping the Karen Diaspora’, The Irrawaddy, 15 February 2011. Christie, N. (1977) ‘Conﬂ icts as Property’, The British Journal of Criminology, 17(1): 1–15. Chusri, D., Rubin, T., Silva, E., Theede, J.D., Wongchalee, S. and Chansawang, P. (2011) Analysis of Donor, INGO/NGO and UN Agency Delivery of Humanitarian Assistance to Displaced Persons from Myanmar Along the Thai-Myanmar Border, Asian Research Center for Migration, Chulalongkorn University: Institute of Asian Studies. Clark, P. (2010) Justice Without Lawyers: Peace, Justice and Reconciliation in Rwanda, Cambridge: Cambridge University Press. Clemmer, D. (1940) The Prison Community, Second Edition, New York: Holt, Rinehart & Wilson. Cohen, A.P. (1985) The Symbolic Construction of Community, London: Ellis Horwood Downloaded by [University of Defence] at 21:28 09 May 2016 and Tavistock Publications. Cohen, S. (2002 [1972]) Folk Devils and Moral Panics: The Creation of the Mods and Rockers, London: Routledge. Colson, E. (2004) ‘Displacement’, in D. Nugent and J. Vincent (eds) A Companion to the Anthropology of Politics, Malden and Oxford: Blackwell Publishing. ____ (2003) ‘Forced Migration and the Anthropological Response’, Journal of Refugee Studies 16: 1–18. Comaroff, J. (2010) ‘Anthropology and Crime: An Interview with Jean Comaroff’, Political and Legal Anthropology Review, 33(1): 133–9. References 169

____ (1996) ‘The Empire’s Old Clothes: Fashioning the Colonial Subject’, in D. Howes (ed.) Cross-cultural Consumption: Global Markets, Local Realities, London: Routledge. Comaroff, J. and Comaroff, J.L. (eds) (2006) Law and Disorder in the Postcolony, London and Chicago: University of Chicago Press. ____ (1989) ‘The Colonization of Consciousness in South Africa’, Economy and Society, 18: 267–96. Comaroff, J.L. and Roberts, S. (1977) ‘Marriage and Extra-marital Sexuality’, Journal of African Law, 21(1): 97–123. Commission on Legal Empowerment of the Poor (2008) Making the Law Work for Everyone: Volume 1, online at: http://www.undp.org/legalempowerment/reports/ concept2action.html (accessed 5 September 2013). Committee for the Coordination of Services to Displaced Persons in Thailand (CCSDPT) (2012) ‘Standards and Principles in Voluntary Repatriation from Thailand to Myanmar’, online at: http://www.ccsdpt.org/resources.htm (accessed 13 September 2013). ____ (2007) CCSDPT/UNHCR Comprehensive Plan 2007/8, online at: http://www. unhcr.org/4691eaa82.html (accessed 12 January 2011). da Costa, R. (2006) The Administration of Justice in Refugee Camps: A Study of Practice, UNHCR Legal and Protection Policy Research Series, Geneva: UNHCR. Cotterrell, R. (1997) Law’s Community: Legal Theory in Sociological Perspective, Oxford: Oxford University Press. Cover, R. (1995) ‘Nomos and Narrative’, in M. Minow, M. Ryan and A. Sarat (eds) Narrative, Violence and the Law: The Essays of Robert Cover, Michigan: University of Michigan Press. ____ (1986) ‘Violence and the Word’, The Yale Law Journal, 95(8): 1601–29. Cowan, J.K., Dembour, M. and Wilson, R. (eds) (2001) Culture and Rights: Anthropological Perspectives, Cambridge: Cambridge University Press. Crawford, A. (2006) ‘Networked Governance and the Post-regulatory State? Steering, Rowing and Anchoring the Provision of Policing and Security’, Theoretical Criminology, 10(4): 449–79. ____ (1997) The Local Governance of Crime: Appeals to Partnerships and Community, Oxford: Clarendon. Crisp, J. (2000) ‘A State of Insecurity: The Political Economy of Violence in Kenya’s Refugee Camps’, African Affairs, 99(397): 601–32. Crisp, J. and Jacobsen, K. (1998) ‘Refugee Camps Reconsidered’, Forced Migration Review, 3: 27–32. Cross, E.B. (1854) ‘On the Karens’, American Oriental Society, 4: 289–316. Downloaded by [University of Defence] at 21:28 09 May 2016 Curwen, P. (1989) ‘The Kawthoolei Women’s Organisation’, Cultural Survival Quarterly, 13(4): 32–3. Cusano, C. (2001) ‘Burma: Displaced Karens. Like Water on the Khu Leaf’, in M. Vincent and B.R. Sorensen (eds) Caught Between Borders: Response Strategies of the Internally Displaced, London: Pluto Press. Danielson, D. (2005) The System of Justice Used in Karenni Refugee Camps 1 and 2 along the Thai-Burmese Border, report prepared for the International Rescue Committee, August 2005, copy on ﬁ le with author. Delang, C.O. (ed.) (2003) Living at the Edge of Thai Society: The Karen in the Highlands of Northern Thailand, London and New York: Routledge Curzon. 170 References

____ (ed.) (2000) Suffering in Silence: The Human Rights Nightmare of the Karen People of Burma, Parkland, F.L.: Universal Publishers. Demusz, K. (1998) ‘From Relief to Development: Negotiating the Continuum on the Thai-Burmese Border’, Journal of Refugee Studies 11(3): 231–44. Donnan, H. (2005) ‘Material Identities: Fixing Ethnicity in the Irish Borderlands’, Identities: Global Studies in Culture and Power, 12: 69–105. Donnan, H. and Magowan, F. (2010) The Anthropology of Sex, Oxford: Berg Publishers. Donnan, H. and Wilson, T. (1999) Borders: Frontiers of Identity, Nation and State, Oxford: Berg Publishers. Donnelly, J. (1984) ‘Cultural Relativism and Universal Human Rights’, Human Rights Quarterly, 6(4): 400–19. Douzinas, C. (2007) Human Rights and Empire: The Political Philosophy of Humanitarianism, Abingdon: Routledge Cavendish. Drum Publications (2006) The History and Traditional Justice System of the Karen People, available in Sgaw Karen and Burmese, online at: http://www.drumpublications. org/download.php (accessed 5 September 2013). Dudley, S. (2010) Materialising Exile: Material Culture and Embodied Experience Among Karenni Refugees in Thailand, New York and Oxford: Berghahn Books. ____ (2003) ‘External’ Aspects of Self-Determination Movements in Burma, Queen Elizabeth House Working Paper Series no. 94, Oxford: Queen Elizabeth House, University of Oxford. ____ (1999) ‘“Traditional” Culture and Refugee Welfare in North-west Thailand’, Forced Migration Review 6: 5–8. Dufﬁ eld, M. (2001) Global Governance and the New Wars, London and New York: Zed Books. Earthrights International (2011) The Burma-China Pipelines: Human Rights Violations, Applicable Law and Revenue Secrecy, Situation Briefer no. 1, March 2011, online at: http://www.earthrights.org/sites/default/files/documents/the-burma-china- pipelines.pdf (accessed 20 April 2011). Edwards, P. (2008) ‘Nationalism by Design: The Politics of Dress in British Burma’, International Institute for Asian Studies Newsletter, 46: 11. Eh Na (2013) ‘Refugee Survey – Most Do Not Want to Go Back to Burma’, Karen News, 15 July 2013. Engel, D.M. and Engel, J.S. (2010) Tort, Custom and Karma: Globalization and Legal Consciousness in Thailand, Stanford: Stanford University Press. Englund, H. (2006) Prisoners of Freedom: Human Rights and the African Poor, Berkeley: University of California Press. Downloaded by [University of Defence] at 21:28 09 May 2016 Enloe, C. (1999) ‘Womenandchildren: Making Sense of the Persian Gulf Crisis’, The Village Voice, 25 September 1999. Ericson, R.V. and Haggerty, K.D. (1999) ‘Governing the Young’, in R. Smandych (ed.) Governable Places: Readings on Governmentality and Crime Control, Aldershot: Ashgate. Eriksen, T.H. (1993) Ethnicity and Nationalism: Anthropological Perspectives, London: Pluto Press. Essed, P., Frerks, G. and Schrijvers, J. (2005) ‘Introduction: Refugees, Agency and Social Transformation’, in P. Essed, G. Frerks and J. Schrijvers (eds) Refugees and the Transformation of Societies: Agency, Policies, Ethics and Politics, New York and Oxford: Berghahn Books. References 171

Falb, K.L., McCormick, M.C., Hemenway, D., Anﬁ nson, K. and Silverman, J.G. (2013) ‘Violence against Refugee Women along the Thai-Burma Border’, International Journal of Gynecology and Obstetrics, 120: 279–83. Falla, J. (1991) True Love and Bartholomew: Rebels on the Burmese Border, Cambridge: Cambridge University Press. Farah, R. (2009) ‘Refugee Camps in the Palestinian and Sahrawi National Liberation Movements: A Comparative Perspective’. Journal of Palestine Studies 38(2): 76–93. Faundez, J. (2010) ‘Legal Pluralism and International Development Agencies: State Building or Legal Reform?’ Warwick School of Law Research Paper 2010/12. Ferguson, J. (2004) ‘Power Topographies’, in D. Nugent and J. Vincent (eds) A Companion to the Anthropology of Politics, Malden and Oxford: Blackwell Publishing. ____ (1990) The Anti-Politics Machine, Cambridge: Cambridge University Press. Ferrante, J. (2008) Sociology: A Global Perspective, Belmont: Thompson Wadsworth. Feyissa, A. and Horn, R. (2008) ‘There is More than One Way of Dying: An Ethiopian Perspective of the Effects of Long Stays in Refugee Camps’, in D. Hollenbach (ed.) Refugee Rights: Ethics, Advocacy and Africa, Washington DC: Georgetown University Press. Fink, C. (2003) ‘Living for Funerals: Karen Teenagers and Romantic Love’, in C.O. Delang (ed.) Living at the Edge of Thai Society: The Karen in the highlands of northern Thailand, London and New York: Routledge Curzon. Fitzpatrick, P. (1989) ‘Crime as Resistance: The Colonial Situation’, The Howard Journal of Criminal Justice, 28(4): 272–81. ____ (1984) ‘Law and Societies’, Osgoode Hall Law Journal, 22: 115–38. Frechette, A. (2002) Tibetans in Nepal: The Dynamics of International Assistance among a Community in Exile, New York and Oxford: Berghahn Books. Freeman, M. (2002) Human Rights, Cambridge: Polity Press. Friedman, E. (1995) ‘Women’s Human Rights: The Emergence of a Movement’, in J. Peters and A. Wolper (eds) Women’s Rights as Human Rights: International Feminist Perspectives, New York and London: Routledge. Fuller, C.F. (1994) ‘Legal Anthropology, Legal Pluralism and Legal Thought’, Anthropology Today, 10(3): 9–12. Furnivall, J.S. (1939) Netherlands India: A Study of Plural Economy, Cambridge: Cambridge University Press. Galanter, M. (1981) ‘Justice in Many Rooms: Courts, Private Ordering and Indigenous Law’, Journal of Legal Pluralism, 19: 1–47. Garland, D. (2002) The Culture of Control: Crime and Social Order in Contemporary Society, Chicago: University of Chicago Press. ____ (1993) Punishment and Modern Society: A Study in Social Theory, Chicago: Downloaded by [University of Defence] at 21:28 09 May 2016 University of Chicago Press. Geertz, C. (2000 [1983]) Local Knowledge: Further Essays in Interpretive Anthropology, New York: Basic Books. ____ (1973) The Interpretation of Cultures: Selected Essays, New York: Basic Books. Geneva Call (2013) Burma/Myanmar: The KNU/KNLA Commits to the Protection of Children and the Prohibition of Conﬂ ict-related Sexual and Gender-based Violence, online at: http://www.genevacall.org/news/press-releases/f-press-releases/2001-2010/ 2013_Communique_KNU.pdf (accessed 3 August 2013). ____ (2011) Armed Non-State Actors and the Protection of Internally Displaced People, conference organised by Geneva Call and the Internal Displacement Monitoring Centre of the Norwegian Refugee Council, Geneva: Geneva Call. 172 References

____ (2010) In Their Words: Perspectives of Armed Non-State Actors on the Protection of Children from the Effects of Armed Confl ict, online at: http://www.genevacall.org/ resources/research/f-research/2001-2010/2010_GC_CANSA_InTheirWords.pdf (accessed 20 February 2011). Gerassi, J. (1989) Jean-Paul Sartre: Hated Conscience of His Century: Volume 1, Chicago: University of Chicago Press. Ghosh, A. (2005) Incendiary Circumstances, New York: Houghton Miffl in. Giddens, A. (1984) The Constitution of Society, Cambridge: Polity Press. Gluckman, M. (1965) Politics, Law and Ritual in Tribal Society, Oxford: Blackwell. Golub, S. (2003) Beyond Rule of Law Orthodoxy: The Legal Empowerment Alternative, Working Paper, Rule of Law Series, Democracy and Rule of Law Project, no. 41, Washington: Carnegie Endowment for International Peace. Goode, E. and Ben-Yahuda, N. (2009) Moral Panics: The Social Construction of Deviance, Second Edition, Oxford: Blackwell. Gravers, M. (ed.) (2007) Exploring Ethnic Diversity in Burma, Copenhagen: Nordic Institute of Asian Studies. ____ (1999) Nationalism as Political Paranoia in Burma, Second Edition, London and New York: Routledge Curzon. Griek, I. (2006a) ‘Traditional Systems of Justice in Refugee Camps: The Need for Alternatives’, US Committee for Refugees and Immigrants Refugee Reports, 27(2): 1–5. ____ (2006b) ‘Access to Justice in Kenyan Refugee Camps: Exploring the Scope of Protection’, Masters thesis, copy on fi le with author. Griffi ths, J. (1986) ‘What is Legal Pluralism?’ Journal of Legal Pluralism and Unoffi cial Law, 24: 1–55. Grundy-Warr, C. and Dean, K. (2003) ‘The Boundaries of Contested Identities: “Kachin” and “Karenni” Spaces in the Troubled Borderlands of Burma’, in E. Berg and H. van Houtum (eds) Routing Borders between Territories, Discourses and Practices, Aldershot: Ashgate. Gupta, A. (2001) ‘The Song of the Nonaligned World: Transnational Identities and the Reinscription of Space in Late Capitalism’, in A. Gupta and J. Ferguson (eds) Culture, Power, Place: Explorations in Critical Anthropology, Durham, N.C. and London: Duke University Press. Gupta, A. and Ferguson, J. (eds) (1997) Anthropological Locations: Boundaries and Grounds of a Field Science, Berkeley: University of California Press. Haddad, E. (2008) The Refugee in International Society: Between Sovereigns, Cambridge: Cambridge University Press. ____ (2007) ‘Danger Happens at the Border’, in P.K. Rajaram and C. Grundy-Warr (eds) Borderscapes: Hidden Geographies and Politics at Territory’s Edge, Minneapolis Downloaded by [University of Defence] at 21:28 09 May 2016 and London: Minnesota University Press. Hafner-Burton, E. and Tsutsui, K. (2005) ‘Human Rights in a Globalizing World: The Paradox of Empty Promises’, American Journal of Sociology, 110(5): 1373–1411. Hampshire, K., Porter, G., Kilpatrick, K., Kyei, P., Adjaloo, M. and Oppong, G. (2008) ‘Liminal Spaces: Changing Inter-Generational Relations among Long- term Liberian Refugees in Ghana’, Human Organization 67(1): 25–37. Hanafi , S. (2008) Palestinian Refugee Camps: Disciplinary Space and Territory of Exception, CARIM Analytic and Synthetic Notes 2008/44, European University Institute/ Robert Schuman Centre for Advanced Studies. References 173

