Ethno-Religious Violence in Indonesia: From Soil to God [1 ed.] 0415453801, 9780415453806, 9780203928981

Table of contents :
Book Cover......Page 1
Title......Page 6
Copyright......Page 7
Dedication......Page 8
Contents......Page 10
Illustrations......Page 11
Acknowledgements......Page 12
Glossary......Page 13
Introduction......Page 16
1 The study of violent communal conflict......Page 30
2 North Maluku in context......Page 44
3 Initiation – Malifut......Page 64
4 Escalation – Ternate and Tidore......Page 86
5 Dispersion – Tobelo and Galela......Page 111
6 Political exploitation – the Putih–Kuning......Page 145
7 Killing in the name of God......Page 162
Conclusion......Page 192
Notes......Page 212
Bibliography......Page 246
Index......Page 254

Citation preview

Ethno-religious Violence in Indonesia

Ethno-religious Violence in Indonesia illustrates in detail how and why two previously harmonious religious communities can descend into violent conflict. Lasting from 1999 until 2000, the conflict in North Maluku, Indonesia, saw the most intense communal violence of Indonesia’s turbulent period of democratization. For almost a year, militias waged a brutal religious war which claimed the lives of nearly 4,000 people. The conflict culminated in ethnic cleansing along lines of religious identity, with approximately 300,000 people fleeing their homes. Based on four years of research, including almost one year living in North Maluku interviewing combatants, politicians and security personnel among others, the book provides the first comprehensive account of this violence. The accounts of participants and witnesses give the reader the opportunity to better understand the tensions and fears involved in the conflict and begin to grasp the motives of those who kill large numbers of men, women and children. The book provides numerous examples of how different conflict theories can be applied in the analysis of real situations of tensions and violence, illustrating the mutually reinforcing nature of mass level sentiment and elite agency, and the rational and emotive influences on those involved. This book will be of interest to researchers in Asian studies, conflict studies and religious violence. Chris Wilson completed his PhD at the Research School of Pacific and Asian Studies, Australian National University.

Routledge Contemporary Southeast Asia Series

1

Land Tenure, Conservation and Development in Southeast Asia Peter Eaton

2

The Politics of Indonesia-Malaysia Relations One kin, two nations Joseph Chinyong Liow

3

4

Governance and Civil Society in Myanmar Education, health and environment Helen James Regionalism in Post-Suharto Indonesia Edited by Maribeth Erb, Priyambudi Sulistiyanto and Carole Faucher

5

Living with Transition in Laos Market integration in Southeast Asia Jonathan Rigg

6

Christianity, Islam and Nationalism in Indonesia Charles E. Farhadian

7

Violent Conflicts in Indonesia Analysis, representation, resolution Edited by Charles A. Coppe

8

Revolution, Reform and Regionalism in Southeast Asia Cambodia, Laos and Vietnam Ronald Bruce St John

9

The Politics of Tyranny in Singapore and Burma Aristotle and the rhetoric of benevolent despotism Stephen McCarthy

I0 Ageing in Singapore Service needs and the state Peggy Teo, Kalyani Makta, Leng Leng Thang and Angelique Chan 11 Security and Sustainable Development in Myanmar Helen James 12 Expressions of Cambodia The politics of tradition, identity and change Edited by Leakthina Chau-Pech Oilier and Tim Winter 13 Financial Fragility and Instability in Indonesia Yasuyuki Matsumoto

14 The Revival of Tradition in Indonesian Politics The deployment of adat from colonialism to indigenism Edited by Jamie S. Davidson and David Henley 15 Communal Violence and Democratization in Indonesia Small town wars Gerry van Klinken

16 Singapore in the Global System Relationship, structure and change Peter Preston 17 Chinese Big Business in Indonesia The state of the capital Christian Chua 18 Ethno-religious Violence in Indonesia From soil to God Chris Wilson

Ethno-religious Violence in Indonesia From soil to God

Chris Wilson

First published 2008 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Avenue, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business

This edition published in the Taylor & Francis e-Library, 2008. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2008 Chris Wilson 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. 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 Wilson, Chris, 1969– Ethno-religious violence in Indonesia: from soil to God / Chris Wilson p. cm. — (Routledge contemporary Southeast Asia series: 18) Includes bibliographical references and index. 1. Ethnic conflict—Indonesia—Maluku. 2. Communalism—Indonesia—Maluku. 3. Muslims—Indonesia—Maluku. 4. Christians—Indonesia—Maluku. 5. Maluku (Indonesia)—Ethnic relations. 6. Maluku (Indonesia)—Social conditions. I. Title. DS646.67.W55 2008 959.803¢9–dc22 2007039080

ISBN 0-203-92898-9 Master e-book ISBN

ISBN10: 0-415-45380-1 (hbk) ISBN10: 0-203-92898-9 (ebk) ISBN13: 978-0-415-45380-6 (hbk) ISBN13: 978-0-203-92898-1 (ebk)

For Elly, Aislin and Ciara. And in memory of the thousands who lost their lives in North Maluku.

Contents

List of illustrations Acknowledgements

x xi

Introduction

1

1

The study of violent communal conflict

15

2

North Maluku in context

29

3

Initiation – Malifut

49

4

Escalation – Ternate and Tidore

71

5

Dispersion – Tobelo and Galela

96

6

Political exploitation – the Putih–Kuning

130

7

Killing in the name of God

147

8

Conclusion

177

Notes Bibliography Index

197 231 239

Illustrations

Tables 1 Phases of the North Maluku conflict 2 Chronology of events before violence in Tobelo City, 1999

4 103

Maps 1 2 3 4 5 6 7

Indonesia North Maluku Kao/Malifut Ternate Tidore Tobelo Galela

xiv 30 53 75 87 100 116

Acknowledgements

A large number of people provided me with assistance during the completion of this book. In North Maluku, hundreds of people gave generously of their time, assistance and friendship in helping me in the field. In particular, I would like to acknowledge the friendship and hospitality of Tot Duan, Ibu and Echa. Samsu Rizal Panggabean of the Center for Security and Peace Studies at Gadjah Mada University supported my research in Indonesia. A number of people in Canberra and elsewhere gave generously of their time to comment on drafts at various stages of disarray, including Elly Lawson, Juliet Taylor, Catherine Mann, Dave McRae, Dr Michael Schiavone, Professor Ben Kerkvliet, Dr Ron May, Dr Ed Aspinall and Professor Paul Brass. I would also like to thank an anonymous reviewer at Routledge for comments on an earlier draft of the book. I would like to express my utmost gratitude to Professor Harold Crouch. From the first until the last day of my research process, Harold provided me with insightful advice and constructive criticism in a friendly and interested manner. His advice has played a major role in anything that is good about this book.

Glossary

adat FKPKHU FPI GMIH

tradition North Halmahera Christian Youth Communication Forum Front Pembela Islam, or Islamic Defenders Front Gereja Masehi Injili Halmahera, Evangelical Church on Halmahera GPM Gereja Protestan Maluku, or Maluku Protestant Church Hibua Lamo System of binding cultural ties between families in North Halmahera IDP Internally Displaced Person Laskar Jihad Islamic militia formed on Java in early 2000 NHM Nusa Halmahera Mineral (mining company operating in Malifut area) Pasukan Jihad Islamic militia formed in North Maluku in early 2000 PDI–P Partai Demokrasi Indonesia – Perjuangan, or Indonesian Democratic Party – Struggle PP42 Government Regulation 42 creating Malifut Sub-District PPP Partai Persatuan Pembangunan, or United Development Party RMS Republik Maluku Selatan, or Republic of South Maluku separatist movement TNI Tentara Nasional Indonesia, or Indonesian National Military

Map 1 Indonesia

Introduction

Behind a mosque in the remote coastal village of Kao in North Maluku, eastern Indonesia, there lies a large stone grave. The remains of seven warriors rest inside the tomb, all members of the Kao ethnic group who died in a clash with Dutch soldiers in the first decade of the twentieth century. The gravesite holds special significance for Kaos. It is the resting place for both Christians and Muslims, a symbol for the Kaos over the past century of their strong ethnic solidarity, regardless of religious differences. For most of the twentieth century this concord was also representative of the North Maluku community as a whole. In 1999, however, this society, which had long prided itself on inter-religious harmony, descended into bloody religious war. Devastating communal violence often begins with a series of events that in hindsight could easily have been contained. In August 1999 a small ethnic dispute erupted in a village on the large island of Halmahera, to the east of Sulawesi, marking the beginning of a series of clashes that brought horrific consequences to the entire region. Throughout North Maluku, over 3,500 men, women and children were killed by opposing mobs armed with machetes, spears and bows and arrows. Several hundred thousand people were displaced from their homes, and the fighting destroyed much of the region’s housing and infrastructure. This book attempts to account for this violence. In particular it seeks to explain five aspects of the conflict: why violence began; how it evolved from a small dispute into a religious war; why it reached such a frightening level of intensity; why it spread across the entire province; and why it ceased. This introductory chapter begins with an overview of the violence and its consequences. The second section gives an overview of the existing literature on the conflict.

The conflict: August 1999 to June 2000 Before August 1999, North Maluku had seen half a century of peace and stability. The region, which until September 1999 comprised two relatively remote districts in the north of Maluku Province, was far removed from national politics. While it did not benefit significantly from Indonesia’s economic boom, North Maluku enjoyed relative self sufficiency and good relations between ethnic and religious groups. Even the onset of terrible religious violence in Ambon in the southern part

2 Ethno-Religious Violence in Indonesia of the province in January 1999 (discussed in Chapter 2) caused little damage to inter-religious relations in the region. Yet, in August of that year, just as North Maluku prepared to become an independent province and reap the benefits of Indonesia’s new era of political liberalization and regional autonomy, this harmony was shattered. On the night of 18 August 1999 two ethnic communities – the Makians, long-term migrants to the area, and the indigenous Kaos – clashed in a village in remote Malifut on Halmahera Island (see Map 3.1).1 As dawn broke and the Kao village of Sosol lay in ruins, thousands of Makians moved on to and quickly overran neighbouring Wangeotak village. Three Kao men perished in the fighting. Two months later, just after the inauguration of North Maluku Province, the Kaos retaliated, driving all Makians from Malifut and destroying their villages. Because the Kaos were predominantly Christian and the Makians Muslim, the stage was set for the large-scale religious conflict that was to sweep through North Maluku, despite the fact religion had been inconsequential to their initial dispute. Thousands of Makians fled to the provincial capital, Ternate, where retaliatory violence erupted, as well as on the neighbouring island of Tidore and some areas of Halmahera. Mobs targeted not just Kaos, but any member of the Christian minority. Rioters destroyed Christian houses and churches, while Christians fled to local police and military compounds. Dozens of people were killed, including a Protestant pastor on the island of Tidore (his body dismembered and burned). These riots had a far greater impact than the Malifut incident, sending shockwaves across the province. From this moment, almost the entire North Maluku community divided along religious lines. In a classic ‘security dilemma’, communities began preparations either to defend themselves or, in some cases, to launch pre-emptive attacks against their neighbours. Over the following months, violence spread out across the region, affecting Halmahera, Bacan, Morotai and all the islands in the archipelago. The violence reached its brutal peak in December 1999 in Tobelo and Galela Sub-Districts in north Halmahera, which had almost equal populations of Muslims and Christians. After fighting began in Tobelo City, Muslims took control of the town for just one night before thousands of Christians, armed with homemade weapons and bombs, flooded in from rural areas. Over the next two days, these Christians expelled the entire Muslim community from the town and surrounding villages. Perhaps a thousand people died in the carnage, including women and children. This violence immediately flowed into the sub-district of Galela, where militia attacks quickly divided the area into exclusively Muslim and Christian enclaves. During this inter-religious violence in north Halmahera, Ternate also descended again into fighting, this time between Muslims. After these intra-Muslim clashes ended, the opposing Muslim factions set aside their differences in order to retaliate against Christian militias on Halmahera. They sought to legitimize these assaults by reference to the principles of jihad. The violence finally ended in July 2000 after the fall of the Christian village of Duma in Galela, which had resisted attacks for six months. Shortly afterwards, in response to continuing violence in Maluku and North Maluku, President

Introduction 3 Abdurrahman Wahid declared a Civil Emergency in the two provinces. Combatants in North Maluku thereafter faced a more effective response from the security forces. Perhaps more importantly, the violence had segregated Christians and Muslims into separate areas of the province, divided by sizeable military contingents, and continued attacks were untenable for the exhausted militia members. After almost one year of violence, over 3,000 people lay dead. The populations of many villages had been decimated, and Halmahera was littered with mass graves. Innumerable people from both communities remained in hiding in the dense and vast forests of the island. Statistics distributed in March 2000 by the North Maluku Governor’s Office stated that 2,083 people had by that date died during the conflict. Although the local government restated this figure in 2003 as a final estimate, the actual number of deaths was almost certainly higher. Further incidents between March and June saw large loss of life, most notably attacks on the villages of Duma, Makete, Dokulamo and Mamuya in Galela, and the tragic sinking of the Cahaya Bahari passenger ship, in which approximately 500 people drowned. The official statistics do not include an estimated 250 casualties from violence in the sub-district of Payahe in November 1999, nor do they include casualties among the local Islamic militia, the Pasukan Jihad (Jihad Force), who were killed outside their home sub-districts.2 According to one leader of this militia, approximately 200 mujaihid died in North Maluku.3 By July 2000 between 3,000 and 3,500 had died. The large number of missing persons cited in March 2000 (2,315) suggests an even greater death toll. The Governor’s Office reported that the violence displaced approximately 250,000 people.4 Of these, 199,605 were displaced within North Maluku Province and 48,015 were forced to flee to elsewhere in Indonesia. Given that this figure was compiled several months before the end of hostilities, the final figure was probably higher, although most people had already fled their homes by the time these figures were collated. Most Muslim Internally Displaced Persons (IDPs)5 were evacuated to Ternate, almost doubling that city’s population to approximately 180,000. Some also sought refuge outside North Maluku in South Sulawesi. Perhaps 15,000 Christian IDPs fled to the Christian-dominated city of Tobelo in north Halmahera. The majority of Christians, however, went to the province of North Sulawesi (see Map 0.1) and were settled by the local government in camps in the cities of Manado and Bitung. In 2005, many IDPs remained in North Maluku and North Sulawesi. Many have settled around their displacement camps in areas such as Tobelo, rather than returning home. Others feel unable to return home because of a lack of government assistance, or because they fear reprisals or new outbreaks of violence. North Maluku’s infrastructure, particularly on Halmahera, was devastated. According to the March Governor’s Office report, 18,022 houses, 97 mosques, 106 churches and 110 schools were destroyed. The final number of buildings destroyed was undoubtedly much higher. Most villages located in a sub-district dominated by the opposing group were destroyed entirely. Combatants destroyed bridges, telephone poles, fishing boats, warehouses, gardens and other infrastructure crucial to North Maluku’s economic and social life.

4 Ethno-Religious Violence in Indonesia To some extent, the conflict in North Maluku also affected national politics and society. The intense violence in Tobelo in late December 1999, which involved the deaths of almost a thousand Muslims and was described in often inflammatory terms in the national media, stimulated the formation in Java of a large Muslim militia, the Laskar Jihad. This organization, which gained national and international notoriety, became involved in communal violence in Maluku Province, although not, as this study will show, in North Maluku.

Studies of the conflict and remaining questions This book seeks to account for this tragedy. What caused violence to start and to escalate is a question not just for Indonesia, although the social, economic and political costs of the conflict were severe for the immediate region and archipelagic nation, as discussed above. The causes of a violent communal conflict of this magnitude are of relevance to the global community. In addition, the fact that the violence occurred during Indonesia’s transition from an authoritarian political system is a reminder of the pitfalls of rapid democratization and decentralization. Similarly, the fact that most of the violence occurred between Muslims and Christians presents lessons for the relationship between these two communities. In order to capture the complexity, temporal dynamism and geographical variation within the North Maluku conflict this study is divided into five chronological phases. The phases are essentially new outbreaks of violence, each involving several unique political, economic and social variables. In some cases, these events involved riots by a majority group against local minorities. In others, clashes took place between relatively evenly balanced forces. While each new outbreak was influenced by those preceding it, each also had its own specific dynamics and causes. The events in these five phases defined the trajectory of the conflict as a whole, determining how it began, escalated, spread, was exploited by political actors and ended in religious war. The phases are listed in Table 1. The following section discusses each phase in turn, beginning with a summary of the existing literature on the events in question. Commentators, the majority of whom are from North Maluku, have compiled a substantial body of literature on Table 1 Phases of the North Maluku conflict Phase

Location

Date

Initiation

Malifut

August and October 1999

Escalation

Ternate and Tidore

November 1999

Dispersion

Tobelo and Galela

December 1999

Political Exploitation

Ternate

December 1999

Religious War

North Halmahera

February to June 2000

Introduction 5 the conflict.6 However, most studies are short in length and analyze the entire conflict in just a few pages, leaving the reader confused as to which issues were central to the first outbreak of violence, and which to later stages in its development. For this reason, the following discussion of the published literature on the conflict is divided into individual scholars’ analyses of each main event in the conflict, rather than presenting their conclusions as a whole. The majority of studies also suffer from a lack of objectivity: they are written on the whole by local commentators and in many cases by individuals directly affected by the violence. Most present either a Christian or Muslim perspective, allocating blame solely to the opposing religious community.7 The discussion of each phase below will then outline the shortcomings in existing published accounts of the events and point to questions that remain unanswered. Each section will also outline the theoretical frameworks used to help analyze each main development. I have utilized these theories where relevant throughout the study and will introduce them briefly below. Initiation – Malifut All commentators on the conflict point to the fact that the first violence in North Maluku, that in Malifut in August 1999, erupted just as the local government approved a new sub-district in the area. However, most disagree on how and why the formation of this new sub-district led to conflict. For some, the destruction that took place in August and October was the culmination of growing tension between the two local ethnic communities – the indigenous Kaos and the Makians, who had been moved to the area by the local government two decades before. M. Kordi writes that the economic success of the migrants relative to that of the Kaos caused inequality, jealousy and inter-ethnic tension.8 The Indonesian sociologist and native of North Maluku, Tamrin Tomagola, takes the long-term tensions argument further, stating that the area of Malifut had become the centre of a struggle between Christianity and Islam on Halmahera.9 According to Tomagola, Christians believed that, in relocating the Makian community from Makian Island to Malifut in 1975, the local district government had deliberately aimed to halt the southward expansion of Christianity. According to Tomagola, this perception contributed to the level of friction in 1999.10 Others claim that the creation of the new sub-district was in itself a sufficiently contentious issue to provoke hostilities. The presence of a large goldmine in the area, operated since 1997 by the Australian company Newcrest, has led many analysts to conclude that economic competition played a crucial role.11 Most analysts argue that the two communities fought for control of this resource and the Kaos eventually destroyed Malifut as a means of eliminating competition.12 Indeed, the Ternate academic, Smith Alhadar, ignores the initial riot in August, claiming that the first violence was that in October, when the Kaos, jealous of the Makians for dominating employment at the mine, attacked and destroyed Malifut.13 The North Malukan Christian academic, Jan Nanere, asserts, however, that neither gold nor jealousy was the primary contentious issue involved in the creation of the new sub-district. More

6 Ethno-Religious Violence in Indonesia importantly, the new sub-district violated long-established ethnic boundaries recognized by the indigenous Kaos. 14 Given that the first Malifut incident occurred while the North Maluku elite jockeyed for political power in the new province, most analysts see the initial outbreak in Malifut, and the conflict in general, as connected to this wider political competition. Alhadar writes that ‘just as the administrative wheels began to turn in mid-1999 to split off North Maluku as a province of its own, the conflict began to escalate’, and that the ‘riots must be seen in the context of a government plan at the time to hold local elections for a new provincial parliament in June 2000’.15 The International Crisis Group also concludes that the ‘separation of North Maluku … stimulated rivalry by creating the need to elect a new governor and this seems to have been one of the driving factors behind the initial outbreak in Halmahera’.16 Nevertheless, most analysts disagree on how this political competition caused violence on Halmahera, and which individuals were most culpable. Tomagola claims that the conflict resulted in part from the increasingly desperate attempts of the Sultan of Ternate to become governor of the new province.17 The Sultan faced a strong challenge for the governorship from the District Head of Central Halmahera, Bahar Andily, who, according to Tomagola, was assured of the support of the majority of the Muslim community, which, according to Tomagola, constituted 87 per cent of the provincial population. The Sultan, although a Muslim, was largely considered to rely on Christian support. Tomagola relates that the Sultan of Ternate sought to shore up this Christian support by promising Christians on Halmahera that the Makian migrants would be removed from Malifut, and claims that this promise encouraged the Kaos to turn to violence when the new sub-district was created in Malifut.18 Nanere also argues that the violence in Malifut was provoked by members of the elite in Ternate. However, he does not lay the blame on the Sultan of Ternate, pointing more to provocative statements by the Mayor of Ternate and the head of the bureaucracy (Regional Secretary).19 He does not make clear why these individuals sought to cause violence on Halmahera. Several analysts have argued that political competition was made more volatile by a revival of the long-standing rivalry between the two historic and powerful sultanates of Ternate and Tidore, and that, in supporting the Kaos in their dispute with the Makians, the Sultan of Ternate was thereby attempting to maintain support in the sultanate’s traditional area of influence.20 The International Crisis Group argued that ‘on the provincial level, seeds of conflict can be found in the centuries-old political rivalry between the Sultans of Ternate and Tidore’.21 Alhadar agreed that ‘Tidore people began to worry that their traditional enemies on Ternate were preparing to revive the cultural dominance they had enjoyed in the past in order to justify a resurgence of their political power.’22 Because North Maluku was still part of Maluku Province in August 1999, most analysts have seen the Malifut clash as a consequence of the inter-religious violence that had gripped the provincial capital, Ambon, for most of that year.23 For example, the International Crisis Group reports that, after a resurgence of conflict in Ambon, the violence then ‘spread in October to Malifut on Halmahera Island’.24 Jacques Bertrand also claims that the conflict in North Maluku was a

Introduction 7 consequence of the ongoing violence in Maluku.25 Citing Tomagola, Bertrand suggests that not only was Malifut the site of competition between Muslims and Christians for control of Halmahera, but that Christians were concerned about the success of Muslims in national and local politics.26 According to Bertrand, stories of the violence in Ambon aggravated these tensions. He concludes that ‘it is not coincidental that the violence erupted after months of conflict between Christians and Muslims in other parts of the region’ and ‘thus local issues, while important, provided the trigger and the filter through which tensions at the national and regional level were expressed.’27 There is therefore a range of, often conflicting, explanations of the start of the conflict. Despite these conclusions, a clear picture has not been provided as to why the dispute had a violent, as opposed to a non-violent outcome. Under the process of pemekaran (literally ‘blossoming’, but more appropriately ‘division’), the Indonesian government has, since 1999, created a large number of new sub-districts and districts and several provinces, but very few have descended into violence as a result.28 It is not clear what lent this case such a highly affective character. Why did violence become accepted as necessary and/or legitimate by large sections of each community, when it had been largely absent from North Maluku before this incident? Did the creation of the new province of North Maluku, and the ensuing competition for political power, play a major role in stimulating violence, as most analysts have suggested? It is not apparent how important, comparatively, were the political changes occurring at the national and local levels. To what extent did several decades of authoritarian rule lay the foundations for the violence in North Maluku in 1999 to 2000? How important was the transition from an authoritarian to a democratic political system in 1998 and 1999? For example, it is not clear whether the military or other individuals or groups central to the New Order regime of President Suharto provoked violence in order to retain power, as has been suggested in the case of conflicts in other areas of Indonesia. Answering these questions may provide an answer to the question of why the violence occurred when it did. The Kao–Makian relationship itself requires examination. Did the two communities have a history of antagonism and/or violence and if so for what reason? Did inequality or other long-standing social, political and economic structures such as segregation and prejudice create the conditions for violence? It is unclear how important religion was in this incident that was to eventually ignite religious war in the region. For example, it remains conjecture that Malifut had become the front line in ongoing religious competition between the Protestant Church and Muslims in North Maluku. In what way did the ongoing religious violence in nearby Ambon (discussed in Chapter 2) contribute to the dispute? The human agency involved in these riots has not yet been adequately uncovered. Analysts have discussed the broader socio-political environment without identifying the actions and motivations of those involved. The existing literature on the conflict does not make clear why the Makians sought an autonomous sub-district in Malifut and why the Kaos mobilized and resisted in such an emotive manner. Were only local people involved in the initial incident, or were people

8 Ethno-Religious Violence in Indonesia from outside the area present, and, if so, did they intentionally or otherwise provoke violence? It remains uncertain whether district government officials incited violence in Malifut or why they would have an interest in doing so, particularly given that the area produced a large amount of revenue for the local government. I will argue in Chapter 3 that the initial violence in North Maluku was not connected in any significant way to either competition for political power at the provincial level or the conflict in Ambon and religious animosity. The violence centred around two primary issues – the control of territory and government partiality. Several theoretical frameworks are used in this analysis. The literature on social movements, particularly that on counter-mobilization, explains how a conflict might unfold when an under-represented community perceives its rights or interests to have been undermined by the mobilization of a rival community which has apparent government support.29 I will refer to Monica Toft and others to elucidate why and how competition over territory can lead to violence, and the theory of Donald Horowitz on the impact of invidious group comparisons.30 The chapter highlights a weakness in the ‘economic causes of civil war’ or ‘greed’ thesis, and other models of conflict based on quantitative datasets.31 The analysis demonstrates that, while lucrative natural resources may be present in a conflict area, they are not necessarily central to the outbreak of violence. This illustrates the danger in imputing too much weight to the presence of objective conditions for conflict without allowing for the importance of subjective understandings and human agency. Even in those cases with clear economic agendas, material considerations alone do not to lead to violence, but only do so in combination with issues more commonly classified as ‘grievance’. Escalation – descent into religious conflict In just a few weeks, the violence in North Maluku escalated from a localized dispute over land into violence targeted at any member of the opposing religion. Most analysts agree the violence took on a religious character only after the distribution of a forged letter ostensibly sent from the Protestant synod in Maluku to the Protestant Church in North Maluku. The letter, entitled ‘Bloody Sosol’ after one of the Christian villages destroyed in Malifut, appeared to be planning for the Kaos’ attack on Malifut. Further, it purportedly proved that the attack was part of wider a strategy of ‘Christianization’ of North Maluku. Most commentators conclude that this letter provoked rage among Muslims in Ternate and Tidore, leading them to attack Christians alongside whom they had lived for years.32 For example, Alhadar states the letter ‘was signed by Rev. Sammy Titaley … it urged Christians to convert Muslims, who were described as “ignorant”. Little wonder people on Tidore were provoked.’33 The assumption that this letter provoked anti-Christian rioting in November has not been questioned. While the letter was clearly forged, its provenance has remained uncertain, and therefore the intentions behind it are unclear. Alhadar raises the possibility that the letter was disseminated by individuals connected with the former New Order

Introduction 9 regime of President Suharto seeking to derail the process of political reform.34 Yet he and most other analysts give only scant consideration to the letter’s origin. Indeed, Bubandt maintains that it is important not to analyze the letter just as a means of riot instigation.35 He claims that the origin of the letter is not as significant as understanding how and why it was believed by so many. In addressing this question, Bubandt stresses the significance of the ‘social and discursive universe within which ordinary people are mobilized during conflict’. In post-Suharto Indonesia, there prevailed a discourse of conspiracy and paranoia, with which the contents of the letter resonated. In Bubandt’s opinion the first violence in Tidore was a ‘spontaneous reaction to the rumour about a Christian conspiracy rather than a long planned assault’.36 Yet the origins of and intentions behind this letter are surely crucial to understanding the escalation and trajectory of the conflict. If the ‘Bloody Sosol’ letter was not from the Protestant Church, who wrote and disseminated it and for what reason? The apparent impact of the letter must also be critically re-examined. Was, as most analysts claim, the wider Muslim community really provoked into violence by it and other forms of propaganda? If so, it is not clear whether intervention by parties with an interest in a wider dispute was necessary to achieve this outcome. Was there either an existing animosity between Muslims and Christians in North Maluku, or a more radical religious ideology that played a role in this widening of the violence? No commentator has yet adequately explained why the large numbers of national security personnel present in the district capitals did not halt the riots. Several analysts have documented how and why small incidents of relevance to only a very restricted group of people (such as individual disputes or crime) can rapidly evolve into sectarian violence. Detailed case studies in other regions demonstrate that the descent into sectarian violence often follows intervention by actors with an interest in a wider, more emotive conflict. Stanley Tambiah’s concepts of ‘transvaluation and focalization’ and Paul Brass’s ‘institutionalized riot systems’ shed much light on the transformation of the Malifut clash into a religious conflict.37 As conceived by these scholars, certain actors portray minor incidents as having major communal significance in order to gain political or other advantage. The use of such concepts will assist in demonstrating how the North Maluku conflict changed in character and escalated to a much higher level of intensity. However, the analysis presented in Chapter 4 also cautions against overemphasizing the importance of propaganda in the escalation of conflict. Propaganda is sometimes disseminated to make communal rioting appear as a sudden eruption of anger, as my analysis demonstrates was the case in the anti-Christian rioting in Ternate. The rioting was carried out by those seeking retaliation for the destruction of Malifut, facilitated by individuals with a political interest in sectarian tension, rather than by those provoked by propaganda. Once rioting started in Ternate, numbers swelled for several reasons: the security forces failed to act against rioters, convincing people that they could act with impunity; excitement spread, particularly among young men; and for some, the targeting of

10 Ethno-Religious Violence in Indonesia churches and other elements of the violence appeared to confirm that this was indeed religious conflict. Dispersion – spread across the province In late 1999 and early 2000, following the riots in Ternate and Tidore, violence broke out in almost every area of North Maluku. Chapter 5 examines how violence spread throughout the province to previously unaffected areas, by focusing on the cases of Tobelo and Galela Sub-Districts. There are several reasons for this focus: Tobelo witnessed the most intense violence in the entire conflict; the area was the location of the largest population centres on Halmahera; and the area possessed relatively equal Christian and Muslim populations. During the terrible violence in Tobelo and Galela, combatants frequently targeted defenceless people, including women, children and the elderly. In many cases, corpses were mutilated and disembowelled. This practice was sometimes followed by the consumption of the body parts, particularly the hearts, of victims. Most local analysts concur that Christians initiated attacks against Muslims in Tobelo after being provoked by the preceding months of violence elsewhere. Tomagola writes that the ‘initial attacks (in Tobelo) were launched simultaneously by the Christians’.38 Alhadar agrees: ‘in response to Muslim attacks on Christians in Ternate, a coalition of Christian tribes in northern Halmahera around Tobelo and Galela on 26 December attacked Muslims living there, eventually resulting in the loss of probably thousands of innocent lives.’39 Ahmad and Oesman relate that local Christians, helped by thousands of Christian IDPs from the violence in November, attacked and drove Muslims from Tobelo. According to these authors, Christians launched the attack so as to remove any obstacle to their control of the valuable economic potential of the sub-district.40 The International Crisis Group also claims that Christians initiated the violence, although motivated by security concerns rather than anger. The group’s report states that in Tobelo ‘local Christians went on the offensive against the local Muslim minority … fuelled by rumours of planned “cleansing operations” on both sides’.41 Bubandt also concludes that ‘conspiratorial fears that the opposing side was planning their wholesale eradication motivated the Christian attacks on Muslim villages in Tobelo.’42 Contrary to this majority view, Jan Nanere claims that local Muslims initiated the violence in Tobelo Sub-District, having planned the attack after the successful expulsion of Christians from Ternate.43 According to Nanere, because they were the minority in Tobelo, Muslims had arranged for reinforcements to arrive from their co-religionists in Ternate and Tidore. However, this assistance did not arrive (with tragic consequences for Tobelo Muslims) because of violence that broke out at the same time in Ternate. As most analyses of the violence in Tobelo are derived from sources in Muslim-dominated areas such as Ternate, the widespread conclusion that Christians initiated violence in Tobelo must be reassessed. Did Christians launch pre-emptive attacks against Muslims and for what reason? To what extent were the attacks a consequence of the ‘reframing’ of the conflict in religious terms by actors

Introduction 11 in Ternate? It is uncertain how important were local animosities and interests in places such as Tobelo and Galela in the spread of violence compared to the religious tension engulfing the region. It is also necessary to uncover who played a role in the spread of violence to Tobelo and Galela, and how and why the violence was so intense in this area, which was previously characterized by inter-religious harmony. Chapter 5 demonstrates that a feeling of insecurity prevalent among both religious communities was crucial to the outbreak of violence. The chapter begins with a discussion of two bodies of conflict theory known as the (physical) Security Dilemma and the Societal Security Dilemma. A synthesis of these two theories provides several insights into the spread of violence to Tobelo and Galela Sub-Districts. Concerns among both communities exacerbated one another, causing increased belligerence between Muslims and Christians and increased legitimacy for the more militant actors in society. My analysis suggests it was these concerns that pushed the two communities towards violence, and neither community deliberately sought to launch pre-emptive attacks. That said, in my discussion of the violence in these sub-districts I have attempted to move away from attributing cause solely to the security dilemma. While this situation certainly stimulates rising militancy, human agency is essential to any complete explanation: it is the acts of militants and other individuals, some of them unintended, which explain the terrible events. It is harder to explain the atrocities witnessed in the two areas. In attempting to do so I will use the arguments of Natalie Zemon Davis, R Scott Appleby and Mark Juergensmeyer. I do not agree with the apparent conclusion of these and other authors that the nature of religion necessarily facilitates extreme violence more than ethnic, class or other ideologies. However, I do conclude that religious sentiment exacerbated the atmosphere of anger and fear that characterized Tobelo in December and did play a role in much of the brutality that occurred. Political exploitation – intra-community conflict Commentators have paid little attention to the apparent paradox of intra-Muslim violence which took place in Ternate at the same time as inter-religious conflict engulfed the rest of the province. Those commentators who have addressed these events conclude that they were the result of political competition, and assert that the Sultan of Ternate and his palace guards were directly responsible for this clash. Tomagola maintains that, like the conflict in Malifut, the violence in Ternate also resulted from the Sultan’s failing attempt to assume dominance in the new province.44 Other commentators suggest that a rising swell of animosity had built up against the Sultan largely because of the aggressive behaviour of his traditional guards towards migrants. The International Crisis Group asserts that the violence against the Sultan and his traditional guards occurred because he had been labelled anti-Islamic for protecting Christians during the earlier November riot.45 Alhadar recounts that, after hearing that Christians had massacred Muslims in Tobelo, Muslims in Ternate attacked the Sultan’s guards, whom they accused of siding

12 Ethno-Religious Violence in Indonesia with Christians in the earlier riots.46 Muslims from Tidore, angered at the destruction of houses owned by Tidore migrants in Ternate, and fearful of a resurgent Ternate sultanate, also mobilized against the Ternate Sultan. According to this argument, the Sultan of Tidore was directly responsible for this mobilization that overthrew the Sultan of Ternate. This intra-Muslim violence clearly demonstrates the existence of several layers to the North Maluku conflict. Religion was certainly not the only divide along which violence was fought. Yet the responsibility and agency involved in the riots requires further examination. The almost universal blame placed on the losing party in this conflict, the Sultan of Ternate, and the total absolution of those who subsequently assumed power in the province as a result, suggest a critical reappraisal of the clashes is required. Was this conflict a direct outcome of the Sultan of Ternate’s support for Kaos and Christians and the way his traditional guards treated Makians, Tidores and other migrants? Or was it directly connected to the struggle for political power in the province? If the violence was a symptom of elite-level competition, to what extent was it a manifestation of the long-standing rivalry between the two sultanates? Contrary to most analyses, I conclude that several political rivals of the Sultan of Ternate exploited the emotions aroused by the wider conflict in the region in order to undermine the Sultan’s increasing political dominance. The chapter will also utilize the claims of Jack Snyder, Paul Brass and Steven Wilkinson that members of the elite often stimulate sectarian animosity for political purposes. Religious war – ethnic cleansing in the name of God In an era in which militant Islam has become such a focus of attention, the large Muslim militia formed in North Maluku in 2000, and its campaign against Christian villages on Halmahera, have received surprisingly little analysis. Most observers of the conflict agree the militia was formed in response to the killing of Muslims in Tobelo and elsewhere. Several analysts claim the militia was joined by members of the Laskar Jihad, the Java-based militia formed after the Tobelo violence. Tomagola asserts that a fourth surge of violence began in May 2000 ‘with the arrival of around 8,000 Lasykar Jihad from Ambon, South Celebes and Java’.47 The International Crisis Group claims that Laskar Jihad advisers ‘introduced a centralized command for the local jihad’.48 No analyses of the North Maluku conflict provide detailed information regarding the formation, constitution, funding and leadership of this militia, however. Was the militia assisted by large numbers of fighters from outside the province or was it primarily or exclusively local? Which national or regional radical Muslim organizations were involved in the militia, if any? For what reason did the militia’s two main operations in the areas of Malifut and Galela have such different outcomes, with a marked lack of success in the former and the successful destruction of several large Christian villages in the latter? Finally, the literature on North Maluku contains few explanations of why the militia disbanded in June 2000 even though Christians continued to control the large sub-districts of Tobelo and Kao.

