ESCAPE FROM VIOLENCE This page intentionally left blank ESCAPE FROM VIOLENCE Conflict and the Refugee Crisis in the Developing World Aristide R. Zolberg Astri Suhrke Sergio Aguayo New York Oxford OXFORD UNIVERSITY PRESS 1989 Oxford University Press Oxford New York Toronto Delhi Bombay Calcutta Madras Karachi Pctaling Jaya Singapore Hong Kong Tokyo Nairobi Dar es Salaam Cape Town Melbourne Auckland and associated companies in Berlin Ibadan Copyright © 1989 by Oxford University Press, Inc. Published by Oxford University Press, Inc., 200 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Zolberg, Aristide R. Escape from violence: conflict and the refugee crisis in the developing world / Aristide R. Zolberg, Astri Suhrke, Sergio Aguayo. p. cm. Includes bibliographies and index. ISBN 0-19-505592-6 1. Refugees—Developing countries. 2. International relief— Developing countries. I. Suhrke, Astri. II. Aguayo, Sergio. III. Title. HV640.4.D44Z65 1989 362.8'7'091724--dc19 88-25156 CIP 2468 10 9 7 5 3 1 Printed in the United States of America on acid-free paper PREFACE Widely perceived as an unprecedented crisis, the number of refugees originating in the developing world since the 1970s has generated urgent concern throughout the West. Such concern is an ambiguous mixture of compassion for the plight of the unfortunates who have been cast adrift and of fear that they will come pouring in. But not only does that fear constantly threaten to undermine the exercise of compassion, it also shows that the affluent countries of the West will neither admit all who seek entry nor give sufficient relief to those who find havens in the developing world itself. This is equally true of neighboring countries in Asia, Africa, and Latin America, which in fact bear the brunt of the crisis. This book seeks to foster a more critical and realistic understanding of the refugee phenomenon, so as to clarify the obligations of the more fortunate of the world toward others in great need, and the ways in which these are best implemented. We shall attempt to explain why the developing world today is producing so many refugees; why they sometimes come in a flood and sometimes in a trickle; where they go; and why they sometimes return and sometimes do not. This is the first attempt to provide a comprehensive, theoretically grounded expla- nation of refugee flows. Social scientists who analyze the causes and consequences of international migrations generally exclude refugee movements, because they believe that the two types of population movement are fundamentally different. It has long been recognized that migration is governed by social and economic forces that themselves are somewhat regular and thus are amenable to theoretical analysis. By contrast, however, refugee flows are unruly in that they result from events such as civil strife, abrupt changes of regime, arbitrary governmental decisions, or international war, all of which are generally considered singular and unpredictable occurrences. This is reflected, for example, in the differing articles, "Migrations" and "Refugees," in the International Encyclopedia of the Social Sciences (1968). The first consists of sections by a sociologist and an economist, who review theories seeking to explain population move- ments generally; the second consists of two sections: one, on "World Problems," was written by a political scientist specializing in international law and organization, and was focused on the development of the Office of the United Nations High Commissioner for Refugees; another was on "Adjustment and Assimilation" of refugees.1 Vi PREFACE When concern over refugees began to mount in the late 1970s, the only analytic framework dealing with such movements was Egon F. Kunz's "Kinetic models," which are, however, formalistic and very abstract.2 Several publications followed, but it is noteworthy that in the two special issues of journals devoted to the subject, one ignores the question of causation altogether, and the other devotes only one article to it—inci- dentally, a preliminary formulation by A. R. Zolberg of the approach used in this book.3 Indeed, the assumption in this book is that refugee flows, like other international population movements, are patterned by identifiable social forces and hence can be viewed as structured events that result from broad historical processes. Although the flows are irregular, the events that trigger them are themselves manifestations of persis- tent trends in the developing world. A further step toward their analysis is to map both the different situations that produce refugees and the characteristics of the flows and to organize them into typologies. We have done this here by surveying the major refugee- producing regions and discussing in more detail a variety of case studies. On the basis of our findings, we demonstrate that different types of social conflict give rise to different types of refugee flows and that the patterns of conflict are themselves intimately related to more general economic and political conditions, not only in the countries from which the refugees originate, but also in the world at large. This emphasis on conflict patterns differs from that on typologies in political science, which generally focus on types of political regime. Our approach is more suitable in this case, however, because in the developing world, regimes come and go (e.g., civilian or military, democratic or authoritarian), whereas the conflicts that produce this instability are themselves enduring. The use of conflict patterns also makes it possible to analyze a given country as a field of social activity within which a number of distinct conflicts are taking place at the same time, possibly generating diverse types of refugee flows con- currently. We attribute considerable weight to factors external to the country in which the refugees originate, thereby challenging the prevailing "internalist" view of the root causes of such flows—that the crucial factors are internal to the state of origin. In contrast with this, our regional studies indicate that usually the causes are more complex and include numerous instances of egregious external intervention. Beyond this, the internal factors themselves often are part of patterns of social change determined by a combination of closely intertwined external and internal processes. In addition, as we shall show, the refugee and immigration