Hanafi , S. and Long, T. (2010) ‘Governance, Governmentalities and the State of Exception in the Palestinian Refugee Camps in Lebanon’, Journal of Refugee Studies, 23(2): 134–59. Hanafi , S. and Tabar, L. (2003) ‘The Intifada and the Aid Industry: The Impact of the New Liberal Agenda on the Palestinian NGOs’, Comparative Studies of South Asia, Africa and the Middle East, 23(1): 205–14. Hansen, T.B. and Stepputat, F. (eds) (2005) Sovereign Bodies: Citizens, Migrants and States in the Postcolonial World, Princeton and Oxford: Princeton University Press. Harding, J., Scanlon, S., Lees, S., Beker, C. and Lim, A.L. (2008) ‘Access to Justice and the Rule of Law’, Forced Migration Review, 30: 28–30. Harding, J. and Varadan, S. (2010) ‘A Community-based Approach to Refugee Protection in a Protracted Refugee Situation’, Humanitarian Exchange Magazine 46: 8–10. ____ (2009) New Approaches to Refugee Protection in a Protracted Refugee Situation, paper presented at the Oxford Refugee Studies Centre conference Protecting People in Confl ict & Crisis, 22–24 September 2009. Harper, E. (2011) Traditional Justice: Practitioners’ Perspectives on Engaging with Customary Justice Systems, Geneva: IDLO. Harrell-Bond, B.E. (2004) ‘Weapons of the Weak’, Journal of Refugee Studies, 17: 27–8. ____ (1999) ‘The Experience of Refugees as Recipients of Aid’, in A. Ager (ed.) Refugees: Perspectives on the Experience of Forced Migration, London: Cassell. ____ (1986) Imposing Aid: Emergency Assistance to Refugees, Oxford: Oxford University Press. Harriden, J. (2002) ‘“Making a Name for Themselves”: Karen Identity and the Politicization of Ethnicity in Burma’, Journal of Burma Studies, 7: 84–144. Harris-Short, S. (2003) ‘International Human Rights Law: Imperialist, Inept and Ineffective? Cultural Relativism and the UN Convention on the Rights of the Child’, Human Rights Quarterly, 25: 130–81. Hathaway, J. (2005) The Rights of Refugees under International Law, Cambridge: Cambridge University Press. ____ (1991) ‘Reconceiving Refugee Law as Human Rights Protection’, Journal of Refugee Studies, 4(2): 113–31. Hayami, Y. (2004) Between Hills and Plains: Power and Practice in Socio-Religious Dynamics Among the Karen, Kyoto: Kyoto University Press and Transpacifi c Press. ____ (2003) ‘Morality, Sexuality and Mobility: Changing Moral Discourse and Self’, in C.O. Delang (ed.) Living at the Edge of Thai Society: The Karen in the Highlands of Northern Thailand, London and New York: Routledge Curzon. Downloaded by [University of Defence] at 21:28 09 May 2016 Heppner, K. (2006) We Have Hands the Same as Them: Struggles for Local Sovereignty and Livelihoods by Internally Displaced Karen Villagers in Burma, Karen Human Rights Group, online at: http://www.khrg.org (accessed 10 December 2010) (currently unavailable). Hinton, P. (1983) ‘Do the Karen Really Exist?’ in J. McKinnon and W. Bhruksasri (eds) Highlanders of Thailand, Kuala Lumpur and Oxford: Oxford University Press. Hobsbawm, E. (1983) ‘Introduction: Inventing Traditions’, in E. Hobsbawm and T. Ranger (eds) The Invention of Tradition, Cambridge: Cambridge University Press. Holzer, E. (2012) ‘A Case Study of Political Failure in a Refugee Camp’, Journal of Refugee Studies 25(2): 257–81. 174 References

Home Offi ce (2002) Protecting the Public: Strengthening Protection Against Sex Offenders and Reforming the Law on Sexual Offences, Home Offi ce: CM 5668, HMSO. Horn, R. (2010a) ‘Responses to Intimate Partner Violence in Kakuma Refugee Camp: Refugee Interactions with Agency Systems’, Social Science & Medicine 70: 160–168. ____ (2010b) ‘Exploring the Impact of Displacement and Encampment on Domestic Violence in Kakuma Refugee Camp’, Journal of Refugee Studies, 23(3): 356–76. Horst, C. (2006) Transnational Nomads: How Somalis Cope with Refugee Life in the Dadaab Camps of Kenya, New York and Oxford: Berghahn Books. Horstmann, A. (2011a) ‘Ethical Dilemmas and Identifi cations of Faith-based Humanitarian Organizations in the Karen Refugee Crisis’, Journal of Refugee Studies 24(3): 513–32. ____ (2011b) ‘Sacred Networks and Struggles among the Karen Baptists on the Thai-Burma Border’, Moussons, 17: 81–104. Houtman, G. (1999) Mental Culture in Burmese Crisis Politics: Aung San Suu Kyi and the National League for Democracy, Study of Languages and Cultures of Asia and Africa Monograph Series no. 33, Tokyo: Institute for the Study of Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies. Huguet, J.W. and Punpuing, S. (2005) International Migration in Thailand, Thailand: International Organisation for Migration, Regional Offi ce Bangkok. Human Rights Documentation Unit (2008) Burma Human Rights Yearbook 2007, online at: http://www.burmalibrary.org/docs5/HRDU-archive/Burma%20 Human%20Righ/ (accessed 12 January 2011). Human Rights Watch (2013a) All You Can Do is Pray: Crimes Against Humanity and Ethnic Cleansing of Rohingya Muslims in Burma’s Arakan State, Washington: Human Rights Watch. ____ (2013b) Hostages of the Gatekeepers: Abuses against Internally Displaced in Mogadishu, Somalia, Washington: Human Rights Watch. ____ (2012) Ad Hoc and Inadequate: Thailand’s Treatment of Refugees and Asylum Seekers, Washington: Human Rights Watch. ____ (2004) Out of Sight, Out of Mind: Thai Policy toward Burmese Refugees, Washington: Human Rights Watch. Hunt, N.R. (1991) ‘Noise Over Camoufl aged Polygamy: Colonial Morality Taxation and a Woman-Naming Crisis in Belgian Africa’, Journal of African History, 32(3): 471–94. Hunter, M. (2009) ‘The Failure of Self-Reliance in Refugee Settlements’, POLIS 2, online at: http://www.polis.leeds.ac.uk/assets/fi les/students/student-journal/ma- Downloaded by [University of Defence] at 21:28 09 May 2016 winter-09/meredith-hunter-winter-09.pdf (accessed 10 September 2013). Hutchinson, Y. (2013) ‘The Transference of Gender-Based Norms in the Law Reform Process: A Refl ection on my Work in Thailand’, Querelles 16, online at: http://www.querelles.de/index.php/qjb/article/view/6/17 (accessed 10 September 2013). Huxley, A. (2001) ‘Precolonial Burmese Law: Conical Hat and Shoulder Bag’, International Institute for Asian Studies Newsletter, online at: http://www.iias.nl/ iiasn/25/theme/25T7.html (accessed 20 September 2010). Hyndman, J. (2004) ‘Refugee Camps as Confl ict Zones: The Politics of Gender’, in W. Giles and J. Hyndman (eds) Sites of Violence: Gender and Confl ict Zones, Berkeley and London: California University Press. References 175

____ (2000) Managing Displacement: Refugees and the Politics of Humanitarianism, Minneapolis and London: University of Minnesota Press. ____ (1997) ‘Refugee Self-Management and the Question of Governance’, Refuge, 16(2): 16–22. Hyndman, J. and de Alwis, M. (2009) ‘Reconstituting the Subject: Feminist Politics of Humanitarian Assistance’, in M. Hajdukowski-Ahmed, N. Khanlou and H. Moussa (eds) Not Born a Refugee Woman: Contesting Identities, Rethinking Practices, New York and Oxford: Berghahn Books. ____ (2000) ‘Capacity building, accountability and humanitarianism in Sri Lanka’, Forced Migration Review 8: 16–19. Ibhawoh, B. (2003) ‘Defi ning Persecution and Protection: The Cultural Relativism Debate and the Rights of Refugees’, in N. Steiner, M. Gibney and G. Loescher (eds) Problems of Protection: The UNHCR, Refugees and Human Rights, New York and London: Routledge. Ignatieff, M. (2001) ‘Dignity and Agency’ in Amy Gutmann ed. Human Rights as Politics and Idolatry, Princeton: Princeton University Press. Integrated Regional Information Networks (IRIN) (2009a) Addressing Sexual Violence in Mae La Refugee Camp, online at: http://www.irinnews.org/report/82372/ (accessed 10 December 2010). ____ (2009b) Myanmar-Thailand: Bringing the Law to Burmese Refugee Camps, 23 April 2009, online at: http://www.unhcr.org/refworld/docid/49f55dd814.html (accessed 26 November 2010). Inter-Agency Standing Committee (2005) Guidelines for Gender-based Violence Interventions in Humanitarian Settings, Geneva: Inter-Agency Standing Committee. International Institute for Strategic Studies (2007) The Military Balance 2006–2007, London: Europa Publications. International Organisation for Migration (IOM) (2012) Resettlement Milestone Reached as 100,000th Refugee Leaves Thai Camps, online at: http://www.iom.int/cms/en/ sites/iom/home/news-and-views/press-briefing-notes/pbn-2012/pbn-listing/ resettlement-milestone-reached-a.html (accessed 10 August 2013). ____ (2008) 9 Lives. Refugee Stories: From Burma to America, Mae Sot: International Organisation for Migration. International Rescue Committee (IRC) (2010) Legal Assistance Center (LAC) Mid Term Assessment, October 2010. ____ (2009) Sub-Project Monitoring Report – Narrative Report. Reporting period: January 1st–June 30th, 2009, copy on fi le with author. ____ (2007) Assessment of Protection Issues, with a Focus on Access to Justice and the Rule of Law, copy on fi le with author. Downloaded by [University of Defence] at 21:28 09 May 2016 Irwin, J. and Cressey, D. (1962) ‘Thieves, Convicts and the Inmate Culture’, Social Problems 10: 142–55. James, H. (2010) ‘Resources, Rent-seeking and Reform in Thailand and Myanmar (Burma)’, Asian Survey, 50: 426–48. Johnston, L. and Shearing, C. (2003) Governing Security: Explorations in Policing and Justice, London: Routledge. Kaiser K. (2002) ‘The UNHCR and Withdrawal from Kiryandongo: Anatomy of a Handover’, Refugee Survey Quarterly 21 (1&2): 201–27. Karashvili, J. (2001) ‘Georgia: Coping by Organising. Displaced Georgians from Abkhazia’, in M. Vincent and B.R. Sorensen (eds) Caught Between Borders: Response Strategies of the Internally Displaced, London: Pluto Press. 176 References