Introduction 13 What were the primary motivations of the militia personnel? To what extent were the Pasukan Jihad and Christian militias in Kao and Galela motivated by religion? To what extent did sentiments of jihad motivate Muslim militia members (for example a desire to defend Muslims and/or spread the Islamic faith in North Maluku)? The chapter begins with a theoretical discussion of the capacity of religion to mobilize people for violence. It then demonstrates that the militia was almost entirely local in composition and goals and no external militias became strongly involved in the conflict. The fact that no external militia became involved in the conflict is explained by the overwhelming numerical dominance of Muslims in the region. The analysis will conclude that, while many other influences are perhaps more important than religion in provoking violence, religion cannot be discounted completely as a motivation for those involved, or as a means of facilitating collective action. Additional questions Several other general questions will be addressed throughout the book. In particular, a further question concerns why the conflict was possible. In a state long characterized as authoritarian in its response to security threats, why did national security forces not manage to control the violence? This question is particularly pertinent with regard to phases of the conflict during which small numbers of rioters faced large numbers of armed security personnel, and particularly when the latter possessed adequate intelligence that rioting was imminent. It is unclear whether these forces lacked the necessary capacity or political will to deal effectively with the riots. To what extent did national factors such as democratization and the military’s loss of influence nationally explain their lack of resolve? It also remains to be seen whether the security forces provoked, became complicit in, or exploited the violence in North Maluku as commentators have claimed they did elsewhere in Indonesia, and if so, for what reasons. A further question considered throughout this study is what is known as the ‘free-rider’ dilemma.49 Why did individuals and communities across North Maluku choose to risk their own death and the destruction of their property, especially when collective action by others was likely to provide benefits to them even without their participation? Did coercion, financial incentives or other factors play a role in their decisions to take part in the conflict? How important were ethnic and religious loyalties in mobilizing people to risk their lives?

Structure of the book The rest of the study examines in depth these peaks of violence across the region, some involving one-sided pogroms against minorities, others more evenly balanced clashes between two heavily armed militias. I discuss in detail the factors behind each phase, their interconnectedness and also the way each new event expanded the issues being fought over and drew in a greater range of participants. The book will explore the conditions in the region before the conflict and the

14 Ethno-Religious Violence in Indonesia causes of the initial violent incident. It will also illustrate how and why the violence rapidly evolved from a local border dispute into province-wide religious conflict. Each chapter critically evaluates prevailing assumptions about the phase of the conflict under discussion. Chapter 1 outlines the analytical approach taken in the study. It discusses two lines of analytical debate that are often implied in the conflict literature. The first involves the relative importance of elite leadership and mass-level factors in violent conflict, and the second concerns the relative role of rationality on one hand and emotion and identity on the other in motivating combatants. The chapter suggests that in sometimes presenting these phenomena in dichotomous terms, many current theoretical models do not adequately capture the complexity of internal conflicts. The chapter also discusses how, when analyzing an entire conflict involving a number of events and spanning a considerable time period and geographical space, the analyst must consider a great deal of temporal and geographical variation.

1

The study of violent communal conflict

This book seeks to account for the bloodshed and destruction in North Maluku from 1999 to 2000. It explores why violence broke out in a region which had seen decades of peaceful coexistence between ethnic and religious communities. It also seeks to explain the trajectory of the conflict: why it transformed from a local dispute into religious violence; how and why the initial incident sparked death and mayhem across an entire province; and, finally, why it ended as abruptly as it had begun. Within these primary foci are several secondary, yet crucial questions. The study examines what motivated those involved and how organized the violence was. It asks whether religion and religious identity played a major role in the violence. It seeks to explain how such devastation could happen in a state which for generations had harshly prevented large-scale domestic disturbances, as well as where responsibility lies for this tragedy. This chapter outlines the analytical and methodological approach taken to answering these questions. The analysis presented throughout this book is based on two main considerations – the need to recognize first the complex interaction of structural influences and human agency; and second the dynamism and geographical variation involved in violent conflict. A detailed account of a single case of large-scale violent conflict requires an inclusive approach – one that recognizes the mutually influential role of elite agency and socio-political conditions, and of instrumentalism, emotion and identity, and that pays close attention to changes over time and to local detail.

Lines of debate in conflict study Elite agency vs. mass phenomena A central debate in conflict study concerns the relative importance of elite instigation versus mass-level phenomena in the onset and trajectory of violence. Within this debate are several key questions relevant to this study: is violent conflict most often spontaneous or coordinated? How causally important is any provocation and organization that does occur before riots and other violent events? Why do ordinary community members follow their leaders into conflict? Time and again commentators have uncovered what appears to be intentional

16 Ethno-Religious Violence in Indonesia instigation of violent conflict by powerful individuals seeking to gain or retain power or economic advantage.1 V. P. Gagnon asserts that in the former Yugoslavia during the 1990s, a wide coalition of ‘conservatives in the Serbian party leadership, local and regional party elites … orthodox Marxist intellectuals … and parts of the nationalist army provoked a conflict along ethnic lines’.2 He concludes that ‘violent conflict along ethnic cleavages is provoked by elites in order to create a domestic political context where ethnicity is the only politically relevant identity.’3 John Mueller has argued that supposedly ‘ethnic’ war is more often the action of thugs employed by members of the elite to instigate conflict.4 When several theorists pointed to what they saw as an upsurge in New Wars since the end of the Cold War, a key component of this new type of conflict was the economic predations of elites. These elites do not enjoy wide societal support, unlike ‘old wars’ which were characterized by widespread support for the ideologies espoused by leaders.5 Some elite-based theories of violent conflict have strong intellectual foundation in the Resource Mobilization/Social Movements literature. Social Movements theorists claim that because discontent and grievance are almost always present in society, they cannot adequately explain mobilization.6 More important to the emergence and success of a social movement is an increase in available resources. Members of the elite who are able to organize money and labour efficiently, and frame the group’s goals, are crucial to the emergence and sustainability of a viable social movement.7 John McCarthy and Mayer Zald quote Turner and Killian, that ‘there is always enough discontent in any society to supply the grass-roots support for a movement if the movement is effectively organized.’8 Doug McAdam argues that this focus on the importance of the elite in the Social Movements school stemmed from a belief that politics, in America as elsewhere, is fundamentally oligarchic. The vast majority of people enjoy little influence or power, while a small group of the elite control political, social and economic life.9 For some analysts, therefore, elite agency can be considered the main cause of violence, with little explanatory weight resting with the conditions prevalent before violent conflict. Benjamin Valentino asserts that ‘the remaining permissive conditions necessary for mass killing have been relatively common across states, culture and time. Even when leaders with an interest in violence have found such conditions absent, it has been remarkably easy to create them.’10 In his thesis, small numbers carry out the violence and little societal support is necessary – all that is required is for the wider population not to physically oppose the killings.11 For Valentino, ‘the search for the causes of mass killing should begin with the capabilities, interests, ideas, and strategies of groups and individuals in positions of political and military power.’12 Conversely, many analysts of conflict eschew this focus on the behaviour of leaders to show the importance of mass-level phenomena. In this view, violence is seen as largely spontaneous, stemming from the strain built up by unjust, unequal or otherwise contentious structural conditions. This tension is eventually ignited into violence, often by small incidents. Some of the social structures commonly seen as causing, or at least facilitating, conflict include: economic and political inequality;13 a lack of networks that span the ethnic or religious divide;14 and a

The study of violent communal conflict 17 country’s reliance on extractable natural resources.15 Sudden changes in long-standing socio-economic and political structures are also frequently seen as important causes of conflict. Examples include: the feelings of insecurity associated with the collapse of national order;16 and changes in the relative socio-economic situations of two ethnic or religious communities.17 The motives and actions of the participants in conflict are largely determined by prevailing structural conditions. The Security Dilemma concept is a particularly influential structuralist explanation of collective violence.18 With origins in International Relations Realist theory, the Security Dilemma concept claims that in the absence of any overarching authority communities may become wary of the intentions of other groups. Accordingly, one or more of the groups may take measures to ensure their own security, thereby threatening the other group, which in turn takes similar measures. Despite the absence of any real intention to initiate violence, this insecurity spiral may none the less lead to conflict. Another prominent general explanation of conflict based on the presence of mass-level phenomena is the concept of Relative Deprivation. Associated most closely with Ted Robert Gurr’s Why Men Rebel, Relative Deprivation asserts that aggression results from frustration.19 Gurr argued that two main forms of deprivation are important causes of frustration – decremental relative deprivation (from declining capabilities) and aspirational relative deprivation (where aspirations increase but the capabilities for achieving these goals do not rise accordingly).20 The greater the intensity of deprivation, the greater the violence. Taking a more synthetic approach to the role of elites and mass-level phenomena, but still very much emphasizing the importance of the latter, is the work of Donald Horowitz. Horowitz writes that riots can be located anywhere along a spectrum of organization from highly coordinated to spontaneous.21 However, he generally gives causal precedence to mass-level sentiment. In his two classic texts of conflict study, Ethnic Groups in Conflict and The Deadly Ethnic Riot, Horowitz presents a common theme – that conflict stems ‘above all from the struggle for relative group worth’.22 Riots will invariably enjoy wide support and legitimacy within the community.23 ‘What the evidence shows is that most riots seem to be unorganized, partially organized and partially spontaneous, or organized by ephemeral leadership that springs up to respond to events as they happen, often suddenly. Most riots, in other words, consist of angry violence.’24 ‘Violence thus produces organization, perhaps more than vice versa … Most of the time, organization seems to take a back seat to passionate killing.’25 ‘The amalgam of purpose and brutality reflects the spontaneous quality of riot behaviour, which proceeds not in response to government orders but in response to the heat of the moment and the feelings of the participants.’26 Reflecting the organization–spontaneity divide within conflict study, some theorists have asserted that different cases of communal violence can be characterized as either elite-led or mass-led. Stuart Kaufmann concludes that mass-led conflict results when mass hostility and fear triggers spontaneous outbreaks of violence.27 In elite-led violence, elites ‘intentionally cause both mass hostility and

18 Ethno-Religious Violence in Indonesia a security dilemma’.28 David Keen also distinguishes between ‘top-down’ and ‘bottom-up’ conflicts.29 Underlying much of the literature on the role of elite instigation and mass-level conditions in causing violence is an assumption that elites are removed from the ‘irrational’ influences operating at the societal level. When elites organize violence, they do so for rational political or economic reasons, not because of rage, injustice or prejudice. Mass-level conflicts stem from less rational sources, however – frustration, hostility and fear. The motivations of those who organize and engage in violence represent a second major analytical question informing this study. Rationality vs. affect and identity When considering what factors are most likely to motivate participants to carry out violence, analysts often emphasize one of two types of motivation – the rational calculation of interests (utility maximization or instrumentalism) or emotional and/or identity-related factors. Proponents of rational choice sometimes recognize the role of emotion in conflict, and those espousing affective explanations allow a degree of instrumentalism in the violence. Yet most analysts give causal precedence to one or other form of motivation. A number of analysts have presented persuasive accounts of violence grounded in the theory of Rational Choice. Despite the spontaneous and irrational appearance of conflict, these commentators point to the rationality of those involved, suggesting that political, economic or other interests are concealed by the rhetoric of grievance and identity. While recognizing the role of ethnic solidarity and emotion, some conclude that these phenomena are ultimately the product of material competition. Susan Olzak asserts that competition increases as migration or other factors bring groups into contact in an environment of declining resources.30 In this situation, ethnic identity becomes more salient, events seemingly attached to ethnicity take on a far greater significance and spontaneous collective action along ethnic lines is promoted.31 When explaining why some communities seem to be targeted more than others in collective violence, even when other groups compete more strongly for economic goods, Olzak asserts that this paradox is due to different forms of competition between different groups. 32 A major question facing the proponents of rational choice is what is known as the ‘free-rider dilemma’. Even if one accepts that the decision to go to war is based on a calculation of one’s personal interests, why would individuals and communities choose to risk their own death and the destruction of their property when collective action by others will provide them with the same benefits even without their participation? James Fearon and David Laitin provide several explanations, strongly grounded in instrumentalism, of why actors sometimes engage in violent conflict when it is clearly of low utility maximization. First, they point to the fact that violent behaviour is rare – inter-ethnic cooperation is the norm.33 This stability is maintained by ‘inter-ethnic policing’ in which communities monitor the behaviour of their own members so as to avoid the catastrophe of violent conflict with another group. Any conflict that does occur is largely explained by information

The study of violent communal conflict 19 failures and commitment problems between two parties meaning they are unable to reach a negotiated settlement.34 As discussed above, several observers have concluded that, in many cases of conflict, elites have provoked tension and violence with clear goals. Elites are seen as provoking or organizing violence so as to undermine political or economic rivals, reach or retain political office or prevent fractionalization of their community during elections. Gagnon asserts that the ethnic violence in the former Yugoslavia during the 1990s was ‘a purposeful and rational strategy planned by those most threatened by changes to the structure of economic and political power’. Rational motives also appear to drive many ordinary participants in violence. Stathis Kalyvas claims that ‘a key motive (in civil war violence) is settling private scores.’35 For some, including Benjamin Valentino and Kalyvas, even the intense violence and atrocities common to civil war have strategic or instrumental roles. Kalyvas asserts that much of the violence in civil war is coercive, carried out by governments and insurgent groups to control populations and eliminate defection.36 Similarly, Valentino has concluded that ethnic cleansing and genocide are often last resorts designed by a small group of powerful leaders to deal with minorities seen as threats to national security.37 He provides the example of how Hitler and the Nazi leadership, believing Jews to be a threat to Germany and allied with Communism, only attempted their mass extermination after attempts to remove them from the country by emigration and deportation had failed.38 For other observers, however, the barbarism of violent conflict, as well as the high risk and low return involved in participation, indicate that more emotional and psychological influences are at work than rational utility maximization. Chaim Kaufman claims that Rational Choice faces a problem explaining ethnic conflict because it does not recognize the power of communal attachments.39 Many theories of conflict assert that violence results from the fear or anger caused by insecurity, inequality or injustice. The fear central to the Security Dilemma concept and the frustration associated with Relative Deprivation are two examples discussed above. Roger Petersen has identified four emotions which commonly lead to violence – fear, hatred, resentment and rage – each emerging in different situations and motivating attacks on different groups..40 For example, fear predicts that a threatening group will be attacked. Hatred and resentment predict a long-held enemy and a higher status group respectively will be targeted.41 Some analysts have claimed that questions of identity play the pre-eminent role in eliciting the emotional responses necessary for violence. In this view, members are driven to conflict by threats or insults to their ethnic group’s identity and will put themselves at risk in order to benefit the wider community. Ole Waever et al. asserted that threats to a community’s distinctiveness caused by government policies, immigration, or the actions of another cultural group, are often the primary cause of rising tension and conflict.42 For Stuart Kaufman, the source of conflict lies in the ‘myth symbol complex’ of a group’s identity. ‘A group mythology that justifies hostility is a precondition for violent ethnic conflict’ and ‘ethnic appeals

20 Ethno-Religious Violence in Indonesia are successful in producing extreme violence only if the group also fears that its existence is threatened.’43 A number of quantitative studies of civil war have recently taken a strong position in the instrumentalism–affect divide. In 2001, Paul Collier and Anke Hoeffler suggested that two contrasting explanations for civil war existed – the ‘political science explanation’, based on the presence of sufficiently acute grievances, and an economic theory explanation based on the opportunity to rebel.44 After carrying out econometric modelling on a large data set of civil wars they concluded that ‘objective grievances’ (measured by ethnic and linguistic diversity, political repression, political exclusion and inequality) had little bearing on the outbreak of civil war.45 Far more important were opportunities to rebel, particularly ‘economic characteristics’ (such as the availability of finance most commonly from primary export commodities, a high proportion of unemployed young men and slow economic growth).46 In Collier and Hoeffler’s terminology, greed is more important than grievance in causing civil war.

Synthesis The section above identified two questions central to the study of conflict – how important, comparatively, are elite leadership and mass-level factors in causing conflict?; and are those involved in violence motivated by instrumental calculation or emotional influences and identity factors? As discussed, to some extent two debates have crystallized around these questions in the conflict literature. While few theorists exclude the role of either elite instigation, group-level factors, instrumental calculation or emotion in violent conflict, many give strong causal precedence to one form of phenomenon.47 Yet, as anyone who has approached the inductive analysis of a single case of conflict will attest, theories based too strongly on the role of either structures or autonomous actors fail to capture the complexity of large-scale violence. Both conditions and human agency play some role in leading to violent conflict. Some violent events are clearly more organized than others. Yet giving overwhelming analytical weight to either elite leadership or mass-level factors leads to several deficiencies. Without detailing the human actions involved, intended and unintended, it is difficult to explain why one situation of, for instance, inequality or insecurity leads to violence while others do not. While influenced by the surrounding context, it is the motives and actions of individuals and groups that translate that situation into violent conflict. Perhaps more importantly, neglecting the human agency in the tragedy of violent conflict removes accountability from the actors most responsible. Human agency is necessary to translate structures into conflict. For example, political inequality may exist for decades without causing violent conflict until the exploitation of that inequality by certain actors stimulates a violent response. Yet denying the influence of ideational, political and economic structures on the actions of individuals and groups also precludes a complete understanding of a conflict. The provocations of elites, or the violent actions of a small number of militants or criminals, are

The study of violent communal conflict 21 unlikely to cause large-scale conflict in the absence of material and ideational structures of insecurity, injustice or prejudice. This is not to attempt to undermine those theories outlined above. All have identified social mechanisms that undoubtedly shed light on conflict processes in a range of cases. But the detailed analysis of any one conflict requires a more synthetic approach that recognizes the mutually constitutive nature of elite agency and structural conditions and instrumentalism and emotion.48 Members of the elite are as much a part of their societies and influenced by the same prejudices, loyalties and other ideational and material structures as their followers. There is no clear demarcation between society and the state. In Indonesia, district and sub-district government officials, as well as national security personnel, represent not only the state but also their own ethnic and religious community. In addition, individuals from various levels of authority have a major influence on conflicts, further confusing any neat demarcation between elites and society. In many conflicts, therefore, it is difficult to distinguish between leadership and mass-level influences in the genesis and trajectory of violence. Indeed, as Paul Brass, Donald Horowitz and others have demonstrated, many conflicts are best understood as a combination of orchestration and chaos.49 Organized mobilization often descends into a mélange of spontaneous acts, and, similarly, random acts of violence often become more coordinated as influential individuals take charge. Before and during violence, much takes place in this grey area between state and society and between leadership and impulse. By investigating this area, and the interaction between mass sentiment and elite leadership, a more complete understanding of events is possible. It also becomes possible to answer the question of why people follow their leaders into violence. Attempting to explain a conflict by reference to either rational interest or emotional outburst alone is also problematic. Conflicts often involve a striking complexity of issues. While Chaim Kaufman and others are correct in criticizing a focus on pure instrumentalism for missing the ties and emotions of identity, it must also be recognized that rational calculation does motivate some actions in conflict. Yet other actions do seem simply to be the venting of rage, and have a consummatory nature, being ends in themselves. Any group will contain different sub-groups with varying identities and motives. In most violent events, various sections of a crowd will have quite different motives for participating, some affective in nature, others more interest-based. Certain sub-groups will have a much greater attachment to the larger group identity than others. Some may be acting with reference to the past (i.e. taking revenge for a past affront) and others acting with a future goal in mind (i.e. to eradicate economic competitors). In many cases, some individuals and groups will have known of planning behind a riot or clash, while others will believe they are part of a spontaneous event. Finally, as discussed further below, as the riot or conflict goes on, new interests and emotions develop, and identities change and harden. Further, the motivations of two opposing groups, both of which must be taken into account when analyzing a conflict, are likely to involve different mixes of

22 Ethno-Religious Violence in Indonesia rationality and emotion. Conflict is fundamentally the result of interaction, and two opposing groups will almost certainly be driven by different motives. One indicator of the maelstrom of issues usually involved in conflict situations is that members of distinct communities and even different members of the same community often have different conceptions of the real causes of the violence. Some members of one community may claim that it was their economic marginalization that led to conflict, while other members blame rising radicalism. Meanwhile, members of the opposing community may point to entirely different causes, such as political rivalry. Yet the analysis of a conflict cannot stop at ascertaining which actions are instrumental and which affectual. Individuals approach a conflict situation with mutually reinforcing rational and ‘irrational’ motives. Case studies invariably demonstrate that actors in a conflict situation simultaneously face a range of economic, political, emotional and identity-related influences upon decision making. These influences do not exist independently of each other but constantly reiterate or intensify the importance of one other. For example, Anthony Regan writes that economic, political and identity-related factors were ‘mutually reinforcing’ in motivating Bougainvilleans in their separatist conflict with Papua New Guinea. While mining revenue was important in this struggle for independence, these economic interests cannot be separated from a political struggle stimulated by years of marginalization and a sense of separate identity among the people of Bougainville.50 Nor in many cases is it feasible to give causal precedence to either form of motivation. The direction of causation between the two phenomena is more multi-directional than many theorists propose. Identities and passions constantly shape interests and vice versa. It is therefore necessary to illustrate the ways in which social conditions shape the identities, interests and actions of actors, the manner in which their actions alter the surrounding structure and in turn how this changed structure changes the interests, identities and interactions of actors. As Horowitz writes, ‘an amalgam of apparently rational-purposive behaviour and irrational-brutal behaviour forms the leitmotiv of the ethnic riot.’51 Several theorists have presented accounts of violent conflict that synthesize elite agency and social structures. In a wide-ranging study on the connections between the transition from authoritarianism to democracy, Jack Snyder concludes that nationalist conflict is likely when elites feel threatened by democratization.52 In an attempt to survive the democratic transition, elites appeal to ethnic solidarity and mobilize followers through small networks and the state bureaucracy. However, he also suggests some prevailing conditions that facilitate this elite instigation. Exclusionary nationalism is likely to occur and succeed when economic development is low, when citizens lack the skills for political participation and when democratic institutions, such as political parties and a professional media, are weak but the bureaucracy remains strong.53 In addition to recognizing the mutually constitutive role of elites and mass level phenomena, it is also necessary to consider the interaction of emotional and rational influences on both elites and masses. Badredine Arfi presents a strong argument regarding the connection between the identities and interests of elites and prevailing structures in the onset of violent conflict. He asserts that it is the

The study of violent communal conflict 23 rapid construction of a belligerent communal identity by elites that causes violent conflict.54 Changes in social identities destabilize patterns of inter-ethnic relations. It is identities that determine whether an actor seeks to reach a negotiated settlement or risk conflict.55 Arfi argues that three social structures are important in the construction of a belligerent social identity. Salient historical memories play an important role in the definition of ethnic groups and the nature of inter-ethnic relations. The structure of ethnic cleavages, such as territorial distribution, reinforces identities. State institutional arrangements constrain actors and empower those with resources. While asserting that elites construct aggressive social identities, crucially Arfi argues that ‘the above social structures are not just tools in the hands of self serving elites but also constitute the agency of elites.’56 The ‘existence of these social structures enables and constrains such a role.’57 Only by recognizing the importance and interplay of both elite agency and societal factors, as well as emotion, identity and interest, can the analyst answer several central questions in the study of conflict: to what extent are those elites actually creating or merely following mass-level sentiment?; why did they come to believe that organizing violence was a profitable means of achieving their agendas?; and why do members of society uncritically accept the provocations of their leaders? This last question is particularly pertinent in those situations where involvement in clashes holds a high risk of substantial cost and little likelihood of reward. Anthony Lake has asserted that the question of why constituents follow their leaders into violent conflict is the most central question in conflict study today.58 Therefore this study proceeds with the assumption that conflict must be understood as the result of human agency, in many cases provocative and Machiavellian, which both shapes and in turn is influenced by a range of mutually exacerbating political, economic, emotional and identity-related factors. Conflict situations are influenced at various levels spanning agency and structure – individual, subcommunal, communal and inter-communal or structural.59 The individual level involves the various personal interests, identities, emotions, psychology and prejudices that contribute to violent outcomes. The sub-communal level involves those collectivities of actors such as paramilitary groups, militias, criminal networks, ethnic or religious organisations, political factions and powerful economic actors. These actors often pursue competing agendas and hold varying attitudes towards neighbouring communities as well as ideas about the legitimacy, efficacy or necessity of violence. Their actions invariably play a central role in rising tension and the outbreak of violence. The communal level of analysis concerns the impact of communal solidarity and community identity, as well as the threats to the community that cause or escalate conflict. The structural level involves those systems and patterns of political power, economic distribution and inter-communal relations (and changes in those structures) that influence a social outcome. It is necessary to examine the interplay between actions and patterns at all these levels. Rather than disaggregating and studying separately the emotional and material aspects of violence and the role of organization and spontaneity, as recommended by some theorists,60 I believe a full explanation of a conflict will be found in the interaction of these elements. What I am suggesting is an approach to the study of

24 Ethno-Religious Violence in Indonesia conflict that best explains the devastation involved in a protracted conflict such as that in North Maluku, as opposed to a single theory of conflict that synthesizes all the factors described above. This approach allows us to witness how instrumental and identity or affective motivations help set the stage for one another, and how elites and other important actors and mass-level structures shape each other. In providing a number of examples of these dynamics this approach holds wider relevance for the study of conflict.

Dynamism and geographical variation Accounting for this complexity becomes more difficult when dealing with a conflict that involved several different events across a wide geographical area and lasted a substantial period of time. By their very nature, conflicts are highly volatile. Analysts sometimes fall prey to adjudging prominent characteristics at the apex of a conflict, such as the economic exploitation of local populations or religious prejudice, to self-evidently indicate the ‘causes’ of that conflict. Often the analyst may be on steady ground in this judgement. Even conflicts that involve several different events in different areas may follow the same logic throughout their duration. Often, conflict appears to take on a self-perpetuating character as refugees flee to neighbouring areas with stories of barbarity that anger their ethnic kin. Subsequent riots are often a direct consequence of preceding violence, as revenge attacks are launched and rumours and misperceptions abound. As Stanley Tambiah puts it, ‘ethnic riots form a series, with antecedent riots influencing the unfolding of subsequent ones.’61 However, many communal conflicts do not follow the same logic throughout their course. Communal conflicts often transform over time in terms of the actors involved, the issues being fought over, the goals of the participants and the strategies and means used to achieve these goals. Conflicts sometimes evolve from clashes that have criminal origins into wider sectarian violence or vice versa. In some cases this transformation reflects the multitude of interests and issues that are present in the conflict situation. A particular issue that did not play a direct role in triggering conflict may come to the fore after the violence has started, sometimes superseding the initial contentious issues.62 New issues sometimes emerge during a conflict as a result of actors exploiting existing chaos to pursue economic or political agendas. Communal or ethno-nationalist conflicts sometimes evolve into criminal, rent-seeking operations. Several case studies of inter-religious violence in India demonstrate how members of the elite have portrayed relatively minor incidents as sectarian in order to obtain political advantage, thereby dramatically widening the scope of a conflict.63 The different riots and clashes that constitute a conflict may also differ in degree of spontaneity – one or several may be spontaneous eruptions of anger, while the other(s) may be more planned and organized. Issues central to violence may also vary depending on locality. Ethnic, religious or class cleavages at the local level may influence whether different areas descend into violence. These local factors, therefore, play an important role in determining

The study of violent communal conflict 25 the path of the conflict as a whole. As Stathis Kalyvas suggests, ‘actions “on the ground” often seem more related to local or private issues than to the war’s driving (or master) cleavage.’64 Kalyvas claims that, on close inspection, civil wars often reveal themselves as numerous local conflicts, with local rivalries becoming enmeshed in the wider dynamic. In turn, these local cleavages have ‘a substantial impact on … the content, direction, and intensity of violence’.65 Ken Young has demonstrated this in his study of the violence in 1965 in Kediri, Java, which appeared to be driven by the anti-communist sentiment that had swept across the country following the apparent Communist coup attempt in the capital. Following the dramatic events in Jakarta, members of the national Islamic organization, Nahdlatul Ulama, attacked members of the Indonesian Communist Party and their sympathizers in many areas of Indonesia. The Kediri violence appeared to be no exception. However, Young asserted that numerous other local dynamics were also at play, including tensions between migrants and local communities, between devout Muslims (santri) and Muslims more open to traditional practices (abangan), as well as grievances over land reform and class-based issues.66 As Kalyvas states ‘analysis of the dynamics of civil war (how and why people join or defect, how violence takes place, et cetera) is impossible in the absence of close attention to local dynamics.’67 To fully grasp the true nature of a conflict one must therefore identify and analyze the points of transition within it. For example, at what point did the violence become sectarian in nature, and, more importantly, was this transition intentional and if so what motivated those who were instrumental in it? Above all, it is necessary to differentiate between the issues and actors which are foremost in the initial stages of a conflict and those that only become salient later. Likewise, it is necessary to account for the differences as well as interconnections between the violence in different areas in a conflict zone. Comparative theories of conflict based on secondary sources and not on close qualitative examination are at risk of confusing these factors and thereby drawing erroneous conclusions about causation. The following study of the cause and trajectory of the North Maluku conflict therefore divides the analysis into the five phases mentioned in the introduction: initiation; escalation; dispersion; political exploitation; religious war. This approach goes some way to disentangling the many factors influencing a protracted conflict involving a series of events across a wide area. By dividing the analysis into these phases, one can separate those factors crucial to the outbreak of violence from those that only came to be central to the violence over time. In addition it better explains why this conflict escalated into widespread religious war. While these phases may not transfer directly to other conflicts, the general framework is likely to be broadly applicable elsewhere.