policies of outside countries and organizations play a poorly understood but significant role in triggering the flows and shaping their course. Why then has the internalist conceptualization dominated both scholarly and public discourse about refugees? Because international law is founded on the notion that the world is divided into a finite set of states with mutually exclusive jurisdiction over segments of territory and clusters of population, there is a tendency to assume that social causation can also be located precisely within or among particular states. For reasons that will be elucidated in Chapter 1, this has generally suited the purposes of the receiving countries of the West, which collectively also shaped the international institutions that deal with refugees and that in turn operate under political constraints that foreclose attribution of responsibility for causing the flows. The research literature also reflects these conventional legal assumptions and diplomatic obfuscations, a neglect that has contributed to the inadequacy of proposals to deal with the "refugee crisis." We therefore hope that the reconceptualization proposed here, simultaneously struc- tural and transnational, will not only stimulate further discussion among academic spe- PREFACE Vii cialists but will also spill over into the policy arena so as to broaden the terms of reference of the current debates. To some extent, this began to happen even before the publication of this book. Initial efforts were summarized by Charles B. Keely, who as far back as 1981 commented that "the structural analyses of Zolberg and Suhrke reduce the feeling of helplessness and support an approach to refugee policy and programs that departs from recent practices."4 We hope to have lived up to his generous characterization. Another encouraging sign is Leon Gordenker's Refugees in International Politics (1987), which is concerned primarily with the scope and limits of the international "refugee system" but which devotes some attention to "the novel nature of recent refugee incidents" as well as their "causal elements" and, in so doing, takes into account some of our earlier formulations.5 The term refugee is ambiguous and has given rise to protracted controversies, and thus, before analyzing refugee flows, we must identify as precisely as possible the category of persons with whom we are concerned. Although few question that "ref- ugees" deserve special care, even a perfunctory survey of the contemporary scene will reveal widespread disagreement within and among countries on the criteria for determin- ing refugee status. And even when the criteria are agreed on, the very nature of social reality renders it difficult to objectify them so as to provide unambiguous guidance for officials. Our first chapter is therefore entitled "Who Is a Refugee?" We answer the question by retracing the European historical origins of contemporary legal definitions and practices and then assessing them in the light of current realities. We then provide a definition that is both realistic and theoretically valid, as well as firmly grounded in humanitarian concern. This is followed by a number of detailed regional studies, covering most of sub- Saharan Africa, Asia, and Latin America, from approximately 1960 to the present (Part Two, Chapters 2 to 8). The only major region not covered is the Middle East, which lies outside our collective competence. We do, however, refer at various points to the Pa- lestinians, because their case is historically important to the development of international institutions and because, at a more analytic level, it is relevant as a distinctive pattern of response by the refugees to their predicament. Each of the chapters is constructed around case studies selected either because of their intrinsic importance to the region in question or because they provide special insights into the relationships with which we are concerned. From a social science perspective, these regional studies are designed to validate our hypotheses, but they also are meant to serve a practical purpose. The public's fear of refugees is closely related to the notion that flows have rapidly increased in recent years and will probably continue to do so in the future. Although nobody can comfortably predict future numbers, a more precise identification of those patterns of social conflict that are characteristic of different regions of the developing world, together with an assessment of their incidence and bearing on the generation of refugee flows, can yield sober projections more useful for decision makers and the concerned public. These studies provide the foundation for the more general analyses presented in Part Three. Our examination of the relationships between social conflict and the formation of refugee flows is elaborated in Chapter 9, in which we also assess the likely incidence of various patterns. Finally, in Chapter 10, we discuss the policy implications of our findings for individual states and international organizations. Keely's comments, published in 1981, reveal the long period of gestation required to complete this study. In fact, he identified theoretical affinities in the work of Suhrke and Viii PREFACE Zolberg even before they decided to collaborate. They became acquainted in 1981 while preparing papers for a "state of the art" conference organized by Mary M. Kritz of the Rockefeller Foundation.6 Suhrke came to the subject by way of research on ethnic conflict in Southeast Asia and takes this opportunity to thank Charles Morrison, at the East-West Center in Hono- lulu, who in 1979 as a staff member in the U.S. Senate, was instrumental in commis- sioning research on the impact of the Indochinese refugees on the ASEAN countries. As one of the few academic experts on Southeast Asia on the U.S. East Coast, Suhrke was asked to conduct the study. Zolberg, a former Africanist, came to refugees by way of research on the role of states in determining and controlling international population movements and takes this opportunity also to thank Gerhard Casper of the University of Chicago Law School, who involved him in that subject (with the promised book still to be completed). In 1982, Zolberg and Suhrke drew up a formal research proposal, focusing on Asia and Africa, which obtained generous funding from the Ford and Rockefeller foundations. Special thanks are due also to Paul