Karen Human Rights Group (KHRG) (2013) Losing Ground: Land Confl icts and Collective Action in Eastern Myanmar, online at: http://www.khrg.org (accessed 10 August 2013). ____ (2011) ‘All the Information I’ve Given You, I Faced it Myself’: Rural Testimony on Abuse in Eastern Burma since November 2010, online at: http://www.khrg.org (accessed 10 August 2013). ____ (2010). Functional Refoulement: Camps in Tha Song Yang District Abandoned as Refugees Bow to Pressure, online at: (accessed 10 August 2013). ____ (2008) Village Agency: Rural Rights and Resistance in a Militarised Karen State, online at: http://www.khrg.org (accessed 10 August 2013). ____ (2006) Surviving in Shadow: Widespread Militarization and the Systematic Use of Forced Labour in the Campaign for Control of Tatun District, online at: http:// www.khrg.org (accessed 10 August 2013). Karen National Union (KNU) (2006) The Karens and their Struggle for Freedom, Revised Edition, online at: http://www.ibiblio.org/obl/docs3/Karen-Struggle-ocr.pdf (accessed 5 September 2013). Karen News (2013a) ‘Irrawaddy Delta Holds First Ever Wrist-Tying Ceremony’, online at: http://www.youtube.com/watch?v=uL07zFwMRb0 (accessed 10 August 2013). ____ (2013b) ‘Talk to Us and Listen to our Voices…’, 24 June 2013. Karen Refugee Committee (KRC) Karen Refugee Committee Monthly Reports, reports from March 1984 to March 2013, online at: http://www.burmalibrary.org/show. php?cat=1831&sl=0 (accessed 5 September 2013). Karen Women Organisation (KWO) (2011a) Analysis Paper by KWO on SGBV Services in Karen Refugee Camps, copy on fi le with author. ____ (2011b). ‘Request for Camp Management Working Group to Discuss Justice System in Camps on Nov 22nd 2011’, copy on fi le with author. ____ (2010) Walking Amongst Sharp Knives, online at: http://www.karenwomen.org/ Reports/WalkingAmongstSharpKnives.pdf (accessed 5 September 2013). ____ (2007) State of Terror, online at: http://www.karenwomen.org/Reports/state%20 of%20terror%20report.pdf (accessed 5 September 2013). ____ (2004) Shattering Silences, online at: http://www.ibiblio.org/obl/docs/Shattering_ Silences.htm (accessed 5 September 2013). Keck, M.E. and Sikkink, K. (1998) Activists Beyond Borders: Advocacy Networks in International Politics, Ithaca: Cornell University Press. Keenan, P. (2012a) The Karen National Union Negotiations 1949–2012, Chiang Mai: Burma Centre for Ethnic Studies. ____ (2012b) Burma’s Ethnic Ceasefi re Agreements, Burma Centre for Ethnic Studies, Downloaded by [University of Defence] at 21:28 09 May 2016 Peace and Reconciliation Briefi ng Paper no. 1, Chiang Mai: Burma Centre for Ethnic Studies. Kelly, T. (2006) Law, Violence and Sovereignty among West Bank Palestinians, Cambridge: Cambridge University Press. Kennedy, D. (2002) ‘The International Human Rights Movement: Part of the Problem?’ Harvard Human Rights Journal, 15: 101–25. Kenny, P. and Lockwood-Kenny, K. (2011) ‘A Mixed Blessing: Karen Resettlement to the United States’, Journal of Refugee Studies, 24(2): 217–38. Keogh, S. and Ruijters, Y. (2012) Dangerous liaisons? A Historical Review of UNHCR’s Engagement with Non-state Armed Actors, UNHCR Policy Development and Evaluation Service, Geneva: UNHCR. References 177

Keyes, C. (2003) ‘Afterword: The Politics of ‘Karen-ness’ in Thailand’, in C.O. Delang (ed.) Living at the Edge of Thai Society: The Karen in the Highlands of Northern Thailand, London and New York: Routledge Curzon. ____ (ed.) (1979) Ethnic Adaptation and Identity: The Karen on the Thai Frontier with Burma, Philadelphia: Institute for the Study of Human Issues. Kibreab, G. (2005) ‘Refugeehood, Loss and Social Change: Eritrean Refugees and Returnees’ in P. Essed, G. Frerks and J. Schrijvers (eds) Refugees and the Transformation of Societies: Agency, Policies, Ethics and Politics, New York and Oxford: Berghahn Books. ____ (2004) ‘Pulling the Wool over the Eyes of the Strangers: Refugee Deceit and Trickery in Institutionalized Settings’, Journal of Refugee Studies, 17(1): 1–26. ____ (1993) ‘The Myth of Dependency among Camp Refugees in Somalia 1979–1989’, Journal of Refugee Studies, 6 (4): 321–49. Kramer, T. (2009) Neither War nor Peace: The Future of the Cease-fi re Agreements in Burma, Amsterdam: Transnational Institute. van Krieken, R. (2001) ‘Legal Informalism, Power and Liberal Governance’, Social & Legal Studies, 10(1): 5–22. Kuroiwa, Y. and Verkuyten, M. (2008) ‘Narratives and the Constitution of a Common Identity: The Karen in Burma’, Identities, 15(4): 391–412. Kyaw Kha (2011) ‘Thailand tightens control of Burmese groups on border’, Mizzima News, 23 February 2011. Lacey, N., Wells, C. and Quick, O. (eds) (2003) Reconstructing Criminal Law: Texts and Materials, Third Edition, Cambridge: Cambridge University Press. Lah Khin Moo (2013) ‘Still not Safe to Send back Refugees’, Karen News, 26 May 2013. di Lampedusa, T. (2007 [1961]) The Leopard, London: Vintage. Landau, L. (2012) ‘Wither Policy? Perspectives on Protecting the Displaced Beyond “Refugee Policy”’, paper prepared for the workshop ‘The Making, Methods and Movement of Global Refugee Policy’, Refugee Studies Centre, Oxford, 5 December 2012. Lang, H. (2002) Fear and Sanctuary: Burmese Refugees in Thailand, Cornell: Cornell University Press. Lattu, K. (2008) To Complain or not to Complain: Still the Question. Consultations with Humanitarian Aid Benefi ciaries on their Perceptions of Efforts to Prevent and Respond to Sexual Exploitation and Abuse, Humanitarian Accountability Partnership Report, online at: http://www.hapinternational.org/pool/fi les/bbc-report-lowres. pdf (accessed 10 December 2010). Lawyers Committee for Human Rights (2002) Refugees, Rebels and the Quest for Justice, Downloaded by [University of Defence] at 21:28 09 May 2016 New York: Lawyers Committee for Human Rights. Lehman, F.K. (1979) ‘Who are the Karen and if so Why? Karen Ethnohistory and a Formal Theory of Ethnicity’, in C. Keyes (ed.) Ethnic Adaptation and Identity: The Karen on the Thai Frontier with Burma, Philadelphia: Institute for the Study of Human Issues. Lieberman, V.B. (1978) ‘Ethnic Politics in Eighteenth-century Burma’, Modern Asian Studies, 12(3): 455–82. Lintner, B. (1999) Burma in Revolt: Opium and Insurgency Since 1948, Chiang Mai: Silkworm Books. Lischer, S. (2005) Dangerous Sanctuaries: Refugee Camps, Civil War and the Dilemmas of Humanitarian Aid, Ithaca: Cornell University Press. 178 References

Loader, I. and Walker, N. (2007) Civilizing Security, Cambridge: Cambridge University Press. Loescher, G. and Milner, J. (2005) Protracted Refugee Situations: Domestic and International Security Implications, New York and London: Routledge. Loescher, G., Milner, J., Newman, E. and Troeller, G. (2007), ‘Protracted Refugee Situations and the Regional Dynamics of Peacebuilding’, Conﬂ ict, Security and Development, 7(3): 491–501. Lubkemann, S.C. (2008) Culture in Chaos: An Anthropology of the Social Condition in War, Chicago and London: University of Chicago Press. Lui, R.N. (2007) ‘Such a Long Journey: Protracted Refugee Situations in Asia’, Global Change, Peace and Security, 19(3): 185–203. MacLachlan, H. (2006) ‘The Don Dance: An Expression of Karen Nationalism’, Voices: The Journal of New York Folklore, 32(3–4): 26–32. Malinowski, B. (1926) Crime and Custom in Savage Society, New York and London: Harcourt Brace Kegan Paul. Malkki, L. (2002) ‘News from Nowhere. Mass Displacement and Globalized ‘Problems of Organization’, Ethnography, 3(3): 351–60. ____ (1995a) ‘Refugees and Exile: From “Refugee Studies” to the National Order of Things’, Annual Review of Anthropology, 24: 495–523. ____ (1995b) Purity and Exile: Violence, Memory and National Cosmology Among Hutu Refugees in Tanzania, Chicago: University of Chicago Press. ____ (1992) ‘National Geographic: The Rooting of Peoples and Territorialization of National Identity among Scholars and Refugees’, Cultural Anthropology 7(1): 24–44. Malseed, K. (2008) ‘Where there is no Movement: Local Resistance and the Potential for Solidarity’, Journal of Agrarian Change, 8(2): 489–514. Margulies, P. (1996) ‘The Violence of Law and Violence against Women’, Cardozo Studies in Law and Literature, 8: 179–202. Marks, M. (2001) Young Warriors: Youth Politics, Identity and Violence in South Africa, Johannesburg: Witwatersrand University Press. Marlowe, D. (1979) ‘In the Mosaic: The Cognitive and Structural Aspects of Karen- Other Relationships’, in C. Keyes (ed.) Ethnic Adaptation and Identity: The Karen on the Thai Frontier with Burma, Philadelphia: Institute for the Study of Human Issues. Marshall, H.I. (1997 [1922]) The Karen People of Burma: A Study in Anthropology and Ethnology, Bangkok: White Lotus Press. Maruna, S. (2001) Making Good: How Ex-Convicts Reform and Rebuild Their Lives, Washington: American Psychological Association. Downloaded by [University of Defence] at 21:28 09 May 2016 Mason, E. (1862) Civilising Mountain Men: Or, Sketches of Mission Work among the Karens, Philadelphia: American Baptist Publication Society. Mason, F. (1847) The Karen Apostle, or, the Memoir of Ko Thah-Byu the First Karen Convert, Boston: Gould, Kendall and Lincoln. ____ (1868) ‘On Dwellings, Works of Art, Laws &c. of the Karens’, Journal of the Asiatic Society, 3: 125–168. Mathiesen, T. (1965) The Defences of the Weak: A Sociological Study of a Norwegian Correctional Institution, London: Tavistock. Matthews, R. (1988) Informal Justice, London: Sage. References 179

McConnachie, K. (2012). ‘Rethinking the “Refugee Warrior”: The Karen National Union and Refugee Protection on the Thai-Burma Border’, Journal of Human Rights Practice, 4(1): 30–56. McEvoy, K. (2007) ‘Beyond Legalism: Towards a Thicker Understanding of Transitional Justice’, Journal of Law and Society, 34: 411–40. ____ (2001) Paramilitary Imprisonment in Northern Ireland, Oxford: Oxford University Press. McEvoy, K. and Eriksson, A. (2008) ‘Who Owns Justice? Community, State, and the Northern Ireland Transition’, in J. Shapland (ed.) Justice, Community and Civil Society: A Contested Terrain across Europe, Cullompton: Willan. McEvoy, K. and Mika, H. (2002) ‘Restorative Justice and the Critique of Informalism in Northern Ireland’, British Journal of Criminology, 42(3): 534–62. Merry, S.E. (2006a) Human Rights and Gender Violence: Translating International Law into Local Justice, Chicago: University of Chicago Press. ____ (2006b) ‘Transnational Human Rights and Local Activism: Mapping the Middle’, American Anthropologist, 108(1): 38–51. ____ (2003) ‘Rights and the Talk and the Experience of Law: Implementing Women’s Human Rights to Protection from Violence’, Human Rights Quarterly, 25(2): 343–81. ____ (2000a) Colonizing Hawai’i: The Cultural Power of Law, Princeton: Princeton University Press. ____ (2000b) ‘Comments’, Current Anthropology, 41(1): 90–1. ____ (1999) ‘Pluralizing Paradigms: From Gluckman to Foucault’, The Political and Legal Anthropology Review, 22(1): 115–22. ____ (1991) ‘Law and Colonialism’, Law and Society Review, 25: 889–922. ____ (1988) ‘Legal Pluralism’, Law and Society Review, 22: 869–96. ____ (1982) ‘The Social Organization of Mediation in Nonindustrial Societies’, in R. Abel (ed.) The Politics of Informal Justice: Volume 2, New York and London: Academic Press. Meyer, S. (2006) The ‘Refugee Aid and Development’ Approach in Uganda: Empowerment and Self-reliance of Refugees in Practice, UNHCR Working Paper. Meyerstein, A. (2007) ‘Between Law and Culture: Rwanda’s Gacaca and Postcolonial Legality’, Law & Social Inquiry 32(2): 467–508. Milner, J. and Loescher, G. (2011) Responding to Protracted Refugee Situations: Lessons from a Decade of Discussion, Forced Migration Policy Brieﬁ ng 6, Oxford: Refugee Studies Centre, University of Oxford. Mischung, R. (2003) ‘When it is Better to Sing than to Speak: The Use of Traditional Verses (Hta) in Tense Social Situations’, in C.O. Delang (ed.) Living at the Edge of Downloaded by [University of Defence] at 21:28 09 May 2016 Thai Society: The Karen in the Highlands of Northern Thailand, London and New York: Routledge Curzon. Miyajima, K. (2010) ‘Karen Refugees from Myanmar Prepare for New Life in Japan’, The Asahi Shimbun, 28 August 2010. Moerman, M. (1965) ‘Ethnic Identiﬁ cation in a Complex Civilisation: Who Are the Lue?’ American Anthropologist, 67(5): 1215–30. Mon Forum (2008) ‘DKBA Executes Couple Accused of Being Witches’, The Mon Forum 12(2008): 3, online at: http://www.burmalibrary.org/docs6/MF2008-12. pdf (accessed 5 September 2013). 180 References

Montgomery, H. (2001) ‘Imposing Rights? A Case Study of Child Prostitution in Thailand’, in J.K. Cowan, M. Dembour, and R. Wilson (eds) Culture and Rights: Anthropological Perspectives, Cambridge: Cambridge University Press. Moore, S.F. (2000 [1978]) Law as Process: An Anthropological Approach, Oxford: James Currey. ____ (1989) ‘History and the Redeﬁ nition of Custom on Kilimanjaro’, in J. Starr and J.F. Collier (eds) History and Power in the Study of Law: New Directions in Legal Anthropology, Ithaca: Cornell University Press. ____ (1973) ‘Law and Social Change: The Semi-Autonomous Social Field as an Appropriate Subject of Study’, Law and Society Review, 7(4): 719–14. Moses (2013) ‘The Rainbow Group in Mae La Camp’, Forced Migration Review 42. Moulin, C. and Nyers, P. (2007) ‘“We Live in a Country of UNHCR” – Refugee Protests and Global Political Society’, International Political Sociology, 1(4): 356–72. Muggah, R. (ed.) (2006) No Refuge: The Crisis of Refugee Militarization in Africa, London and New York: Zed Books. Muggah, R. and Mogire, E. (2006) ‘Arms Availability and Refugee Militarization in Africa: Conceptualizing the Issues’, in R. Muggah (ed.) No Refuge: The Crisis of Refugee Militarization in Africa, London and New York: Zed Books. Muncie, J. (2004) Youth and Crime, Second Edition, London: Sage. Mundy, M. (ed.) (2002) Law and Anthropology, Second Series, Aldershot: Dartmouth Ashgate. Murray, D.E. (1961) ‘Ancient Laws on Adultery: A Synopsis’, Journal of Family Law, 1: 89–104. Mutua, M. (2002) Human Rights: A Political and Cultural Critique, Philadelphia: University of Pennsylvania Press. ____ (2001) ‘Savages, Victims and Saviours: The Metaphor of Human Rights’, Harvard International Law Journal, 42(1): 201–45. Myint-U, T. (2007) The Making of Modern Burma, Cambridge: Cambridge University Press. Nader, L. (2005) The Life of the Law: Anthropological Projects, Berkeley: University of California Press. ____ (1990) Harmony Ideology: Justice and Control in a Mountain Zapotec Village, Stanford: Stanford University Press. ____ (1979) ‘Disputing Without the Force of Law’, Yale Law Journal, 88(5): 998–1021. ____ (1969a) Law in Culture and Society, Chicago: Aldine Publishing. ____ (1969b) ‘Up the Anthropologist: Perspectives Gained from Studying Up’, in Downloaded by [University of Defence] at 21:28 09 May 2016 D. Hynes (ed.) Reinventing Anthropology, New York: Pantheon Press. Nanuam, W. (2008) ‘Muslims from Burma: Samak Wants Rohingya Put on an Island’, Bangkok Post, 29 March 2008. Nashif, E. (2008) Palestinian Political Prisioners: Identity and Community, London and New York: Routledge. Nelken, D. (2010) Comparative Criminal Justice: Making Sense of Difference, London: Sage Publications. ____ (2009) ‘Corruption as Governance: Law, Transparency and Appointment Procedures in Italian Universities’, in F. von Benda-Beckmann, K. von References 181