Researching violent conflict The considerations regarding the study of communal violence discussed above also determined the research methodology adopted in this study. I have focused solely on the case of North Maluku rather than attempting a comparative study

26 Ethno-Religious Violence in Indonesia with other conflicts in Indonesia or elsewhere. This approach enabled me to present a more detailed and thorough case study and better uncover the range of issues and the agency involved in the conflict. I make limited reference to other cases of conflict so as to avoid taking them out of their context, which might result in misleading comparisons and conclusions. I have attempted through qualitative research, primarily field interviews, to obtain insight into the perspectives of those involved in the events leading up to the violence and of the combatants themselves. I conducted interviews with members of almost all major ethnic groups and both religious communities involved in the violence.68 Interview respondents included ordinary villagers, townspeople, politicians, businesspeople, community and religious leaders, militia members and leaders, and military and police personnel. All interviews were carried out in the Indonesian language (Bahasa Indonesia). I carried out nine consecutive months of fieldwork in North Maluku. This research was focused on four sub-districts (Tobelo, Galela, Kao and Malifut) and one municipality (Ternate), and the capital of Central Halmahera District (Tidore). While violence took place in many areas of the province, these sites were chosen for their centrality and importance to the wider conflict. The Kao/Malifut area was the location of the initial outbreak of violence. Tobelo and Galela saw the most intense violence and had the largest Christian populations in the province. Ternate and Tidore were chosen because of their status as district (and in the case of Ternate, provincial) capitals. Ternate was also the main economic and political centre and most populous city in the province and most heterogeneous in terms of ethnic groups. Importantly, it was also because of events in Ternate and Tidore that the conflict developed from a local ethnic dispute into a province-wide religious conflict. In addition, I carried out several months of interviews and searches of secondary literature in other areas of Indonesia, including Manado, North Sulawesi, Bandung, Yogyakarta and Jakarta. Underlying this qualitative methodology is a focus on the importance of the social interaction of individuals, groups and communities. In this sense, I follow the Symbolic Interactionism school of Herbert Blumer.69 In its simplest form Symbolic Interactionism has three main premises.70 First, Symbolic Interactionism suggests that people act towards things based on the meanings those things hold for them. The second and third premises of Symbolic Interactionism suggest that the meaning of such things arises from, and subsequently changes as a result of, social interaction. Therefore, rather than focusing on the apparently obvious meaning of issues and objects central to the violence, I have attempted to ascertain the real import of these factors to the people involved. These different foci lead to quite different conclusions about conflict. For example, while the goldmine in Malifut was highly sought-after by both parties because of its material benefit, and therefore may appear to the outside observer to have been integral to the outbreak of violence, close examination of the motives of those involved indicated this material competition was less important than a range of other factors. Likewise, the ‘Bloody Sosol’ letter disseminated in Ternate appeared to be important in instigating the ensuing anti-Christian rioting. Yet

The study of violent communal conflict 27 interviews with those involved showed that this propaganda had little impact on Muslims in the city. I have used newspaper reports infrequently. Most local newspapers, such as the Ternate Post, carried very subjective reports of the violence and in many cases emotional and provocative accounts of events. National newspapers have been used occasionally, although often their accounts of events in North Maluku were based upon local newspaper reports or telephone interviews and in many cases reported what appears to be incorrect information. While newspaper reports do have some discursive impact on local communities during conflict, this influence was far smaller than the spread of rumours, movement of refugees, propaganda of elites and other more direct manifestations of the conflict situation. Several methodological difficulties come into play when researching violent conflict. The rhetoric of participants and members of the elite often masks more than it illuminates. Because of the sensitive nature of the topic, respondents are often reluctant to discuss their experiences. Most members of society are seeking to assign the conflict to the past, are fearful of reigniting inter-communal tension, suffering retribution or prosecution, and are often not inclined to talk about the conflict with strangers. People often portray events during the conflict in a manner that is favourable to themselves and to their ethnic or religious community. Carrying out research several years after the events in question also presents an aditional problem in that people have often finalized their version of events and in some cases now believe that version fervently. To circumvent such problems I have tried to develop relationships within all relevant ethnic and religious communities and assure people of my objectivity regarding the events. In addition, I have attempted to ‘triangulate’ sources, obtaining several different accounts of the same event from opposing sides in the conflict.

Conclusion This study provides a more detailed explanation of the tragic events in North Maluku than can be found elsewhere. It explores the conditions in the region before the conflict and the causes of the initial violent incident. It also illustrates how and why the violence rapidly evolved from a local border dispute into province-wide religious conflict. The study critically evaluates prevailing assumptions about the outbreak and trajectory of the violence. I argue that it is crucial to analyze this conflict at four different levels, individual, sub-communal, communal and structural. The study does not focus solely on the political, economic or identity-related aspects of the conflict but seeks to demonstrate the ways in which numerous agendas, concerns and relationships and, crucially, human agency interacted to shape outcomes. I have divided the conflict into phases to capture its temporal dynamism and geographical variation. My analysis of each phase sheds light on the trajectory of the conflict as well as salient issues and agency. The examination of successive phases of the conflict, each with distinct characteristics, allows the identification and use of the most relevant and helpful theoretical concepts from the wider study of violence.

28 Ethno-Religious Violence in Indonesia Chapter 2 examines the background to the conflict, including the history and social structure of North Maluku since the fifteenth century and the major changes that occurred in Indonesia and in the immediate region in the years immediately prior to the conflict. The chapter discusses the history of inter-religious relations in Indonesia and North Maluku more specifically, illustrating how tension had risen over the previous two decades between Muslims and Christians. It discusses the factors behind this development and covers the acceleration of inter-religious violence across Indonesia, including the terrible carnage in Ambon which began in January 1999.

2

North Maluku in context

Introduction There are few more breathtaking sights than North Maluku from the air as your plane begins its descent towards Ternate. A chain of small volcanic islands rises sharply out of the water. The city of Ternate clings to the side of one of these volcanoes, massive Halmahera lies to the east and Tidore broods opposite. Once the site of intense international competition over the lucrative spice trade, its waters alive with European gunships, this place then slumbered for centuries in relative anonymity. Its ancient sultanates, once fierce rivals, fell redundant. The remote islands were relatively untouched by the tumultuous events of Indonesian politics in the twentieth century. But just before the turn of the millennium North Maluku once more rose to prominence, yet again a scene of political competition and bloody rivalry. In 1999, Indonesia was in a state of flux. The Asian financial crisis forced millions into unemployment and poverty, triggering nation-wide protests. In turn, the protests led to widespread demands for political reform, culminating in President Suharto’s resignation in May 1998. His replacement, B. J. Habibie, initiated a process of democratization as well as decentralization of authority to the regions. The Indonesian armed forces, beset by factionalism and poor operational capacity, were not untouched by this turmoil, facing growing criticism and pressure to withdraw from economic and political life. While these national developments affected all regions of Indonesia, North Maluku faced additional local changes. In early 1999, the central government in Jakarta announced that two remote districts in northern Maluku Province were to form a new and separate province. This stimulated competition among the region’s elite for political power. Religious violence in Ambon City also influenced North Maluku society. Many public servants, students and workers returned from Ambon with horrific stories, and the entire community witnessed on television religious violence on a scale unprecedented in Indonesia.

Overview of North Maluku North Maluku comprises dozens of beautiful and heavily forested volcanic islands in Indonesia’s east. The region lies between Sulawesi to the west, Papua (formerly

30 Ethno-Religious Violence in Indonesia

Map 2 North Maluku

Irian Jaya) to the east and the southern Philippines to the north (see Map 1). For several centuries, the area was the only source of a highly valued commodity on the world market, the clove. Because of the wealth and influence afforded by trade in this spice, several regional powers emerged in North Maluku, most notably Ternate and Tidore, which exercised influence as far as Sulawesi, Papua and the southern Philippines. However, the ‘Spice Islands’ as they became known, like most areas possessing valuable resources, also became the site of competition and turmoil, particularly after the arrival of European powers. North Maluku’s role in the spice trade began to decline in the seventeenth century, after which the region began a period of decline. The past few decades have nevertheless seen a moderate upsurge in economic development. Until October 1999, North Maluku was part of Maluku Province, governed from the provincial capital, Ambon City. The region was divided into two districts

North Maluku in context 31 – North Maluku District and Central Halmahera District.1 The boundaries of the two districts coincided approximately with the traditional spheres of influence of the two most prominent sultanates – Ternate and Tidore. The main villages on the small volcanic islands of Ternate and Tidore were also the district capitals: Ternate City on Ternate, and the quiet village of Soasio on the eastern side of Tidore. North Maluku District comprised Ternate, the northern half of Halmahera Island and a number of islands, including Makian and Morotai. Central Halmahera District comprised Tidore, Bacan and the southern half of Halmahera. In October, the region became a province in its own right. Ternate City was declared the temporary provincial capital, while a small village in Central Halmahera District was to become the permanent capital. The North Maluku region is populated by a large number of related ethnic groups, each closely associated with a particular area. According to local ethnologist Yusuf Abdurrahman, there are 30 ethnic communities with distinct languages.2 The largest ethnic communities, according to official statistics, are the Makians, Ternates and Tidores although other communities, such as the Tobelos, form sizeable minorities. No one ethnic community numerically dominates the region.3 The region is also the long-term home to a minority of Arabs, Chinese and other migrants, as well as small numbers of Javanese who were settled in Halmahera under the central government’s ‘transmigration’ programme.4 The region is demographically dominated by Muslims. Between 75 and 80 per cent of the North Maluku population identify themselves as Muslim, the remaining population (20–25 per cent) predominantly as Protestant Christian.5 Muslims constitute the vast majority on most islands and sub-districts. This religious constitution is reflected in the district parliaments, where Muslims greatly outnumber Christians. Christianity is strongest in northern Halmahera, particularly in Kao and Tobelo Sub-Districts, where Protestant Christians constitute large majorities, but Christians also comprise sizeable minorities in Galela, Jailolo, Ibu, Bacan and several other sub-districts. In Ternate, approximately 10 per cent of the population was Christian before the violence in 1999, mostly migrants including Chinese Indonesians, Ambonese, Minahasans and other ethnicities from elsewhere in Indonesia. Chinese Christians are highly visible in the Ternate economy, so to some extent religious and class differences coincided.6 This chapter discusses the above context to the outbreak of violence in North Maluku in 1999. It discusses relevant aspects of the area’s early modern history – the struggle between the main sultanates for power in the region and the development of inter-religious relations. It then provides an overview of North Maluku during the New Order period from 1966 until 1998, followed by the dramatic changes that occurred in Indonesia from 1997 until 1999. Finally, the chapter explores the inter-communal violence in Ambon and the quest by the Ternate elite to create the nation’s newest province.

32 Ethno-Religious Violence in Indonesia

The spice islands Rival sultanates North Maluku’s distinctive and, in many ways, troubled history is to a great extent due to the region’s wealth of valuable spices.7 Ternate, Tidore and Bacan, the small volcanic islands along the western edge of the region, were home to the clove, renowned among Chinese and European societies for its medicinal properties.8 Several other valuable spices, such as nutmeg and mace, grew on these and other islands of North Maluku. By the fourteenth century, these spices were traded along networks passing from North Maluku, through northern Java, Malacca, Gujarat, the Persian Gulf and the Mediterranean. Based largely on this spice trade, by the fourteenth century four powerful kingdoms had emerged in North Maluku – Ternate, Tidore, Bacan and Jailolo. After a period of competition between these four kingdoms, two eventually became predominant.9 By the sixteenth century, the rulers of Ternate and Tidore had established influence over relatively large areas. Ternate’s authority extended southward to Ambon and more tenuously to North Sulawesi.10 Andaya writes that in the seventeenth century Ternate exercised influence over and enjoyed a close and strongly hierarchical relationship with the alifuru or indigenous and animist inhabitants of northern Halmahera.11 By the seventeenth century the populations of Kao, Tobelo, Galela and other areas of Halmahera were forced to pay tribute to the sultanate in the form of crops, pearls and labour, and provided personnel for the sultanate’s military campaigns.12 Kaos, Tobelos and other Halmaheran peoples carried out military service for the Sultan, in some cases fighting against the Dutch colonial authorities (1679–81) and in others enlisted by the Dutch to fight against other communities in the region, for example during the Pattimura rebellion in Ambon. The Sultanate of Tidore exercised power over the southern half of Halmahera, Bacan and as far as the Raja Ampat area on the western edge of Papua, where local inhabitants paid tribute to the Sultan of Tidore and provided personnel for military campaigns. The two kingdoms competed for authority over the region and occasionally took up arms against each other. Competition was focused on the island of Makian to their south,13 the most populous island in the region and a productive source of cloves. Makian was for a long time powerful in its own right, considered a fourth North Malukan kingdom as well as the origin of the Bacan kingdom’s royal line. The island gradually lost power to its two rivals to the north and many of Makian’s inhabitants moved to neighbouring islands after periodic volcanic eruptions. 14 Competition between Ternate and Tidore became the accepted order for North Malukans. Andaya noted that ‘As long as the two pillars remained, all was well with Maluku.’15 Ternate was the more powerful of the two kingdoms, partly because of that sultanate’s closer relationship with the Europeans and the greater revenue it received from international trade. Yet its predominance did not go unchallenged. As Haire writes, ‘Jailolo and Tidore tended to be the sources of opposition and even revolt.’16 For many North Malukans, the collapse of Jailolo

North Maluku in context 33 and Bacan, and Ternate’s growing dominance over Tidore, had disrupted the correct social order of the region.17 The colonial powers utilized and exacerbated these regional rivalries. Almost all European powers who ventured into North Maluku made alliances with one of the two main sultanates. In turn, the local kingdoms exploited the European presence in their own power struggle, fighting several battles during the sixteenth and seventeenth centuries.18 From the latter part of the Dutch era, this competition between Ternate and Tidore fell dormant. Both sultans declared loyalty to the Dutch administration, which paid them each an annual stipend.19 Like Bacan and Jailolo, the sultanate of Tidore eventually fell vacant during the twentieth century. It was co-opted by the central government, although, as discussed below, it retained a great deal of cultural influence and formal political power. As will be seen, this rivalry between Ternate and Tidore re-emerged to some extent in 1999, both during efforts to establish North Maluku as a separate province and during the communal conflict that engulfed the region. Other social cleavages formed during the early modern era also came to the fore in 1999. The following section discusses the introduction of Islam and Christianity to North Maluku and the subsequent events that shaped inter-religious relations.

Islam and Christianity It is thought that Javanese traders introduced Islam to Ternate and the North Maluku region sometime in the late fifteenth century.20 After the rulers of Ternate, Tidore, Bacan and Jailolo converted to Islam, the religion spread rapidly throughout the region. Most historians agree that, initially, these rulers adopted Islam to facilitate trade with the Muslims who dominated the spice trading networks. Villagers on Ternate, Halmahera and elsewhere who adopted Islam did so in a syncretic way, melding Islamic practices with traditional religions. However, by the time Europeans arrived in the area, Islam was firmly entrenched. In 1486, the ruler of Ternate, Djainal Abidin, assumed the Islamic title of Sultan, a title soon adopted by other regional leaders.21 Christianity was first introduced to North Maluku by Catholic priests connected with the Portuguese trading mission on Ternate. The missionaries gained their first converts in the village of Mamuya in north Halmahera in 1533.22 The Sultan of Jailolo and the Sultan of Bacan also converted from Islam to Christianity, largely to gain Portuguese protection from Ternate and Tidore.23 However, these Christian kingdoms were quickly destroyed by their more powerful Islamic rivals. The sixteenth century was a period of religious turmoil as communities converted back and forth between religions depending on political and economic exigencies. According to Andaya, most North Malukans did not see a fundamental contradiction between the two religions, often using the practices and symbols of both, along with traditional beliefs, to accrue additional spiritual advantages.24 Yet relations between the two religious communities were not always peaceful.25 On numerous occasions, the Portuguese brutally punished North Maluku’s Islamic rulers for perceived insubordination, in several cases kidnapping or killing sultans.

34 Ethno-Religious Violence in Indonesia This tension reached a tipping point during the reign of Sultan Hairun (or Khairun) of Ternate in the sixteenth century. Hairun was initially tolerant of Portuguese Christian evangelism in North Maluku but, after his arrest at the hands of the Europeans, he became more protective of Islam and opposed to the Portuguese. 26 In response, the Portuguese arranged his assassination. Hairun’s son, Baabullah, then led a failed military campaign against the Portuguese in Ambon, before turning his attention to Portuguese and Malukan Christians in north Halmahera, where his forces destroyed several settlements and killed large numbers of both communities.27 Christianity subsequently began to lose its following and largely disappeared from North Maluku for two-and-a-half centuries. James Haire concludes that this decline had several sources, including the superficiality of earlier Christian conversion and the killing of Sultan Hairun by the Portuguese.28 Islam remained strong in the region, although existing alongside traditional spirituality. Christianity was reintroduced to Halmahera in 1864 when the Protestant Dutch Reformed Church established a mission at the village of Duma in inland Galela.29 Opposition from Muslim communities and from the Sultan of Ternate, who ordered the arrest of Muslims who converted to Christianity, meant missionaries proselytized only among animist communities in north Halmahera.30 During this period, most pastors in north Halmahera came from Ambon, causing resentment among local Christians. After World War II, Halmaherans established a local Protestant Church – the Evangelical Church on Halmahera or GMIH (Gereja Masehi Injili Halmahera). When, partly to undermine support for communism, the new government of President Suharto declared that all Indonesians must join one of five officially recognized religious traditions (Islam, Protestantism, Catholicism, Hinduism and Buddhism), the number of Christians affiliated with GMIH jumped from 60,000 in 1965 to 106,630 in 1979.31 The majority were animist converts. Historically, both Islam and Christianity in North Maluku have been suffused with traditional spiritual practices, beliefs and ceremonies. In north Halmahera in the early twentieth century, Dutch missionaries considered local traditions (adat) useful for the introduction of Christianity.32 However, religious identity has always been very important to both Muslims and Christians. North Maluku Muslims retained a strong sense of their Islamic identity even after the decline in political influence of the sultans, traditionally the central pillars of the Islamic community. For most of the twentieth century, Muslims and Christians in the region lived in relative harmony. Events of the nineteenth and twentieth centuries affected both communities equally and in some cases drew them closer together. The Dutch authorities did not favour North Maluku’s Christians over Muslims as they did in Ambon.33 In some cases, Christians and Muslims cooperated in fighting against the Dutch. During World War II, occupying Japanese soldiers treated members of both communities with equal brutality.34 Both Muslims and Christians joined the 1958–60 Permesta rebellion, seeking greater autonomy for North Maluku.35 Although the rebellion was soon quelled, according to local historian Herry Nachrawy, this cooperation strengthened inter-religious ties. 36

North Maluku in context 35 The southern part of Maluku Province saw one of Indonesia’s most serious regional rebellions in the 1950s. In 1950 a group of predominantly Christian Ambonese declared their own state – the Republic of South Maluku or RMS (Republik Maluku Selatan).37 This separatist movement gained control of the island of Ambon (see Map 1) from April to December 1950, when it was overthrown by a massive attack by Indonesian forces, resulting in numerous casualties.38 As Richard Chauvel wrote, ‘The RMS and its suppression was perhaps the worst possible way for Ambon to be incorporated into independent Indonesia. Ambonese society emerged deeply divided.’39 North Maluku remained detached from the RMS rebellion.40 Clashes between Indonesian forces and the RMS were confined to central Maluku. According to local historian Herry Nachrawy, RMS leader Soumokil invited the Sultan of Ternate to join the separatist movement.41 The sultan declined because the apparent Christian character of the movement was at odds with North Maluku’s predominantly Islamic identity.42 Anti-Jakarta sentiment had never taken hold in North Maluku as it had among some sections of Ambonese society. Haire writes that, during the war of independence, a wave of Indonesian nationalism and anti-Dutch sentiment swept Halmahera, where people thought the Ambonese considered themselves ‘half-European’.43 While the RMS rebellion had little direct impact on North Maluku in the 1950s, the memory of it nevertheless lingered and re-emerged to affect inter-religious relations in the late 1990s. In the second half of the twentieth century, many Muslims in North Maluku continued to believe that resident Ambonese Christians supported RMS goals.44 Claims of RMS involvement became common during the 1999 conflict and were used by some Muslim leaders as a pretext to attack Christian villages on Halmahera. Some of Ternate’s Muslim political and economic elites, who had tried to separate North Maluku from Ambon during the previous decades, also wanted to be free of control by a perceived Christian elite based in Ambon. The other major rebellion of Indonesia’s first decade of independence, Darul Islam, had a minor impact on inter-religious relations in North Maluku. The movement began as a series of rebellions in Java, Aceh, Southeast Sulawesi and Kalimantan involving Muslim units that had fought against the Dutch from 1945 to 1949. By 1953, these units adopted the common aspiration of forming an Islamic state, the Islamic State of Indonesia or NII (Negara Islam Indonesia).45 According to local Christians, a handful of Muslims established a North Maluku unit of the movement, which included several people from Southeast Sulawesi.46 The group once reportedly attempted to attack the church in Dokulamo village in Galela but was repelled by police. Christians in Tobelo believe that stories of these events continued to stimulate fervour among young Muslims in Galela and Tobelo in the 1990s, particularly among some of the children and grandchildren of the men involved in the 1950s.47

36 Ethno-Religious Violence in Indonesia

Suharto’s New Order 1966–98 Economy During the first two decades of independence, North Maluku, as with most peripheral areas in Indonesia, received minimal developmental assistance.48 Little effort was made to develop Maluku as a whole, and, located a great distance from the capital Ambon, North Maluku fared worse than the rest of the province. Schools, roads and most other infrastructure deteriorated during this period and the region produced little in the way of exportable goods. The community was required to import most commodities from Java and elsewhere.49 President Suharto’s New Order brought increased development to the region, including improved roads, communications, ports and warehouses as well as a vast increase in the number of educational institutions. Government infrastructure also improved markedly under the New Order, with the construction of large district and sub-district government buildings. The region became more integrated with the rest of Indonesia with the opening of daily flights between Ternate and North and South Sulawesi and Ambon, and the improvement of shipping services from both Ternate and Tobelo. From a net importer of goods during the ‘Old Order’, Maluku became in the 1980s, along with Irian Jaya (Papua), one of only two net exporters among the eastern Indonesian provinces.50 As a result of the abundance of local fish and agricultural products, by the mid-1980s poverty was low. 51 North Maluku’s economy is largely driven by the exploitation of the region’s abundant timber resources. About three-quarters of the region is forested.52 Logs are processed at the large Barito Pacific plywood factory at Sidangoli in Jailolo Sub-District, which also serves as the main transit port between Halmahera and the rest of the region. North Maluku is also rich in mineral resources, including nickel deposits on Gebe Island and Halmahera. During the 1980s, Central Halmahera district’s GDP was twice the Maluku provincial average, growing 10.7 per cent annually from 1975 to 1983.53 In June 1999, PT Nusa Halmahera Mineral (NHM), a joint venture between the Australian company Newcrest Mining Ltd and the Indonesian company PT Aneka Tambang, commenced an open-cut gold-mining operation at Gosowong, about 40 km to the south-west of Malifut in northern Halmahera.54 With 350 local employees, NHM was the largest single employer in the immediate area. The fishing industry is central to the North Maluku economy. Agricultural products provide a small income for many farmers and the bulk of the subsistence needs of most villages. Copra production (dried coconut flesh for the production of oils) is particularly important. In 1991, the Indonesian company PT Global Agronusa Indonesia (GAI), in a joint venture with prominent American food corporation Del Monte, established a 2,000 ha banana plantation in the north of Galela Sub-District employing approximately 3,000 people. The socio-political situation During the New Order period, North Maluku experienced sustained political stability. The community largely accepted the provincial government’s appointment

North Maluku in context 37 of district and sub-district government heads. Yet, as will be seen, the appointment of Ambonese Christians as sub-district heads in Tobelo contributed to inter-religious tension in 1999. During the twentieth century Ternate City remained the administrative, cultural and economic centre of North Maluku and is still the main centre for processing and exporting wood, fish and other products. A large service sector has developed in the city catering to the district government and the forestry, mining and other industries. Employment in these industries, as well as the educational facilities in Ternate, has attracted large numbers of migrants, predominantly Muslim, from neighbouring islands such as Tidore, Makian and Sanana.55 Migrants from Sulawesi have also settled in the city, engaged largely in low-skilled occupations. Most migrants have settled in the southern areas of the city, while the indigenous Ternate ethnic community lives predominantly in the north of the city around the Sultan of Ternate’s Palace or Kedaton (see Map 4). In the decade preceding the 1999 conflict, three ethnic communities in North Maluku had assumed political and social influence greater than their relative size – the Makians, Tidores and Sananas. They were disproportionately represented in the parliaments and bureaucracies of both districts. The Makians in particular enjoyed substantial political representation and social influence.56 Over preceding decades, Makians had migrated to Ternate for its employment and educational opportunities, as well as to escape the harsh environment and volcanic eruptions on their home island. According to one elderly Christian man, Makians who attended school in Ternate in the 1960s and 1970s were considered backward because they were poorly educated and unfamiliar with local culture.57 But by the 1990s, Makians were the most educated and successful ethnic group in North Maluku, heavily represented in local parliaments.58 Makian representation in the bureaucracy was even greater. Many were employed in local government departments, and in 1999 a Makian, Thaib Armain, became the first head of the bureaucracy for North Maluku Province.59 Makians were also prominent in other sectors of society. In 1999, a Makian, Yusuf Abdurrahman, was both Rector of Khairun University and chairman of the Indonesian Council of Islamic Scholars for North Maluku (MUI).60 He resided on the campus and had substantial influence in the university and in Ternate. A high proportion of university lecturers and students were also Makians. The success of Makians and Tidores was in stark contrast to the other large ethnic group, the Ternate, who were less likely to undertake tertiary education and therefore to assume prominent positions in the bureaucracy or to enter local politics. The current Sultan of Ternate, Mudaffar Syah, no longer commands the political influence the sultanate enjoyed during the pre and early colonial era. Indonesian governments, particularly during the New Order, have reduced the sultan’s role to that of traditional leader. The Suharto government co-opted the sultan, who became a Golkar party representative in the local district parliament in 1971. However, it is important not to underestimate his influence in North Maluku. At the time of the conflict he continued to exercise great cultural influence over several ethnic communities in North Maluku District, particularly the Ternates but

38 Ethno-Religious Violence in Indonesia also the Kaos and Christian Tobelos. This was largely a consequence of the sultanate’s historical influence over these regions, and the fact that many Christians considered the sultan to be their traditional protector. The sultan still maintains a palace guard which he has used on occasion to intimidate political rivals.61 In 1999, Sultan Mudaffar Syah was the Chairman of the North Maluku District Parliament. As discussed, the other three historical sultanates had no such influence. All fell vacant before or during the New Order period, although as will be discussed below, the sultanate of Tidore was revitalized in 1999. Religion Religious communities coexisted in North Maluku without major confrontation before 1999. Inter-religious marriages were rare but did occur, particularly in areas where ethnic ties cut across religious differences, as in Tobelo.62 Fights had broken out intermittently over the previous decade in Ternate and in Tobelo and elsewhere on Halmahera, often between Muslims and Christians, but most were not over religion, but rather over girlfriends, insults and similar issues. All older North Malukans described relations between the two communities as good prior to 1999. Respondents from both religious communities said inter-religious relations had deteriorated in the decade prior to the conflict. Tobelo Christians suggested that traditionally close ties with local Muslims began to deteriorate following the in-migration of Muslims from Makian and Tidore who held more conservative attitudes toward Islam and inter-religious relations. Christian Kiem asserts that conservative ‘modernist’ Islam spread in some areas of North Maluku, particularly Ternate, after the establishment of Islamic schools and foundations associated with the national Islamic organization, Muhammadiyah. He claims young Muslims increasingly opposed a perceived westernization of Indonesian society and the continued participation by Muslims in pre-Islamic practices associated with local traditional culture.63 In Ternate, this trend was strongest among migrants from Tidore, Halmahera, Sulawesi and elsewhere. Their criticism of traditional practices generated tension with the Sultan of Ternate’s supporters, who most commonly engaged in these practices.64 But none of these tensions gave forewarning of the large-scale religious violence that would soon erupt in the region.

A nation in flux 1999 After several decades of political and economic stability, Indonesia entered a period of profound crisis in 1997. The Asian economic crisis sent the national economy into freefall, in turn stimulating large-scale political and social change, commonly known as reformasi (reform). While the crisis did not have significant economic repercussions in North Maluku, the ensuing political change certainly did. This section will briefly cover the dramatic changes that occurred in Indonesia from 1997 until 1999.

North Maluku in context 39 Financial crisis 1997 For three decades Indonesia was seen as an economic ‘tiger’. An oil and gas boom from the late 1960s, high levels of foreign direct investment, the successful development of the manufacturing sector and increased rice production all contributed to the country’s impressive growth. Poverty decreased remarkably, from 40 per cent in 1976 to 15.1 per cent in 1990 and 11.3 per cent in 1996.65 All provinces experienced improvements in most development indicators.66 However, nepotism and corruption undermined governance and economic efficiency. Capital was also highly mobile.67 Therefore when the Thai, Malaysian and Philippines currencies collapsed, international and domestic investors withdrew capital from Indonesia, prompting a run on the Indonesian currency. The rupiah decreased in value from 2,200 to the US dollar in July 1997 to 6,000 in December 1997, and fell as low as 17,000 in January 1998. In 1997, a long-running drought led to a 4 per cent drop in rice production, exacerbating price rises associated with the lower value of the rupiah. Unemployment increased, and the subsequent exodus of people back to rural areas seeking employment depressed agricultural wages further. In addition, from 1997 to 1998 annual inflation rose from 6 per cent to 78 per cent.68 As a result, poverty levels peaked at 27 per cent in 1999. Riots over the removal of government subsidies and the price of fuel, food and other necessities broke out across the country. Democratization 1998 The economic crisis triggered massive protest and pressure for governance reform from a wide range of civil society actors. Students led mass protests on the streets of all major cities and, along with leaders of opposition parties, academics and other leaders, called for President Suharto’s resignation and the withdrawal of the military or TNI (Tentara Nasional Indonesia)69 from political and civilian life. Facing overwhelming opposition and declining support, and with his health failing, Suharto resigned on 21 May 1998, and was succeeded by his vice-president, B. J. Habibie. After assuming the presidency, Habibie scheduled national elections for June 1999. The government also removed the restriction on the number of political parties allowed to contest elections, after which dozens of new parties were formed. Competition for political power suddenly assumed an intensity unknown for several decades. Opposition groups were presented with the political freedom to criticize the old regime, and local communities with the ability to lobby for changes, such as the recognition of indigenous rights. After centuries of almost uninterrupted authoritarian rule, Indonesia was on the brink of democracy. 70 Decentralization Democratization was accompanied by political and financial decentralization. This process was an attempt to address regional grievances, particularly in places that had experienced separatist rebellions, such as Aceh and Papua (Irian Jaya),

40 Ethno-Religious Violence in Indonesia and to increase public participation in politics, government accountability and to provide better health, infrastructure and other social services.71 Decentralization was based on two laws signed by the president in May 1999 – Law 22/1999 on Regional Government and Law 25/1999 on Fiscal Balance between the Centre and the Regions. Law 22 devolved to the district level authority over all functions of government except defence and security, international relations, justice, monetary and fiscal affairs and religion. The law also gave local parliaments the power to elect and dismiss their own district heads and provincial governors. Law 25 established a new financial relationship between the centre and the regions. The law left expenditure largely to the discretion of local governments, and allowed them to collect taxation as long as there was no duplication of the national tax regime. More importantly for many resource-rich districts, a large percentage of revenue raised from local resources was to be returned to the region of extraction, including the ‘producing’ district. A regulation accompanying Law 25 stated that 32 per cent of revenue from mining operations had to be returned to the producing district.72 The legislation also allowed for the creation of new sub-districts, districts and provinces through the process of pemekaran. Through pemekaran, the number of districts in Indonesia rose from under 300 in 1999 to over 400 by 2001. The central government also created several new provinces, including North Maluku, as discussed in more detail below. Hence by mid-1999 Indonesia’s regions were faced with a range of dramatic changes in political, social and economic life. The changes associated with reformasi, as well as providing the foundations for far greater political participation by civil society, also provided economic and political opportunities for powerful individuals. The security sector During the New Order, the TNI played a central role in the social, economic and political life of Indonesia. The institution enjoyed considerable political representation, holding 100 out of 500 seats in the national parliament or DPR (Dewan Perwakilan Rakyat).73 The military also maintained substantial influence in the regions, through the appointment of military officers as governors and district heads and through representation in provincial and district parliaments. In the late 1990s, former military officers occupied at least 50 per cent of gubernatorial posts and 40 per cent of district positions. In addition, two-thirds of the army was stationed throughout the country in units structured in parallel with the civilian administration, from village to regional levels, giving it control over numerous aspects of economic and political life throughout Indonesia.74 However, both the military and police were handicapped financially. For the entire post-independence period, the military raised much of its operating expenses independently of the government. In recent times the central government has provided only 30 per cent of the funding required by the military.75 Military and police personnel were insufficiently paid to meet accommodation, schooling, health and other costs. 76 In order to make up this shortfall, the armed forces

North Maluku in context 41 developed substantial business interests.77 Following the onset of the economic crisis in 1997, however, many of these business interests ceased to be profitable. It is estimated that the purchasing power of the military decreased by 30 per cent from mid-1997 to mid-1998.78 The military has allegedly compensated itself for its budgetary shortfall through involvement in illegal activities such as smuggling, extraction of protection money, extortion and illegal mining and forestry.79 By 1999, this lack of funding had weakened the capacity and motivation of the security forces to respond effectively to serious internal security disturbances. This was particularly the case in regions located far from Jakarta where budgets were stretched most thin.80 Many regional military and police units lacked the transport, communications and equipment necessary to respond effectively to serious conflict. Personnel numbers were also low, widely dispersed in small units of only several men.81 Poor living conditions and meagre salaries diminished morale and reduced soldiers’ willingness to act effectively in dangerous situations. Corruption was rife; moreover police inaction (after receiving bribes) sometimes helped to exacerbate violence because communities felt the need to take the law into their own hands.82 Lower-ranking military and police were poorly trained, and both the military and police possessed low levels of field intelligence, particularly in remote regions. Military involvement in local businesses also impaired the institutional command structure and institutional coherence, creating the opportunity for a divergence of interests between regional and local commanders and personnel on the one hand, and the central military hierarchy on the other. 83 Factionalism in the upper ranks of the military also undermined the TNI’s ability to respond to internal disturbances. While some leading officers called for and accepted the military’s reduced political and social influence, others resisted these changes. As the likelihood of investigations into alleged human rights abuses increased, a number of senior officers openly challenged the reform process. This factionalism further weakened the chain of command between field commanders and their superiors. There are also strong indications that some military officers allowed or even instigated violence in order to derail the reform process, particularly after the reform-minded Abdurrahman Wahid was elected president. In the late 1990s, the Indonesian military also faced declining popular legitimacy.84 As human rights abuse allegations heightened, ordinary officers and soldiers became less willing to respond to civilian protests and other unrest. At the same time, TNI’s apparent inability to prevent Suharto’s fall gave students and other societal groups more confidence to challenge the security forces. This also had a bearing on conflict areas, where combatants perceived a lack of willingness on the part of TNI and police personnel to use force. The separation of the military and police on 1 April 1999 also disrupted the security sector. In separating the two institutions, the government assigned to the police primary responsibility for internal security, including response to communal conflict and insurgency. Some army personnel were reluctant to relinquish this responsibility.85 Some analysts have suggested that the military delayed coming to the assistance of the police during conflict and other disturbances to demonstrate the army’s indispensability to internal security. In addition, since

42 Ethno-Religious Violence in Indonesia their separation, the military and police have competed for business and other economic interests, in some cases leading to armed clashes. These rivalries greatly undermined the state’s effectiveness in upholding internal security. Thus by 1999 the capacity of the police and military to respond effectively to the kind of large-scale conflict that occurred in North Maluku had been greatly reduced.