Balaran (Ford) and again Mary M. Kritz, who not only recommended approval of the project but also urged that it be broadened to cover Latin America. On the advice of Patricia Weiss Fagen, the original duo enlisted Sergio Aguayo, who was writing a book on Central American refugees.7 As is evident from any "before" and "after" comparison, inclusion of the Latin American refugee experience led to a revision of the theoretical arguments previously developed on the basis of Asia and Africa alone. The resulting work is therefore genuinely the output of a trio. Our research began in late 1983. Over the next two years, we proceeded along two tracks, carrying on research on our respective case studies while simultaneously elabo- rating the common framework, which at times took on the air of a flying circus (with a bow to Monty Python). Although our grant did not allow us to do extensive field work, Suhrke and Aguayo were able to take advantage of various opportunities to make firsthand observations. In March 1984, Suhrke and Zolberg presented a preliminary version of the frame- work of this book, based on case studies of Ethiopia and Afghanistan, at the Center for Migration and Population Studies, Harvard University (Oded Stark, director).8 Following especially constructive comments from Myron Weiner, they prepared a revised version for delivery at the American Political Science Association meetings the following fall, where they benefited further from the insights of Crawford Young and Art Hansen. In the summer of 1984 the trio embarked on a combined research and lecture tour in Western Europe. Particularly stimulating were the seminar discussions organized by Barbara Har- rell-Bond, director of the Refugee Research Center at Oxford University, and the late Hedley Bull; Goran Melander at the University of Lund; and Catherine de Wenden at the Centre d'Etude des Relations Internationales in Paris. Other helpful visits included the British Refugee Group (London), the Refugee Documentation Center (Bonn), the Peace Research Institute (Frankfurt), the Office of the UNHCR (Geneva), and the Independent Commission on International Humanitarian Issues (Geneva), whose report parallels the theme of this book.9 It was at this time that the transnational emphasis—stimulated by closer examination of the Latin American experience—emerged as a major theme. This new orientation led to a substantial recasting of the original framework and was formulated in a point paper prepared for a special issue of the International Migration Review, edited by Dennis Gallagher, director of the Refugee Policy Group in Washington, D.C.10 PREFACE ix We further had occasion to discuss our approach in a seminar at the Colegio de Mexico (Mexico City) and to observe refugee settlements in the state of Chiapas. We are grateful to our hosts, who must remain anonymous. Further versions were presented at various scholarly conferences, and a final, memorable week was spent in a mountain retreat, hammering out the last version of the framework, and preparing a detailed outline of this book.11 Its completion was delayed by the difficulties of long-distance coauthorship. The regional chapters were written separately but subjected to mutual criticism; the analytical chapters were outlined collectively but written by Zolberg (1 and 9) and Suhrke (10). With additional assistance from the sponsoring foundations, we were able to meet for final workshops in the spring and fall of 1987 and to present our draft manuscript to colleagues at an invitational conference convened at the New School for Social Research in New York City. All three authors acknowledge their special gratitude to forbearing spouses who put up not only with repeated absences but also, even more patiently, with recurrent visits from one or the other coauthor dropping by for yet another protracted round-the-clock working session. Vera Zolberg bore the brunt of this with unswerving kindness and good humor, often at the expense of her own work. A few good friends also provided moral support at critical junctures, as well as hospitality; among them Christine Birnbaum, University of Paris XIII—Villetanneuse, deserves special mention. Finally, we also thank our respective institutions, the Graduate Faculty of the New School for Social Research, which managed the grants and provided additional research assistance; the School of International Service, American University; and the Colegio de Mexico. Throughout, our project also benefited from the suggestions of an advisory group whose participants included at various times Senator Dick Clark (former U.S. coordinator for refugees); Richard Day (Subcommittee on Immigration and Refugees, U.S. Senate); Arthur Dewey, deputy assistant secretary, Bureau for Refugee Programs, U.S. Depart- ment of State; Patricia Fagen (then Refugee Policy Group); Dennis Gallagher (Refugee Policy Group); Albert Hirschman (Institute for Advanced Study); Philip Johnston (CARE); Charles Keely; David Martin (University of Virginia Law School); Doris Meissner (then executive associate commissioner, U.S. Immigration and Naturalization Service); Michael Posner (Lawyers' Committee for International Rights); Zia Rizvi (then regional representative, UNHCR Regional Office); Saskia Sassen-Koob (City University of New York); Ambassador Walter Stoessel; Dale Swartz (National Immigration Refugee and Citizenship Forum); Julia Taft (consultant); Jerry Tinker (Subcommittee on Immi- gration and Refugees, U.S. Senate); and Jennifer Whitaker (Council on Foreign Rela- tions). Aguayo and Zolberg also profited from contact with the project on Latin American and Caribbean Immigration and U.S. Foreign Policy conducted by Christopher Mitchell at New York University. Beyond these collective acknowledgments, Sergio Aguayo wishes to thank Laura O'Dogherty, Mexican Academy of Human Rights; Rodolfo Stavenhagen; Maria Moore; Alfredo Witsschi (UNHCR); Gordon Hutchison, Belela Herrera, Felipe Tomic, and Eli- zabeth Odio of Project Counseling Service for Latin American Refugees; colleagues at La Jornada, especially Carlos Payan and Carmen Lira; Dr. Roberto Gomez Alfara and Rene Morin of SEDEPAC; Juan Juarez, refugee from Guatemala; and colleagues at El Colegio de Mexico. Astri Suhrke expresses her gratitude for the patient assistance of the refugees and those aiding them during her fieldwork in South and Southeast Asia since 1979, partic-