Benda-Beckmann, and J. Eckert (eds) Rules of Law and Laws of Ruling: On the Governance of Law, Farnham: Ashgate Publishing. Norrie, A. (1999) ‘From Law to Popular Justice: Beyond Antinomialism’, in E. Darian-Smith and P. Fitzpatrick (eds) Laws of the Postcolonial, Ann Arbor: University of Michigan Press. Norwegian Refugee Committee (2007) Camp Management Toolkit, online at: http:// www.nrc.no/?aid=9380323 (accessed April 2013). Nyers, P. (2006) Rethinking Refugees: Beyond States of Emergency, New York and London: Routledge. Ofﬁ ce for National Statistics. (2013) Statistical Bulletin: Crime in England and Wales, Year Ending March 2013, online at: http://www.ons.gov.uk/ons/dcp171778_ 318761.pdf (accessed 5 September 2013). Oh, S. (2013) Competing Forms of Sovereignty in the Karen State of Myanmar, Institute of South East Asian Studies, Working Paper no.1, online at: http://www.iseas. edu.sg/documents/publication/ISEAS_Working%20Papers-25mar13.pdf (accessed 5 September 2013). Oh, S. and van der Stouwe, M. (2008) ‘Education, Diversity, and Inclusion in Burmese Refugee Camps in Thailand’, Comparative Education Review, 52(4): 589–617. O’Kane, M. (2007) ‘Blood, Sweat and Tears: The Political Agency of Women Activist-Refugees of Burma’, Intersections: Gender, History and Agency in the Asian Context, online at: http://intersections.anu.edu.au/issue15/okane.htm (accessed 17 January 2011). Olivius, E. (2013) ‘(Un)Governable Subjects: The Limits of Refugee Participation in the Promotion of Gender Equality in Humanitarian Aid’, Journal of Refugee Studies 26(4). O’Neill, M. (2010) Asylum, Migration and Community, Bristol: Policy Press. Otlhogile, B. (1993) ‘Criminal Justice and the Problems of a Dual Legal System in Botswana’, Criminal Law Forum, 4(3): 521–33. Palmary. I. (2008) ‘Gender, Race and Culture: Unpacking Discourses of Tradition and Culture in UNHCR Refugee Policy’, Annual Review of Critical Psychology, 6: 125–33. Paris, R. (2002) ‘International Peacebuilding and the “Mission Civilisatrice”’, Review of International Studies, 28(4): 637–56. Paung, S. (2006) ‘KNLA Commanders Accuse Unnamed KNU Leaders’, The Irrawaddy, 31 July 2006, online at: http://www.irrawaddy.org/article.php?art_ id=6034 (accessed 10 December 2010) (currently unavailable). Peteet, J. (2005) Landscape of Hope and Despair: Palestinian Refugee Camps, Philadelphia: University of Pennsylvania Press. Downloaded by [University of Defence] at 21:28 09 May 2016 Peters, J. and Wolper, A. (eds) (1995) Women’s Rights as Human Rights: International Feminist Perspectives, New York and London: Routledge. Physicians for Human Rights (2012) Under Siege in Kachin State, Burma, Washington: PHR. Pickering S. and O’Kane, M. (2002) ‘Policing, Exile and Gender in States’ Borderlands’, Current Issues in Criminal Justice, 14(1): 106–10. Pimentel, D. (2010a) ‘Rule of Law Reform without Cultural Imperialism? Reinforcing Customary Justice through Collateral Review in South Sudan’, Hague Journal on the Rule of Law, 2(1): 1–4. 182 References

____ (2010b) ‘Legal Pluralism and the Rule of Law: Can Indigenous Justice Survive?’ Harvard International Review, 32(2): 32–9. Pirie, F. (2007) ‘Order, Individualism and Responsibility: Contrasting Dynamics on the Tibetan Plateau’, in K. von Benda-Beckmann and F. Pirie (eds) Order and Disorder, New York and Oxford: Berghahn Books. ____ (2006) ‘Legal Autonomy as Political Engagement: The Ladakhi Village in the Wider World’, Law and Society Review, 40(1): 77–104. Po Chit, S. (1947) Karens and Karen State, Burma: Karen National Union. Poe Kwa Yay (2012) ‘Two Mae La Camp Refugees Shot and Killed’, Karen News, 12 November 2012. Pouligny, B. (2006) Peacekeeping Operations Seen From Below, London: Hurst Publishing. Pratt, J. (2002) Punishment & Civilization, London: Sage. Rajagopal, B. (2009) ‘The International Human Rights Movement Today’, Maryland Journal of International Law, 24: 56–62. Rajah, A. (2002) ‘A “Nation of Intent” in Burma: Karen Ethno-nationalism, Nationalism and Narrations of Nation’, The Paciﬁ c Review, 15(4): 517–37. ____ (1986) ‘Remaining Karen: A Study of Cultural Reproduction and the Maintenance of Identity’, Ph.D. thesis, Australian National University. Rajaram, P.K. (2007) ‘Locating Political Space through Time: Asylum and Excision in Australia’, in P.K. Rajaram and C. Grundy-Warr (eds) Borderscapes: Hidden Geographies and Politics at Territory’s Edge, Minneapolis and London: Minnesota University Press. ____ (2002) ‘Humanitarianism and Representations of the Refugee’, Journal of Refugee Studies, 15(3): 247–64. Ramadan, A. (2013) ‘Spatialising the Refugee Camp’, Transactions of the Institute of British Geographers, 38(1): 65–77. ____ (2009) ‘Destroying Nahr el-Bared: Sovereignty and Urbicide in the Space of Exception’, Political Geography, 28: 153–63. Refugees International (2012) Syrian Women & Girls: No Safe Refuge, online at: http:// refugeesinternational.org/policy/ﬁ eld-report/syrian-women-girls-no-safe-refuge (accessed 16 July 2013). Renard, R.D. (1990) ‘The Karen Rebellion in Burma’, in R. Premdas, S.W.R. de A. Samarasinghe and A.B. Anderson (eds) Secessionist Movements in Comparative Perspective, London: International Centre for Ethnic Studies. Riles, A. (2006) ‘Anthropology, Human Rights, and Legal Knowledge: Culture in the Iron Cage’, American Anthropologist, 108(1): 52–65. Robinson, W.C. (2000) Terms of Refuge: The Indochinese Exodus and the International Response, London and New York: Zed Books. Downloaded by [University of Defence] at 21:28 09 May 2016 Rogers, B. (2004) A Land Without Evil, Grand Rapids Michigan: Monarch Books. Rosen, L. (2006) Law as Culture: An Invitation, Princeton and Oxford: Princeton University Press. ____ (1989) The Anthropology of Justice: Law as Culture in Islamic Society, Cambridge: Cambridge University Press. Rosenfeld, M. (2004) Confronting the Occupation: Work, Education, and Political Activism of Palestinian Families in a Refugee Camp, Stanford: Stanford University Press. Rothkegel, S., Poluda, J., Wonani, C., Papy, J., Engelhardt-Wendt, E., Weyermann, B. and Hennig, R. (2008) Evaluation of UNHCR’s Efforts to Prevent and Respond to References 183

Sexual and Gender-based Violence in Situations of Forced Displacement, Policy Development and Evaluation Service, Geneva: UNHCR. Safman, R. (2007) ‘Minorities and State-building in Mainland Southeast Asia’, in N. Ganesan and Kyaw Yin Hlaing (eds) Myanmar: State, Society and Ethnicity, Singapore: Institute of South East Asian Studies. Said, E. (1984) ‘Refl ections on Exile’, Granta Magazine, 13: 159–72. Sakhong, L. (2012) The Dynamics of Sixty Years of Ethnic Armed Confl ict in Burma, Burma Centre for Ethnic Studies, Peace and Reconciliation Analysis Paper no. 1, online at: http://burmaethnicstudies.net/pdf/Analysis%20Paper%20No%201. pdf (accessed 5 September 2013). Samuels, K. (2006) Rule of Law Reform in Post-Confl ict Countries: Operational Initiatives and Lessons Learned, World Bank Social Development Papers no. 37, online at: http://siteresources.worldbank.org/INTCPR/Resources/WP37_web.pdf (accessed 2 September 2010). San C. Po (2001 [1928]) Burma and the Karens, Bangkok: White Lotus. Santos, B. de Sousa. (2006) ‘The Heterogeneous State and Legal Pluralism in Mozambique’, Law & Society Review, 40(1): 39–76. ____ (1987) ‘Law: A Map of Misreading: Toward a Postmodern Conception of Law’, Journal of Law and Society, 14(3): 279–302. ____ (1984) ‘From Customary Law to Popular Justice’, Journal of African Law, 28(1– 2): 90–8. ____ (1982) ‘Law and Revolution in Portugal: The Experiences of Popular Justice after the 25th of April 1974’, in R. Abel (ed.) The Politics of Informal Justice: Volume 2, New York and London: Academic Press. ____ (1977) ‘The Law of the Oppressed: The Construction and Reproduction of Legality in Pasargada’, Law & Society Review, 12(1): 5–126. Save the Children (2007) Child Protection in Emergencies: Priorities, Principles and Practices, online at: http://www.unhcr.org/refworld/docid/47fb94fe2.html (accessed 11 March 2011). Scheper-Hughes, N. (2008) ‘Illegal Organ Trade: Global Justice and the Traffi c in Human Organs’, in R.W.G. Gruessner and E. Benedetti (eds) Living Donor Organ Transplantation, New York and London: McGraw-Hill Medical. ____ (2006) ‘Death Squads and Democracy in Northeast Brazil’, in J. Comaroff and J.L. Comaroff (eds) Law and Disorder in the Postcolony, London and Chicago: University of Chicago Press. Schmiechen, M.M. (2003) ‘Parallel Lives, Uneven Justice: An Analysis of Rights, Protection and Redress for Refugee and Internally Displaced Women in Camps’, St Louis University Public Law Review, 22: 473–520. Downloaded by [University of Defence] at 21:28 09 May 2016 Schnoebelen, J. (2009) ‘Witchcraft Allegations, Refugee Protection and Human Rights: A Review of the Evidence’, UNHCR: New Issues in Refugee Research. Research Paper no. 169. Schouten, P. (2009) ‘Theory Talk no. 34: James Ferguson on Modernity, Development and Reading Foucault in Lesotho’ Theory Talks, online at: http://www.theory- talks.org/2009/11/theory-talk-34.html (accessed 2 November 2010). Scott, J.C. (2012) Two Cheers for Anarchism, Princeton: Princeton University Press. ____ (2009) The Art of Not Being Governed: An Anarchist History of Upland South-east Asia, New Haven and London: Yale University Press. 184 References

____ (1998) Seeing Like a State, New Haven and London: Yale University Press. ____ (1985) Weapons of the Weak: Everyday Forms of Peasant Resistance, New Haven and London: Yale University Press. Shacknove, A (1993) ‘From Asylum to Containment’, International Journal of Refugee Law, 5(4): 516–33. Shan Women’s Action Network (SWAN) (2011) Burma Army Gives Rapists Free Rein in Northern Shan State Offensive, SWAN Report, online at: http://www. burmapartnership.org/2011/07/burma-army-gives-rapists-free-rein-in-northern- shan-state-offensive (accessed 5 September 2013). ____ (2002) License to Rape, SWAN Report, Chiang Mai: Shan Women’s Action Network. Shearing, C. (2001) ‘Punishment and the changing face of governance’, Punishment and Society, 3(2): 203–20. Shearing, C. and Wood, J. (2003) ‘Nodal Governance, Democracy and the New “Denizens”’, Journal of Law and Society, 30(3): 400–19. Sherman, L.W. and Strang, H. (2007) Restorative Justice: The Evidence, London: The Smith Institute. Shukla, K. and Olson, C. (2007) ‘Thailand: Burmese Resettlement Offering New Opportunities and Creating Complications’, Refugees International Bulletin, 21 May 2007. Sieder, R. (2001) ‘Rethinking Citizenship: Reforming the Law in Postwar Guatemala’, in T. B. Hansen and F. Stepputat (eds) States of Imagination: Ethnographic Explorations of the Postcolonial State, Durham and London: Duke University Press. Simon, J. (2007) Governing Through Crime: How the War on Crime Transformed American Democracy and Created a Culture of Fear, Oxford: Oxford University Press. ____ (2001) ‘The Vicissitudes of Law’s Violence’, in A. Sarat (ed.) Law, Violence and the Possibility of Justice, Princeton: Princeton University Press. ____ (2000) ‘The “Society of Captives” in the Era of Hyper-incarceration’, Theoretical Criminology, 4(3): 285–308. ____ (1998) ‘Refugees in a Carceral Age: The Rebirth of Immigration Prisons in the United States’, Public Culture, 10(3): 577–607. Slaughter, A. and Crisp, J. (2009) A Surrogate State? The Role of UNHCR in Protracted Refugee Situations, UNHCR Research Paper no. 168, Geneva: UNHCR. Sluka, J. (1989) Hearts and Minds, Water and Fish: Support for the IRA and INLA in a Northern Irish Ghetto, Greenwich: JAI Press. Small, M.L. (2004) Villa Victoria: The Transformation of Social Capital in a Boston Barrio, London and Chicago: University of Chicago Press. Smeaton, D.M. (1887) The Loyal Karens of Burma, electronic version/Project Downloaded by [University of Defence] at 21:28 09 May 2016 Gutenberg. Smith, M. (2005) ‘Warehousing Refugees: A Denial of Rights, a Waste of Humanity’, World Refugee Survey 2004: 38–56. Smith, M.J. (2007) State of Strife: The Dynamics of Ethnic Conﬂ ict in Burma, Washington: East-West Center Washington. ____ (2003) ‘The Enigma of Burma’s Tatmadaw: A “State Within a State”’, Critical Asian Studies, 35(4): 621–32. ____ (1999) Burma: Insurgency and the Politics of Ethnicity, Second Edition, London and New Jersey: Zed Books. References 185