Rising inter-religious tension Religious sectarianism has been the exception rather than the rule in Indonesia. The vast majority of the population is religiously moderate and, at least until the onset of reformasi, the authoritarian government in Jakarta kept radicalism firmly in check.86 Throughout the 1970s and 1980s, most senior military officers, both Christian and Muslim, identified with the nationalist ideology – the pluralist pancasila – and stayed out of religious politics.87 This is not to say, however, that the country has been completely free of religious tension. Most decades have witnessed violent incidents that were ostensibly religious in character. The 1980s saw an increase in Islamic schools which espoused radical teachings and served as bases for militant action in Indonesia. Muslims trained in the Middle East introduced conservative doctrines into the country, including the Saudi Wahhabist doctrine. These schools and institutions found a growing number of supporters when Indonesian Islamic society underwent a process of revival or santrification during the late 1970s and 1980s, evidenced by greater mosque attendance, the wearing of Islamic dress and other factors.88 Over the last few decades many conservative Muslims have also become concerned at a perceived process of ‘Christianization’ in Indonesian political and economic life. Much of this sentiment stemmed from a relatively high Christian representation in the government and the economic success of (often Chinese) Christian Indonesians. Some Muslims contrasted the success of Christians with the way the government suppressed political Islam. In addition, following the legal prescription by the Suharto government in the late 1960s that all Indonesians must declare their membership of one of the five major religious traditions, some Muslims became concerned at the growing tide of Christian conversion and the increasing proselytization by foreign Christian missionaries. Muslims protested at the construction of new churches, in some cases destroying them along with schools and other buildings associated with Christian communities.89 The New Order regime’s suppression of political activity and dissent also partly explains the rise of Islamic radicalism. Political Islam had, since the Darul Islam rebellion, been a potential locus of political opposition to the central government.90 The New Order regime’s steps to eradicate this threat radicalized some Muslims. Much of this unrest centred around the government’s pressure on political parties to accept the pluralist ideology of pancasila91 as a ‘sole ideological principle’ (asas tunggal) in the 1980s. Many Muslims saw this policy as a means to eradicate religion as a basis for political opposition, and were angered at the capitulation of the main Islamic political party, PPP, to this pressure.92 This

North Maluku in context 43 tension reached a climax when the security forces opened fire on a group of Muslim protestors associated with a mosque in the Tanjung Priok area of Jakarta which had been involved in protesting against the asas tunggal policy. The general considered to have been responsible for the shooting of these protestors, Gen. Benny Murdani, was a Christian. Subsequently, a series of bombings took place in Chinese districts in Jakarta, of Christian buildings in Malang (on Christmas Eve) and at the Borobodur temple in Central Java.93 While some Muslims had been concerned at ‘Christianization’ in the 1970s and 1980s, by the late 1980s some Christians became concerned at what they considered to be the growing ‘Islamization’ of Indonesian society and politics. These concerns were exacerbated by the fact that President Suharto had increasingly turned to Muslims as a means of widening his support base. The president replaced several high-ranking Christians in the government and armed forces (including Benny Murdani) with Muslims. In 1990, the president sponsored the formation of the Indonesian Muslim Intellectuals Association (ICMI), an organization with wide-ranging political and social influence. The mid-to late 1990s saw an upsurge in violent incidents that seemingly reflected growing inter-religious tension. On the island of Flores, Catholics rioted in 1993, 1994 and 1995 after accusing either Muslims or Protestants of desecrating the eucharist during mass.94 In 1996, anti-Christian rioting occurred in several locations on Java. While these riots were not triggered by incidents involving Christians, they and their property and churches were the primary targets in all three.95 The rioting, in particular the destruction of churches, harmed inter-religious relations in Indonesia. A sentiment developed among some Christians in North Maluku that if Muslims ‘try that’ in north Halmahera Christians ‘will not retreat’.96 In late 1998, in the Ketapang area of Jakarta, Muslims attacked Ambonese security guards working in front of local gambling parlours and destroyed several churches. Subsequently, Christians in the city of Kupang in West Timor (NTT Province) rioted and destroyed ten mosques. One month later, in December 1998, rioting between Protestants and Muslims broke out in the town of Poso in Central Sulawesi Province.97 Conflict in Maluku 1999 Inter-religious violence of a scale and intensity unprecedented in recent Indonesian history began in Ambon City, the capital of Maluku Province (which at that time included the North Maluku region, see Map 1). On 19 January, on the Islamic festival day of Idul Fitri (or Lebaran) marking the end of the fasting period of Ramadan, a riot began in the Batumerah and Mardika areas of Ambon City.98 It was sparked by an altercation on a bus between a Bugis migrant from South Sulawesi and the Christian Ambonese driver.99 A series of tit-for-tat attacks developed into large-scale rioting. Ambon quickly divided along religious lines. Churches and mosques were attacked and Christians and Muslims fought in the city for days, the security forces seemingly unable to halt the rioting. The violence then began to spread beyond Ambon Island to other areas

44 Ethno-Religious Violence in Indonesia in central and south-east Maluku.100 By mid-1999 the conflict had developed into the worst religious violence in Indonesia in living memory. This turmoil was preceded by rising tension in Ambon between local Protestants and Muslims. A steady influx of Muslim migrants, mostly from South Sulawesi, was central to this rising tension. From 1969 to 1999, 25,319 households or 97,422 people were resettled in Maluku under the central government’s transmigration programme.101 A large number of independent migrants also settled in Ambon. A report by the national newspaper, Kompas, stated in early February that 225,000 migrants from South Sulawesi lived in Maluku, 30,000 of them in Ambon City.102 A combination of economic competition, crime and cultural differences caused rising tension between local Ambonese and Bugis, Butonese and Makassarese migrants.103 These migrants began to dominate many low-skilled sectors of the city’s economy including the service sector and markets. Following the 1997 economic crisis, many Ambonese Christians were forced to compete for jobs which they had previously considered too menial.104 Crowded areas in Ambon City, such as Batumerah and Mardika, saw frequent fights between Christian Ambonese and Muslim Butonese and Buginese youths.105 In rural areas, the question of land ownership also became contentious between Butonese migrants and local Ambonese.106 Migration was accompanied by increasing political and bureaucratic representation among Muslims. In 1992, President Suharto appointed a Muslim Ambonese, Akib Latuconsina, as the first local Muslim governor since 1965. The next governor, appointed in 1997, Saleh Latuconsina, was also a local Muslim. After the appointment of these two governors, educated local and migrant Muslims came to be more highly represented in the local bureaucracy, challenging the traditional position of the local Christian Ambonese elite.107 Local Christians, who had initially welcomed migrants because they were willing to undertake low-skilled employment, came to resent their increased power and status.108 The enhanced status of Muslims in Maluku as well as nationally seems to have created extreme anxiety on the part of Malukan Christians.109 Bertrand argues that the national influence of ICMI, the appointment of Muslim governors in Maluku, the transmigration programme and the increasing appointment of Muslims to the bureaucracy and other institutions in Ambon throughout the 1990s were interpreted by Christians as an Islamization of Indonesia and Maluku.110 Other analysts have argued that the conflict was more the result of provocation than societal tension. Van Klinken says that when in December 1998 the central government set a date for national elections to be held in June 1999, many local politicians and other members of the elite in Ambon mobilized members of their patronage and religious networks in order to gain or retain political or bureaucratic office.111 Other analysts have argued that the violence in Ambon was instigated by senior military officers and ex-New Order officials as a way of disrupting the process of reformasi.112 According to the Indonesian academic George Aditjondro, these figures employed gangs of Ambonese thugs, many of them sent from Jakarta to spark rioting in the city.113 Little evidence has as yet been provided to substantiate these claims.

North Maluku in context 45 Security concerns appear to have mounted to very high levels in Ambon in the months before the violence. Deteriorating Muslim–Christian relations in Indonesia since the mid-1990s and riots in other areas of Indonesia in late 1998 magnified tensions in Maluku. The International Crisis Group reported that communities in Ambon had been arming themselves for weeks before the outbreak of violence, as rumours forecasting impending violence circulated throughout the city.114 North Malukans watched as the southern part of their province burned. The dramatic stories emanating from Ambon and the regular media coverage of the violence caused tension in most areas of North Maluku. Several thousand North Malukans who had been employed or studying in Ambon fled northwards in early 1999, primarily to Ternate. Many North Malukans became suspicious of any Ambonese from the other religion, even if the individual in question had been living in the area for a long period. As discussed in the previous chapter, a number of analysts have argued that the violence that began in North Maluku in August 1999 was a result of the devastation in Ambon. However, the following chapter will demonstrate that, while the violence in Ambon to some extent affected inter-religious relations in North Maluku, North Malukans saw Ambon as an external issue. Some of the historical reasons for this, such as a lack of affinity between North Malukans and Ambonese, have been discussed in this chapter. More importantly, other local developments occupied North Malukans during this period. The initial clash in North Maluku was triggered by local issues and disputes and was not a direct consequence of the Ambon conflict.

North Maluku becomes a province The North Maluku political and economic elite made many attempts since World War II to seek independent provincial status for North Maluku. These individuals argued that North Maluku was too far from Ambon to allow for effective administration.115 Religion was also a factor behind this campaign: some Muslims pushing for a separate province wanted to break away from a perceived Christian elite in Ambon. However, religious issues were a minor consideration. The push to form North Maluku as a province in its own right was driven by members of both religious communities, united in the goal of obtaining a greater portion of the region’s resource wealth for North Maluku as a whole. The financial benefits of becoming an autonomous province were substantial, with the former spice islands the home to abundant natural resources such as forestry and fisheries. The Sultan of Ternate convinced President Sukarno in 1951 that the region should become an independent province. The president offered to create a Daerah Istimewa, or Special Region, in the style of Yogyakarta. According to North Maluku historian Herry Nachrawy, the North Maluku political elite refused the president’s offer because they felt such a status would have given the province a feudal character.116 In the early 1960s, several members of the North Maluku elite gifted 1000 tons of copra to President Sukarno to facilitate the creation of a new province, this time with normal status. This attempt to separate from Maluku

46 Ethno-Religious Violence in Indonesia Province, and a later attempt in 1971, foundered on opposition from the provincial government in Ambon.117 The onset of reformasi in 1998 stimulated a renewed effort among members of the North Maluku economic and political elite to secede from Maluku Province. Representatives from almost all ethnic and religious communities were involved in this effort. While reform at the national level provided a profound opportunity to create the new province, the motivations behind the political momentum remained largely the same as those behind previous efforts: the logistical difficulties associated with the distances in the region; the perception that the economic potential in the region was being expropriated by people outside North Maluku; and concern at being governed by a ‘Christian elite’ in Ambon.118 In 1997, a youth group, Maloko Kie Raha Students Association, organized a national symposium on the division of North Maluku and the development of Halmahera Island. The symposium was attended by leading North Maluku political figures such as Mudaffar Syah (the Sultan of Ternate), Syamsir Andili (the Mayor of Ternate), Bahar Andili (the district head of Central Halmahera) as well as central government representatives. Crucially, the governor of Maluku Province, Saleh Latuconsina, was also present and supported the creation of a new province. This momentum continued in 1998, those involved taking advantage of the widespread political and social change occurring throughout Indonesia.119 Representatives from political parties considered to be on different sides of the political spectrum such as Golkar and PPP, and from both districts (North Maluku and Central Halmahera), formed two groups, the Team of Nine from North Maluku and the Team of Five from Central Halmahera. Both groups pressured Indonesian legislators, cabinet ministers and officials of the Department of Home Affairs in Jakarta. A student group called the North Maluku Youth Student Forum or FPPMU (Forum Pelajaran Pemuda Maluku Utara), sponsored by these political leaders, travelled to Ambon. Along with approximately 100 students from North Maluku studying at Ambon’s Pattimura University, the group demonstrated outside the provincial parliament requesting that the provincial parliament agree to the division of the province.120 The governor, Saleh Latuconsina, and representatives of the parliament again publicly agreed with the students’ goals. Around twelve of these students also travelled to Jakarta, along with leading North Maluku parliamentarians Rusdi Hanafi, Rustam Conoras and Saiful Bakhri. In February 1999, President Habibie promised the group that his government would create a new province of North Maluku. As it became clear in early 1999 that North Maluku was indeed to become an independent province, the individuals and groups involved began to divide into political factions surrounding candidates for the governorship. There were four main candidates: Mudaffar Syah, Bahar Andili, Syamsir Andili and Thaib Armain (who later that year became the head of the North Maluku provincial bureaucracy). Each of the four main candidates held a great deal of power in the province and therefore had very strong claims to become North Maluku’s first governor.

North Maluku in context 47 Mudaffar Syah was in a particularly powerful position, serving not only as Sultan of Ternate but also as North Maluku Chairman of the Golkar party, which dominated both North Maluku and Central Halmahera District Parliaments. He had also recently been elected as Chairman of the North Maluku District Parliament. The Sultan received most of his support from the Ternate ethnic group situated mostly in the northern parts of Ternate City and in the villages of Ternate Island. The Sultan also enjoyed a great deal of support on Halmahera, particularly among Christian communities in northern Halmahera. Likely to receive the support of most representatives of the secular, nationalist parties which would dominate the provincial parliament when it was formed (Golkar, PDI-P as well as representatives of TNI), in 1999 Mudaffar Syah was expected to win the governorship. Mudaffar Syah’s main rival for the governorship was Bahar Andili, the District Head of Central Halmahera. As district head, Bahar Andili exercised an enormous amount of influence in that district. The two other gubernatorial candidates were also important in this competition. As the Mayor of Ternate, Syamsir Andili enjoyed widespread support in that city, particularly in the southern areas of the city, such as Bastiong, where large numbers of Tidores had settled.121 As a Makian, and having two years before held the highest position in the district bureaucracy, Thaib Armain enjoyed a great deal of support from the large numbers of Makians and Tidores within the bureaucracy, as well as in the south of Ternate.122 These candidates differed on several issues concerning the character of the new province, including its status, and the location of the capital. Mudaffar Syah lobbied for the provincial capital to be located in the city of Ternate, arguing that it already housed the province’s major infrastructure, including an airport and port. He also pushed for some recognition of the cultural traditions of the region. He argued that the province should, like Yogyakarta (also strongly associated with a local sultanate), hold the status of Special Region (Daerah Istimewa).123 He proposed that the province be called Maloko Kie Raha, the name for the area of North Maluku during the era of the four sultanates. The sultan eventually compromised by suggesting the capital should be located in Sidangoli, in Jailolo Sub-District on Halmahera, but still in the traditional sphere of influence of the Ternate sultanate. Bahar Andili and his younger brother Syamsir together took a different position on all of these issues. They sought normal provincial status for North Maluku, with the province’s capital to be located provisionally in Soasio, Tidore, and permanently in Sofifi in Central Halmahera, within the traditional territory of the sultanate of Tidore and easily seen across the bay from Soasio.124 This would bring immense economic benefits to the district of Central Halmahera. Thaib Armain was less visible in this struggle but reportedly supported the Sultan’s goal of locating the provincial capital in Ternate or alternatively on Halmahera. However, Thaib Armain opposed the adoption of the name Maloko Kie Raha and the status of Daerah Istimewa. Supporters of Thaib Armain, many of whom were civil servants, opposed a ‘Special Region’ status for North Maluku

48 Ethno-Religious Violence in Indonesia because they believed it would enhance the political and cultural power of the Sultan of Ternate and alter employment practices in the provincial bureaucracy. It is important to note that, while none of the above candidates was Christian, this fact did not cause tension within the Christian community. In early 1999, religion had not become an important marker for mobilizing political support by any of the gubernatorial candidates, even with the ongoing violence in Ambon. Throughout the first half of the year several prominent Christians from Tobelo and elsewhere on Halmahera supported the efforts of Thaib Armain to become governor, primarily because of his stated goal of locating the provincial capital on that island. These forms of cross-cutting alliances only began to fragment after the initial violence in the area of Malifut on Halmahera.

Conclusion Historically, the remote ‘spice islands’ of North Maluku have long been a site of competition and turmoil, not only between the ancient sultanates of Ternate and Tidore, but also as pawns in the struggle between global powers for control over their abundant aromatic spices. Relative stability as well as anonymity descended on the region during the New Order period of President Suharto, allowing a certain degree of economic growth. From 1997 to 1999, North Maluku, along with the rest of Indonesia, went through a period of rapid and exciting change as Indonesia set out on the path to democratization. The economic crisis did not affect North Maluku in the way it affected Java and other areas. Yet other national changes, in particular the decentralization of financial and political authority to the regions, provided political and economic opportunities, particularly for the elite. While this change was felt all across Indonesia, the North Maluku region faced additional disruption. In 1999, inter-religious conflict broke out in Ambon City and elsewhere in central Maluku. Yet local politics, specifically the push to make North Maluku an independent province, were far more pertinent to the trajectory of the North Maluku conflict than external religious influences. After several months of lobbying, the central government announced that the region would be inaugurated as a province in its own right. After decades of political appointments from Ambon, North Maluku was suddenly presented not only with new-found democracy but also with the political competition associated with elections for the governorship. While the political elite was initially united in the quest to secede from Maluku Province, it subsequently splintered into several competing political factions. These factions reflected not only networks of political affiliation and patronage but also, to some extent, ethnicity. All of these issues served as background to the North Maluku conflict and many of them influenced its trajectory and spread in one way or another. Yet the original trigger for violence lay within the traditional and emotional heart of North Maluku society. As we will see in the next chapter, the fighting began with a dispute over that most emotive of resources – land – the control of which was brought into question by the creation of a new sub-district in Malifut, North Halmahera.

3

Initiation – Malifut

Introduction Many Makians did not want to leave their homeland of Makian Island when government officials arrived in 1975. Despite increased seismic activity over the preceding months, and the warnings that Kie Besi volcano was ready to erupt, the people of Makian wanted to stay with their homes and gardens. Many believed that the government was simply trying to move them to less crowded and unproductive land. When officials began forcibly removing many families to Malifut on Halmahera, this distrust turned to anger. Yet by the 1990s the Makians had accepted they would remain in Malifut and sought to legitimate their permanence in the area by having their new sub-district formally recognized. However, in 1999, just as they were about to achieve their goal, the Makians found their way blocked by their neighbouring community, the Kaos. The indigenous Kao ethnic community refused to accept a new sub-district for migrants on what they considered to be their traditional land. The boundary of the new sub-district cut through the Kao community, leaving some members in a Makian-dominated sub-district. Goldmining operations in the area since 1998 added material incentive to this opposition. After the central government mandated the new sub-district, rising tension between the Kaos and the Makians culminated in a Makian attack on two Kao villages that refused to be incorporated. Two months later, several thousand Kaos attacked Malifut, driving the entire community from Halmahera. By 26 October 1999 Malifut had been completely destroyed with only mosques and schools still standing. This chapter seeks to explain why a territorial dispute became so emotive. It concludes that the territory in question was simultaneously an economic, political and identity issue in the Kao–Makian relationship. The land was not just an important resource to the communities but also defined the way each perceived themselves in relation to the other. By 1999 land was the one remaining aspect of the relationship over which the Kaos felt a source of pride and dominance vis-à-vis the Makians. The biased and nepotistic response of the local government made this issue more volatile. Despite the religious differences between the two communities and the fact the North Maluku conflict subsequently became religious in character, religion played little part in this initial incident.

50 Ethno-Religious Violence in Indonesia In order to better understand the Kao–Makian riots this chapter begins with a brief discussion of three bodies of theory: on the mobilization of social movements, on the role of territory and on inter-group comparisons in conflict. The chapter then discusses in detail the period before the first incident in August, and the subsequent events leading up to the October attack. The final section then analyzes why the creation of a new sub-district led to violence in Malifut, when similar developments elsewhere in Indonesia passed peacefully.

Mobilization for violence Mobilization and counter-mobilization Some, perhaps many, cases of communal violence are preceded by a relatively powerful group mobilizing for greater control of political office, an economic niche or lucrative natural resources. These groups often have the capacity to mobilize quickly and effectively in response to increased opportunity. When another community mobilizes in response, the conditions for conflict are laid. In such cases, Resource Mobilization theory provides a number of insights into when mobilization is likely to occur, when it is likely to be successful and when it is likely to stimulate counter mobilization.1 This approach can also shed light on when one or more parties are likely to resort to action outside institutional channels. According to the Resource Mobilization school, social movements arise out of a convergence of adequate resources and increased opportunity to mobilize. Strong community networks allow easier recruitment, but influential individuals must also sustain the movement through organizing money and labour and promoting common purpose among participants. The likelihood of a movement emerging will also depend on the potential costs and rewards of mobilization, termed the ‘political opportunity structure’.2 People will refrain from mobilizing others or participating if the risks of repression appear too great or if the likelihood of substantial returns is minimal. Mobilization is far more likely if participants anticipate minimal repression and a high return. Other studies within the Resource Mobilization tradition focus on the reaction of other groups to mobilization, providing further insight into the period preceding communal violence.3 To be effective, counter-movements also require a mobilizing ideology, adequate resources and the ‘political opportunity’ to emerge.4 Mobilization by one group creates further opportunity not only for that group but also for other actors to counter-mobilize. The initial movement may often stimulate counter-mobilization by demonstrating to others that their interests, too, could be served by acting collectively. However, counter-mobilization is most likely to occur if the initial movement or mobilization appears to threaten the interests of other groups.5 Perceived government support for the initial movement may also appear as a threat to other groups, stimulating counter-mobilization.

Initiation – Malifut 51 From mobilization to violence Mobilization and counter-mobilization can polarize two communities by severing any ties between them. In turn, this polarization can cause other pre-existing contentious issues between the two communities, not connected to the original mobilization, to come to the fore. Leaders of each community are quick to play on long-standing grievances or prejudices in order to recruit new members to the movement. In addition, both the initial movement and the counter-movement are more likely to adopt non-institutionalized strategies if institutionalized action appears likely to be ineffective. David Meyer and Suzanne Staggenborg assert that actors are more likely to engage in ‘direct’ (non-institutionalized) action, such as protest, if they have failed to counter the success of another group through institutional channels.6 Doug McAdam says that the basic problem facing insurgents is to overcome their powerlessness in the face of the authorities. To do this, they must disrupt established structures of power, forcing their opponents to make concessions.7 Yet this form of competition and attendant polarization occurs in many situations without resulting in violence and destruction. Movements often find institutional avenues blocked but do not resort to violence. There is also a great deal of difference between peaceful ‘non-institutionalized’ action such as protest or sabotage, and large-scale rioting and violence. The question remains: when and how does mobilization become violent? Many theorists of conflict have claimed that violence becomes more likely if the issue in dispute holds not just material benefit but emotional resonance for one or more parties.8 The character of communal violence itself seems to support this conclusion. Communal violence is often incredibly intense, involving massive destruction and the targeting of women, children and the elderly. Much of the violence often appears expressive or consummatory as opposed to goal-based (instrumental). In addition, communal violence often occurs rapidly, with little time for premeditation, and is often counter-productive for participants, who ultimately bring on themselves destruction and loss of life. It is clear that emotional factors, as well as purely rational interest, play a role in communal violence. A perusal of the wider literature on conflict suggests that certain types of issues are more likely to cause a dispute to take on such an affective character. Conflict may be more likely to lead to violence if it involves not only conflicting interests but also issues that are important to a group’s ethnic identity and central to how each group perceives itself in relation to another. Donald Horowitz asserts that the ‘sources of ethnic conflict reside, above all, in the struggle for relative group worth’.9 He maintains that, largely because of colonialism, ethnic communities have become ‘ranked’ into what he terms backward and advanced groups in many countries. The sensitivities associated with this ranking lie behind much ethnic conflict. The humiliation that accompanies a sense of lower status and a realization that the other group has ‘mastered modern skills’ makes a group determined not to relinquish any power that it manages to obtain. A

52 Ethno-Religious Violence in Indonesia fear of subordination ensures that even minor issues can appear relevant to the survival of one’s group.10 Many less advanced groups make claims to being the indigenous community to an area, the ‘sons of the soil’, as a means of achieving recognition in this struggle for relative group worth. Territory is often much more than an economic resource to communities, the attachment felt by its residents and past residents going much deeper than the economic or subsistence benefits associated with its ownership. Affective ties to territory are often central to ethnic identity and nationalism.11 Toft concludes that a long-standing occupation by another state or ethnic group, a history of struggle for, and blood spilt on, the land and a name reflecting ownership by one’s group are all likely to increase emotional ties to a particular tract of territory.12 Claims to territory are central to a number of internal conflicts, both inter-communal and separatist in character, for example those between Israel and the Palestinians, and between Protestants and Catholics in Northern Ireland.13 As will be seen in this chapter, territory can be central not only to the identity of one or more parties to the conflict, but also to the mutual understandings between the two communities. Yet inter-group comparisons and attachments to territory are merely conditions conducive to conflict. They cannot in themselves provide complete explanations of violence. Such conditions exist in numerous situations around the globe but do not result in violence. It is necessary to uncover the actions and motivations that bring two communities to the brink of tragedy and ultimately send them into the carnage of violent conflict. This synthesis of agency and structure, of emotion and interest, as well of mass and elite action, is demonstrated in what follows.

The sources of tension Kao Sub-District is located on the northern peninsula of Halmahera. In 1999 the area was part of the district of North Maluku.14 The sub-district is largely an alluvial plain with a heavily forested and mountainous interior. The capital of Kao is a large village of the same name, housing most infrastructure, including the office of the sub-district head and schools. The population of Kao Sub-District, officially 26,704 in 1999, is concentrated along the coast, although there are also many villages in the more remote interior of West Kao.15 The majority of the population is from the Kao ethnic group, which itself is composed of four sub-ethnic groups, the Kao, Modole, Pagu and Boeng. The vast majority of Kaos are Protestant Christian (affiliated with the Tobelo-based GMIH), although there is a small Muslim Kao minority. based mainly in the capital.16 Perhaps several dozen Ambonese also live in Kao. Several transmigration settlements populated primarily by Javanese are located in the interior.17 Most Kaos are involved in subsistence farming and fishing, and most paid employment is in the large copra industry. In 1975 the government of North Maluku District, based in Ternate, began a local transmigration programme from Makian Island, located west of the large island of Halmahera (see Map 2). The programme was officially prompted by an

Map 3 Kao/Malifut

54 Ethno-Religious Violence in Indonesia anticipated volcanic eruption on the island.18 The government moved the majority of the population of Makian, in many cases by coercion, to the area of Malifut on Halmahera, about 15 km south of the capital of Kao Sub-District.19 The Makian community was, and remains, exclusively Muslim. The location of the resettlement site is a subject of some controversy today, with people on both sides of the conflict suggesting that the district government chose this site to either halt the southward spread of the Christian community of north Halmahera, and/or to place a more progressive community in the area to stimulate development.20 The Makian community began to arrive in the area of the Pagu villages of Malifut, Wangeotak, and Balisosang in the centre of Kao Sub-District (see Map 3) in 1975. The local Pagu and wider Kao population welcomed the Makians, offering them food and established banana, cassava and coconut gardens, as well as land for building houses and cultivating gardens.21 From 1975 until 1980, approximately 6,000 Makians were moved to Malifut. In 1999, the local government calculated that there were approximately 16,000 Makians living in 16 villages in Malifut, plus a population of 26,704, predominantly from the Kao ethnic group, living in the surrounding Kao Sub-District.22 During the second half of the 1970s, the district government also moved Makian’s sub-district government office. Entire villages were moved, retaining their Makian names (e.g. Tahane, Ngofagita). Standing village leaders were also retained.23 In official government documents the community was subsequently referred to as Malifut/ Makian Sub-District. However, the sub-district never received formal recognition in its new location, which required both the consent of the district parliament and legal recognition by the Department of Home Affairs in Jakarta. The community of Malifut had therefore enjoyed the infrastructure of the sub-district of Makian without the administrative unit ever being granted any legal status and with no official recognition of Makian land-ownership or formal institutions of self-government. The government had also never formally established the border of the new sub-district, an issue that gained significance after gold mining began in the area. The Kaos and Makians also held fundamentally different perceptions of the status of the land on which the Makians had settled. As discussed further below, the Kaos understood the area of Malifut to be Kao traditional land on which they had permitted the Makians to reside, as victims of a natural disaster. The Makians, however, considered the area of Malifut to be government land, and did not recognize traditional Kao ownership. One Makian complained that the Kaos ‘constantly said we lived on their land, but they did not realize it was government land and had been given to us by the Department of Transmigration’. Much of this confusion appears to have stemmed from the manner in which the transmigration programme was carried out in 1975. The North Maluku District government explained the programme in different ways to each community. In order to overcome the reluctance of many Makians to leave their home island, the government reassured them that their relocation was part of an official transmigration programme and that they would therefore receive legal ownership of land and accompanying documentation. The Kaos, however, agreed to the relocation of the

Initiation – Malifut 55 Makians on the understanding that the Makians were guests on Kao land. For the Kaos, their traditional or customary ownership of the entire sub-district was proved by their long history of settlement, place names in the area and important landmarks, such as a gravesite containing both Christian and Muslim Kaos (as discussed further below). However, no official recognition had ever been given to Kao customary ownership of the territory in question. The district government appears not to have made explicit the legality of either position. For many Makians in Ternate and Malifut, the uncertain status of the sub-district in Malifut was a major source of frustration. Many had been forced by the government to leave Makian Island and resettle in north Halmahera, and had been told that the resettlement programme was permanent and not a temporary emergency relocation. In addition, the district government closed the sub-district on Makian Island in 1995 on account of frequent volcanic activity, precluding any return there by those Makians who wished to leave Halmahera.24 This grievance was exacerbated by the fact that the Makians, like the Kaos, had a strong emotional attachment to their homeland. Writing just after the transmigration of the community to Malifut, Ronald Lucardie concluded ‘the unattractiveness of the resettlement at Malifut to the Makianese thus cannot be adequately explained by their practical objections … the emotional dimension – the persistent, strong attachment to home villages on Makian – is the key.’25 Although having eventually resigned themselves to remaining in Malifut, the lack of any sense of ‘ownership’ of their new home area compounded their anger at having lost these close ties to Makian Island. Other aspects of the Kao–Makian relationship also caused tension. Despite two decades of living in close proximity, the two groups lacked any meaningful integration. There were no cases of intermarriage, even between Muslim Kaos and Makians. In 1999, although the Makian villages of Tahane and Ngofagita were immediately contiguous with the Pagu villages of Sosol and Wangeotak, the communities remained socially segregated. The two communities were also highly unequal in terms of status and access to resources. In 1999, the Kao ethnic community of 27,000 had only one member in the North Maluku District parliament. Kao representation in the district bureaucracy was almost non-existent. Few Kaos study in the provincial universities in Ternate and do not usually hold positions in the education sector. Conversely, as discussed in Chapter 2, Makians were highly influential in many sectors of North Maluku life. A large proportion of young Makians had undertaken tertiary education and as a result Makians were heavily represented in the government, bureaucracy and educational sector. Strong mutual stereotypes also characterized the Kao–Makian relationship. The Kaos considered the Makians to be greedy, and accused them of constantly extending the territory allocated to them during their relocation to Malifut.26 They believed that the Makians dominated the government and bureaucracy, which in turn strongly favoured Makians over members of other groups for employment and government funding. The Kaos perceived the greater economic success of Malifut compared to their own villages to be a result of this disproportionate allocation of funding.27 Many Makians stereotyped the Kaos as lazy and backward,