____ (1991) Burma: Insurgency and the Politics of Ethnicity, London and New Jersey: Zed Books. Snajdr, E. (2005) ‘Gender, Power and the Performance of Justice: Muslim Women’s Responses to Domestic Violence in Kazakhstan’, American Ethnologist, 32(2): 294–311. Soguk, N. (1999) States and Strangers: Refugees and Displacements of Statecraft, Minneapolis and London: University of Minnesota Press. Sommers, M. (2006) Fearing Africa’s Young Men: The Case of Rwanda, World Bank Social Development Paper no. 32, Washington: World Bank. Sontasombat, Y. (2004) ‘Karen Cultural Capital and the Political Economy of Symbolic Power’, Asian Ethnicity 5(1): 105–20. Sorenson, B.R. and Vincent, M. (2001) ‘Conclusion’, in M. Vincent and B.R. Sorensen (eds) Caught Between Borders: Response Strategies of the Internally Displaced, London: Pluto Press. South, A. (2012a) ‘The Politics of Protection in Burma’, Critical Asian Studies, 44(2): 175–204. ____ (2012b) Prospects for Peace in Myanmar: Opportunities and Threats, Oslo: Peace Research Institute Oslo. ____ (2008) Ethnic Politics in Burma: States of Confl ict, Abingdon: Routledge. ____ (2007a) ‘Karen Nationalist Communities: The “Problem” of Diversity’, Contemporary South-East Asia, 29(1): 55–76. ____ (2007b) ‘Ceasefi res and Civil Society: The Case of the Mon’, in M. Gravers (ed.) Exploring Ethnic Diversity in Burma, Copenhagen: Nordic Institute of Asian Studies. ____ (2003) Mon Nationalism and Civil War in Burma: The Golden Sheldrake, New York and London: Routledge. South, A., Perhult, M. and Carstensen, N. (2010) Confl ict and Survival: Self-protection in South-east Burma, Chatham House Asia Programme Paper 2010/04, London: Chatham House. S’Phan Shaung (2013) ‘Donor Support Drops for Refugees – Food Rations to be Cut’, Karen News, 28 June 2013. Spector, M. and Kitsuse, J.I. (2001 [1977]) Constructing Social Problems, New Brunswick: Transaction Publishers. Speed, S. and Collier, J.F. (2000) ‘Limiting Indigenous Autonomy in Chiapas, Mexico: The State Government’s Use of Human Rights’, Human Rights Quarterly, 22(4): 877–904. Starr, J. and Collier, J.F. (1989) ‘Introduction: Dialogues in Legal Anthropology’, in J. Starr and J.F. Collier (eds) History and Power in the Study of Law: New Directions Downloaded by [University of Defence] at 21:28 09 May 2016 in Legal Anthropology, Ithaca: Cornell University Press. Stedman, S.J. and Tanner, F. (2005) Refugee Manipulation: War, Politics and the Abuse of Human Suffering, Washington: The Brookings Institution. Steinberg, D. (2010) Burma/Myanmar: What Everyone Needs to Know, Oxford: Oxford University Press. Steiner, N., Gibney, M. and Loescher, G. (2003) Problems of Protection: The UNHCR, Refugees and Human Rights, New York and London: Routledge. Stevens, J. (2006) ‘Prisons of the Stateless: The Derelictions of UNHCR’, New Left Review, 42: 53–67. 186 References

Suleiman, J. (1999) ‘The Current Political, Organizational and Security Situation in the Palestinian Refugee Camps of Lebanon’, Journal of Palestine Studies, 29(1): 66–80. Sunstein, C. (2010) On Rumours: How Falsehoods Spread, Why We Believe Them, What Can Be Done, London: Penguin. Suttles, G.D. (1968) The Social Order of the Slum, Chicago: University of Chicago Press. Sykes, G. (2007 [1958]) The Society of Captives: A Study of a Maximum-Security Prison, Princeton: Princeton University Press. Tamanaha, B. (2008) Understanding Legal Pluralism: Past to Present, Local to Global, Legal Studies Research Paper Series no.07-0080, New York: St John’s University, School of Law. ____ (2001) A General Jurisprudence of Law and Society, Oxford: Oxford University Press. Tangseefa, D. (2007) ‘“Temporary Shelter Areas” and the Paradox of Perceptibility: Imperceptible Naked-Karens in the Thai-Burma Border Zones’, in P.K. Rajaram and C. Grundy-Warr (eds) Borderscapes: Hidden Geographies and Politics at Territory’s Edge, London: University of Minnesota Press. ____ (2006) ‘Taking Flight in Condemned Grounds: Forcibly Displaced Karens and the Thai-Burmese In-Between Spaces’, Alternatives 31(4): 405–29. Taussig, M. (2003) Law in a Lawless Land, Chicago and London, University of Chicago Press. Taw, D. (2005) Choosing to Engage: Strategic Considerations for the Karen National Union. Conciliation Resources, online at: http://www.c-r.org/our-work/accord/engaging- groups/karen-national-union.php (accessed 2 April 2011). Taylor, R. (2009) The State in Myanmar, London: Hurst. ____ (2007) British Policy towards Myanmar and the Creation of the ‘Burma Problem’, in N. Ganesan and Kyaw Yin Hlaing (eds) Myanmar: State, Society and Ethnicity, Singapore and Japan: Institute of Southeast Asian Studies and Hiroshima Peace Institute. TBC (The Border Consortium formerly TBBC/Thai–Burma Border Consortium) (2013) Camp statistics, including population and ethnic composition, online at: http://theborderconsortium.org/ (accessed 3 September 2013). ____ (2012) Changing Realities: Poverty and Displacement in South East Burma/Myanmar, Bangkok: TBBC. ____ (2011) Burmese Border Displaced Persons: December 2011, online at: http:// theborderconsortium.org/camps/2011-12-dec-map-tbbc-unhcr.pdf (accessed 5 September 2013). Downloaded by [University of Defence] at 21:28 09 May 2016 ____ (2010a) Nine Thousand Nights. Refugees from Burma: A People’s Scrapbook, Bangkok: TBBC. ____ (2010b) 3 Sides to Every Story: A Proﬁ le of Muslim Communities in the Refugee Camps on the Thailand Burma Border, online at: http://www.tbbc.org (accessed 26 September 2010). ____ (2010c) Terms of Reference: Baseline Livelihood Vulnerability Analysis of Refugee Camps along the Thailand-Burma Border, online at: http://www.alnap.org/pool/ vacancies/tor-baseline-livelihood.pdf (accessed 5 September 2013). ____ (2008a) Observations of Karen Resettled in the USA, July 2008, online at: http:// www.tbbc.org (accessed 26 September 2010). References 187

____ (2008b) Responses to Observations of Karen Resettled in the USA November 2008, online at: http://www.tbbc.org (accessed 26 September 2010). ____ (2008c) Internal Displacement and International Law in Eastern Burma, online at: http://www.tbbc.org (accessed 26 March 2011). ____ (2004) Between Worlds: Twenty Years on the Border, Bangkok: TBBC. Terry, F. (2002) Condemned to Repeat? The Paradox of Humanitarian Action, Ithaca and London: Cornell University Press. Thawnghmung, A.M. (2011a) Beyond Armed Resistance: Ethnonational Politics in Burma (Myanmar), Singapore: East-West Centre. ____ (2011b) The Other Karen in Myanmar: Ethnic Minorities and the Struggle Without Arms, Lanham: Lexington Books. ____ (2008) The Karen Revolution in Burma: Diverse Voices, Uncertain Ends, Singapore: East-West Centre. Thompson, S. (2008) ‘Community-based camp management’, Forced Migration Review, 30: 26–8. Tonkin, E. (2000) ‘Autonomous Judges: African Ordeals as Dramas of Power’, Ethnos: Journal of Anthropology 65(3): 366–386. Turcot, P. (2012) Adaptation, Resilience and Transition: Report of the Formative Evaluation of Camp Management in the Burmese Refugee Camps in Thailand – Long Report, Report prepared for CIDA, USAID, AUSAID and Act for Peace by E.T. Jackson and Associates Ltd, online at: http://www.etjackson.com/images/2012-10-burma- refugee-camp-management-evaluation-long-report.pdf (accessed 5 September 2013). Turner, S. (2010) Politics of Innocence: Hutu Identity, Conﬂ ict and Camp Life, New York and Oxford: Berghahn Books. ____ (2005a) ‘New Opportunities: Angry Young Men in a Tanzanian Refugee Camp’, in P. Essed, G. Frerks and J. Schrijvers (eds) Refugees and the Transformation of Societies: Agency, Policies, Ethics and Politics, New York and Oxford: Berghahn Books. ____ (2005b) ‘Suspended Spaces: Contesting Sovereignties in a Refugee Camp’, in B. Hansen and F. Stepputat (eds) Sovereign Bodies: Citizens, Migrants and States in the Postcolonial World, Princeton and Oxford: Princeton University Press. ____ (2004) ‘Under the Gaze of the “Big Nations”: Refugees, Rumours and the International Community in Tanzania’, African Affairs 103: 227–47. Twining, W. (2000) Globalisation and Legal Theory, Cambridge: Cambridge University Press. Tyler, T.R. (1990) Why People Obey the Law, New Haven and London: Yale University Press. Downloaded by [University of Defence] at 21:28 09 May 2016 Ubink, J. and McInerney, T. (2011) Customary Justice: Perspectives on Legal Empowerment, Leiden: International Development Law Organization. UK Ofﬁ ce for National Statistics (2013) Statistical bulletin: Crime in England and Wales, Year Ending March 2013, online at: http://www.ons.gov.uk/ons/ dcp171778_318761.pdf (accessed 5 September 2013). United Nationalities Federal Council (UNFC) (2013) Statement of the Ethnic Nationalities Conference July 29–31 2013, UNFC, online at: http://www. burmapartnership.org/2013/08/statement-of-the-ethnic-nationalities-conference/ (accessed 5 September 2013). 188 References

United Nations General Assembly (2011) Situation of Human Rights in Myanmar: Report of the Special Rapporteur on the Situation of Human Rights in Myanmar, document no. A/66/365. ____ (2010) Progress Report of the Special Rapporteur on the Situation of Human Rights in Myanmar, document no. A/HRC/13/48. ____ (2009) Situation of Human Rights in Myanmar: Report of the Special Rapporteur on the Situation of Human Rights in Myanmar, document no. A/64/318. ____ (2004) Report of the Secretary General on the Rule of Law and Transitional Justice in Conﬂ ict and Post-conﬂ ict Societies, document no. S/2004/616. United Nations High Commissioner for Refugees (UNHCR) (2013) UNHCR Global Appeal 2013, online at: http://www.unhcr.org/50a9f82d20.html (accessed 5 September 2013). ____ (2012) UNHCR Global Trends Data 2012, online at: http://unhcr.org/ globaltrendsjune2013/UNHCR%20GLOBAL%20TRENDS%202012_V05. pdf> (accessed 5 September 2013). ____ (2011) UNHCR Global Appeal 2010–2011, online at: http://www.unhcr. org/4b0506219.html (accessed 5 September 2013). ____ (2009) Conclusion on Protracted Refugee Situations, UNHCR Executive Committee Conclusions no. 109, Geneva: UNHCR Executive Committee. ____ (2008a) Protracted Refugee Situations: Revisiting the Problem, Executive Committee of High Commissioner’s Programme, Standing Committee, 42nd Meeting Report, document no. EC/59/SC/CRP.13, Geneva: UNHCR. ____ (2008b) A Community-based Approach in UNHCR Operations, Geneva: UNHCR. ____ (2008c) Handbook for the Protection of Women and Girls, Geneva: UNHCR. ____ (2007) Conclusion on Children at Risk, 5 October 2007, Executive Committee Conclusion no. 107 (LVIII). ____ (2006a) The State of the World’s Refugees 2006, Clarendon: Oxford University Press. ____ (2006b) Operational Protection in Camps and Settlements: A Reference Guide of Good Practices in the Protection of Refugees and Other Persons of Concern, Geneva: UNHCR. ____ (2006c) Note on International Protection, 12 July 2006, online at: http://www. unhcr.org/44c9cb212.html (accessed 5 September 2013). ____ (2006d) Handbook for Self-Reliance, Geneva: UNHCR. ____ (2006e) Operational Guidelines on Maintaining the Civilian and Humanitarian Character of Asylum, Geneva: UNHCR. ____ (2004a) Protracted Refugee Situations, Executive Committee of High Commissioner’s Programme, Standing Committee, 30th Meeting Report, Downloaded by [University of Defence] at 21:28 09 May 2016 document no. EC/54/SC/CRP.14, Geneva: UNHCR. ____ (2004b) UNHCR Code of Conduct: Guidelines for Managers, online at: http:// www.un.org/en/pseataskforce/docs/code_of_conduct_guidelines_for_managers_ english.pdf (accessed 5 September 2013). ____ (2003a) Agenda for Protection, Geneva: UNHCR. ____ (2003b) Discussion Paper on the Strategic Use of Resettlement, UNHCR Working Group on Resettlement, online at: http://www.unhcr.org/protect/ PROTECTION/3ee6dc6f4.pdf (accessed 5 September 2013). ____ (2002) Conclusion on the Civilian and Humanitarian Character of Asylum, 8 October 2002, Executive Committee Conclusion no. 94 (LIII). References 189