56 Ethno-Religious Violence in Indonesia and considered the differences in development between the two communities to be the result of their own greater industriousness. 28 Prior to 1999 several violent incidents broke out between the Malifut and Kao communities, although none presaged the extent of destruction that would occur in 1999. In 1982, a Christian Kao was beaten and killed during fighting between Makian and Kao youths. The young man had been invited to a family meal during Ramadan by a Muslim girl.29 In 1989 community leaders and police prevented a potential riot following a ‘mysterious death’. Most Kaos and Makians stated, however, that these incidents were not connected to religion and that relations between the two communities were generally ‘good enough’.30 One Christian leader claimed that, before the conflict, Christian Kaos had better relations with Muslim Makians than with Muslims from their own ethnic group, whom the respondent considered to be more ‘fanatic’.31 According to this respondent, fights were more common between Christian and Muslim Kaos than between Kaos and Makians, although these generally concerned football, girlfriends and other non-religious issues. As discussed in the previous chapter, in June 1999 the mining company PT Nusa Halmahera Mineral (NHM) commenced gold-mining operations at Gosowong, near Malifut.32 NHM paid a number of ‘honorariums’ to local officials on Halmahera and elsewhere, including the sub-district heads of Malifut and Kao.33 The company also constructed several schools in Malifut and Kao villages. With 350 local employees, NHM was the largest single employer in the immediate area. By mid-1999 the composition of the workforce at the mine had become a contentious issue. The Australian managers of the mine attempted a policy of employing 50 per cent from each community, but by August 1999 90 per cent of employees were Makians and 10 per cent were Kaos, largely because of the higher levels of education and better organizational skills among Makians.34 Small fights occurred between Kao and Makian mine labourers before larger riots erupted in August. 35

A powerful ethnic group mobilizes The wider Kao–Makian relationship began to deteriorate in 1999 following efforts by Makians in Malifut and Ternate to have the Makian Sub-District in Malifut legally recognized. As discussed above, the uncertain status of the community in Malifut was a source of potential conflict, as the Kaos and Makians held opposing views about the territory on which the community was founded. The two sub-district governments appeared to operate in parallel, although the Kaos continued to consider Malifut as legally part of Kao Sub-District. This uncertainty frustrated many Makians, who considered that the community deserved sub-district status after two decades of residence on land which was (in their understanding) allocated to them through an official transmigration programme. In addition, with the start of gold-mining operations at Gosowong in June 1999, it became apparent that the Malifut area was a particularly resource-rich area. The employment opportunities offered by NHM would be highly beneficial to a community the size of Malifut. NHM also offered

Initiation – Malifut 57 regular funding to both the local Kao and Malifut communities for the building of schools and health facilities such as local clinics. Many Makians saw the era of reformasi in late 1998 and early 1999 as an opportunity to formalize the status of the sub-district of Malifut. Makians in the government, the bureaucracy and the educational sector, along with Makian students studying in Ternate, Ambon and elsewhere, launched a powerful political initiative to achieve this goal. While the Makian elite outside Malifut led this drive, Makian community members within Malifut also pressed to have their sub-district formally recognized. In December 1998 community leaders from Malifut and Kao met to discuss the possibility of formally separating Malifut Sub-District from Kao. Several Makians claim that community leaders from the Kao villages immediately adjacent to Malifut (Sosol and Wangeotak) agreed to the plan.36 However, all Kaos deny having ever agreed to the establishment of a new sub-district. In December 1998, Makian students studying at Pattimura University in Ambon pressed the provincial parliament to legalize the status of Makian Sub-District in Malifut. In January a team of six Makian university students from Ternate and Ambon travelled to Jakarta to pressure the Department of Home Affairs to issue a government regulation (Peraturan Pemerintah, PP) making Malifut a formal sub-district.37 The mobilization of these students appears to have been facilitated, and their travel to Jakarta paid for, by the Makian elite in Ternate. When meeting government officials in Jakarta, these students requested that this new sub-district be called Kecamatan Makian Daratan di Malifut (The Sub-District of Mainland Makian in Malifut).38 The group subsequently became known as the Makayoa, after the islands of Makian and Kayoa (a neighbouring island with strong ethnic connections to Makian). The group, including those who had been studying in Ambon but could not return after rioting broke out in that city on 19 January 1999, then returned to Ternate. Several months later the former rector of Khairun University, Yusuf Abdurrahman, allegedly contacted Makian students in Ujung Pandang (Makassar) and requested they return to North Maluku to pressure first the government to inaugurate the sub-district, and then, as will be seen, the Kao community to accept this decision.39 Many among the political elite in Ternate also campaigned for legal recognition of Malifut Sub-District. In early 1999, the chairman of the district parliament, Suleiman Adam, presented a letter to the district head of North Maluku, Abdullah Assagaf, stating the consensus of the parliamentary assembly to formalize the sub-district in Malifut.40 Assagaf himself agreed with creating a new political unit in Malifut, believing it would increase administrative efficiency.41 The district head subsequently travelled to Jakarta to request that officials of the Department of Home Affairs issue a government regulation giving the sub-district legal status. Most Makian students and members of the Makian elite involved in this movement were motivated by ethnic solidarity. Many, although living in Ternate, had been born and raised in Malifut, where, in many cases, their families continued to live. Following the closure of the island of Makian, Malifut was the one area in North Maluku that remained strongly Makian. In addition, most felt little affinity

58 Ethno-Religious Violence in Indonesia with the Kaos. Malifut’s legal status as merely a part of Kao Sub-District had long been a source of discontent among those with connections to the Malifut area. Legal recognition of the sub-district would also legitimize the Malifut community’s control over the area’s valuable economic resources. Prior to the formalization of Malifut, the community was required to share the benefits of the NHM gold mine with the larger Kao community. Particularly important in this regard were the hundreds of employment positions at the mine, but the assistance paid by the company to the community for the construction of schools and other infrastructure was also important. If NHM were to be located within a newly independent Malifut Sub-District, the Malifut community would monopolize these benefits. However, such material and legal benefits of a new sub-district were probably insufficient to inspire the political momentum that occurred among the elite. Many in North Maluku suggest that the recognition of Malifut as an autonomous sub-district was only a first step towards achieving a long-held goal for many connected with Malifut – establishing the town as the capital of a district.42 In 1998 the district head, Ret. Lt Col. Abdullah Assagaf, had advocated the creation of a new North Halmahera District covering the whole northern peninsula of Halmahera.43 As the likelihood increased in early 1999 that North Maluku would become a province in its own right, it became clear that the province would indeed be divided into more than the existing three districts.44 The location of the capital of a North Halmahera District was a sensitive issue for many in the area. As the oldest and largest town in north Halmahera, Tobelo was the obvious choice as capital of a new district in northern Halmahera. But as a successful, relatively affluent group of villages with an increasingly developed infrastructure, and located closer to Ternate than Tobelo, many also considered Malifut an ideal candidate for capital. In early 1999, Abdullah Assagaf asked the Tobelo Christian community leader Hein Namotemo if he would obtain agreement among the Tobelos for Malifut to become the capital of this new district.45 Namotemo refused, saying that it was very unlikely Halmaherans would agree. As a series of villages officially within a sub-district (Kao) of which it was not capital, Malifut had little chance of assuming this position when compared to Tobelo, already the capital of its own sub-district. It seems likely that the prospect of Malifut coming under the jurisdiction of Tobelo as district capital was perceived unfavourably by many Makians. Makians were unlikely to agree to being part of a new district dominated not only by Halmaherans but also by Christians. In addition, following the new national laws on regional financial autonomy, the choice of district capital had major financial implications.46 Under Law 25/1999, 80 per cent of the royalties from mining was to be returned to the region in which the mine was located, and 20 per cent held by the central government. The government of the district in which the mine was located would receive approximately 32 per cent of the production revenue.47 In 2001, Nusa Halmahera Mineral paid to the central government US$2,064,594 in Exploitation Rent and in 2002, US$911,715.48 Of that figure, the central government was required by Law 25/1999 and Government Regulation 104/2000 to return to the North Maluku District Government approximately US$700,000 and US$315,000

Initiation – Malifut 59 for 2001 and 2002 respectively. The economic benefits of having the local government located in Malifut would also be great in terms of employment, infrastructure construction and development. As the capital of a district, therefore, the community of Malifut, particularly members of the government and bureaucracy located there, would receive considerable financial benefit.49 As a sub-district, however, Malifut would not necessarily receive any of the royalties of the NHM gold mine, a factor that the large numbers of well-informed Makian elite would clearly have been aware of.50 Most respondents suggested that if it were eventually to become the capital of a district, Malifut first had to become an autonomous sub-district. The desire of many Makians to achieve this goal quickly may be attributed either to their concern that it should be accomplished in advance of the district elections in July 1999, coinciding with the national election, or to the need to formalize the status of the sub-district before North Maluku was inaugurated as a province in October 1999. It is possible that many Makians were concerned that the support they enjoyed from the incumbent government might not be repeated by a new district parliament.51 In short, Makians living in Malifut, Ternate and elsewhere pressed for the legal recognition of Malifut Sub-District for several reasons. Makians outside Malifut, including students studying in Ambon and elsewhere, felt a strong sense of ethnic solidarity with the Malifut community since many had originated from or had family living there. Many wished to see the creation of a successful Makian ‘homeland’ within North Maluku following the closure of Makian Island. Most of those involved sought to afford Malifut exclusive access to the employment and other benefits presented by NHM. It also appears likely that many in the Makian elite sought to establish Malifut as a sub-district as a first step towards installing it as the capital of a new district likely to be formed after the separation of North Maluku from Maluku Province. The Makian students who actively advocated the creation of a Malifut Sub-District were influenced and assisted in their efforts by members of the Makian elite in Ternate, who covered the students’ travel costs to and from Ambon, Jakarta and elsewhere.

Counter-mobilization On May 26 1999, following the visit of Abdullah Assagaf to Jakarta, the Department of Home Affairs in Jakarta released Government Regulation No. 42 (PP42),52 which legally recognized the Sub-District of Makian Daratan di Malifut. Several people involved in pushing for PP42 suggested that officials in central government departments in 1999 were very keen to support local programmes seen or portrayed as being in the spirit of reformasi.53 However, the Kao community had not been consulted since December 1998 about the creation of this new sub-district within Kao. The Kaos objected strongly when they first heard of the release of PP42 in May 1999. Their objections intensified when Kao community leaders became aware of the new sub-district’s borders and the implications of the sub-district for their own interests.54 Their objections were based on a number of factors, both identity-related

60 Ethno-Religious Violence in Indonesia and resource-related. Makian di Malifut Sub-District included the entire area covered under the Contract of Work (COW) held by NHM for mining on Halmahera. Many Makians state that Kao officials, such as the sub-district head, Mohtar Sangaji, provoked ordinary Kaos into opposing PP42 because these officials did not want to lose NHM funding.55 Kaos denied this, as did one Australian employee of NHM.56 In fact, the Kaos denied that the mine was important in their objections.57 Yet the mine cannot be discounted completely as a motivating factor. The Kaos’ denial that the mine was important in their opposition to PP42 most likely stemmed from a desire to emphasize cultural rather than economic motives, the latter often seen in less sympathetic terms by the wider North Maluku community. As the primary employer in the area, the mine was an important source of livelihood for many ordinary Kaos, and it is reasonable to assume that many Kaos would have seen the creation of a new Makian Sub-District as exacerbating the unequal distribution of employment between Kaos and Makians. However, issues of ownership of territory and the unity of the Kao ethnic group were more important in these objections. The Kaos reiterated that the Makians had moved to the area because of a natural disaster, not through an official transmigration programme. They were welcome as long as they respected Kao tradition (adat).58 To the Kaos, this entailed the Makians recognizing that they lived on Kao (especially Pagu) land. The Kaos argued that Makians who had moved to other areas on Halmahera, such as Central Halmahera District and Ibu in northern Halmahera, had not established new sub-districts.59 They also argued that Javanese transmigrants relocated to Kao had not established new sub-districts.60 Kao objections to the name ‘Makian Daratan’ (Mainland Makian) were strongly connected to the issue of land ownership. To the Kaos, the suggested name was a violation of their traditional ownership of Kao territory. Many Kaos did not object to the creation of a new sub-district in the area, as long as it was a division (pemekaran) of Kao Sub-District, with its name reflecting this. Indeed, many Kao community leaders told Makian leaders and government officials that they held no objections to the creation of a new sub-district with the name ‘Kao Selatan’ (South Kao) or ‘Pagu’.61 Makian leaders agreed to compromise by changing the name to ‘Makian di Malifut’ but refused to drop the word ‘Makian’ from the name, because, according to one respondent, to do so would mean the Makians had been effectively assimilated into an alien culture.62 Perhaps the most strongly articulated Kao objection to PP42 was that Makian di Malifut Sub District would divide the Kao population. The boundary of the sub district created by PP42 incorporated five Kao villages: Sosol, Wangeotak, Balisosang, Gayok and Tabobo, meaning that a portion of the Kao community would be separated from the majority in Kao Sub-District.63 The Kaos are very proud of the integrity of the four sub-groups of their community and point to a long history of these groups living and fighting together. A strong feeling of ethnic solidarity prevailed among both Christian and Muslim Kaos. This feeling of solidarity was symbolized by a well-maintained grave in the centre of Kao village containing Kaos from both religions, and also by the principle ‘hidup bersama, mati sekubur’ (live together, dead in one grave).64 According to local oral history, in 1904 the

Initiation – Malifut 61 Kaos travelled from Gosowong to the area of present-day Kao village to attack a Dutch military base. In the clash between the Kaos and the Dutch army, seven Kaos – four Christians and three Muslims – were killed. Following a consultation between the community leaders of the various sub-ethnic groups and religious leaders, the decision was made to bury the seven casualties in one grave. Several Kao community leaders said cultural differences between the Kaos and the Makians were so strong that it was not feasible for any Kaos to live in Makian di Malifut Sub-District.65 As well as fearing that the Kao community would be divided, many Kaos held concerns about being ‘controlled’ by a Makian-dominated sub-district government. Villagers from Sosol and Wangeotak often described the tensions surrounding PP42 by stating that the Makians ‘wanted to control us, but we are the indigenous people of this land’.66 Therefore, both the Kaos and the Makians had similar concerns about becoming subordinate to another ethnic group. There were two major Kao protests against the creation of Makian di Malifut Sub-District. The Kaos protested outside the Kao and Malifut Sub-District government offices in May after hearing of the release of PP42, and a delegation visited the district head, Abdullah Assagaf, in Ternate in the same month. The primary objection put to the district head was that the inhabitants of some villages included in the new sub-district were Kaos and that this would undermine Kao unity and land ownership.67 Assagaf told the Kaos that the creation of the sub-district was an administrative issue, not an emotional one, and that they should accept the government’s decision As Kao protests mounted, pressure from some Makians also grew. The Makayoa student group held demonstrations in Ternate to pressure the district parliament not to accede to the Kaos protests and to inaugurate the new sub-district. On 13 August several dozen members of the Makayoa organization travelled to Malifut to bolster support among the Makian community for PP42 and to pressure those Kao villages located within the new sub-district to accept the regulation. When the Kaos voiced their objections, the Makayoa stated that a government regulation had been released and that to change it the Kaos would have to go to Jakarta, a task beyond most Kaos.68 When Kao leaders protested about PP42 to the government in Ternate, Yunus Abbas, the first assistant to the head of the bureaucracy, himself a Makian, also stated that the regulation was already legally binding and it was too late to change.69 As the Kaos continued to refuse to accept PP42, the Makians assumed a more belligerent and confrontational approach towards the Kaos. Many Makians, particularly the Makayoa student group, were angered that the Kaos were attempting to prevent them from being officially recognized as the permanent residents of what was, to their minds, government-owned land. By this time, a high proportion of Makians were convinced that it was the inhabitants of Sosol, in particular, who were resisting inclusion in Makian di Malifut Sub-District.70 Officials from the district government in Ternate, including Abdullah Assagaf, asked Kao leaders to convince the Kao community to accept PP42. Those approached included the local Protestant minister (pendeta), Pastor Salamena,

62 Ethno-Religious Violence in Indonesia who refused to cooperate.71 When, because of continued Kao opposition, Assagaf began to publicly express doubts that the situation in Malifut and Kao was conducive to the inauguration of the new sub-district, the position of the Makayoa student group in Malifut became more antagonistic.72

The August riot At around 9 p.m. on 18 August, stones were thrown at houses in the Pagu (Kao) village of Sosol, according to some accounts following a dispute at a party. Ten minutes later, violence broke out between residents of Sosol and the neighbouring Makian village of Tahane. A major attack was subsequently launched from Malifut as shouts of ‘Allahu Akbar’ were heard from the Tahane mosque.73 In the initial attack from Tahane, the father of the village head of Sosol, Yordan Moumou, was killed. The attack appears to have largely stopped overnight and restarted the following morning. At 6 a.m. both communities regathered, and at 8 a.m. there was a further Sosol casualty, Erasmus Dodowol. The community of Sosol was able to resist the attack for several hours, when the entire community fled to the beach and was evacuated by boat to the capital, Kao. The entire village of Sosol was destroyed, including the church and school. Still in Malifut, the Makayoa students were involved in the attack, and the leader of the group was killed in Sosol.74 On the morning of 19 August Makians from Tahane, Ngofagita and other villages attacked the neighbouring Pagu village of Wangeotak.75 One man was killed in the attack, and the village resisted for only a few hours, after which the community was evacuated to Kao with the help of local military personnel and reinforcements that had arrived from the military company based in Tobelo. As with Sosol, every house and the church in Wangeotak were destroyed. The isolation of Sosol and Wangeotak from other Kao villages meant there was little chance of either village repelling the attacks. By the afternoon, the two villages at the centre of opposition to being included in Malifut lay smouldering. The five security personnel present in Malifut until reinforcements arrived from Tobelo were completely overwhelmed. The commander of the unit stated that, given there were over a thousand Kaos and Makians fighting, it would not have been a ‘good idea’ for his personnel to shoot at the crowds.76 The military merely shot in the air, to little effect. Reinforcements did not arrive from Tobelo until 11 a.m. that morning and were only able to assist in the evacuation of Wangeotak. The Kaos, including villagers from the northern part of the sub-district, attempted several counterattacks on 20 and 21 August. They destroyed the infrastructure in the eastern part of Malifut, including the large Malifut market and bus terminal. However, military personnel from Tobelo eventually repelled these attacks. The exact intentions of the two communities involved in the first hours of fighting on the evening of 18 August cannot be known with certainty. Unsurprisingly, members of both communities claim the other initiated the violence. It seems unlikely that the Kaos of Sosol and Wangeotak, vastly outnumbered by surrounding Makians, would have launched an attack. The two Kao villages are approximately

Initiation – Malifut 63 15–20 km away from the closest reinforcements and would surely suffer defeat in any conflict with the Makians unless they had first arranged assistance from their wider community. It is possible that the fighting may have begun spontaneously in the midst of a party involving alcohol and loud music. Yet the scale of the Makian attack against the two villages, along with the surrounding tension over the refusal of these two villages to accept PP42, suggests that a degree of organization was also involved on the part of the Makians. The attack also eliminated the two villages seen as posing the greatest obstacle to the viability of the new sub-district. It seems likely that it was members of the Makayoa student group who provoked and organized this attack.

Government partiality On Saturday 21 August the new district head of North Maluku, Rusli Andiaco, the district head of police, Lt Col. Didik Prijandono, and the Sultan of Ternate, Mudaffar Syah, travelled from the district capital, Ternate, to pacify the situation. The sultan, at that time the chairman of the North Maluku District parliament, was highly respected by the Kaos as a traditional leader.77 The delegation met with community leaders from both Kao and Malifut. Kao leaders say they demanded two things: that the government rebuild Sosol and Wangeotak and that PP42 be cancelled. However, several more belligerent community leaders pressed for a further demand, that the entire Makian community leave Malifut.78 The Sultan of Ternate managed to pacify the Kaos. He agreed with the Kaos that the Makians had violated local traditions and that the four sub-ethnic groups of the Kaos should not be separated. But he also insisted that the problem had to be resolved through traditional forms of resolution, including meetings between community leaders, and not through violence.79 As will be seen in later chapters, different perceptions of the sultan’s response to the Kaos’ demands and to the 18 August riot were central to the development of violence later in 1999. Many members of the Makian elite believe that the sultan told the Kaos that PP42 would be cancelled and that he supported retaliatory action. They believe the Kaos therefore considered that they had permission to carry out further attacks against Malifut. Yet the Kaos initially attempted diplomatic efforts to resolve the conflict, which suggests that the sultan did not recommend retaliation. Kao community leaders formed a delegation called the Team of Nine to make representations to the government in Ternate.80 The Team of Nine made two trips to Ternate, the first to meet with the Sultan of Ternate, the second to meet with members of the North Maluku parliament. The delegation told the sultan that since the Makians had violated hak adat (traditional rights) and destroyed the houses and churches of the indigenous population, they had to leave Malifut.81 The sultan again ordered the Kaos not to break tradition and told them they were forbidden to expel the Makians.82 The Team of Nine’s second visit to Ternate in late September was to the district parliament. The Kaos demanded the cancellation of PP42 and the rehabilitation of

64 Ethno-Religious Violence in Indonesia Sosol and Wangeotak. According to one member of the Team of Nine, there was already a feeling among some members of the district parliament that the rioting was caused by religious animosity, a claim the Kao representatives strongly denied.83 The members of the Team of Nine were also forced to refute an accusation from a member of the assembly that it was Muslim Kaos who had destroyed churches in Sosol and Wangeotak, not the Makians. 84 Following the return of the Team of Nine in late August, the Kaos waited for funding to reconstruct the two destroyed villages, while IDPs from those villages lived in temporary accommodation in the capital. There was little response from the district government, and neither the communities of Sosol and Wangeotak nor the Kao Sub-District government received any funding for the reconstruction of the villages. The government did not rescind PP42 nor was any investigation carried out into the attack and no charges were laid against anyone involved. There are several explanations for this lack of response from the district government. It is likely that many officials were busy with preparations for the inauguration of the province, to take place on 12 October. In addition, the Team of Nine was not particularly influential in Ternate, partly because there was only one Kao in the district parliament. When compared to the political influence enjoyed by the Makians in North Maluku, this weakness was all the more apparent. The Sultan and the several Christian members of parliament from Halmahera were in a substantial minority in the parliament and unable successfully to represent the Kaos. Indeed, the Sultan of Ternate did little to press the Kaos claims in Ternate, perhaps focusing more on his political campaign to become governor and unwilling to jeopardize this campaign by becoming involved in the incident. More importantly, most members of the parliament, especially Makian members, believed that since PP42 was already authorized by the Department of Home Affairs, the Kao community would therefore be forced to accept it. It appears that they also felt that the Kaos were not in a position either to lobby Jakarta directly to have PP42 cancelled or to retaliate for the destruction of Sosol and Wangeotak. It is likely that Makians in Ternate, and the government in general, considered that the military unit stationed on the north-eastern border of Malifut was adequate to resist small attacks like those that had come from Kao after the destruction of the two villages. There was clearly little foresight of the events that were to come in Malifut.

The October riot When the Kaos received no response from the local government throughout October, their perception of unjust treatment at the hands of a Makian-dominated government increased. The Kaos’ anger at the Makians for destroying Sosol and Wangeotak was compounded by this perceived bias of the government, a government that had failed to provide funds for the villages’ rehabilitation or prosecute anyone involved in the incident. For this reason, the Kaos’ focus shifted from diplomatic efforts to preparations for physical retaliation, led by a highly respected and feared Kao, Benny (Bernard) Bitjara.85 The Kaos made a large number of

Initiation – Malifut 65 traditional weapons such as bows, spears and machetes. Many also made a powerful variant of the Molotov cocktail from sulphur extracted from bombs found on the several Japanese World War II battleships sunk off the coast of Kao. On the morning of Sunday 24 October a clash occurred between Muslim Kaos and Makians in coconut fields in Kalijodo, the area bordering the two sub-districts. Versions of this incident from the two sides differ sharply.86 What is clear is that a large group of Makians requested permission from the military officer in charge of guarding the eastern edge of Malifut to enter their coconut gardens, which were located in no man’s land between that military post and the border of Kao. At that time, as many Christian Kaos were attending church, the border was guarded by approximately 12 Muslim Kaos.87 The intention of this group of Makians is a matter of disagreement. Makians involved maintain that they were merely planning to gather coconuts.88 All Kaos interviewed, however, including several who were guarding the border at that time, say that at least a hundred armed Makians in trucks, having passed through the military line, directly attacked the Kao guards, and obviously intended to attack the capital, Kao.89 The Kaos claim the men guarding the border held back the Makians using bows and arrows until reinforcements arrived following the church service. A number of factors nevertheless suggest that the Makians did not intend to attack that day. After two months of tension and restricted movement, the Malifut community was very short of food.90 Therefore it is highly likely the Makians were gathering coconuts and the dangers of working so close to Kao may explain the large number of armed men who entered the field. It is also unlikely that 12 Kao men with bows and arrows could have stopped the advance of over a hundred men in trucks. According to the Kaos, the Makians had only machetes, not bombs or petrol, which are the usual weapons for attacking a village. In addition, a senior military officer present at the time of the clash stated that the Makians were definitely just gathering coconuts, and were then attacked by a large number of Kaos.91 The information available suggests that the Makians did not intend to attack the Kao capital, at least on 24 October. There are, however, much stronger indications that at least some Kaos had planned a large-scale assault on Malifut and used the clash in the coconut field as a pretext to launch this attack, namely the massive and rapid mobilization of a huge militia and the great preparation demonstrated by the attack. The man who had led Kao preparations for military action over the past month, Benny Bitjara, was also present in Kao that weekend, having travelled from his home in south Tobelo Sub-District and led the large Kao attack the following day. Following the initial fight in the coconut field, during which, around five Makians were killed (and no Kaos), the Makians were driven back to Malifut, and the Kaos reassembled in the capital along with others from the large northern villages of Pediwang and Gamlaha. Reportedly notified by messengers travelling in cars, Kaos from the west of the sub-district joined them in the capital.92 In the early morning of 25 October, Benny Bitjara led perhaps 5,000 armed Kaos in a massive attack on Malifut.93

66 Ethno-Religious Violence in Indonesia In the attack, the Kaos used sulphur bombs, a small number of homemade firearms, bows and arrows and spears. Almost all men carried machetes. The Makians attempted to defend their villages with only stones, and were completely overwhelmed. Because of the rising tensions associated with PP42 and the initial riots, the majority of the Malifut population had already fled over the preceding weeks, leaving only approximately 5,000 (mostly men) in Malifut.94 The military unit guarding the border abandoned any attempt to prevent the two sides from clashing because the number of Kaos attacking Malifut was far too large. Among the Makians there were hasty deliberations over the best response. The village head of Tahane argued for attempting to hold out and guard Malifut. However, the majority of community leaders decided the Kao force was too great and opted to evacuate the population to the sub-district military compound in Malifut.95 The Makians had resisted the attack for less than an hour. The speed with which the Makians were driven from Malifut meant that only three people were killed.96 Several Kao leaders stated that their goal was to drive the Malifut community from the area, not to kill large numbers. However, the scale of destruction that took place as the Makians were evacuated was massive. Every house in all 16 villages was either bombed or burned. The entire infrastructure of the sub-district was destroyed, including the sub-district government office, although several schools were left untouched.97 Before the attack, Benny Bitjara and other Kao leaders had ordered militia members not to destroy mosques. Following the expulsion of the Makians, Benny Bitjara took a military officer to each mosque in order to confirm that it had not been destroyed.98 The Kaos also raised an Indonesian flag in front of each one. Bitjara gave the order to protect the mosques because, first, Muslim Kaos were involved in the attack and, second, he sought to demonstrate that this conflict was not ‘about religion’.99 Representatives of the district government arrived in Malifut to assess the situation, meeting Makian leaders in the military base and also calling the Team of Nine from Kao. The members of the Team of Nine demanded the Makians be taken from the area, and Makian community leaders stated their unwillingness to return to their villages. The remaining Malifut population was then evacuated by truck to Sidangoli and from there by speedboat to Ternate.

Why did violence come to North Maluku? After two decades of coexistence and economic cooperation, the relationship between the Kaos and Makians quickly deteriorated in 1999. In August this animosity spilled over into fighting leading to the destruction of two Kao villages, and in October to the razing of Malifut to the ground. The case demonstrates the complexity of civil conflict, illustrating the interaction of structure and agency, emotion and reason, and leadership and mass sentiment that plays a role in causing violence. The following section discusses the importance of structural conditions such as political inequality, attachments to territory and religious sentiment, along

Initiation – Malifut 67 with the agency involved in the mobilization to form a new sub-district and in the August and October riots themselves. Mobilization and counter-mobilization Events in Malifut before and during this violence concur closely with the patterns suggested in Resource Mobilization theory. The relationship between the Kaos and Makians on Halmahera had not been particularly intimate before the conflict and occasional brawls had broken out between youths from the two communities. Yet, while largely segregated, the two communities had never experienced large-scale violent conflict, and the relationship had not been affected to any great degree by the conflict in Ambon. Successful mobilization by members of the provincial Makian community to divide Kao Sub-District and provide the Makians in Malifut with their own territory changed the relationship, perhaps irrevocably. This campaign was stimulated by changes in the prevailing ‘opportunity structure’. The impending pemekaran of the province and the creation of a new North Halmahera District threatened to place Malifut under the control of Tobelo, at a time when new decentralization laws had increased the influence of district governments over sub-district or village-level governments. At the same time, the new decentralization laws, along with the presence of a lucrative gold mine, presented the community of Malifut with the opportunity to gain substantial financial benefits if it were to become the capital of one of the new districts created in place of the old North Maluku District. The strong Makian ethnic network throughout the province and the influence of the Makian elite, allowed the Makians to achieve their goal without consulting the Kaos. The influence of the Makians in the government and bureaucracy also led to these institutions ignoring the Kaos’ subsequent opposition. Threatened by the success of the Makians, the Kaos launched a strong grass-roots counter-mobilization to delegitimate the new sub-district. The loss of the gold mine from Kao territory caused concerns among the community about the loss of employment and other revenue. Their conviction that they were being deprived of resources that had traditionally belonged to them increased a feeling of frustration among many members of the community. The economic benefits of the territory of Malifut, particularly the gold mine at Gosowong, were obviously a central motivation in this early stage of the spiral of mobilization and counter-mobilization. Attachment to territory But the rational element of this competition alone does not explain why the two communities resorted to violence. For the Kaos, territory was not just a source of economic livelihood in the form of copra crops and more recently gold, but was also central to their sense of ethnic identity. The Kaos had all the connections with territory identified by Toft: they had lived for centuries on the land; the territory bore their name; and they had shed blood in their defence of their land. The boundaries of the new sub-district also threatened another important element of Kao

68 Ethno-Religious Violence in Indonesia ethnic identity, the unity of all four sub-ethnic groups. The Kaos’ counter-mobilization was therefore infused with emotion. It was nevertheless non-violent and institutionalized in the first instance. The issue of territory was also volatile in this case because it was central to the inter-subjective understandings between the two communities. Control over the territory of Malifut was the one advantage the Kaos still held over the more politically powerful and economically successful Makian migrants. While the Kaos suffered from a lower level of development, education and representation in the local government, their sense of deprivation stemming from this inferior position had been assuaged by the knowledge that the Makians lived, as their guests, on Kao land. This status was now at stake – a highly emotional issue for the Kaos. The Kaos felt that the creation of a separate sub-district in Malifut violated their status as the indigenous community of the area by reassigning ownership of their land without their agreement, separating members of the Kao community and using the name of a migrant group for the new sub-district. As such, in line with Horowitz’s theory, perceptions of relative group worth played a central role in the violence in Malifut. For the Makians, too, territory was central to their understanding of their own status relative to the Kaos. Many had become frustrated by uncertainty over their right to live in Malifut and by the Kaos’ continuing claim that they lived on Kao land. Many Makians, particularly the Makayoa student group, were angered that the Kaos were attempting to prevent what was to them a long-overdue process: the recognition of the Malifut community as the official owners of the territory on which they had resided for 25 years.100 Their resentment was intensified by their perception that the Kaos might prevent them from profiting from the lucrative gold mine. The strength of these emotions led to a more belligerent response to Kao opposition (especially by the Makayoa student group) and eventually the destruction of the two Kao villages that posed the greatest obstacle to the viability of the new sub-district. Therefore, as in most, if not all, cases of communal conflict, the interaction of both material and non-material factors increased the likelihood of violence. Government partiality A further crucial factor in the riots in both August and October was government partiality. The Malifut case strongly suggests that violence can become more likely if both communities perceive the authorities to be biased in their treatment of the dispute. In such cases, one party may feel aggrieved at the partiality of the authorities, while the other may feel a sense of empowerment from the apparently assured government support. In the Malifut case, violence was initiated by a group, the Makian, which considered itself to have political and economic weight behind it. In ignoring Kao concerns and finally in attacking Sosol and Wangeotak, the Makians were motivated not just by economic interest and frustration at the uncertain status of Malifut, but a sense that they could act with impunity because of the backing of the district government.