____ (2001) Prevention and Response to Sexual and Gender-Based Violence in Refugee Situations, Geneva: UNHCR. ____ (1995) Sexual Violence Against Refugees: Guidelines on Prevention and Response, Geneva: UNHCR. ____ (1994) Refugee Children: Guidelines on Protection and Care, Geneva: UNHCR. ____ (1991) Guidelines on the Protection of Refugee Women, Geneva: UNHCR. ____ (1987) Note on Refugee Children, Geneva: UNHCR. ____ (1950) Statute of the Offi ce of the United Nations High Commissioner for Human Rights, Geneva: UNHCR. UNHCR Field Offi ce Mae Hong Son, copy on fi le with author. ____ (2006d) Administration of Justice in Tham Hin Camp, copy on fi le with author. ____ (2006e) Administration of Justice in Ban Don Yang Camp, copy on fi le with author. ____ (2006g) SGBV Statistics for the Period January to June 2006, copy on fi le with author. UNHCR & International Labour Organization (ILO) (2007) Strengthening Protection Capacity Project: Livelihoods Component. Phase 1: Mae Hong Son Province, copy on fi le with author. UNHCR: Thailand (2011) UNHCR Bangkok Newsletter. ____ (2008) UNHCR Strengthening Protection Capacity Project (SPCP) Update–Thailand 2008, online at: http://www.unhcr.org/4884ae662.html-Thailand Update (accessed 5 September 2013). ____ (2007) Thailand: Strengthening Protection Capacity Programme Matrix 31st December 2007, online at: http://www.unhcr.org/473082062.pdf (accessed 5 September 2013). ____ (2006a) Protection Program for Thailand in 2006, copy on fi le with author. ____ (2006b) Analysis of Gaps in Refugee Protection Capacity, Thailand, online at: http://www.unhcr.org/refworld/docid/472897020.html (accessed 31 January 2011). ____ (2006c) Brief Evaluation Report: SGBV Committees in MRM and MLO Refugee Camps, United Nations Human Rights Council (2007) Report of the Special Rapporteur on the Situation of Human Rights in Myanmar, by Paulo Sérgio Pinheiro, document no. A/HRC/4/14, online at: http://www.ibiblio.org/obl/docs4/ HRC2007-SRM-rep.pdf (accessed 5 September 2013). United Nations Human Rights Council. (2007) Report of the Special Rapporteur on the Situation of Human Rights in Myanmar, by Paul Sérgio Pinheiro, document no. A/ HRC/4/14, online at: http://www.ibiblio.org/obl/docs4/HRC2007-SRM-rep.pdf (last accessed 5 September 2013). Downloaded by [University of Defence] at 21:28 09 May 2016 United Nations Offi ce for the Coordination of Humanitarian Affairs (UN OCHA) (2013) ‘Kachin: Unimpeded Access to All People in Need is Essential’, OCHA Humanitarian Bulletin Myanmar, 6(1–31 July 2013): 4. Verdirame, G. (2011) The UN and Human Rights: Who Guards the Guardians? Cambridge, Cambridge University Press. Verdirame, G. and Harrell-Bond, B.E. (2005) Rights in Exile: Janus-Faced Humanitarianism, New York and Oxford: Berghahn Books. Veroff, J. (2010) Crimes, Confl icts and Courts: The Administration of Justice in a Zambian Refugee Settlement, New Issues in Refugee Research Paper no. 192, Geneva: UNHCR. 190 References

Voutira, E. and Harrell-Bond, B.E. (1995) ‘In Search of the Locus of Trust: The Social World of the Refugee Camp’, in E.V. Daniel and J.C. Knudsen (eds) Mistrusting Refugees, Berkeley: University of California Press. Vungsiriphisal, P., Bennett, G., Poomkacha, C., Jitpong, W. and Reungsamran, K. (2011) Sustainable Solutions to the Displaced Person Situation on the Thai-Myanmar Border: Analysis of Royal Thai Government Policy, Asian Research Center for Migration, Chulalongkorn University: Institute of Asian Studies. Wacquant, L. (1997) ‘Three Pernicious Premises in the Study of the American Ghetto’, International Journal of Urban and Regional Research, 21(2): 341–53. Wald, P. (1974) ‘Making Sense out of the Rights of Youth’, Human Rights, 4: 13–17. Wilde, R. (1998) ‘Quis Custodiet Ipsos Custodes?: Why and How UNHCR Governance of “Development” Refugee Camps Should be Subject to International Human Rights Law’, Yale Human Rights and Development Law Journal, 1: 107–28. Wilkinson, R. and Pickett, K. (2009) The Spirit Level: Why More Equal Societies Almost Always Do Better, London: Allen Lane. Wilson, A. (2010) ‘Democratising Elections without Parties: Reﬂ ections on the Sahrawi Arab Democratic Republic’, The Journal of North African Studies, 15(4): 423–38. Wilson, R.A. (2000) ‘Reconciliation and Revenge in Post-Apartheid South Africa: Rethinking Legal Pluralism and Human Rights’, Current Anthropology, 41(1): 75–98. ____ (ed.) (1997) Human Rights: Culture and Context: Anthropological Perspectives, London: Pluto Press. Wintin, T. and Brown, I. (2005) ‘Colonial Burma’s Prison: Continuity with its Pre- colonial Past?’ International Institute for Asian Studies Newsletter, 39: 5. Wojkowska, E. (2006) Doing Justice: How Informal Justice Systems Can Contribute, Oslo: United Nations Development Programme. Women Exchange (2003) Automatic Response Mechanism (ARM): What to do in case of Sexual Violence for Migrant and Refugee Women, copy on ﬁ le with author. Women’s League of Burma (2004) System of Impunity: Nationwide Patterns of Sexual Violence by Military Regime’s Army and Authorities in Burma, Chiang Mai: Women’s League of Burma. Women’s League of Chinland (2007) Unsafe State: State-sanctioned Sexual Violence against Chin Women in Burma, Women’s League of Chinland. Yawd Serk (2010) ‘The Weak Points of Burma’s Ethnic Resistance Groups’, The Nation, 21 July 2010. Zetter, R. (2007) ‘More Labels, Fewer Refugees: Remaking the Refugee Label in an Era of Globalization’, Journal of Refugee Studies, 20(2): 172–92. Downloaded by [University of Defence] at 21:28 09 May 2016 ____ (1991) ‘Labelling Refugees: Forming and Transforming a Bureaucratic Identity’, Journal of Refugee Studies, 4(1): 39–62. Zion, J.W. (1999) ‘Monster Slayer and Born for Water: The Intersection of Restorative and Indigenous Justice’, Contemporary Justice Review, 2(4): 359–82. Index

abortion 66, 77n11 Ban Mae Surin (Site 2) 5, 19n2, 87, 101n5, ABSDF see All-Burma Students Democratic 118, 119, 123; ethnic composition 20n5 Front Ban Mai Nai Soi (Site 1) 5, 19n2, 101n5, administration 3, 7, 11, 12–15, 39, 96, 118; 101n13, 123, 119; legal assistance centre see also governance in 118 administration of justice 1, 12, 17, 18, 59, banishment 73, 74, 110, 157 71, 95, 96, 117, 159, Baptists 24, 25, 36n7, 49, 70 adultery 18, 68, 70, 72, 73–6, 78n23, Bo Mya 28–9, 71 79n28, 79n30, 79n31, 103, 106, 120, Border Consortium, The (TBC) 6, 12, 15, 121, 130n18, 153n10; and marriage 19n1, 32, 34, 35, 38n18, 81, 87, 91, 93, 78n23, 78n25; 137–9; see also 94, 97, 101n1, 154n26, 157, 158, 161, extramarital sex; marriage 163 agency (of refugees) 8–15, 82, 91, 162 bribery 32, 77n4, 84 agencies (international humanitarian) 1, 2, Buddhism 23, 25, 26, 28, 30, 36n7 6, 7, 8, 9, 10, 11, 12, 18, 45, 54, , 76, Burma: as source of refugees 5; British 80–2, 88–91, 94–8, 110, 114, 117, colonial occupation 22; civilian 124–5, 126, 136, 142, 146, 155, 158, government 163; community management 159; aid 2, 8, 22, 33, 38, 41, 90, 91, 97, 128; conditions inside 7; corruption 85; 157, 159 criminal justice 61; demilitarization 22; alcohol 75, 79n30, 103, 106, 112, 145, 146, diarchic rule 27; economic development in 147, 154n26; see also drugs 22; education in 30; ethnic conﬂ icts in 5; All-Burma Students Democratic Front ethnic politics 20n4, 22–31; government (ABSDF) 43, 57n24 163; health care in 30; human rights 5, alternative dispute resolution 105, 121, 35; independence 5, 21; liberalization 2, 130n21; see also restorative justice 21, 22; maintenance of order 58–79; animism 26, 28, 36n7, 65–9, 71, 74, oppression 2; politics 3, 6, 19, 22; prison 77n15, 106 sentences in 75; pro-democracy movement

Downloaded by [University of Defence] at 21:28 09 May 2016 Approval Committee 119, 130n21 5; repatriation to 7–8, 19, 162; slow arrival arbitration 121, 122, 147 of refugees 155–6; social practices 65; armed groups 82, 161; see also KNLA, see also repatriation; Union of Burma militarization and ‘non-state actors’ Burma Army 5, 32, 34, 38n17, 42, 48, 59, arrests 7, 33, 101n5 60, 77n1, 82, 83, 134, 137, 163 assault 3, 5, 75, 103, 107, 108, 115, 121, Burma Lawyers’ Council 71, 114 131n32 Burman (ethnic group) 22, 23, 43, 47, Aung San Suu Kyi 22 57n24, 70 authority 3, 9, 10, 17, 80, 155 Burmanization 43, 49 Burmese (language) 44, 49, 52, 84 Ban Don Yang 5, 149; ethnic composition Burmese Border Consortium see Border 16, 20n10 Consortium, The 192 Index

camp commander (Palat) 85, 113; 137, 150, 160; rights 18, 134, 144, 147, see also Palat 160; vulnerability 145; see also young camp committees 7, 81, 85, 92, 93, 101n13, people 112, 113, 130n21, 140, 155, 161 child soldiers 110, 146 camp justice 75, 106–11, 126, 130n19; Christianity 24–7, 30, 36n7, 49, 59, 60, see also justice 68–71, 75, 99 camp leaders 19, 63, 72, 76, 93, 110, 111, churches 2, 7, 47, 48, 49, 79n28, 97 113, 122, 133, 146, 147, 161 civil litigation 65, 120 camp management programme 93 claims making 149–52 camp rules 13, 72, 79, 98, 106, 108, 110, clothing 17, 46, 48, 50–2; Karen shirt 50–1 119, 120, 129, 137, 146, 148, 154 codes of conduct 133, 136 camps: and cross-border relationships 33; and colonialism 105, 128; see also colonization social change 9; as anomic spaces 9; as colonial legal systems 14, 77n4 , 116 communities 17, 39–57; as complex colonization 25–8, 58, 60; see also cultural/political worlds 3; as functional colonialism societies 155; as places of quarantine 87; as Committee for the Coordination of Services short-term policy 1; as sites of anarchy 2, to Displaced Persons in Thailand 3; as tabula rasa 159; by-laws 79n31; (CCSDPT) 81, 100n1, 159, 164 character 5, 20n6; civilian nature 98; communities: and resettlement 53–5; camp closure 163; community structures 3, 8; 17, 39–57; defi nition 41; displaced 44–5; crime rates 19n2; culture 4, 8; design 87, dynamic 52–3; political 46–53; 88; differences from prisons 4; effects of life situational 41–2 in 1; effects of resettlement on 7; elections community-based justice 67, 106, 116, 128; 102n16; ethnic composition 5, 16, 20n5, see also customary justice 20n10, 43, 149, 156; ethnic diversity in community-based organizations 3, 18–19, 55n1; ideal type 11; in Gaza 19n1; in 141, 161 Jordan 103; in Kenya 3, 40, 87–9, 94, community peace teams 140, 141; see also 102n15, 108, 136, 158; in Lebanon 1, SGBV Committees 19n1, 40, 89, 91; in Somalia 40; in community service orders (CSOs) 121, 147 Tanzania 47, 57n26, 89, 90, 109, 136, compensation 74, 76, 118 158; international agencies’ view of 1; in confl ict 9, 66, 109; avoidance 63 the West Bank 1, 19n1, 40, 60; in Uganda Convention on the Elimination of 3, 158; in Zambia 107; Karen 140, 141, Discrimination Against Women 156; Karenni 119, 123, 129n11; lack of (CEDAW) 134 confi dentiality in 142; living conditions Convention on the Rights of the Child 148; management 7, 11, 15, 18, 44, 53, (CRC) 133, 134, 144, 151 61, 81, 82, 88–97, 126, 130n19, 140, corruption 40, 85, 99, 117, 126 157, 159; militarization 94, 97, 98, 99; courts: access to 12; formal 13, 115; gacaca normative universe 155; proximity to Thai 115, 127; KNU 78n18; Thai 113, 117, towns 148; remoteness 104; role of KNU 121, 123–6, 140, 141, 153n12, 153n23, in 158; rules 13, 72, 79, 98, 106, 108, 159, 161 110, 119, 120, 129, 137, 146, 148, 154; crime 103, 111, 120, 123, 124, 131n33; Downloaded by [University of Defence] at 21:28 09 May 2016 spatial ordering and ownership 87–8 organized 147; rates in camps 19n2; CCSDPT see Committee for the Coordination resolution by refugees 13; violent 12, 58, of Services to Displaced Persons in 103; see also delinquency Thailand CSOs see community service orders ceasefi re 37n16, 163 cultural relativism 132–3 CEDAW see Convention on the Elimination culture 3, 4, 8, 17, 76, 149; and human of Discrimination Against Women rights 132–4; legal 64; of prisons 3 ceremonies 46, 61, 63, 64, 65, 67, 70, 75 curfew 79n30 child protection 107, 137, 150, 160 customary justice 13, 105, 128; see also children 100, 134; abuse 131n32, 136, 137, informal justice; non-state justice; 146; offences against 121; protection 107, traditional justice Index 193

Dadaab camps 3, 94, 102n15, 108 ethnic minorities 127 death penalty 68, 69, 78n25; see also ethnic plurality 15, 115 execution executions 61, 72, 74, 98; see also death decision-making 108, 109, 112, 122 penalty Deeds of Commitment 100 expulsion 62 delinquency 117, 144–9, 160; see also crime; extortion 5, 31, 32, 37n16, 98 youth offending extramarital sex 59, 66, 70, 137; see also Democratic Karen Buddhist Army (DKBA, adultery also known as Democratic Karen Benevolent Army) 30, 32, 34, 38n17, fi ghting 66, 79n30, 107, 110, 145, 147 38n20, 42, 68, 83 fi nes 61, 76, 106, 121 Department of Homeland Security (US) forced labour 5, 32, 37n16, 98 57n24 forced marriage 136, 137, 154n33 detention 75, 76, 106, 110, 146, 147, 150; forced military recruitment 31 arbitrary 21, 32; centres 61; immigration forced relocation 32 19n3; see also jail, prison foreign infl uences 146, 148–9 de-territorialized sovereignty 10–12, 82 forestry offences 120, 121, 123, 131n32 deviance 147–9 forum shopping 14, 129n9, 149–52 discourses: humanitarian 8; human rights fraud 89, 120 116; of danger 82–8; of human rights freedom of movement 1–4, 19n3, 33, 90, 160; of protection 88 157, 161–2 discrimination 28, 44, 93, 123, 151, 152, 161 gacaca courts 115, 127 displacement 5, 17, 31–5, 44–5; 163; and gambling 66, 67, 103, 106, 120 violence 32 gangs 145, 148 dispute resolution 13, 14, 58–60, 62, 65, GBV see gender-based violence and sexual 66, 79n30, 96, 103–9, 112–14, 116, 121, and gender based violence 124, 126–7, 129, 159 gender: and status 134; equality 19, divorce 12, 75, 76, 78n27, 79n27, 101n9, 46, 114, 160; expectations 9; 121, 131n32, 140, 153n10 programming 95; representation DKBA see Democratic Karen Buddhist 115; rights 135, 144 Army gender-based violence (GBV) 137, 152n2 domestic abuse 141, 143, 153n10; see also gossip 53, 63, 99 domestic violence governance 80–102; and authority 155; and domestic violence 75, 103, 107, 111, 117, corruption 85; and ideology 18; and 121, 123, 125, 131n32, 135–9; diffi culty maintenance of order 58–79; and the in policing 139; see also domestic abuse; Royal Thai Government 18, 30; and the spousal assault UNHCR 11, 18, 80; beyond the state donors 7, 11, 97, 98, 163 129; camp 107, 112; community 91, 92, drugs 79n30, 103, 106, 120, 121, 124, 125, 128, 161; devolved 17; ecosystem 131n32, 145, 147, 153n23; see also 94; new forms of 10; non-state 11; alcohol Palestinian 91; patrimonial 59; political Downloaded by [University of Defence] at 21:28 09 May 2016 due process 114–17 1; pre-camp 61; role of international agencies 11; role of KNU 99, 100; role of early marriage 136, 146; see also marriage non-state actors 97–100; role of refugee education 9, 30, 32, 39, 45, 83, 93, 98, organizations 11; role of refugees 4; role 100n1, 106, 120, 133, 140, 145 of Thai authorities 11; role of UNHCR elders 13, 63 11; shared 44–6; structures 45; village 24, elections 93, 102n16 60–1, 63, 156 English (language) 52, 77n6, 145, 149; governance palimpsest 17, 58–79, 152, 157, classes 4 159; see also self-governance ethnic diversity 55n1, 161 Grandfather Longlegs 36n11 ethnicity 42, 44, 50, 52, 59, 135 group conferencing 76, 121, 147 194 Index

halal rations 93 International Rescue Committee (IRC) 18, harm 73–6, 123, 131n32, 137–9, 159 76, 100n1–101n1, 101n5, 106, 123, 125, harmony ideology 109 129n4, 129n5, 129n10 hate speech 44 health 32, 65, 83 jails 109; see also prison health care 30, 100n1; access to 32 jokes 33, 47, 63 homosexuality 149 judges 63, 71 human rights 89, 114–17, 127, 130n21; justice: access to 13, 14, 18, 106, 125; advocacy by KWO 143; and delinquency administration of 1, 12–15, 71, 95, 96, 148; and international organizations 18; 106, 112, 117, 118; and control over and local justice 132–54; and NGOs 128; refugees 96; and violence 110; and women and social transformation 133; and 117; camp 75, 106–11, 126, 130n19, women’s rights 135–44; guarantees 159; community 18, 116, 119; criminal 37n16; impact 149–52; imposition 19, 61, 117; customary 13, 105, 128; 132–3; in Brazil 134; in Burma 5, 35; devolution of 104; indigenous 14, 116; language 150; law 14–15, 105, 118, 134, informal 104, 115; local 18; non-state 142; organizations 32; reception of 149; 104, 105, 106, 115, 126; ownership of standards 129; universality 132–3; 18; parallel systems 125; reform 119; vernacularization 141; violations 21, 32, restorative 109, 122, 128; social 15; state 35, 36n10, 55n7, 163 104; struggle for ownership 103–31; Thai human trafﬁ cking 120, 121, 136, 137, system 117, 126, 131n33, 138, 143, 157; 153n23 traditional 64, 125 justice committees 61, 79, 107, 134, 150 identities: assertion of 9; Burman 47; community 46; construction 40; ethnic Kachin 22, 44; Independence Army 23; 50; Karen 23–5, 46–53; national 83; Independence Organization 37n14; State political 50; pre-camp 40; religious 50 30, 163 IDPs see internally displaced people Kakuma Camp 54, 89, 94, 108 illegal presence 123, 131n32 Karen: as an ethno-national community income generation 7, 32, 34, 83 42–4; camps 140, 141, 156; communities indigenous justice 14, 116 92; constitutional position in Burma 23;; informal justice 115; see also community- experiences of 21–38; identity 23–5, based justice, non-state justice 46–53, 157, insurgency 28–31, 34, informal legal processes 105 37n13, 54, 97; KNU as representative informers 99 of 31; nationalism 15, 24, 27–8, 42, 60, inheritance 12, 78n26 69, 99; persecution of 47; relationship institutional pluralism 128, 132; see also with Burmans 28, 43; relationship with legal pluralism state 25 insurgency 28–31, 34, 54, 37n13, 97 Karen Baptist Committee 26 inter-ethnic marriage 151 Karen Baptist Convention 29 intergenerational relations 144–9 Karen Elders Advisory Board 106 internally displaced people (IDPs) 32 Karen National Association (KNA) 27 Downloaded by [University of Defence] at 21:28 09 May 2016 international agencies 46, 88–97, 124, 126, Karen National Defence Organization 27, 142, 162; and governance 11; and justice 37n15, 48 159; engagement with 1; policies 3, 155; Karen National Liberation Army (KNLA) view of camps 1; view of refugees 159; 23, 27, 29, 32, 43, 48, 54, 57n24, 72, see also international organizations 143, 146 international organizations 98, 117, 133, Karen National Union (KNU): and dispute 135, 143, 146, 159, 162; and human resolution 110, 146; and Kawthoolei law rights 5,18; denial of women’s agency 9; 71–4; and militarization of camps 94, 97, foreign staff 149; policies 10; political 98; and the Karen Women Organization involvement 30; role in governance 11; 45, 142, 143; and the Royal Thai see also international agencies Government 34; armed resistance 23, 27; Index 195

as a non-state actor 97–100; as language: and community 49–50; and representative of the Karen 31; human rights 152; and invention of Christianity in 30; courts 78n18; decline tradition 46; and sexual violence 152n4; 30; Education Department 98; executive as an ethnic marker 52; as problem in committee 71; fi nancial support for 29; resettlement 35; barriers 123, 161; of formation 27; jurisdiction 72; leadership rights 150; see also Burmese (language); 30, 97; links with refugee organizations English (language) 46; negotiations with Burmese law: and animist belief systems 65; and government 31, 163; pseudo-state refugees’ agency 10; codifi ed 157; structures 157; removal from US ‘terrorist’ defi nition 59; fragmented nature of 17; list 54, 57n24; role in camp governance human rights 14–15, 105, 118, 134; in 18, 99, 100, 157, 158; state-building refugee camps 12–15; international 90, project 18, 59; support for 38n17 159; Kawthoolei 71–4, 78n17; KNU 72, ‘Karenness’ 46, 53, 63, 71, 93, 144 73; reform 14, 119, 120, 128, 130n21; Karen New Year 27, 48, 50, 52 spirit 65–70, 106, 157; Thai 79n31, 105, Karenni 5, 22, 44, 139; camps 119, 123, 118, 119–22, 126; traditional 73; 129n11; and Panglong Agreement 23; UNHCR 95 women 51; National Progressive Party leaders: and camp management capacity 53; 37n14, 97; Refugee Committee; Women and power sharing 80; camp 19, 63, 72, Organization 95, 12, 20n5, 119 76, 93, 110, 111, 113, 122, 133, 146, Karenni State 5, 30 147, 161; community 18, 161; cultural Karen Peace Force 32 13; household 81; KNU 97; refugee 80, Karen People’s Liberation Army 27 82, 84, 88, 96, 103, 122, 139, 146; Karen Refugee Committee (KRC) 5, 12, 72, religious 13, 45, 130n19; section 93, 113, 81, 86, 87, 92, 106, 107, 158, 163, 164 122, 124, 130n19; spiritual 63, 66; Karen Resistance Day 48 village 9, 32, 60–1, 72, 73, 85; women 9, Karen State 5, 9, 24, 29, 32, 36n4, 46, 49, 60–1; zone 130n19 58, 59, 71, 72, 77n1, 78n25, 100, 110, leadership 4, 9, 26, 29, 30, 32, 37, 45, 60, 128 93, 111, 112, 161, 163 Karen Women Organization (KWO) 9, 29, Lebanon 1, 19n1, 40, 89, 91, 97 45, 46, 61, 56n8, 86, 107, 140–3, legal assistance centres (LACs) 18, 106, 153n14 117–19, 123–7, 129n5, 129n10, 130n19 Karen Women Organization Day 48 legal centralism 14 Karen Youth Organization (KYO) 29, 45, legal cultures 58, 64, 103 46, 61, 86, 107, 144 legal empowerment 14, 105, 117, 118, Karen Youth Organization Day 48, 51 119, 126 Kawthoolei 29, 56n15 legal formalism 109, 127 Kawthoolei Day 48 legal pluralism 8–15, 105, 152, 160 Kawthoolei law 71–4, 60, 78n17 legitimacy 9, 111–17, 143, 159; of refugee- Kayah 36n1; see Karenni led committees 13; sociological 113–14 Kayin 36n1; see Karen Lesbian Gay Bisexual Transgender Intersex Kenya 3, 40, 87–9, 94, 102n15, 108, 136, (LGBTI) community group 154n27 Downloaded by [University of Defence] at 21:28 09 May 2016 158; see also Dadaab camp; Kakuma camp liberalism 15, 159 Khin Nyunt 35n1 logging 30, 33, 36n10 KNA see Karen National Association Lukole Camp 57n26, 90, 99 KNLA see Karen National Liberation Army lynching 77n1 knowledge brokers 19, 142, 160 KNU see Karen National Union Mae Hong Son 6, 16 KNU law see Kawthoolei Law Mae La 19n2, 86, 119, 123, 131n32, 136, KRC see Karen Refugee Committee 145, 154n27; arrests in 101n5; KRC rules see camp rules cooperation between ethnic groups in 44; KWO see Karen Women Organization dispute resolution in 107; ethnic KYO see Karen Youth Organization composition 16, 20n10, 43; legal 196 Index

assistance centre in 118; resettlement Ministry of Interior (Thailand) 11, 38, 80, from 56n22; surveys on crime 113; wrist- 123, 126 tying ceremony 77n15 missionaries 24–7, 58, 62, 70, 71, 73, 159 Mae La Oon 5, 6, 16, 42, 64, 69, 87, 113, Mon (ethnic group) 20n10, 22, 23, 24 129n10, 137, 139, 140, 149, 154n24; moral guidance 108; see also moral narratives ban on alcohol in 146; camp justice in 75; morality 63, 70, 76, 93, 104, 106, 108, 116 celebration of national days in 48; moral narratives 62–5, 157 Christianity in 70; dispute resolution in moral panic 147–9 79n30, 107; effects of resettlement 53; mosques 2 elections in 102n16; ethnic composition Mozambique 16, 58, 104 16, 20n10, 43; fear of UNHCR in 95; murder 12, 72, 74, 77, 106, 120, 121, 123, justice committee 134; moral instruction 129, 131n32, 153n23 in 63; multiple communities in 41; Muslims 9, 26, 43, 55n7, 93, 141, 163, 160 mythico-history production in 47; Muslim Women Association 9 planning dispute 101n14–102n14; Myanmaﬁ cation 43; see also Burmanization relationships with host community 85; Myitsone Dam 22 religious groups in 36n7; use of Burmese language in 52; youth delinquency in Nai Amphur 80 145, 147 narratives 46–9, 60; moral 62–5 Mae Ra Ma Luang 5, 6, 16, 96, 129n10, national anthem 50 139, 140–1; baptisms 26; character of National Convention 35n1 camp 20n6; dispute resolution in 107; national days 48 ethnic composition 16, 20n10; nationalism 15, 24, 27–8, 42, 68, 92, 99; relationships with host community 85; constructions of 46; Karen 15, 24, 27–8, use of Burmese language in 52 42, 60, 69, 99 Mae Sariang 16, 143 nationality 135 Mae Sot 16, 143, 145 National League for Democracy 22 Manerplaw 30, 43, 148 National Security Council (Thailand) 80 manslaughter 123 nation-states 10, 59, 80, 82; see also state, marijuana 154n23 the marriage 121, 150, 151; and adultery 73–6, natural justice 116 78n23, 78n25; early 136, 146; forced NDUF see National Democratic United 136, 137, 154n33; in Karen animist Front villages 67; monogamous 70; registration negotiation 32, 84, 85, 86 12; strategic 89; see also adultery; early networks: between refugees and other actors marriage; remarriage 4; church 97; community 44, 91, 160; material culture 40 family 44; internal 9; of support 156; material objects 50 social 2, 8, 44 material traditions 46 Ne Win 21, 37n15 media 37n16, 46 non-discrimination 120, 160 mediation 76, 101n9, 108, 121, 122 non-governmental organizations (NGOs) 11, Mediation and Dispute Resolution 12, 15, 42, 46, 50, 81, 89, 94–7, 100n1, Downloaded by [University of Defence] at 21:28 09 May 2016 Guidelines 76, 120–3, 129n4, 131n30, 117–26, 128, 130n19, 140, 143, 146, 159 137, 140, 147, 153n12, 153n23 non-state actors 10, 59, 97–100, 161 medical facilities 4, 85; see also clinics non-state justice 115, 126 medics 7, 45, 53, 100n1 norms 17, 62, 152 memorialization 17, 40, 48 Nu Po 5, 19n2; ethnic composition 20n10; mental illnesses 96 legal assistance centre in 118 methodology 15–17 militarization 94, 97, 98, 99, 157 Or Sor (Thai camp rangers) 41, 81, 84, 85, military uniform 79n30, 99 86, 101n13, 113, 129n11, 136 Ministry of Defence (Thailand) 80 order 1–4, 8, 12–15, 17–18, 58–79, 114, Ministry of Foreign Affairs (Thailand) 80 125, 157, 159 Index 197