Initiation – Malifut 69 A lack of government impartiality also partly explains the actions of the Kaos in October. When the district government failed to respond to any of their demands following the August riot, Kao perceptions of bias on the part of that government increased. In line with Meyer and Staggenborg’s argument about counter-movements, this prompted members of the Kao elite to employ a ‘non-institutionalized’ strategy to counter the success of their opponents – namely large-scale preparation for conflict and eventually massive violent retaliation against Malifut. Therefore the manner in which the sub-district was formed acted as a catalyst to transform the tension concerning the territory into violence. Religious sentiment Given that the conflict in Ambon had been raging for seven months by August, it is necessary to consider the role of religion in this first large violent incident in North Maluku. There is little doubt that, for some among the Makayoa, the experience of religious conflict in Ambon increased their antagonism towards the predominantly Christian Kaos. For the Kao community, too, stories of the conflict in the south of the province probably increased their suspicion of, and animosity towards, the Makians throughout this period. Inter-religious tension was perhaps increased by the presence of a small number of Ambonese IDPs in Kao.101 Shouts of ‘Allahu Akbar’ from the Tahane mosque in Malifut and the destruction of the churches in Sosol and Wangeotak demonstrated some degree of religious tension in the Makian attack in August. However, the evidence strongly suggests that religious sentiment or tension was not the primary cause of either the August or the October attacks and motivated only a minority of the individuals involved. The violence in Malifut can be seen as the result of months of tension arising from the creation of the new sub-district in Malifut. The initial dispute was related primarily to those factors addressed above: ethnic solidarity and competition; the importance of traditional land to the Kaos; and the Makians’ frustration at the refusal of the Kaos to recognize what they saw as their rights. Illustrating the non-religious character of the conflict, Muslim and Christian Kaos maintained ethnic solidarity in the face of what was ultimately ethnic antagonism, Muslim Kaos joining in retaliatory attacks against the Muslim Makians. No other Muslim or Christian ethnic groups assisted the Makians or Kaos until members of the Makian elite subsequently reframed the conflict in terms of religion, as discussed in the next chapter.102 Kao relations with other Islamic communities in the area, including Javanese transmigrants, remained civil throughout 1999 until religious violence elsewhere forced them to flee the island.103 Agency The actions of several influential individuals brought the emotional impact of these macro conditions (ties to territory and government partiality) to the fore. These actions translated the tension in the sub-district into violence. The presence and rhetoric of the Makayoa student group in Malifut in August played a crucial

70 Ethno-Religious Violence in Indonesia role in the attack on 18 August. Responsibility for the violence also lies with those elites who mobilized students both to engender support among the Malifut community and to intimidate the Kaos. Most respondents interviewed regarding this mobilization claimed it was carried out by Yusuf Abdurrahman, the Makian former rector of Khairun University. In pressing through PP42 without any consultation of the Kaos and then ignoring and delegitimating Kao objections and claims for compensation, many members of the district government paved the way for the August and October riots. The leadership of Benny Bitjara was also crucial in the latter event. His determination to launch forceful retaliation against the Makians for violating traditional rights, and his charismatic leadership, played a large part in preparing the Kaos for the October attack.

Conclusion By the end of October all the villages that were to constitute the new sub-district of Malifut lay in ruins. The Makians of Malifut were again IDPs, compelled to leave this time not by force of nature or by government decree, but by overwhelming physical assault. This chapter has asserted that what made the creation of a sub-district in Malifut such a volatile issue was the importance of the territory in question to all parties: it was simultaneously economically valuable; important to the ethnic identity of one group; and a crucial element of the inter-subjective understandings of the two communities. Government partiality exacerbated this tension, as did the actions of several individuals among the Ternate elite. The ethnic clash between the mainly Christian Kaos and Muslim Makians subsequently sparked religious war throughout almost all areas of North Maluku, as discussed in the following chapters. However, while the North Maluku conflict became religious in character, this chapter has shown that it did not begin as such. The initial conflict stemmed from local issues. For the Kaos in particular, the dispute was ethnic in character, a fact underscored by the assistance given by Muslim Kaos to Christians in the October attack on Malifut. As will be seen, Muslim and Christian Kaos continued to maintain ethnic solidarity throughout the subsequent inter-religious North Maluku conflict.104 Yet Christians still living unaware in Ternate and Tidore were to become the next victims as the province moved tragically towards religious war.

4

Escalation – Ternate and Tidore

Introduction It was in Ternate and Tidore, North Maluku’s historic centres of political and religious power, that the violence took a religious turn. For centuries, Christians had enjoyed relative security in the regional capitals despite the islands’ strong Islamic heritage. But in November 1999, in the wake of the Malifut conflict, they appeared to be caught up in a wave of Muslim anger. Rioting mobs targeted Christians regardless of ethnicity, and destroyed their homes and churches. The violence in these two cities set off a wave of killing and destruction that engulfed the entire province. This chapter considers not only why violence spread to the two major provincial centres, but how and why the conflict escalated from a border dispute between ethnic groups into an inter-religious war involving all ethnic communities in the province.1 It examines the common claim that the riots were the inevitable result of a flood of IDPs from Malifut and the rumours that accompanied them. It concludes that while the condition of these IDPs elicited sympathy from Muslims in Ternate this was not the primary reason for the spread of violence to the capital, nor for the dramatic emergence of religion as the primary marker in the conflict. By the time the violence had spread to Tidore and Ternate it displayed a marked degree of organization. Members of the Makian elite portrayed the earlier clashes in Malifut as religious in character so as to displace blame from their own community for these events and to obtain sympathy from the wider Islamic community for retaliation against the Kaos on Halmahera. When prevented by the security forces from returning to launch retaliatory attacks, the group focused their attentions on Christians on Tidore and Ternate. To make these attacks possible, powerful members of the elite immobilized the only institutions capable of halting the violence, the military (TNI) and police (Polri). In this, these elites found ready allies among other high-ranking politicians with a different goal – to undermine the daunting political support enjoyed by their political rival, the Sultan of Ternate. The following section will outline several theoretical considerations of how violence can escalate; the remainder of the chapter then describes and analyzes the riots themselves.

72 Ethno-Religious Violence in Indonesia

Leadership and mass sentiment in the escalation of conflict A central question in the study of violent conflict concerns the relationship between elite leadership and mass-level sentiment and action. Is conflict most often created from above, as a means of achieving the interests of political or economic elites, or do interests and prejudices among, and the actions of, ordinary community members constitute the predominant cause of violence? As the following discussion on the escalation of violence and the case addressed in this chapter demonstrate, this relationship is often complex. It indicates that, often, neither can be considered in isolation from the influence of the other – leadership and mass-level factors are invariably mutually constitutive. The escalation of violence from a minor, non-sectarian dispute into widespread communal conflict often appears to occur almost naturally. Even isolated criminal acts or traffic accidents can eventually escalate into large-scale sectarian violence when ethnic or religious sympathies attract large numbers to the dispute. Charles Tilly asserts that when parties to a minor incident call on their friends and family to support them, they can inadvertently redefine the situation as sectarian. Tilly terms this process ‘network-based escalation’.2 Anonymous rumours often exaggerate the extent of minor incidents and provoke anger and, ultimately, sectarian rioting. These rumours, often combined with the movement of refugees and other manifestations of the dispute, produce anger and feelings of insecurity, reduce the apparent feasibility of peaceful solutions and legitimize militant actors within society.3 When escalation is viewed in this manner, an initial violent incident is seen as spontaneously igniting the ‘dry grass’ of long-standing tensions between two or more communal groups. Reframing conflict Yet, in many cases, escalation of a conflict is more intentional. In such cases, individuals deliberately portray an initial dispute as sectarian in nature as a means of achieving some political or economic goal.4 In his study of communal violence in South Asia, Stanley Tambiah shows how immediately before many cases of widespread communal rioting, members of the political and economic elite had publicly misrepresented previous interpersonal, business or other disputes.5 They had removed any acknowledgement of the specific context and local character of incidents, distorting them so that they ‘appeal to larger deeper, more emotive and enduring loyalties and cleavages’.6 Tambiah terms this strategy ‘focalization’ and ‘transvaluation’. Paul Brass agrees, pointing to the fact that incidents similar to those which trigger communal rioting have occurred on numerous other occasions and passed almost unnoticed. According to Brass, the difference in outcome depends on the surrounding political context and who stands to gain and who to lose from reinterpreting or politicizing such events.7 Undoubtedly many cases of communal rioting are a direct consequence of the propaganda efforts of members of the elite. When elites exhort their followers to avenge murders committed within places of worship, the brutal rape of women and other examples of inhumanity,

Escalation – Ternate and Tidore 73 they can provoke fury that is only assuaged by violence against the apparent perpetrators. However, in some cases it is possible to overstate the role of propaganda in explaining why people engage in sectarian rioting. In fact, propaganda, particularly when carried out anonymously, is sometimes used to make rioting appear spontaneous when it was in fact planned and orchestrated. As stated above, Brass demonstrates that communal riots are often preceded by a portrayal of everyday incidents as sectarian in nature. However, he also makes clear that these riots are often instigated and carried out by ‘riot specialists’. These ‘specialists’ are not acting because they have been plied with, and believed, propaganda, but because they have a direct interest in violence. In their review of six books on cases of ethnic violence, James Fearon and David Laitin come to a similar conclusion. They assert that the evidence suggests that in many cases where members of the elite appear to have mobilized a form of ‘false consciousness’ in their followers in order to get them to engage in violence, the masses were ‘not duped at all’. Rather they have acted out of their own interests, ‘such as looting, land grabs, and personal revenge’.8 Horowitz and other scholars demonstrate that once rioting has begun, larger numbers of participants will subsequently join the violence, including a wider section of the community with no direct involvement in preceding incidents.9 Many of these people become involved for a range of reasons, including excitement, the opportunity to loot victims’ houses and a sense that the violence erupting on the streets and the destruction of buildings confirm the content of recent propaganda. Underlying this participation is a sense that they can take part with impunity, having seen that the rioting is not prevented and no punishment is incurred by participants. The reaction of the security forces is therefore crucial to the onset and scale of rioting. The response of the security forces to a particular outbreak of killing and burning may be influenced by the prejudices of commanders or by less malign factors, such as incapacity or incompetence. Yet in many other cases, their response is directly tied to the interests of civilian leaders who exercise authority over them. Tambiah has shown that in Trilokpuri, India, politicians mobilized crowds for violence while simultaneously putting pressure on the security forces in order to ‘immobilize’ them.10 Similarly, considering the incidence of inter-religious violence in India, Steven Wilkinson concludes that parties whose platform is based on ethnic or religious identity sometimes stimulate sectarian tension as a means of winning the votes of members of their target community who would otherwise vote along lines other than ethnic or religious affiliation.11 This is sometimes achieved through holding marches or other activities that provoke counter-mobilization by members of the other ethnic or religious community. Whether these local provocations lead to violence depends on the electoral incentives of the party controlling the state government. If the current government enjoys the support of the minority group at risk or may need that group’s support in the future, it will almost always order the security forces to prevent attacks against that group.12 If it does not require its

74 Ethno-Religious Violence in Indonesia support but is at risk of losing an election to an opposition party that does, incumbent politicians will be far less likely to protect that community. Wilkinson states that ‘abundant comparative evidence shows that large-scale ethnic rioting does not take place where a state’s army or police force is ordered to stop it using all means necessary’ but that the police and military will only intervene and halt violence if ordered to do so by their civilian, political superiors. 13

Ternate and Tidore in 1999 The two volcanic islands of Ternate and Tidore, home to two great historical regional powers, are separated by a deep but narrow tract of water. Having amassed substantial power from the spice trade, these two kingdoms projected influence to far-flung territories such as Ambon, Papua and Sulawesi. Each manipulated global rivalries to further their own interests. Over this period, Ternate and Tidore also came to embody the strong Islamic character of the region. Their rulers converted to Islam in the sixteenth century, assuming the title of sultan, and the religion spread quickly throughout the archipelago. For centuries, the sultans of the two islands were the traditional rulers of the region, respected equally by Christians, Muslims and animists. Both the Ternate and Tidore sultanates have declined in power and political importance over the ensuing centuries. During the colonial era, most sultans in Indonesia were stripped of formal political power – a process which continued into the post-independence era. However, as traditional figures of great political and religious authority, many sultans exercised substantial ‘charismatic’ influence over local communities. Many retained traditional palaces (kraton or kedaton) and numerous staff. Some used this informal influence to assume positions of real political power within the former regime of President Suharto. This was the case with the Sultan of Ternate. Although he held no formal constitutional power by merit of his traditional position, the Sultan of Ternate enjoyed strong loyalty within communities in Ternate and north Halmahera. In 1999, the current sultan, Mudaffar Syah, retained a palace (kedaton) and a large palace guard (colloquially known as the Pasukan Kuning) and regular ceremonies were held in his honour. Ternates continued to believe that the sultan possessed strong spiritual or magical powers which were intimately connected to the volcano that looms imposingly above Ternate City. In addition to this traditional influence, the sultan had assumed actual political authority during the Suharto era. By early 1999, Mudaffar Syah was chairman of the North Maluku branch of the nationally predominant political party, Golkar, and as a result of this party’s dominance in local politics, was elected chairman of the district parliament. The Tidore Sultanate had not retained the same level of traditional influence in North Maluku as its neighbour and historical rival, and it became vacant after independence. Over ensuing decades, the palace of the sultanate on the eastern side of Tidore fell into disrepair. In 1999, however, not long before the new province of North Maluku was to be inaugurated, the district-head of Central Halmahera, Bahar Andili, nominated a new sultan, Djafar Syah. The district parliament

Map 4 Ternate

76 Ethno-Religious Violence in Indonesia supported the nomination and, as the new province was inaugurated, he was installed as sultan.14 In early 1999 Ternate was the capital of North Maluku District, which comprised the northern half of Halmahera, and several islands, including Makian and Morotai. After the inauguration of North Maluku Province in October that year, Ternate became the interim provincial capital.15 In early 1999 the city also assumed municipality status and gained its own district government, bureaucracy and parliament. Following the June 1999 general election the North Maluku District parliament was dominated by the Golkar Party (of which the sultan was chairman) which had won 17 out of 40 seats. The party with the next highest representation was the other major secular nationalist party, the Indonesian Democratic Party – Struggle (PDI-P), with eight seats.16 At that time, Tidore was the capital of Central Halmahera District, which spanned the traditional area of the Tidore Sultanate’s influence, including the southern half of Halmahera, Bacan and other islands in the south of the region.17 In 1999 the Central Halmahera District parliament was also dominated by Golkar, with 11 seats, while the Islamic United Development Party (PPP) and PDI-P had five and four seats respectively.18 The population of the island of Tidore was approximately 52,000, with 25,000 people living in and around the capital Soasio (see Map 4.2). Most of the island’s population was from the Tidore ethnic group but sizeable Makian, Sanana, and Kayoa minorities also lived on the island. Ternate has long been the economic centre of North Maluku. The city possesses the province’s key port and a large quantity of exports passes through en route to Surabaya and North Sulawesi (Bitung), as well as to international destinations including Japan, South Korea, Singapore and Hong Kong.19 Thriving service, retail and transport sectors have developed around this trade and around the district and municipality governments based in the city. It also possesses the region’s two universities – Khairun and Muhammadiyah. In contrast, the Tidore economy revolves solely around the local government and bureaucracy. Over the past decades, large numbers of Tidores have departed for Ternate to seek employment or to attend secondary or tertiary educational institutions. Until late 1999 several hundred migrants, including Christians and Chinese, lived in the suburb of Indonesiana and worked in Soasio. Indeed, not just Tidores but large numbers of migrants from all ethnic groups across the region were drawn to Ternate City in the last decades of the twentieth century. The majority came from the nearby islands of Makian and Sanana, along with Tidore. These migrants mostly settled in the southern suburbs of Bastiong, Mangga Dua and Takoma (see Map 4.1). As a result, Ternate was the most densely populated island in the archipelago, with a total citizenry of almost 100,000 in 1999. As well as swelling the population of Ternate, these communities also became well represented in the North Maluku District government and bureaucracy. As discussed, of these ethnic communities, the Makians were particularly successful, becoming prominent in local government as well as in the region’s educational institutions. The success of the Makians (and to a lesser extent the Tidores) was in stark contrast to the indigenous Ternate ethnic community. Less

Escalation – Ternate and Tidore 77 inclined to undertake tertiary education, the Ternates were under-represented in the government and bureaucracy and few became lecturers, staff members or students in the two universities in Ternate. Ternates lived predominantly in the northern suburbs of Dufa Dufa and Santiong near the sultan’s palace. The city was therefore segregated between the north, dominated by the indigenous Ternates, and the south, where most migrants resided. Both islands were overwhelmingly Muslim, with approximately 90 per cent and 95 per cent of the populations of Ternate and Tidore identifying themselves as Muslim. The Tobelo-based Protestant Church, GMIH, and the Ambon-based GPM (Gereja Protestan Maluku – Protestant Church of Maluku) both had congregations and churches on the islands. A small Catholic community also resided in Ternate, constituted mostly of migrants from Southeast Maluku, East Nusa Tenggara and other areas. The Catholic Church also ran a busy hospital in the centre of the city. Chinese Indonesians, most of whom were Protestant Christians, were highly visible in the economic life of both islands, particularly Ternate. They owned most shops and a large number of restaurants and were often involved in the transportation sector servicing the export trade. Relations between Muslims and the Christian minority had always been good on both islands, particularly on the more heterogeneous Ternate.20 One Christian resident in Ternate stated that while fights broke out occasionally between neighbouring areas (kampung) of the city, they were not usually between followers of the different religions and did not concern religion. At least one commentator has suggested that the introduction of more formal religious education in schools run by the national Islamic organization, Muhammadiyah, led some Muslims to take a more dogmatic stance towards non-Muslims in the decade before the conflict.21 However, most Muslims generally agreed that Islam in Ternate was not ‘ideological’ and Muslims were tolerant of followers of other religions.22 Christians generally concurred, and, as will be seen, most stayed in Ternate right up until rioting broke out, feeling secure despite rising inter-religious tension in the city. In Tidore, too, most respondents considered relations between the two religious communities to have always been clement. Muslim residents of Tidore described a high degree of ethnic pluralism, particularly in the Soasio suburb of Indonesiana. However, several Christian respondents from Tidore suggested that Muslims on that island had long been more exclusive or ‘fanatic’ and less tolerant of non-Muslims than those in Ternate. One former resident of Indonesiana related how Muslims refused to buy food from Christian traders or accept foodstuffs as gifts from Christians.23 The inter-religious conflict in Ambon in early 1999 had a greater impact on Ternate and Tidore than on other areas of North Maluku. Many individuals from the two islands had been working as civil servants in the provincial capital at the time of the outbreak. In addition, most students who had been studying at Ambon’s Pattimura University returned to Ternate after the violence in Ambon to attend Khairun University or seek employment. Inter-religious tension also increased on Tidore. One Muslim leader on Tidore, Abu Bakar Wahid, told me he returned from witnessing the outbreak of violence in Ambon in January 1999 and ‘prepared’ to

78 Ethno-Religious Violence in Indonesia ensure that the local Muslim community did not suffer the same fate as that in Maluku.24 Some Muslims who had returned from Ambon attempted to intimidate Christians by pelting their houses with stones.25 Yet to a large extent concerns in Ternate and Tidore caused by the violence in Ambon were secondary to the political transition taking place in North Maluku in 1999. As discussed in Chapter 2, Ternate and Tidore were once again emerging as regional rivals, the two islands at the centre of the political struggle to separate North Maluku from Maluku Province. When it became clear that North Maluku was finally to assume provincial status, the region’s elite inevitably divided into several factions. These surrounded the main contenders for the governorship: Mudaffar Syah (the Sultan of Ternate); the brothers Bahar and Syamsir Andili (the District Head of Central Halmahera and the Mayor of Ternate City respectively); and Thaib Armain (soon to become the head of bureaucracy for the new province). Originally united in the goal of seceding from Maluku, these factions saw tensions begin to rise over several issues. Two issues in particular caused the most friction between the main political factions – those surrounding Mudaffar Syah and Bahar Andili. First, these two men were the main contenders for the first governorship in the new province, and second, each attempted to have the provincial capital located within his own district. Both issues had major implications for the patronage networks of each. The future governor would have immense influence over the appointment of senior civil servants, which would in turn determine employment throughout the government in general. The location of the capital would also bring immense revenue to the surrounding district through the construction of facilities, allocation of funding from the central government and the employment and revenue associated with local government. The two men rigidly opposed the other’s initial proposals, that the capital be located in Ternate and Soasio, Tidore respectively. Each subsequently proposed locations on Halmahera, again within their own districts. Mudaffar Syah suggested that the port of Sidangoli, directly across the strait from Ternate, be made the capital, while Bahar Andili advanced the location of Sofifi, at that time a quiet village within sight of Tidore. As a compromise, the central government ultimately decided that Ternate would act as a temporary capital, while a permanent seat of government would be constructed in Sofifi. 26 Despite failing to achieve his desired outcomes regarding the location of the capital, and several other initiatives regarding recognition of the regional sultanates, the sultan was still in a powerful political position. Supported by a large proportion of the Ternate population, as well as by most Christians on Halmahera, the sultan wielded considerable authority as chairman of the locally dominant Golkar Party for North Maluku and chairman of the North Maluku District Parliament. In 1999 Mudaffar Syah appeared destined to become the province’s first governor, and the Ternates seemed set to challenge the dominance of the Makians and Tidores in the local government and bureaucracy.

Escalation – Ternate and Tidore 79

Rising tension This political struggle led to rising tension in Ternate. In early September, when the central government announced that the provincial capital was to be constructed in the small village of Sofifi in the district of Central Halmahera, several thousand people protested outside the North Maluku District government building. To the protestors, the fact that a team from Central Halmahera had recently travelled to Jakarta to discuss the location of the capital suggested that corruption had swayed this decision. The protesters accused leading Central Halmahera District politicians, including Bahar and Syamsir Andili, of using ‘money politics’ to influence central government officials. The protestors held banners stating ‘Bahar – Syamsir Andili provokator’ (Bahar and Syamsir Andili are provocateurs), accusing them not only of corruption, but also of attempting to destabilize Ternate and the district of North Maluku so as to undermine its claim to hosting the capital.27 The destruction of the Kao villages of Sosol and Wangeotak in Malifut in August had little impact in Ternate or Tidore. The local newspaper carried several short articles, posing the question of who provoked the destruction and why the police and military had not had adequate field intelligence to prevent it.28 The newspaper provided no answers to these questions and any attention to the Malifut dispute soon waned in Ternate and Tidore. Other issues emerged to cause tension around this time. In August, just before the first clash in Malifut, rumours spread around Ternate, and elsewhere in North Maluku, that the date 9 September (9/9/ 99) would bring catastrophe in the form of natural disaster. The scenario received widespread coverage in Ternate, occupying the front page of the weekly newspaper, the Ternate Post.29 As mentioned, the inter-religious clashes in Ambon also caused some tension in Ternate, although much of this was clearly stirred up by individuals with an interest in inciting Muslim hostility towards Christians. In early September, a map was circulated in the city, purportedly drawn by members of the Ambon-based Protestant Church of Maluku (GPM).30 The rudimentary map appeared to be evidence that Christians in Ternate were preparing to launch a coordinated and pre-emptive attack on Muslims in the city. The map divided Ternate into a number of sectors, detailing the strength in numbers of Christians in each area. The drawing was entitled Map of Attack (Peta Penyerangan). Following the circulation of this map, the atmosphere in Ternate became sufficiently tense for the new district head of North Maluku, Rusli Andiaco, to call an open-air community meeting in front of the district government building. A Protestant pastor spoke to the assembled crowd and explained that the map had originally been the work of a student preparing for the ministry. It had originally been entitled Peta Pelayanan (Map of Service) because the trainee pastor had been attempting to divide Ternate into areas to improve the religious service of GPM. He explained that there was certainly no plan on the part of Christians to attack Muslims in Ternate, adding that, as Christians constituted only 10 per cent of the population of the small island, any such plan would be foolish.31 He stated that he did not know how, or by whom, the title of the map had subsequently been altered

80 Ethno-Religious Violence in Indonesia from ‘Map of Service’ to ‘Map of Attack’ so as to appear to be a plan for an attack on the Muslim community. The crowd was sufficiently mollified and left the district government building without incident. The conflict in Ambon had therefore created an atmosphere of inter-religious tension in Ternate and Tidore. However, it seems likely that if it had not been for the dramatic events that touched the city in October, the cities would have remained calm, just as they had for ten months while violence raged in Ambon. There are several reasons why inter-religious violence had not, and probably would not, have occurred in the two district capitals. While most people in North Maluku were disturbed by the violence in Ambon, they felt it was sufficiently removed in character and geography to be largely irrelevant to them. The relatively small number of Christians in North Maluku as a proportion of the population (approximately 20 per cent), but particularly in Ternate (less than 10 per cent) and Tidore (approximately 2 per cent), also reduced concerns on the part of Muslims that they might come under threat from local Christians. More importantly, Ternate was in the grip of ‘new province fever’.32 Most high-level political leaders in North Maluku had for the past six months been united in the task of achieving provincial status for North Maluku, including both Muslim and Christian politicians. For example, Thaib Armain, who in late 1999 assumed the position of head of bureaucracy for the new province, and the Christian political leader from Tobelo, Hein Namotemo, cooperated until November to ensure that the provincial capital was located on Halmahera.

The aftermath of Malifut The complete destruction of Malifut in October, and the arrival of thousands of IDPs, dramatically changed the atmosphere in the two cities. Shocked Makian IDPs told stories of how the Kaos had stormed their villages in their thousands killing defenceless people. They also claimed that all mosques in Malifut had been destroyed, and that a copy of the Qur’an had been torn up and thrown in the street.33 The sight of thousands of traumatized and destitute Makian men, women and children enraged many of their ethnic kin living in Ternate. Many had family members and friends among the IDPs. Ternate municipality parliamentarian Wahda Zainal Imam, a Makian, formed an organization to provide food, medicine, shelter and donations for the IDPs. 34 The Sultan of Ternate had exacerbated this tension among the Makian community by opposing the immediate evacuation of the Malifut community to Ternate. He had argued instead that the Makians should stay on Halmahera until a settlement could be reached between the two communities. This infuriated many Makians in the city, who already believed the sultan was supporting the Kaos in their efforts to cancel the formation of Malifut Sub-District. Disregarding the sultan’s objections, the Mayor of Ternate, Syamsir Andili, organized the transportation of the IDPs to Ternate, the one area with adequate resources to cope with so many people. The sultan’s apparent support for the Kaos and his attempt to prevent the evacuation of the Makians to Ternate brought to an end his political alliance

Escalation – Ternate and Tidore 81 with Thaib Armain. The sultan’s perceived bias and Armain’s strong sense of sympathy for his fellow Makians eradicated any common ground the two had enjoyed regarding the location of the provincial capital and other issues. On 25 October, after news of the events in Malifut reached Ternate, a group of around a hundred largely Makian students attacked and punched Christian students from Halmahera on the Khairun University campus in northern Ternate City. The crowd chased dozens of Christian students from the campus. The rector of the university, Rivai Umar, subsequently announced that Christian students should not return to the campus, citing concerns over potential future disturbances.35 Sporadic murders and assaults took place over the following days. Several Christians were killed near Bastiong port in the southern part of Ternate City, an area dominated by Makians and Tidores.36 Over this period, some of the city’s more wealthy Christian residents began to leave the island for Halmahera or North Sulawesi. Almost immediately after arriving in Ternate, the IDPs from Malifut, along with Makians living in Ternate, attacked Kao homes.37 On 26 October, the day the majority of the IDPs arrived, a crowd of IDPs attacked and destroyed the house of Jesaja Singa, a Kao member of the North Maluku District parliament. Singa and his wife fled to the district Police Compound (Polres) in central Ternate City. From 26 to 28 October the IDPs and other Makians staged demonstrations expressing outrage at the Kaos’ attack against Malifut. During these demonstrations, Makian youths attacked Kaos’ houses in central and southern areas of Ternate City. The attacks over this period were almost exclusively carried out by Makians and, while several Christians from other ethnic groups were also killed, were targeted primarily at Kaos.

Creating a mob While sporadic attacks against Christians, and in particular Kaos and other Halmaherans were taking place, there were also calculated efforts to mobilize organized violence. The Makian IDPs were initially placed in the suburb of Dufa Dufa, close to the sultan’s palace. Soon, however, several Makian community leaders arrived to organize their transfer to the southern part of the city, an area largely populated by Makians and Tidores. Over the following days, in areas such as Bastiong and Mangga Dua (see Map 4.1), community leaders, imams and politicians held closed meetings with IDPs and other Makians. During these meetings these leaders made provocative speeches attempting to mobilize the audience to take retaliatory action.38 A police intelligence officer present at one meeting in the suburb of Mangga Dua claimed that Wahda Zainal Imam, the Makian member of the Ternate Municipality Parliament from the Islamic United Development Party (PPP), advocated a riot targeting not just Kaos but all Christians.39 Stories also circulated throughout Ternate that a Christian militia, the Pasukan Merah, was gathering on the slopes of the volcano that dominates most of the island. As well as carrying out humanitarian work, the organization formed by Wahda Zainal Imam also attempted to mobilize Muslim youth to join the Makians in

82 Ethno-Religious Violence in Indonesia retaliatory attacks against Kao. One young Tidore man recounted how Makians, including the local youth leader of this organization, put pressure on Tidores and other Muslims to assist in their planned return to Halmahera.40 According to this respondent, the young Makian man declared that if Muslims were not allowed to leave and avenge the destruction of Malifut, Christians in Ternate would bear the brunt of their anger. According to a number of sources, this experience of mobilization was common throughout the southern part of Ternate in the last week of October. On 28 October several hundred Makians did indeed attempt to return to Halmahera and attack Kao. Led by Wahda Zainal Imam, the group assembled near the port on a street in the centre of the city. The crowd was armed with machetes, spears and other weapons and wore white headbands (the first such use of an Islamic symbol). A large contingent of military and police personnel prevented the group from leaving and carried out a sweeping operation confiscating all weapons. The district’s Deputy Chief of Police told Wahda Zainal Imam that the group would not be allowed to depart for Halmahera. According to one Makian, Wahda Zainal Imam then spoke in front of the crowd, proclaiming that the Kaos had attacked the Makians in Malifut because of their faith and, along with other Christians throughout the region, were seeking to dominate north Halmahera. 41 Dispersed by the security forces, the crowd moved through the centre of the city breaking the windows of (largely Chinese) Christian-owned shops and throwing stones at the houses of Christians.42 The house of a Christian Tobelo community leader, Yohannes Namotemo, was destroyed in this disturbance, he and his family forced to run to the police compound. The crowd then proceeded to the suburb of Kalumata in the southern part of Ternate City where they pelted the local Protestant church with stones and set it on fire. Once again, security personnel acted forcefully, halting the riot and arresting approximately 35 people, who were taken into custody at the police headquarters.43 Makian community leaders had thereby successfully mobilized IDPs from Malifut as well as Makians living in Ternate, drawing support from the almost universal anger felt among the Makian community over the destruction of Malifut. This anger led to attacks on Kaos in Ternate, on Christians, churches, houses and shops, as well as an attempt to return to Halmahera in order to attack Kao. The Makians involved in these incidents were not just IDPs or criminals, but included students, bureaucrats and politicians. But, despite stories of the destruction of mosques and the desecration of the Qur’an, few Muslims from other ethnic groups had become involved. Muslims from most other communities, particularly Ternates, considered the Malifut conflict to have been ethnicity-based and the destruction of Malifut to have been a backlash against the Makian’s attempts to monopolize natural resources in North Halmahera. This lack of wider participation in the disturbances afflicting Ternate meant that the security forces were able to prevent the Makians’ return to Halmahera, preclude further rioting and minimize destruction. For this reason, leading members of the Makian community now accelerated their efforts to expand the conflict. Over the first few days of November they took

Escalation – Ternate and Tidore 83 steps to convince the wider Islamic community that the destruction of Malifut was indeed of relevance to all Muslims. Simultaneously, and perhaps more importantly (although not recognized in any existing literature on the conflict), they sought to immobilize the security forces.

Immobilization of the security forces The arrest of the group of Makians for rioting prompted a high-ranking Makian delegation to go to the district police compound to request their release.44 According to a police officer present, the delegation included Thaib Armain and the next highest ranking civil servant, Yunus Abbas, as well as Wahda Zainal Imam and the city mayor, Syamsir Andili.45 This delegation requested the release of the prisoners on the grounds that their incarceration would be likely to heighten tension in the southern part of the city and provoke more rioting. After lengthy debate with these individuals, the police commander requested that they provide him with a written request. After receiving this documentation, the commander ordered the release of the prisoners. From this point, neither the police nor the military made any significant efforts to halt the ensuing violence. The security forces also failed to act against apparently illegal behaviour by the Sultan of Ternate’s traditional guards. The sultan first ordered the Pasukan Kuning to begin patrolling the streets after the first arrival of Makian IDPs from Malifut, prior to the major influx in late October. The sultan described their deployment as a means of guarding vital infrastructure in Ternate and protecting the lives and houses of Christians. The majority of the sultan’s traditional palace guards were Muslim Ternates, although a minority were from Halmaheran ethnic groups, including some Christians. While it is illegal for civilians to carry weapons on the street, the security forces made no effort to prevent patrols by members of this militia, who were armed with spears and machetes. Several politicians requested that the security forces prevent these patrols, but no action was taken.46 Ibrahim Fabanyo, deputy head of the district parliament and also chairman of the National Mandate Party (Partai Amanat Nasional – PAN), requested that the sultan withdraw the Pasukan Kuning from the streets.47 According to Fabanyo, the sultan agreed, but failed to carry out his promise. The Pasukan Kuning’s apparent intent to protect Christians angered Makians, who also took umbrage that the leader (kapita) of the Pasukan Kuning was a Christian from Halmahera. For some Muslims then, particularly Makians, the sultan appeared to be siding with Christians against Muslims in what they took to be a religious conflict. The perceived arrogance of the sultan’s troops, and the fact they carried weapons, also angered Muslims from other ethnic groups, including Tidores. The Pasukan Kuning alienated many Muslims by using force against civilians and demanding to see identity cards, on some occasions denying the validity of these cards, throwing them on the ground and assaulting those they had stopped.

84 Ethno-Religious Violence in Indonesia

‘Transvaluation’: the Bloody Sosol letter In this tense atmosphere during the last few days of October, Muslims began to pass from hand to hand a photocopied letter with potentially catastrophic implications for the region. The letter was ostensibly from the Ambon-based Moderator of the Protestant Church in Maluku (GPM), Pastor Sammy Titaley, to the Chairman of the Evangelical Church on Halmahera (GMIH), Pastor Aesh. The document appeared to be evidence of collusion between the two Churches in the destruction of Malifut. In the letter, Pastor Titaley advocated a campaign of revenge attacks by Christians against the Makians for their attack on the Christian Kao villages of Sosol and Wangeotak. The letter advocated the killing and expulsion of the Makians from Halmahera, and used imagery such as ‘the giant of Christ on the rampage’ and the unity of the ‘golden (Christian) triangle’ of Maluku, North Sulawesi and Irian Jaya. The letter thus both reflected and appealed to concerns among more conservative Muslims over a national process of ‘Christianization’ (see Chapter 2). The letter was not authentic, a fact that became universally recognized in North Maluku shortly after its dissemination. The author used an incorrect letterhead and reference number, greetings that were never used in official Church letters, and signed the letter incorrectly as Semi Titaley instead of Pdt. S. P. Titaley. Several references in the letter were also incorrect. For example, a reference to ‘organizing intellectuals from the Theological University of Ternate’ demonstrated a lack of knowledge of Christian educational institutions in North Maluku. The term appears to have been a mistaken reference to the Theology College (Sekolah Tinggi Teologi), which was in fact located in Tobelo. The letter was also dated 29 July 1999. The Makian attack on Sosol and Wangeotak, for which the letter was ostensibly seeking revenge, did not take place until 18 and 19 August – well after the date on the letter. The author had thus backdated the letter too far. Pastor Aesh, the head of GMIH and supposedly the intended recipient of the letter, first learned of its existence when he received a photocopy from a Muslim friend in Tobelo. 48 The exact origins of the letter are still a subject of debate. However, there was a consensus among many well-informed people with whom I spoke that the letter was composed and typed by one of the instigators of the earlier violence, a local politician and his associate, both Makians.49 The two men reportedly composed and typed the letter at the school where the latter man was a teacher. A large number of people appear to have assisted in distributing the letter throughout Ternate and Tidore, including in the district parliament buildings, local government department offices and schools. A number of politicians and other members of the elite appear to have distributed the letter, suggesting their complicity in organizing the riots that ensued. In one incident, members of the Pasukan Kuning reportedly stopped the car of a Makian leader, Fahri Almari, and found multiple copies of the letter in his car.50 In Tidore, also, several politicians and bureaucrats were involved in distributing the letter. One member of the district parliament alleged that a colleague, along with a district bureaucrat, had handed the letter out in the parliament.51

Escalation – Ternate and Tidore 85 Pastor Aesh and other Church officials from GMIH sent a letter to the local district government stating that the letter was a forgery and listing the large number of errors in the document. According to GMIH, the government did not reply to these points. In addition, no official effort was made to publicly clarify that the letter was a forgery and dispel suspicions within the Muslim community, as the government had done in the case of the previous ‘Attack Map’ forgery. The reasons for this are unclear. However, several respondents claimed that the North Maluku District Head, Rusli Andiaco, who had been in that position for only several months, was intimidated by leading politicians. While he attempted to announce that the letter was false and obviously penned and distributed to provoke Muslims, and further that the Malifut clash had little to do with religion, he was intimidated into remaining quiet. Several politicians and leading figures also made public pronouncements that the Malifut incident had had religious overtones. A senior member of the district parliament allegedly stated during a parliamentary session that the incident in Malifut demonstrated that Christians had brought religious violence to North Maluku as they had to Maluku.52 According to a member of parliament present during the speech, this individual demanded that all Christians leave Ternate.53 As tension rose over the incident, a delegation from the North Maluku District parliament travelled from Ternate to Malifut to investigate claims that mosques had been destroyed. The leaders of this delegation claim that, after visiting Malifut, the team ascertained that mosques had not been destroyed and therefore agreed that the conflict had not been religious in character. However, they claim that the team was too late to convey that message to the parliament back in Ternate.54 Strangely, given the high level of tension in Ternate and Tidore and the wider context of the conflict in Ambon, the ‘Bloody Sosol’ letter did not have an immediate impact. For several days people read and discussed the letter but the violence did not escalate. The letter had a varied impact on Muslims depending on their ethnicity, religiosity, educational level and their relationship with Makians. Those ethnic groups considered to be more devout Muslims while at the same time most closely integrated with Makians – the Sananas, Kayoas and Tidores – were the most ready to believe the letter. People in Ternate were far more likely to be educated and employed and also enjoyed greater access to information about the circumstances of the Malifut clash. Members of the Ternate ethnic community in particular discounted the letter, continuing to believe the destruction of Malifut had been caused by competition over territory and the gold mine. Most non-Makian Muslims, and therefore the majority of those not already taking part in demonstrations and attacks against Christians, were not provoked by the letter. Christian respondents stated that, although the ‘Bloody Sosol’ letter had been circulating for approximately four days and sporadic attacks on Christians had occurred, they had not expected large-scale rioting to break out in Ternate. The majority of Christians remained on the two islands. After five days of little impact in Ternate, those seeking to initiate violence turned their attention to Tidore. There are several possible reasons for this: that Muslims on that island were more receptive to claims of wider Christian involvement and of religious motives in the

86 Ethno-Religious Violence in Indonesia Malifut attack; because the military contingent on Tidore was smaller than that on Ternate or security personnel had greater sympathy for Makians; or because it was home to a far smaller Christian population. Whatever the reason, the ensuing events in Tidore would make a similar incident on Ternate almost unavoidable, by raising tension to unprecedented levels and proving that violence could be carried out with impunity.

Anti-Christian rioting The Tidore riot On the evening of 3 November (several days after the dissemination of the ‘Bloody Sosol’ letter), citing heightened tension on Tidore, the administrative head of the suburb of Indonesiana called a community meeting in his office. All community leaders, including Christian religious leaders, were summoned to discuss the accusations made in the letter.55 People crowded into the local government office to attend the meeting and yet more people gathered outside. When the local Protestant pastor, Arie Risakotta, failed to arrive, the local police chief, Captain Muhar, drove to the pastor’s house and escorted him to the local government office.56 No other Christians attended the meeting. According to several sources present, Pastor Risakotta was forced to read the ‘Bloody Sosol’ letter to the agitated crowd. While asserting that the letter was false and that he, along with the entire Church, had no knowledge of its production, he was punched and ran from the office. A large number of men from the crowd chased and caught the pastor, hacked him to death with machetes and set his corpse alight. The crowd then proceeded to the nearby Protestant church and burned it, and began to attack the houses of Christians in Indonesiana and elsewhere in Soasio. While the meeting at the government office took place, Christians in Soasio began fleeing from their homes, either to the sub-district police station or to the district military compound. This flight continued in heavy rain after the killing of the pastor and throughout the night. Security personnel protected those sheltering in the compounds and prevented the rioting mobs from entering. Several Christian respondents said Muslim neighbours and friends also helped them reach safety or hid them in their houses. Christians who were unable to reach police or military bases hid overnight in gardens pounded by the driving rain. After one terrifying night of violence and destruction, 35 Christians had been killed before they could reach safety.57 Three churches were gutted by fire and all 260 homes on the island owned by Christians were destroyed. Christian men, women and children waited inside the military and police compounds to be evacuated to safety. The riot appears to have been planned and carried out by people from outside Soasio.58 Most witnesses stated that large numbers of Makians were involved, and concluded they were most likely from Ternate. One Christian respondent said that Makians led the rioting crowds, and that local Muslims knew none of those

Map 5 Tidore

88 Ethno-Religious Violence in Indonesia involved. Indeed, many Muslims fled the carnage themselves. The ‘Prime Minister’ for the Tidore Sultanate (a largely ceremonial position), Ridwan Do Tahir, maintains that, following the incident, he led an investigatory team which ascertained that 500 people had crossed from Ternate that day, as opposed to the normal 300. Several witnesses did state, however, that those involved were residents of villages on the opposite, western side of Tidore, such as Tomalou, most of whom were Makian migrants.59 Almost all witnesses stated that many in the crowd arrived carrying petrol cans, suggesting that they intended to start the riot regardless of any explanations offered by the pastor. The Ternate riot Tensions rose markedly in Ternate after the riot on Tidore. Jati writes that in the days after the events in Tidore Christian religious leaders in Ternate received frequent anonymous telephone threats. Under the cover of darkness, crowds of men threw stones at Christians’ houses. Individuals carried out planning to replicate the riot in Tidore. One Makian man recounted that he was warned by other Makians to stay off the streets the following day because rioting was going to occur. Several Christian respondents stated that in the two days after the Tidore riots the fronts of their houses were painted with red crosses to signal the faith of the occupant.60 Oddly, most Christians remained in the city, perhaps believing that inter-religious relationships in Ternate was better than that on Tidore and that the security forces present were adequate to prevent rioting. At around 4 a.m. on Saturday 6 November anti-Christian rioting began in the southern suburbs of Ternate City. Crowds began attacking churches and the houses of Christians in the areas of Kalumata, Tanah Tinggi, Mangga Dua and Jati (see Map 4.1). One Christian resident recounted that the electricity in the area of Tanah Tinggi was turned off and Christian houses were pelted with stones. As large mobs of men moved through the streets attacking the houses of Christian residents, Christians fled north, attempting to reach the district police compound or the guard posts of the sultan’s traditional guards in the centre of the city. Many could only reach neighbourhood churches from where they were evacuated by the military to the police compound. Those who could not reach safety were speared or hacked with machetes. Most witnesses, including young Muslim men in the southern suburbs, said the rioters were primarily Makian men, including university students and IDPs from Malifut. According to people present, the municipality parliamentarian Wahda Zainal Imam and other local Makian leaders led the rioting. The rioting gathered momentum as other groups (such as Tidores and Sananas) in the southern areas began to participate. The rioters spread through other suburbs such as Bastiong, but were blocked from entering the central city by hundreds of the sultan’s guards (the Pasukan Kuning), who were stationed near the provincial government office and just north of the district police compound. Several houses owned by Christians located directly opposite the police compound were destroyed. However, no attempt was made by police officers to prevent this.61

Escalation – Ternate and Tidore 89 The security forces did little to prevent attacks against Christians, focusing on evacuating groups from churches and elsewhere and guarding those who made it to military and police compounds. Personnel took no forceful action against rioters. Several witnesses state that military and police personnel, when they did venture on to the streets to evacuate Christians, were armed only with sticks. One witness claims that military and police commanders removed firearms from their personnel patrolling the streets. These assertions are difficult to verify.62 The military’s inaction in the face of the rioting is demonstrated by the story of one Christian respondent who sought refuge in the Protestant church in Tanah Tinggi. According to this respondent, with the church full of terrified and defenceless people and a mob gathering outside, the pastor phoned the district military base only to be asked whether she and her congregation wanted to be evacuated or to ‘hold out’ against the rioters.63 Thirty-one people were killed in the riot and large numbers injured. During the riot and over the following days, six churches and 353 houses were destroyed.64 In all, 12,763 Christians fled to the police and military bases and to the sultan’s palace. The Pasukan Kuning protected large numbers of fleeing Christians, further angering those Muslims involved in the riot, particularly Makians. Large numbers of Christians sheltered in the neighbourhood of Dufa Dufa, close to the sultan’s palace, and in the palace itself. After several days, the vast majority of these IDPs were evacuated to Christian regions outside North Maluku, such as North Sulawesi. A much smaller number of Christians travelled to Tobelo in north Halmahera. On Sunday 7 November, one day after the Ternate riot, violence broke out once more on Halmahera, this time in Central Halmahera District, in the sub-district of Payahe.65 Christians appear to have been the primary victims of these clashes.66 A series of attacks in the sub-district led to approximately 1,600 Christians fleeing into the forest. These IDPs made their way north to Kao Sub-District and most were eventually housed in the city of Tobelo. As will be seen, the flood of people fleeing from this violence, along with graphic accounts of the Ternate and Tidore riots, played a major role in causing violence to break out elsewhere in the region.

Religion enters the North Maluku conflict This chapter has accounted for the escalation of the North Maluku conflict from a border dispute on Halmahera into extensive and devastating ethno-religious conflict. In doing so it has illustrated how an analytical distinction between the role of elite agency and mass-level factors in causing conflict is problematic. It has also demonstrated how the impact of rational interest and more emotional influences were almost analytically indistinguishable, exacerbating one another as they motivated participants to join the riot. Tension was rising in Ternate throughout 1999. The conflict in Ambon had stimulated anti-Christian sentiment among some Muslims. More importantly, the elevation of North Maluku to provincial status stimulated political rivalry and accusations of ‘money politics’. The destruction of two Kao villages in the Malifut

90 Ethno-Religious Violence in Indonesia area on Halmahera during this period had little impact in Ternate, with the exception of one article in the local newspaper speculating on who had provoked the attack. After the destruction of Malifut by the Kaos and the influx of Makian IDPs into Ternate, however, the atmosphere degenerated considerably. The large numbers of destitute IDPs, and their allegations that the Kaos had insulted Islam, angered Makians and alarmed Christians. Makians from Ternate, along with hundreds of refugees, attempted to return to Halmahera to attack Kao, and carried out sporadic acts of violence against Christians, and Kaos in particular, residing in Ternate. For many Makians the primary means of comprehending catastrophic world events was through the lens of their religion. For this reason many came to believe that the Kaos had somehow been motivated by inter-religious animosity, and that their attack had been led by the Protestant Synod based in Tobelo. For them, this made Protestant churches and Christians in general legitimate targets for their reprisals. I have demonstrated that this influx of refugees and the accompanying increase in tension was nevertheless insufficient to explain the spread of violence to Ternate. For almost two weeks, there was some level of sympathy for the Makians among Muslims from other ethnic groups, but little real anger about the destruction of Malifut or a desire to initiate religious war. Many individuals, particularly Ternates, attributed the clashes on Halmahera to over-zealous attempts by the Makians to obtain power and resources. This lack of sympathy for the Makians among the wider Muslim population meant that the security forces were able to easily control the relatively small numbers of potential rioters and prevent them from returning to Halmahera. In addition, as small-scale acts of violence occurred, the Sultan of Ternate’s traditional troops began to dominate the security situation on the city’s streets. These men, some of whom were Christian, stopped and searched any men they suspected of seeking to cause disturbances and began to control certain areas of Ternate with the apparent acquiescence of the security forces. In order to circumvent these restrictions and generate support for their goal of exacting retribution against the Kaos, several influential members of the Muslim political elite intervened in two ways. First, they sought to create a new understanding of the violence in Malifut as religious in character, and second, they intentionally or inadvertently created the political opportunity for rioting to occur by immobilizing the security forces. One of the most common means by which elites are seen to provoke their followers into violence is propaganda. Makian leaders portrayed the Kaos’ destruction of Malifut as Christian expansionism, both in closed meetings in houses and mosques and in public meetings in the district and city parliaments. This effort was aimed at gaining sympathy for the Makians from the wider Muslim community. Members of the Makian elite in North Maluku were well aware that the one issue likely to gain widespread support from Muslims and to provoke people into violent action was religion. The effort was therefore intended to transform an ethnic dispute into a wider religious conflict. In doing this, those Makians

Escalation – Ternate and Tidore 91 involved sought to facilitate their retaliation against Kao on Halmahera and to displace blame from their own community for the Malifut riots. The success of propaganda in causing violence is most often attributed to its use of highly sensitive and volatile subject matter. This certainly appears to be the case with the ‘Bloody Sosol’ letter, the most visible element of the effort to escalate the conflict. The letter played on several connected issues that its authors hoped would anger and concern Muslims: links between North Maluku Christians and Christians in Ambon; Christian separatism connected with the RMS; and the ‘Christianization’ of North Maluku and a physical threat to the Muslim community. Within the context of the Ambon conflict, and the clashes in Kao and Malifut, the letter was certainly intended to have a provocative effect on Muslims. The fact that all Makians are Muslim, and the Kaos predominantly Christian, no doubt helped convince some Muslims that the Protestant Church was involved in the violence. Bubandt in particular puts a great deal of importance on the letter in the ensuing violence. He argues against seeing the letter as just a means of provocation, concluding that the violence in Tidore was a ‘spontaneous reaction to the rumour about a Christian conspiracy rather than a long planned assault’.67 Bubandt asserts that the letter played strongly into the prevailing ‘social and discursive universe’ of conspiracy and paranoia in North Maluku and Indonesia as a whole in the post-Suharto era. In other words, an understanding of the impact of the letter on ordinary Muslims is more important in explaining the riots than the actual propaganda campaign. This follows Horowitz’s conclusion regarding the role of rumours in conflict situations: while they are often intended to initiate rioting, they are unlikely to do so if there is no ‘market’ for the messages contained in them. ‘What is remarkable is not that an interested agitator starts a rumor but that the rumor is spread, believed, and acted upon.’68 Yet this conclusion is predicated upon the assertion that those who started the rioting did so because they were provoked by the rumour. This chapter has demonstrated that, contrary to most analyses of the anti-Christian rioting in Ternate and Tidore, it was not the ‘Bloody Sosol’ letter that instigated violence. While the Makian leaders allegedly involved in this process had moderate levels of influence over the residents of certain areas of Ternate, this influence did not extend over other ethnic groups across the city or over the security forces. Fears of a Christian attack in Ternate were not credible given the tiny Christian populations in Ternate and Tidore.69 The ‘Bloody Sosol’ letter did little to alter the majority Muslim view that Makians, and not Christians, had provoked the Malifut violence. Most remembered that the Makians had instigated the Kao attack by previously destroying two Kao villages. The letter circulated in Ternate for five days with little impact on the wider Muslim community. When violence exploded it did not result from a neat linear process of propaganda, suggestibility and participation. While at least some sections of the Islamic community in Ternate and Tidore believed the letter was genuine, the majority of those involved in the initial rioting, like those who organized and led it, participated not because they believed in a

92 Ethno-Religious Violence in Indonesia Christian conspiracy, but for reasons of revenge. Most participants in the early stages of rioting were Makians, either refugees from Malifut or those who were living in Tidore and Ternate, some of whom originated from and had family connections to Malifut. Ethnic solidarity and a desire for revenge for what happened in Malifut motivated almost all of those who took part in the early rioting in Tidore and Ternate and many of the community leaders who organized it. Makians had set their sights on establishing Malifut as a sub-district not only throughout 1999, but ever since the local transmigration of the community from Makian Island in 1975. To see this project destroyed in flames on 25 October dismayed Makian elites in Ternate and Malifut. Those who initiated and carried out the violence in Tidore and Ternate acted out of a direct grievance over the destruction of Malifut and not because of the ‘Bloody Sosol’ letter or other forms of propaganda. It was their attacks on Christians and churches, rather than reframing by propagandists, which transformed ethnic conflict into religious conflict. Participation in the rioting quickly broadened, however, as excitement spread among young men, and, importantly, as it became clear they could take part with impunity. While burning churches perhaps appeared to confirm the claimed religious element of the conflict to some Muslims, anti-Christian sentiment was not felt strongly beyond the Makian community. Even after this violence, most Muslims saw little cause to wage war against Christians on Halmahera. No widespread multi-ethnic religious war was launched from Ternate and Tidore until after the clashes in Tobelo and Galela in late December, to be discussed in the next chapter. The ‘Bloody Sosol’ letter did perform one very important function, however – that of making two organized and well-planned riots appear to be spontaneous eruptions of Muslim anger. While the authors of the letter, and their supporters, hoped the letter would provoke the turmoil that would allow the rioting to proceed, their primary goal was to establish a credible reason for the rioting: namely the understandable anger of the Islamic community at a violent Christian conspiracy. By the time the ‘Bloody Sosol’ letter was produced and disseminated, the riot had already become imminent because of the immobilization of the security forces. The letter and other forms of provocation were less instrumental in the onset of rioting in Ternate and Tidore than the pressure brought to bear on the security forces by several of the highest-ranking politicians in North Maluku. After Syamsir Andili, Thaib Armain and several other officials requested that the police free the prisoners arrested for attacking Christian buildings in Ternate, the security forces made little effort to halt subsequent rioting. The security forces also afforded the same political opportunity to the Sultan of Ternate and his traditional guards. The immobilization of the security forces was thus one of the most telling instances of elite agency in the riots and, ultimately in the North Maluku conflict as a whole. It seems likely that the security forces were reluctant to oppose the possible future power holders in the province. While the appointment (and withdrawal) of regional military and police commanders is determined in Jakarta, the influence of local figures on that decision can be substantial. As a Golkar Party

Escalation – Ternate and Tidore 93 functionary, Mudaffar Syah was considered to have considerable influence in Jakarta as well as in Ternate itself. The influence of district heads, mayors and the regional secretary would also be important in terms of the future economic activities of the police and military in the province. Civilian politicians could therefore have an impact on the economic well-being of local commanders. No arrests were made following the riots, despite clear indications of who had planned and led the violence. Whether all members of the elite group that requested the release of the prisoners intended to facilitate increased rioting is uncertain. While it is feasible that the head of the bureaucracy and mayor of Ternate requested the prisoners’ release in order to prevent further disturbance, the release of 35 men who had already attacked churches and shops and tried to instigate inter-religious violence was almost certain to have the opposite effect. According to several witnesses, some of these elites, such as Wahda Zainal Imam, certainly intended further violence. Several days after the release of these prisoners, Wahda led these men and hundreds of others in anti-Christian rioting, this time facing no resistance from the security forces. It seems likely that these individuals seeking to start anti-Christian rioting found allies among other Muslim politicians, such as Syamsir Andili, whose primary goal was to undermine the growing dominance of the Sultan of Ternate.70 Indeed, these two groups were not mutually exclusive – many Makian politicians and civil servants, such as Thaib Armain, were both aggrieved at the destruction of Malifut and sought to undermine Mudaffar Syah. Thaib Armain abandoned any alliance he had with the Sultan of Ternate after the destruction of Malifut, because of the sultan’s alleged support for the Kaos. Important elites were thus subject to the same emotional and psychological influences as the less visible young men in the streets during the riot. By late October there was a growing desire among much of the North Maluku elite to undermine the sultan’s strong, and apparently improving, political position. His strength was a threat to his main rivals for the governorship, Bahar Andili and Thaib Armain. The position of Syamsir Andili, the Mayor of Ternate City, was particularly compromised by the growing strategic and political dominance of the sultan, as his Pasukan Kuning dominated the central city. In addition, the likely elevation of the sultan to the position of governor in elections scheduled to take place sometime in the following year was also a direct threat to the large numbers of Makians, Tidores and Sananas who dominated the bureaucracy.71 It was probable that the sultan, on becoming governor, would appoint Ternates and even Halmaherans to high-level positions in the civil service. The expulsion of Christians, particularly members of the three district and city parliaments, would have reduced some political support for the sultan in upcoming gubernatorial elections. However, the small number of Christians in Ternate and Tidore and the little bearing they had on political outcomes suggest an alternative goal behind the facilitation of rioting. It may have been calculated that the sultan’s gubernatorial ambitions would be checked if he acted to protect Christians in this situation, an action that would be likely to anger many Muslims. Rioting therefore provided an opportunity to unite Muslims against the sultan.

94 Ethno-Religious Violence in Indonesia However, while this political context meant many members of the elite facilitated rioting and immobilized the security forces, it cannot be said that the initial violence was a direct consequence of political competition. This elite-level political struggle between the Sultan of Ternate and his rivals (Bahar and Syamsir Andili and Thaib Armain) merely provided the conditions for the brewing riot to go ahead. If the ‘Bloody Sosol’ letter had been primarily aimed at undermining support for the sultan it would almost certainly have mentioned him and his support for the Kaos. Later rioting in Ternate, in late December, would demonstrate the extent to which political factors had by that time taken a more central role in the conflict. While not ‘riot specialists’, in Brass’s meaning of the term, Makians, led by Wahda Zainal Imam, initiated the rioting in November for clear goals: to exact revenge on Kaos and Christians in general and, second, to deflect blame for the clashes in Malifut away from their own community. They did not do so under the sway of propagandists. The violence launched by the Makians and members of other groups in southern Ternate subsequently initiated a series of violent events that pushed North Maluku almost inexorably towards religious war.

Conclusion Inter-religious relations in North Maluku have been changed, perhaps irrevocably, by the events of early November 1999. The Christian minorities in Ternate and Tidore were violently targeted, treated as scapegoats following a dispute in a distant and remote area. The events were to lead to widespread and far more deadly inter-religious violence across the new province, as innocent followers of both religions were slaughtered. Most existing studies of the conflict explain the rioting in the region’s two main cities as a spontaneous outburst of Muslim anger at anonymous provocation in the form of the ‘Bloody Sosol’ letter. The analysis presented in this chapter suggests that a far more malign explanation of these riots is appropriate. The November Tidore and Ternate riots were not a spontaneous response to the Malifut violence or to the inflammatory letter that attempted to draw a link between that violence and the local Christian Church. The rioting appears to have been organized by local politicians and community and religious leaders. At another higher level, high-ranking bureaucrats and politicians both allowed and facilitated the rioting to occur for more political purposes. Above all, the rioting in both cities was organized and started by Makians who did not require further provocation. The influx of Makian refugees from Halmahera into Ternate and elsewhere had a major emotional impact on many members of the Makian elite. The vast majority of Muslims of other ethnic groups in Ternate and Tidore, perhaps less susceptible to religious propaganda than is often claimed, did not accept the assertions that the Malifut conflict called for inter-religious animosity. While some subsequently joined the rioting, most Muslims did not participate. As demonstrated by the analysis above, leadership and mass action are not entirely separate phenomena. Both elites and ordinary participants reacted to the

Escalation – Ternate and Tidore 95 statements and actions of the other. Both important elites and masses were motivated by emotional/psychological factors as well as by rational interest. What made the rioting possible was a convergence of interests between those who sought revenge for Malifut and political elites who sought to unify Muslim political support against the Sultan of Ternate. A coalition of these individuals then pressured the security forces to refrain from action as the sultan’s political supporters were targeted. Despite their relatively small numbers, the rioters were able to freely target Christians and their property and face no sanction because police and military commanders were hesitant to act against the supporters of the potential future provincial power holders. Within two months, almost the entire province was engulfed in religious violence. As will be seen in the next chapter, this violence, often far more brutal and extensive than that which had preceded it, was in large part a consequence of the riots in Ternate and Tidore. That violence and the targeting of innocent Christians on those islands caused intense anger and distrust across the province. The scapegoating of Christians because of a clash in a remote area of North Maluku, and for political purposes, escalated a small dispute into province-wide religious war.

5

Dispersion – Tobelo and Galela

Introduction Since the sixteenth century, Islam and Christianity have converged in north Halmahera. While North Maluku has long been strongly Islamic, in this one corner of the region the two great faiths coincided and their members mixed in relatively even numbers. The palm-tree-lined coastal road from Galela to Tobelo passes through neighbouring Muslim and Christian villages as well as those where members of both faiths have lived side by side for over a century. In Tobelo and Galela, peaceful coexistence between Christians and Muslims, enshrined in the local cultural tradition of Hibua Lamo, had always been a source of pride. When violence wracked much of the region in late 1999, Tobelos were sure provocateurs would try and undermine this harmony. They knew that ‘if they make Tobelos fight, the rest of North Maluku will follow.’1 Most people in North Maluku saw the Malifut conflict between Kaos and Makians as a local issue, a dispute over territorial borders and the control of natural resources. For this reason, they deemed it unlikely that violence would spread to other areas in the region. However, after the anti-Christian rioting in Tidore and Ternate, and attacks on Christian communities in several villages in Central Halmahera, the violence assumed a religious character that appeared relevant to the entire North Maluku community. Christians, particularly those in north Halmahera, believed that the Makian community had targeted defenceless Christians to avenge their failed attempt to take control of valuable Kao land in Malifut. Muslims also began to feel concerned at this rising sentiment and to take defensive precautions, including making weaponry. This chapter analyzes how and why violence spread throughout the new province by considering in detail the events in the adjacent sub-districts of Tobelo and Galela in north Halmahera. I will argue that two factors – security concerns and religious animosity – in large part caused the bloodshed to spread to these areas. The sight of Christians fleeing to Tobelo and accounts of the murder of a pastor and the destruction of churches spurred anti-Muslim sentiment. Rising tension meant that fear and anger increased within both communities. Unsurprisingly, after the nature of the violence in Tidore and Ternate, many members of both communities in Tobelo and Galela began to assert their religious identity in a more belligerent manner.

Dispersion – Tobelo and Galela 97 To help explain how such concerns can bring individuals and groups to commit violence, the chapter first presents two general explanations for the outbreak of communal violence: the (physical) security dilemma and societal security dilemma. It will then describe the events leading up to the violence in late December 1999 and the clashes themselves. The final section draws some conclusions on how and why two sub-districts with such a long history of inter-religious harmony could descend into violence. While emphasizing the importance of insecurity, in my analysis of these events I attempt to move away from an apolitical focus on the impact of this tension alone and reveal the human agency that was crucial to the violence.

The security dilemma Why do two communities with apparently no interest in initiating devastating civil war nevertheless find themselves fighting that war? To answer this question, several general theorists of conflict have utilized the security dilemma concept central to ‘realist’ thought in international ielations.2 They assert that, simply by taking measures to ensure their own security, an ethnic group can inadvertently threaten others. Barry Posen posits that this often occurs when state authority is removed and ethnic communities are suddenly required to consider the intentions and military capabilities of, and their past relationship with, neighbouring groups.3 If the benign intent of that group is not assured, the community will take steps to ensure its own safety, reaffirming group solidarity, mobilizing militias or constructing weapons. Members of the other community will find it difficult to see these measures as purely defensive and will make similar preparations. A spiral of insecurity and tension will ensue, heightening the tension between the two groups.4 A breakdown in information flows between the communities will exacerbate this situation.5 Given that both have much to lose from violent conflict, they will normally find a negotiated solution that maintains the status quo and prevents chaos.6 The ability to reach a negotiated settlement to a dispute is diminished, however, when one or more parties misrepresent their true position by ‘bluffing … exaggerating their strengths, minimizing their weaknesses’.7 As each party becomes suspicious of the true motives of the other, conflict becomes more likely. If the state and the official security structure appear unable or unwilling to provide protection, militant activity is likely to obtain greater legitimacy in the eyes of the community. Groups may eventually adopt more offensive measures if they appear to be the best way of eradicating the apparent threat to their group and if the opportunity to do so exists now but may not in the future.8 Other theorists have claimed that threats to a community’s identity, rather than purely physical threats, may often be the primary cause of rising tension and conflict. Ole Waever et al. have suggested that perceptions of existential threats can arise from in-migration, changes in education or other mechanisms that erode a group’s distinctiveness.9 When a community perceives that its language, culture or religion is threatened by government policies or the actions of another group, it

98 Ethno-Religious Violence in Indonesia will attempt to protect that cultural identity, sometimes through military means, but often by reasserting that identity. The displays of group solidarity that accompany these actions will often unwittingly threaten another community.10 A synthesis of these two theories provides insights into the period before many outbreaks of communal violence. Material concerns (with physical safety, economic resources, etc.) and identity-related factors (such as cultural freedom or ethno-nationalism) are clearly present and important in many conflict situations. In all social phenomena, identities and motivational factors (interests) are mutually constitutive – influencing, changing and creating each other. Increased communal solidarity is often an effective means of gaining economic advantage. Fear of physical attack will often strengthen and radicalize communal identity.11 This, and the rising militancy that accompanies it, will in turn appear increasingly threatening to neighbouring communities. Indeed, it is often difficult to establish whether an issue central to rising tension has material or identity-related importance. For example, the possibility of attack threatens not only the safety of group members but also the territory, buildings and symbols central to a community’s identity. Certain places and buildings not only house and support communities but are also of a sacred nature. Rumours often play a central role during periods of acute insecurity.12 The horrifying content of some rumours stimulates a sense of urgency and makes non-violent alternatives seem ridiculous and dangerous. In such situations, opposing groups may decide their options lie not between cooperation and aggression, but between aggression and victimhood.13 The use of violence in the self-defence of one’s community therefore becomes legitimate.14 Sporadic confrontations appear to confirm the aggressive character of the other group and the danger to one’s own group. In these situations members of a community are particularly vulnerable to manipulation. In situations involving religious tension, religious and secular leaders sometimes use the teachings of the faith to legitimate violence. When militia leaders use religion to mobilize followers, they often also draw on local traditions and prejudices. Having limited knowledge of religious texts, community members are unable to challenge or even critically evaluate the use of doctrine to legitimate violence.15 Such use of religion appears to create particularly volatile situations, with disputes coming to be perceived as a struggle between good and evil,16 and one’s opponents seen as ‘fanatics’.17 This combination of insecurity, religious fervour and traditional practices may increase the intensity of violence. In her study of religious riots in sixteenth-century France, Natalie Zemon Davis writes that the atrocities involved in Catholic–Protestant riots ‘can be reduced to a repertory of actions, derived from the bible, from the liturgy, from the action of political authority, or from the traditions of popular folk justice, intended to purify the religious community, humiliate the enemy and thus make him less harmful’.18 Fire and water (drowning) were considered sacred means of purification. Catholics committed desecration of Protestant corpses in order to further humiliate the victim, disembowelling them and dragging them through

Dispersion – Tobelo and Galela 99 the streets.19 Davis points to one other function of the terrible atrocities committed during riots. The desecration of corpses, dismemberment and disembowelling with butcher’s implements completes a process of dehumanization of the victims.20 While providing strong insights into many conflict situations, the structural nature of the physical and societal security dilemma concepts gives inadequate explanatory weight to human agency. This means that it is difficult to differentiate between those situations of insecurity that lead to violence and those that do not. It also means that the analyst will miss those actions that are motivated by factors other than physical and existential anxiety. While militants and provocateurs can be ‘as much a product as a producer of ethnic fears’,21 fear is often not their primary motivation. For many individuals and sub-groups, a conflict situation offers opportunity as well as threat. While some act out of fear, other actions central to the outbreak of violence are based on personal vendetta, a quest for status, financial reward, criminality or religious militancy. Indeed, inter-community relations in the period before violence are often complex, individuals and groups in each community having very different interpretations of the situation and the intent of the other community, and considering different interests and strategies. By analyzing in depth the motives and actions of individuals and sub-groups it becomes possible to explain why the absence of strong government and an atmosphere of insecurity sometimes leads to violence.22 While Indonesia in 1999 cannot be considered a failed state, this chapter demonstrates that the security dilemma concept can help explain the worst violence in the conflict – that in Tobelo and Galela. The riots in Ternate and elsewhere stimulated a great deal of insecurity in those areas. Physical and identity-related concerns reinforced and exacerbated each other. As issues of insecurity, material interest and identity interacted and exacerbated inter-communal tension, the conflict drew in increasing numbers of people.