‘Othering’ 147–9 property rights 78n26 overnight guests 106 prostitution 120, 137 Protestants 46–7 PABs see provincial admission boards protests 8, 21, 43, 84 Palat (Thai camp commander) 80, 84, proverbs 77n6 101n13; see also camp commander provincial admission boards (PABs) 34, Palestinian Authority 60 38n18 Palestinian Liberation Organization 97 provincial governor (Nai Amphur) 80 Palestinian refugees 1, 19n1, 40, 91 PSAE see prevention of sexual abuse and Panglong Agreement 22–3 exploitation parallel justice systems 125 public education campaigns 6 participation 89, 90, 106, 157, 161 public meetings 49 patrimonial governance 59 punishment society 122, 128 peace negotiations 23, 31–2, 37n16, 163 Pwo Karen 23 peasant resistance 85 penal sociology 4, 128 Rakhine see Arakanese pernicious premises 2, 9 Rangoon 43, 52 physical harm 123, 131n32 Rangoon Central Jail 77n4 poems 63 rape 32, 61, 72, 77n1, 108, 115, 121, 123, police 134, 149; Thai 8, 97, 113, 117, 124, 131n32, 135, 136, 137, 140, 153n23; 154n23; see also policing attempted 123, 131n32, 136; deﬁ nition policies: aid 8; anti-immigration 82, 83; 137–9; see also sexual assault; sexual national; of containment 88; of violence international agencies 3, 10, 155; of rations 6, 20n7, 84, 91, 93, 163 protection 88; Thai government 82–4, refoulement 84, 162 155; Refugee Convention (1951) 5, 12, 34 policing 59, 67, 70, 103, 104, 128, 129, refugees: agency 8–15, 82, 91, 162; and 130n18, 139; internal 14; of adultery national security 83; and sexual violence 73–6; see also police 135–7; as source of pollution 82; assertion Polisario 97 of identities 9; as threatening 81, 82, 83, political communities 46–53 159; as victims 81, 88–97, 159; political imprisonment 19n3 Bhutanese 40; Burman 43, 57n24; political organization 2, 3, 90, 91 classiﬁ cation 6; communication with political power 3, 11 international agencies 95, 162; political prisoners 22, 37n16, 92, 163 engagement with bureaucracies 9; future polygamy 70, 130n18 of 19; Hutu 47; in Calcutta 12; pornography 106 Indochinese 33, 38n19; in Mozambique positivism 132 16; in South Africa 44; in Sudan 8; postcolonialism 127 integration 35; Karenni 139; leaders 80, post-regulatory state 17, 18 82, 84, 88, 96, 103, 122, 139, 146; power 1, 3, 8, 11, 30, 59, 80, 82, 117, 119, marginalization 8; Muslim 43, 55n7, 93; 145, 149, 160, 162 Palestinian 1, 19n1, 40, 91; Downloaded by [University of Defence] at 21:28 09 May 2016 prayers 48, 75 pathologization 8; preferences 155; premarital sex 71, 73, 103, 120, 130n18, registration 34, 38n18, 89, 94; 137, 149, 151 relationship with host community 85; prevention of sexual abuse and exploitation relations with Thai authorities 86; (PSAE) 152n2 resolution of crimes by 13; Saharawi 40, prisons: culture 3; differences from camps 4; 97; self-management 158; voices 8; in Brazil 134; in Burma 61; overcrowding vulnerability 1, 2, 3, 6, 7, 160; 126; research on 3; Thai 127 warehousing of 1, 4 promissory notes 76, 121, 147 regulations 83, 87, 101n4; and refugees’ property offences 19n2 agency 10; see also rules property restitution 163 rehabilitation 110, 111, 122 198 Index

reintegration 111, 163 section committees 7, 81 reintegrative shaming 111 section leader 93, 113, 111, 122, 124, religion 15, 40, 59, 63, 64, 65 130n19 religious identities 50 security 15, 18, 89, 103, 118; of camps 2; religious leaders 13, 45, 130n19 staff 107, 112, 113, 129n5, 130n19, religious teaching 76 136, 147 remarriage 78n27; see also marriage segregation 3 repatriation 7–8, 19, 35, 38n22, 86, 94, self-governance 92, 160; see also governance 109, 163, 164 self-reliance 19, 89, 91–4, 162; limits ‘repugnancy clause’ 116 160–2 resettlement 20n8, 35, 38n21, 39, 50, 53–5, self-restraint 4 56n22, 87, 88, 89, 95, 96, 107, 117, 140, self-suffi ciency 35, 45 148, 151, 154n33, 164; as driver of semi-autonomous social fi elds 14 change 7 service delivery 156 restorative justice 109, 122, 128 Seventh Day Adventists 36n7, 49, 71 right to work 162 sexual and gender-based violence (SGBV) rituals 61, 65, 67, 68, 74, 77n15, 78n27, 92 50, 136, 137, 141–4, 146 Roadmap to Democracy 21, 35n1 Sexual and Gender-Based Violence Rohingya 83, 163 Committee 107 Royal Thai Government: and border controls sexual assault 106, 153n23; see also rape; 157; and governance 18, 30, 80, 81; and sexual violence outsourced sovereignty 157, 158; and the sexual harassment 137 KNU 34; and the UNHCR 34, 38n19, sexual harm 137–9 157, 163; approach to international sexual misbehavior 75 agencies 97; approach to refugee problem sexual misconduct 79n30 5; attitude to parallel justice systems 125; sexual morality 70 camp consolidations 34; dependence on sexual offences 19n2, 120 refugees’ cooperation 84; diffi culties in sexual violence 5, 32, 101n9, 111, 117, 123, obtaining interviews with 20n12; 135–7, 142; accountability 139–41; and guidelines 101n4, 101n8; in discussions language 152n4; reporting 140; see also on repatriation 8; initial perception of rape; sexual assault refugee problem 33–4; policies 82–4; Sgawization 49 policing 103; rejection of integration 35; Sgaw Karen 23, 41–3, 49, 64, 71, 77n6, 84, restrictions on income generation 7; 99, 133 support for Karen insurgency 34; talks SGBV see sexual and gender-based violence with Burmese government 163; view of SGBV awareness 136, 141 refugees 159 SGBV committees 136, 140, 141, 143, 151 rule of law orthodoxy 105, 117–26 shadow sovereignty 37n14 rules 10, 13, 58, 79n30, 86, 87, 92, 98, Shan 22, 23, 44 106, 108, 119–22; see also regulations Shan Restoration Council 36n13 rumours 53, 54, 57n26 Shan State Army – South 23 Rwanda 115, 127, 144 Shanti Volunteer Association 101n1 Downloaded by [University of Defence] at 21:28 09 May 2016 Shari’a law 141 safe houses (women’s shelter) 140 Shinawatra, Thaksin 83 Saharawi refugees 40, 97 shunning 99 San C. Po 27, 50 Sierra Leone 79n31, 104 sanctions 61, 67, 109–11 situational communities 41–2 sanitation 100n1; experts 45 Smith, Alan 36n12 sarongs 50 social capital 8, 39, 46, 155 Sartre, Jean-Paul 155 social networks 2, 8, 44 Saw Ba U Gyi 28, 47 social norms 17, 62, 152 schools 2, 7, 39, 47, 49, 85, 100n1 social problems 12, 76n1, 149 Scott, James C. 11, 25, 62, 85, 90, 156 sociological legitimacy 113–14 Index 199

Somalia 40 Tha Ka Sa Pa 32 sovereignty 10–15, 80–102; Tham Hin 5, 164; ethnic composition 16, de-territorialized 10–12, 82; local 94–7; 20n10 outsourced 81, 157, 158; shadow 37n14; theft 12, 19n2, 66, 75, 103, 106, 107, 108, village 18, 58, 60–2, 157, 159 131n32; see also stealing SPDC see State Peace and Development torture 5, 32, 61, 115 Council tradition 46, 58, 59, 63, 76, 157 spells 68–9 see witchcraft traditional justice 64, 125 spirit law 60, 65–70, 106, 157 traditional knowledge 71 spirits 65, 66, 67, 69, 74, 75 traditional law 73 spiritual beliefs 18, 65, 67, 157 translation 133, 141–4 spiritual leaders 63, 66 transnational agencies 80 spousal abuse 135; see also domestic abuse; trial by ordeal 109 domestic violence spousal assault 152n4; see also domestic Uganda 3, 158, 162 violence Umpiem Mai 5, 16, 19n2; 87, staff turnover 117, 120 140; character of camp 5; ethnic State Law and Order Restoration Council composition 20n10, 156; legal 21, 37n15 assistance centre in 118 State Peace and Development Council unemployment 82 (SPDC) 21, 30, 37n15, 38n20 UNFC see United Nationalities Federal state, the 10–12, 17, 18, 25, 80, 82, 90, Council 105, 129; and order 60–73; see also UNHCR law 95 nation-states Union of Burma 22, 23, 27; see also Burma stealing 79n30, 110, 120; see also theft United Nationalities Federal Council stigma 19n3, 111, 136 (UNFC) 37n16 stocks 109, 129n5 United Nations High Commissioner for stories 46, 47, 52 Refugees (UNHCR): access to justice 18, story-telling 40, 65 106; administration of justice programme structuration 10, 158 117; and camp management 88–97, 158; subversion 8, 84 and crime reporting 113; and dispute Sudan 8 resolution 111, 129n4; and engagement suicide 66, 77n11, 89, 131n32 with armed groups 161; and gender Sundaravej, Samak 83 equality 114; and governance 11, 18, 80; supernaturalism 18, 58, 59, 60, 157 and justice matters 123, 125–6; and legal Syria 103, 156 plurality 152; and non-state actors 98, 99; and repatriation 7–8. 164; and Tanzania 47, 57n26, 89, 90, 109, 136, 158 resettlement 54; and the Royal Thai taxation 37n16, 98 Government 34, 38n19, 157, 163; as TBC see Border Consortium, The source of advice 119, 130n19; fear of teachers 7, 45, 52, 53 95; funding for legal assistance centres teenage sexual activity 149 118; funding for safe houses 140; Downloaded by [University of Defence] at 21:28 09 May 2016 Thai (ethnic group) 70 guidelines 140; involvement in support Thai (language) 52, 145 101n1; negative comments about 94; Thai army 81 on policing adultery 76, 79n31; policies Thai authorities 130n19 161; priorities 18; protection mandate Thai-Burma Border Consortium see Border 34, 94; protection of women 135; Consortium, The protection programming 136, 137; Thai courts 113, 117, 121, 123–6, 140, recognized solutions to refugee situations 141, 153n12, 153n23, 159, 161 35; role in Kenya 102n15; size of Thai law 79n31, 105, 118–22, 126 population of concern to 19n1; Thai police 8, 97, 113, 117, 123, 124, statist approach 90; withdrawal from 154n23 Uganda 162 200 Index

VAW see violence against women warnings (as sanction) 106, 120, 121, 147 vehicle accidents 131n32 West Bank 1, 19n1, 40, 60 vernacularization 141, 159 White Elephant 36n7 village destructions 5 witchcraft 66, 68–9, 77n13, 79n31 village governance 24, 60–1, 63, 156 women: access to education 9; agency 9; and village leaders 9, 32, 60–1, 72, 73, 85 justice 117; as UK Supreme Court Judges village sovereignty 18, 58, 60–2, 157, 159 116; as village leaders 9, 60–1; clothing village spirits 67 50; Karenni 51; Muslim 9, 141, 160; violence 40, 58, 59, 61, 106, 120, 147, 148; protection of 100,135; rights 18–19, and displacement 32; and divorce 134–44, 159 153n10; and justice 110; and use of wrist-tying ceremony 69, 77n15 detention 147; bureaucratized 127; domestic 75, 103, 107, 111, 117, 121, x-rated movies 120 123, 125, 131n32, 135–9; gender-based 137; numbers of offences 19n2; political Yawd Serk 36n13 9, 40; protection from 111; sexual 5, 32, young offenders 144 101n9, 111, 117, 123, 135–7, 139–42, young people 39, 48, 51, 52, 73, 86, 152n4; threat of 99; see also violence 110, 139, 144–8, 154n24; see also against women (VAW) children violence against women (VAW) 152n2; see youth offending 117, 122, 144–9, 160; also violence see also crime; delinquency

Wacquant, Loic 2; see also pernicious Zionist narratives 47 premises Zipporah Sein 45 warehousing 1, 4 zone leaders 130n19 see leaders, zone Downloaded by [University of Defence] at 21:28 09 May 2016