A region of inter-religious harmony Tobelo Tobelo Sub-District (split in 2000 into Tobelo and South Tobelo Sub-Districts) stretches from Galela in the north to Kao in the south (see Map 5.1). The sub-district is primarily rural, with most farmers growing coconut for the production of copra. There is one major road that runs from the port of Sidangoli in Jailolo Sub-District (the main entry point to Halmahera) along the coast through Kao to Tobelo, and then on to Soasio, Galela. The city of Tobelo is the main centre for the north Halmahera region and the export hub for copra and other commodities. The city is now the capital of the new North Halmahera District, formalized with district elections in 2004. In 1999, the population of Tobelo Sub-District was 55,046 and lived predominantly along the coast.23 The majority of the population is of the Tobelo ethnic group, closely related to the other ethnic groups in north Halmahera, including the

Map 6 Tobelo

Dispersion – Tobelo and Galela 101 Kao and Galela.24 The majority of Tobelos, and the population of Tobelo Sub-District, are Protestant Christians. While the Dutch Reformed Church mission was first established at the inland Galela village of Duma,25 Tobelo City is now the centre of the Evangelical Protestant Church on Halmahera (GMIH), and the location of the Christian Theology College (Sekolah Tinggi Teologi – STT). There is also a small, mostly migrant Catholic population in Tobelo City. Protestant Christians comprise the majorities of most villages in the sub-district, particularly those located on the road between Tobelo and Kao. A substantial minority of the Tobelo ethnic group is Muslim. For the decade before the conflict, Muslims constituted the majority in the sub-district capital, Tobelo City. In 1999 the city was also home to several thousand Muslim migrants from the Tidore, Makian and Bugis ethnic groups, many of whom were successful small traders and shop owners. Large communities of Muslims were also located in the villages of Gorua, 5 km north of Tobelo City, and Togoliua, approximately 30 km to the south of the city.26 Muslims also comprised small proportions of some villages, such as Gamhoku. Several thousand government-sponsored transmigrants, mostly Muslims from East Java, also lived in the sub-district, the largest community located adjacent to the village of Togoliua. Family connections between Protestant and Muslim Tobelos are very close. Almost all Tobelos interviewed claimed that they have a parent, uncle, grandmother or other relative of a different religion. Members of the same extended family often celebrated Christmas, Easter, Idul Fitri and Idul Adha with each other, eating in the houses of the family members celebrating the event. Muslims would assist neighbouring Christian communities in constructing new churches and Christians would provide labour for the erection of new mosques. These connections have been formalized in a cultural structure known as Hibua Lamo. Hibua Lamo is considered similar to the Pela Gandong cultural system in Ambon and central Maluku, which binds Christian and Muslim villages in a pact of mutual assistance and non-aggression.27 One Muslim Tobelo proclaimed that Hibua Lamo was stronger than Pela Gandong as it was built on ‘blood’ or family ties.28

The weakening of ethnic ties However, several Christian respondents noted that, during the 1980s and 1990s, the closeness of this relationship declined. The practice of dining with family or friends during holy festivals had become far less frequent over the past decade. According to some Christians, where Muslim and Christian children had always played together, by the late 1980s bullying and fights had increased. The 1990s saw more serious fighting breaking out or being narrowly averted. The central suburb of Gosoma in particular saw sporadic fights between young men from the two communities. One bout of fighting in this neighbourhood in the early 1990s was serious enough to damage several buildings, including a local school. Rumours and rising tension almost led to rioting in the period leading up to Christmas 1994. There are several possible explanations for the deterioration in relations between indigenous Christian and Muslim Tobelos. Christians blame an increase

102 Ethno-Religious Violence in Indonesia in ‘radicalism’ among local Muslims, due because of the settlement in the city of large numbers of migrants over the preceding two decades. Several Christians considered that these migrants, many from Tidore and Makian and from outside North Maluku, such as South Sulawesi, were far stricter in their observance of Islam and more exclusive and antagonistic towards Christians. Migrants also felt little affinity with the ties of Hibua Lamo.29 The perception that Muslims had become more radical over the two decades prior to the conflict is widespread among Tobelo Christians. Some developments in the city certainly suggested a degree of increased religiosity. For example, from the late 1980s the sale of pigs or pork was banned in the central market located in a Muslim area. But on the whole both communities were equally religious. When external events, such as political turmoil in the Middle East, influenced the atmosphere in the city, they appear to have affected each community equally. During such periods, some individuals and groups from each community used the events to antagonize the other. For example, in 1991 during the first Gulf War against Iraq, provocative slogans and symbols such as the Star of David, ‘PLO’ (Palestine Liberation Organization) and ‘Mossad’ appeared on walls, raising emotions in the city.30 External actors also appear to have affected the relationship between Christians and Muslims in Tobelo. In 1993, an American Christian missionary from the American evangelical organization the New Tribes Mission, Pastor Bouwens, who had long lived in Tobelo, was involved in an altercation with Muslim men that almost started a communal riot. After allegedly asking several Muslim men why they were leaving rubbish near his house ‘when Islam was supposed to be a clean religion’, Bouwens was beaten seriously enough to necessitate medical care in the local hospital, Bethesda.31 A crowd of Christian men gathered outside the hospital and demanded they be allowed to take revenge against the Muslims responsible but were calmed by a Christian community leader. In 1998 a sermon delivered in a mosque by a visiting Muslim medical practitioner caused tension when overheard by nearby Christians. The man told the audience of Muslims that they should no longer dine with Christians nor offer them salutations during the Christmas period. Christians living close to the mosque overheard the sermon, which was broadcast through the mosque’s loudspeaker.32 After complaints from these Christians, the Sub-District Head of Tobelo, Agil Bachmid (a Muslim), asked the man, Dr Husen, for an explanation of his actions and instructed him to write a letter of apology to the Christian community.33 While these incidents caused tension and demonstrations of bravado in the Tobelo community, none instigated rioting. The Ambon conflict also influenced the atmosphere in Tobelo, although it did not lead to violence. Leaders from the Muslim and Christian communities met frequently after violence broke out in the provincial capital in January, stating their continued respect for the cultural ties of Hibua Lamo and their determination to prevent conflict spreading to north Halmahera. Several families seeking refuge from the turmoil in Ambon arrived in Tobelo Sub-District but their presence did not greatly influence the local atmosphere. However, several Tobelo Christians

Dispersion – Tobelo and Galela 103 Table 5.1 Chronology of events before violence in Tobelo City, 1999 19 January

Religious violence begins in Ambon City

22 June

Inter-village clash in Ibu Sub District in north-western Halmahera

18 August

First clash in Malifut – two Kao villages destroyed

25 October

Kao attack and destroy Malifut

3–6 November

Anti-Christian rioting in Tidore and Ternate

8 November

Attacks against Christians in Central Halmahera – IDPs flee to Tobelo

14 November

Christian man throws bomb in Tobelo market

20 November

Christian student group confronts Sub-District Head

26 November

Tobelo Muslim discovered making jihad robes

7 December

Visit of Sultan of Ternate and Governor to Tobelo

24 December

GMIH requests protection for central church

26 December

Rioting starts in Tobelo City

told me the events in Ambon reconfirmed to them what they had perceived as a persecution of Christians throughout Indonesia, citing examples such as the destruction of churches in Java, repression in East Timor and the riot in Ketapang in Jakarta. They recounted that a consensus had developed among many in the Christian community in Tobelo that if ‘they [meaning mainly Muslims, but also the government or the TNI] try that here, we will not step back’.34 Similar sentiments were prevalent among the Islamic community, who viewed the Ambon conflict as caused by aggression by Christians and particularly by the RMS separatist organization based in Ambon in the 1950s. 35 The atmosphere in Tobelo became slightly tenser following a clash in June 1999 in the sub-district of Ibu on the western side of the north Halmahera peninsula. This area is close to Tobelo both geographically and ethnically. A fight between youths from the Christian village of Bataka and the Muslim village of Talaga left several people dead. My interviews with people involved suggested the clash was sparked by a dispute between youths over suspected criminal activity. However, Christians in Tobelo considered it an example of Muslim aggression towards the Christian minority.36 The 18 August destruction of the Kao villages of Sosol and Wangeotak by Makians angered Christians in Tobelo, yet both Muslims and Christians considered the dispute to be ethnically-based, with a very local scope, concerning a sub-district boundary. The lack of response from the district government in Ternate to the Kaos’ demands confirmed Christians’ perception that Makians dominated the government and supported the interests of their co-ethnics at the expense of other groups. But relations between Christians and Muslims in Tobelo remained peaceful. While the Kaos’ subsequent destruction of Malifut on October 25, and the expulsion of the Makians, was still largely considered a local

104 Ethno-Religious Violence in Indonesia problem not relevant to Tobelo, many Muslims considered it a rather threatening development. Makians in particular were concerned at the events and the subsequent intimidation they faced from Kaos as they travelled to Ternate.

Fear and rising radicalism The atmosphere between Christians and Muslims in Tobelo City and surrounding villages changed markedly following the riots in Tidore and Ternate in early November. While most Christian IDPs from these riots were evacuated to North Sulawesi, some returned to their original homes in Tobelo. Most of these were politicians and civil servants who had witnessed the efforts by their Makian colleagues in Ternate to portray the conflict in Malifut as religious in character and had perceived the organized nature of the riots in the city. One related how his house had been painted with a red cross in the days before the rioting. Many of these Christian leaders were angered at the events in Tidore and Ternate and influenced the situation in Tobelo through their rhetoric. What changed the security situation in Tobelo most dramatically was the arrival of several thousand IDPs fleeing violence in Central Halmahera. Attacks on Christian communities had occurred in several villages in Payahe and West Gane Sub-Districts in the days following the Ternate riots. It seems likely that these attacks were launched from Ternate. According to a Makian respondent in that city on 8 November 1999, immediately after the anti-Christian rioting in Ternate, as Christians sheltered around the Sultan’s palace and in police and military bases, Wahda Zainal Imam and other leaders allegedly formed a militia to travel to Payahe and elsewhere to mobilize local Muslims for attacks on neighbouring Christians.37 Wahda Zainal Imam and other leaders referred to this militia as the North Maluku branch of the Islamic Defenders Front (Front Pembela Islam – FPI).38 Muslims appear to have attacked Christians in the village of Lola in Payahe Sub-District on Halmahera opposite Makian Island, home to large numbers of Makian migrants.39 Christians suffered large numbers of casualties in this attack and in a subsequent attack in the sub-district of West Gane. These conflicts forced around 1,600 people from their homes, most fleeing into the surrounding forest. In response to requests from community and church leaders, several pastors of GMIH in Tobelo organized a ship to travel to Central Halmahera to evacuate the Christian community. Large numbers of Christian IDPs were therefore brought to Tobelo and housed in camps in the city and surrounding villages. Christians were angry that neither the North Maluku nor Central Halmahera District governments organized the evacuation of, or provided any assistance to, these IDPs in the same way they had to the Makians after the destruction of Malifut.40 The atmosphere in Tobelo now became highly tense. Christians were outraged at the killing of a pastor and the destruction of churches, and saw these events and the targeting of Christians in general (as opposed to just Kaos) as clear signs that the conflict was now about religion. IDPs also brought stories that the military in Ternate and Tidore, if not complicit in assisting the attacks, had not acted to

Dispersion – Tobelo and Galela 105 prevent the riot. This added to an awareness among many Christians of the need to defend themselves and many turned to militant individuals. When religious violence occurred in several areas of Halmahera in November, the leader of the Christian militia in Kao, Benny Bitjara, sought religious sanction to take his militia to these areas. Bitjara visited the head of GMIH, Pastor Aesh, and requested church sanction for an attack against Muslims in these communities. When Pastor Aesh refused to provide church legitimacy for attacks against Muslims, Bitjara became angry and accused the pastor of not protecting Christians. From this moment the Christian community was divided in its response to the rising animosity. Muslims in Tobelo increasingly feared that Christians might retaliate against them for attacks on Christians elsewhere. Muslims claimed that the IDPs arriving in the city were almost all young men, and were armed with traditional weapons such as bows and arrows. Christians denied this; however, some Muslims saw the fact that the GMIH Synod had sent a ship to evacuate Christians from Central Halmahera to Tobelo as confirmation of their suspicions that Christians were planning an assault on Muslims. Many perceived these IDPs as reinforcements for Christians in Tobelo. To many Muslims, it became apparent that some Christians were planning to attack Muslims in the area. Both Christians and Muslims began to step up preparations for what inceasingly appeared to be an inevitable conflict. Men and women on both sides set about making spears, machetes and bows and arrows. By December the sound of explosions became common as bombs were made and tested. Each side set up security posts in their areas of town. Because of this growing sense of insecurity, both Christian and Muslim shop owners paid militants from their religious community to provide protection. Several Christian shop owners paid Benny Bitjara for protection.41 Christians in turn claimed that a Muslim shopkeeper named Muhammad Albar organized security for shops in the town owned by Muslims. After the arrival of the Christian IDPs, a number of incidents both reflected and increased the level of tension in the town and surrounding villages. November and early December saw several fights between drunken men in the town market. Although not particularly unusual, these fights had a far greater impact on the community than previously. On one occasion in early December, a fight triggered a stand-off between people wearing either white or red headbands (Islamic and Christian symbols) and carrying weapons. Many Muslim residents of Tobelo, particularly non-Tobelos, left the city over this period. On 14 November guards at the Tobelo market expelled the younger brother of Benny Bitjara for being drunk and causing a disturbance. Fleeing Tobelo on the back of a motorcycle, witnesses claim he threw a pipe bomb at the national rice agency (Bulog) building. The bomb failed to explode but lay smoking on the ground and several police officers chased him. At Efi Efi, a village to the south of the city, he was apprehended but then released by the police. Subsequently a rumour circulated throughout north Halmahera that when the police attempted to arrest him, Bitjara had picked up a large rock and eaten it, frightening the police and onlookers and persuading the officers to allow him to continue unhindered.

106 Ethno-Religious Violence in Indonesia Bitjara probably escaped arrest, despite his obvious attempt at provocation, because he was the younger brother of the head of the local Christian militia. Nevertheless, the stone-eating rumour heightened the sense of nervousness in the sub-districts of Tobelo, Galela and Kao. Christian youths associated with the Theology College in Tobelo (STT) formed a group called the Christian Youth Communication Forum or FKPKHU (Forum Kommunikasi Pemuda Kristen Halmahera Utara). On 20 November members of this group approached the sub-district head, Agil Bachmid, and told him that if he did not find a way to prevent conflict in Tobelo, they would do so themselves. They demanded that all Makians, Tidores and Sananas be expelled from Tobelo. Bachmid, himself a Makian, told them that there were several ethnic groups involved in the conflict in North Maluku and that everybody needed to take a measured and careful response. His wording – ‘there are provocateurs within our own community’ – conveyed a different meaning to the predominately Christian audience, however.42 The members of FKPKHU understood his statement to mean that there were certain individuals trying to provoke violence in Tobelo. When he refused their demands to identify who was provoking tension in the city, he was himself accused, and left Tobelo for Ternate in late December, several days before the rioting broke out. On 26 November several Christian men discovered a Muslim Tobelo sewing white robes and headbands in the suburb of Gura, clothing associated with militant Islam. Christian leaders reported the man’s activities to the police, who arrested him. While Christians and the FKPKHU demanded that the police investigate who had commissioned the tailor to make these uniforms, no further action was taken. At least one Muslim community leader, Dr Musriyono Nobio, demanded that the police release the man, but pressure from FKPKHU prevented this. On 7 December a political delegation travelled from Ternate to Tobelo City, made up of the Sultan of Ternate, Mudaffar Syah, the interim governor, Surasmin, and the district head of North Maluku District, Rusli Andiaco. The delegation members addressed a crowd in the grounds of the sub-district government offices, including large numbers of Christian IDPs from Central Halmahera, in an effort to reduce tensions in the town. Groups of people at the meeting displayed banners stating ‘Muslims are our brothers and sisters, but Makians and Tidores are riot instigators who have to leave Tobelo Sub-District’.43 This indicates that, even at this late stage, many people still felt the violence in the province so far had not been about religious tension but had been instigated by individuals from certain ethnic groups. However, the banner was highly provocative, particularly as the sub-district head, Agil Bachmid, was a Makian. Security concerns and religious tension continued to converge in late December with the beginning of Ramadan and the approach of Christmas. A rumour circulated among the Christian community that they faced a ‘Bloody Christmas’ (Natal Berdarah), unsurprisingly, given the events in Ternate and Tidore, causing a great deal of anxiety. The Islamic fasting month of Ramadan also began in mid-December, according to several Muslim respondents, a time when Muslims are more sensitive to apparent insults to Islam. Given that the month of Ramadan

Dispersion – Tobelo and Galela 107 coincided with the noisy Christmas preparations and celebrations typical of Tobelo, tensions were at high levels. Several small disputes arose when Christians sang loudly outside the homes of Muslims or youths from both communities threw stones at houses. Graffiti appeared on the walls of the city, with messages such as ‘Tobelo is the Second Israel’ and ‘PLO’. Small groups took a more aggressive stance towards the other community. There are strong indications that some groups in the Muslim community, such as one centred around the family of Yahir Patty in Gosoma, believed it was better to initiate hostilities before Christians could do so.44 Many leaders from both communities continued to work hard throughout December to prevent conflict. On 21 December, Muslim and Christian community leaders met and agreed that conflict should not occur in Tobelo. They reaffirmed the ethnic relationship between Tobelo Christians and Muslims and the ties of tradition embodied in the cultural structure Hibua Lamo. In some areas, members of both Christian and Muslim communities cooperated in taking physical precautions to prevent conflict. For example in the suburb of Gura, members of both communities stood guard at posts on the main street intersections. Several Muslim respondents said that, even at this stage, they still believed that conflict would not break out, because Christian leaders appeared to be resolved to prevent it. Others believed that conflict might occur but that the security forces, particularly TNI, would bring it under control. However, some efforts at conflict prevention aggravated the situation. One particularly acrimonious community meeting deteriorated into shouting between a pastor, Samuel Ray Ray, and a Muslim leader, Musriyono Nobio. According to witnesses, each shouted that they could mobilize several thousand men to secure the city. Around this time, the sub-district head, Agil Bachmid, made frequent trips between Tobelo City and the major Muslim villages of Togoliua and Gorua, stating he was attempting to alleviate the growing concern in the sub-district. However, the fact that he carried out many of these visits at night and appeared only to visit Muslim villages convinced many Christians that he was in fact preparing the Muslim community for conflict. As mentioned above, after continued accusations that he was attempting to provoke a Muslim attack, Bachmid left Tobelo for Ternate. As Christmas Day approached amid rising tension and with the ‘Bloody Christmas’ rumour fresh in the minds of members of both communities, several officials of the GMIH Church sought to ensure security for the GMIH complex in central Tobelo. The compound was of major religious and administrative importance for the Christian community, containing the central church (Bethesda) and the Synod office of GMIH as well as north Halmahera’s main hospital, Bethesda. Many Christians considered that this compound would be a primary target of any Muslim attack. On 24 December a resident of the GMIH compound, Pastor Charles Kaya, requested that the local military commander provide security for the compound over the Christmas period.45 Receiving no response, he then contacted churches in the predominately Christian villages of Kupa Kupa and Efi Efi in south Tobelo and requested that 40 villagers travel to Tobelo City to guard the church overnight. The

108 Ethno-Religious Violence in Indonesia Synod hired a truck owned by a Christian Chinese Tobelo shop owner to transport the villagers. The truck drove through the centre of Tobelo, a strongly Muslim area, with the men standing in the back wearing red headbands and holding spears. Muslims I spoke to saw this at the very least as a provocative act and some as evidence of a Christian plan to attack Muslims, led by GMIH officials. In response to this development, Haji Ponoh, a Muslim community leader in Tobelo and Gorua, asked several Christian leaders, including Pastor Aesh (the Chairman of GMIH), Pastor Jacob Soselissa (a pastor of Ambonese ethnicity from Kupa Kupa, South Tobelo) and the former sub-district head, Mr Huwae, to explain why armed men had been brought into the city. The Christian leaders assured him it was only to guard the church, and that if no violence occurred during Christmas then they would simply return to their villages. Indeed there appears to be little evidence that GMIH brought these men to Tobelo to either attack or to intentionally provoke the Muslim community. The pastors involved probably did not foresee that the group would enter the city in such a provocative manner. Yet it seems likely that, although the rioting did not start until two days later (Christmas Day, 25 December, passing without incident), the impact this act had on the Muslim community and on the relationship between the two communities meant that this was, ultimately, a primary trigger for the violence.

Violence in Tobelo On 26 December a group of Muslim youths threw stones at the house of a retired Christian Ambonese police officer, Chris Maitimu, in the western part of Gosoma behind the major sports field in Tobelo City, Lapangan Kariangan. When Maitimu came out of his house and accused the youths, a crowd gathered and threw stones at Muslim houses. According to a Christian respondent, a crowd of approximately 80 Muslim men then attacked a Christian neighbourhood guard post in Gosoma. Within an hour, hundreds of Christian and Muslim men were fighting on the street with swords, machetes and homemade bazookas, and, according to some accounts, wearing white and red headbands. Although Muslims blockaded the main southern entry into Tobelo, several dozen Christian militia members were able to enter the city via a road through gardens into the area of Gosoma. There were several casualties from both communities in these initial clashes. In the early morning of 27 December Christians fought running battles with Muslims between Gosoma and the monument on Tugu Street. At around 9.30 a.m., Muslims set fire to the Pentecostal church, and by 10.30 a.m. had pushed the Gosoma Christians back and were attacking Gosoma church, the main church in the south-west of the city. Several witnesses stated that a local (Tobelo) Muslim military officer named Hatalah joined this attack and shot dead Wanto Hohakay, one of the local Christian men defending the church. 46 By mid-morning, Muslims controlled the central area of Tobelo.47 Those Christians who remained in the city were confined to the suburbs surrounding the centre of the city such as Gura and southern Gosoma. Many also fled through the forest to Christian villages to the south. Some Christians, mostly Chinese shop owners, fled

Dispersion – Tobelo and Galela 109 to the military base (Kompi) on the northern edge of the city. Several shop ruins in the centre of the city still bear the message ‘Muslim’ from this period, painted by their Muslim owners seeking to prevent the destruction of their premises. During this period the city’s infrastructure was relatively untouched by the violence. Muslims did not destroy shops owned by Christians in the central city they now controlled. One exception was a complex of five houses near the offices of the sub-district government which were burned to the ground. These houses were almost all occupied by current and former civil servants of Ambonese ethnicity, including the former sub-district head, Mr Huwae. Local Muslim men I interviewed, when questioned as to why they targeted those houses, claimed the residents were ‘RMS’ (Republic of South Maluku) Christian separatists. While Muslims alleged that the RMS was present in North Maluku and in Tobelo in 1999, there is little evidence to support this claim. Prior to September 1999, while North Maluku remained part of Maluku Province, many officials in the local Tobelo government were appointed by the provincial government in Ambon. Many of these officials, often of Ambonese ethnicity, were still in their positions after the creation of North Maluku Province and many others, like Mr Huwae, had retired but remained in Tobelo City. On this occasion, being of Ambonese ethnicity was apparently enough to raise suspicion and animosity. On the morning of 27 December, as violence continued in Tobelo City, clashes erupted to the north of the city in two large neighbouring villages with Muslim majorities, Gorua and Popilo (see Map 5.1). After violence started in the city on the evening of 26 December, Muslim leaders in Gorua offered to guard the church, an offer that local Christians refused. The following morning, Muslims attacked the Christians guarding the church. Because the church and most Christian houses were located at the southern end of Gorua, Christians were able to safely reach Tobelo City and the village of Wari on the city’s northern edge. Muslims in Gorua destroyed the church and all Christian houses after the Christians had fled. Their escape, and the lack of casualties, suggests the attack was carried out to expel the Christian community rather than kill them. Over the next few days, clashes occurred between Muslims from Gorua and Christians from Wari. In the adjacent village of Popilo, Muslims also appear to have launched a pre-emptive attack. According to the Christian pastor in Popilo, the Muslim village head arrived and reassured local Christians that Muslims would not attack them, and they would therefore not require assistance from the neighbouring Christian village of Ruko to guard their church. However, soon after that meeting, Muslims did attack the Christian area of the village and destroyed the church. Christians believe Muslims assured them only in order to ensure that they (Christians) were unprepared for conflict. However, my interviews with the village head and other Muslim leaders make it seem likely that the assurances had been genuine but that the attack had been launched by a group of more aggressive men from the village. The fact that the entire Christian community from Popilo reached Tobelo safely also suggests that, as in Gorua, there was no intention to kill them but only to drive the community from the village.

110 Ethno-Religious Violence in Indonesia During the night of 26 December several senior Christian community leaders telephoned Benny Bitjara in Kupa Kupa and other Christian militia leaders in the major villages south of Tobelo City.48 That night, and the following morning, Benny Bitjara, along with Hersen Tinangon in Pediwang, mobilized thousands of Christian Kaos to travel to the city. The thousands of Kao militia members had maintained a high degree of readiness for conflict since their attack against Malifut in October, with machetes, spears, bows and arrows and sulphur bombs. While Muslim Kaos wished to join the Christians in their attack, the militia leadership refused this request.49 Christians referred to this militia as the Pasukan Merah (Red Troops). As it travelled north, the militia was forced to pass through the large Muslim village of Togoliua, about 30 km to the south of Tobelo City. The 1,365-strong population of the village is Muslim, primarily of the Tobelo ethnic group. Upon reaching Togoliua, the Christian militia was blocked by coconut trees that the local villagers had laid across the road in an effort to prevent the militia from reaching Tobelo.50 This forced Benny Bitjara and the Kao militia to board boats in the village of Kusuri and travel around Togoliua, rejoining the road to Tobelo City. Togoliua Muslims explained their actions by stating that they had family in Tobelo and they attempted to block the road in order to prevent the Kaos travelling there and engaging in what would inevitably become a one-sided massacre. They argued that when Malifut was destroyed in October, they and other Tobelo Muslims had not travelled there to assist the Muslim Makians. Nonetheless, Bitjara arranged with the villagers of Kusuri to attack Togoliua once Christians had regained control of Tobelo City. Muslims in the mixed village of Gamhoku also attempted to block the road laying large rocks across the road to prevent the militia’s trucks from passing through.51 However, Christian community leaders persuaded the Muslims to clear the road of rocks, reassuring them that nobody from Gamhoku would travel to join the violence in Tobelo and they would protect them from the Pasukan Merah. As the Christian militia travelled through Gamhoku towards Tobelo on the evening of the 26 and 27 December, members of the Christian community did indeed protect local Muslims by standing in front of the mosque as the Muslim community sheltered inside. A Christian community leader, Sakeus Odara, led hundreds of Christian men from the village of Tobe towards Tobelo City in vans and other vehicles, destroying several mosques and the houses of Muslims in the villages along the road. The militia killed members of the Muslim minorities in these predominately Christian villages, with the exception of Gamhoku where local Christians protected the small Muslim community. Odara stated to me that it did not matter if the Muslims they met were Tobelos, Makians or Javanese, he ‘did not choose’.52 On the evening of 27 December, this militia gathered in the church in Pitu village on the southern border of Tobelo City which was dominated by Muslims. The militia included men from Pediwang and other villages in Kao and almost all men from the villages south of Tobelo. A Protestant Pastor led the militia in a religious ceremony and blessed the militia’s weapons inside the church.

Dispersion – Tobelo and Galela 111 That evening, as the militia prepared for night in the grounds of the village church, with smoke from the fires of Tobelo City clearly visible against the night sky, Benny Bitjara walked among his troops holding a bible. To the excitement of the men waiting for battle, Bitjara informed them that Jesus had appeared before him and commanded him to read Psalm 91 to his troops, which he duly did. An excerpt from the psalm follows.53 You who live in the shelter of the Most High, who abide in the shadow of the Almighty, will say to the Lord, ‘My refuge and my fortress; my God, in whom I trust’ For he will deliver you from the snare of the fowler and from the deadly pestilence; he will cover you with his pinions, and under his wings you will find refuge; his faithfulness is a shield and a buckler. You will not fear the terror of the night, or the arrow that flies by day, or the pestilence that stalks in darkness, or the destruction that wastes at noonday. A thousand may fall at your side, ten thousand at your right hand, but it will not come near you. You will only look with your eyes and see the punishment of the wicked. Because you have made the Lord your refuge, the Most High your dwelling place, no evil shall befall you, no scourge come near your tent. For he will command his angels concerning you to guard you in all your ways. On their hands they will bear you up, so that you will not dash your foot against a stone. You will tread on the lion and the adder, The young lion and the serpent you will trample under your foot. According to Bitjara the message from Jesus to his troops was that the following day, when they entered the city, it would not be they who would be doing the fighting, but angels.54 The troops then slept in and around the small Pitu church.

112 Ethno-Religious Violence in Indonesia At 5.30 a.m. the following morning the Pasukan Merah attacked Tobelo City. The majority of the militia approached and overran a Muslim blockade at Kaliseratus on the southern edge of the city, while hundreds of other militia moved through the area of Kampung Baru in the south-east of the city, destroying large numbers of houses and a mosque.55 Muslims fled, hid in the dim light or were cut down by the advancing Christians. The militia used sulphur bombs in the attack, which both terrified Muslim crowds into running and almost completely destroyed any buildings into which they were thrown. Throughout that morning, Muslims ran from the central city as the thousands of Christians moved in. By around 10 a.m. on 28 December the majority of Muslims remaining in the city withdrew to the largest mosque in Tobelo, Mesjid Raya. Military personnel from the base in the north of the city surrounded the mosque and held back the Christian militia. Muslims also sought shelter in a smaller mosque several hundred metres from Mesjid Raya where they too were protected by military personnel. Muslims from the suburb of Gosoma also sought protection in the sub-district military and police compounds (Koramil and Polsek). When the Pasukan Merah demanded that a group of about 15 men, women and children sheltering in the small military compound be handed over to them, the predominantly Christian soldiers fled. According to witnesses, lacking firearms, the military personnel had little choice but to hand over the group.56 As the Muslims attempted to run from the building, the Christian militia killed them with spears. The thousands of Christian militia then controlling the city, many of them from outside Tobelo, began to destroy all Muslim houses and loot the shops along the main street of Tobelo.57 Many shops, including some owned by Christians, were destroyed during the chaos and looting in this period. Some Christian shop owners I spoke to said they did not blame the militia, as the goods they took were just reward for expelling Muslims from the city.58 As Christians returned to their homes, they began to mark their houses with crosses to prevent their destruction. Approximately 100 Muslims died in the attack on the city by the Pasukan Merah. Many of their bodies were burned in the street. One particularly large pile of corpses was burned in the middle of Jalan Pelabuhan near the port.59 On the afternoon of 28 December the military evacuated the Muslims remaining in the city’s mosques to the military base on the northern edge of the city. As these Muslims entered the base, Christians already sheltering there left and returned to the city. One Christian woman recounted how the return to the city was terrifying even for Christians, as the Christian militia was looting and burning shops and intimidating those attempting to return to their businesses. She said she witnessed militia stop a vanload of Muslims trying to flee the city and burn the vehicle without allowing the passengers to get out. Upon the evacuation of the Muslims sheltering in Mesjid Raya, Christian militia entered and destroyed the building. Following the expulsion of all Muslims from Tobelo, a pastor ‘baptized’ Benny Bitjara as panglima or commander of the North Maluku Christian militia.

Dispersion – Tobelo and Galela 113

Attacks on remaining Muslim villages Having expelled all Muslims from Tobelo to the military base or northwards to Galela Sub-District, leaders of the Pasukan Merah met in the Elim church in the suburb of Gura in the north of the city on the evening of 28 December. They were joined by Christian religious and community leaders. As well as being a religious service to celebrate their control of Tobelo, the meeting was called to discuss what to do about several large Muslim villages to the north and south of Tobelo. A large Muslim community remained in the two adjacent villages of Gorua and Popilo, 10 km to the north of Tobelo. As discussed above, the two villages had already experienced rioting on 27 December. Approximately 30 km to the south was the Muslim village of Togoliua, where villagers had attempted to block the Christian militia arriving from Kao. The leaders of the Christian community, including pastors, local politicians and militia leaders, decided to attack these three villages. This decision was taken for several reasons. First, most Christians from Tobelo and Kao were angry that Muslims in Gorua and Popilo had attacked the Christian minority and destroyed the churches in those villages. Christians in Tobelo considered the Muslims in Gorua in particular to be ‘fanatic’ with regard to their religion and in their relations with Christians. The presence of Christian IDPs from those two villages among the militia undoubtedly exacerbated this anger. Similar sentiment was felt regarding Togoliua. Christians believed that in the attempt to block the Christian militia, Muslims in the village had shown themselves to be part of a Muslim plan to isolate Tobelo City from Christians in Kao. Second, Christians were concerned that after the Pasukan Merah had returned to their homes, Tobelo would once again be vulnerable to attack from these Muslim villages. A large portion of the Christian militia was from Kao, south of Togoliua. In addition, it was believed that the vast majority of the military personnel that would inevitably arrive from Ternate or elsewhere to reinforce local contingents would be Muslim and would assist these villages in attacking Tobelo. With many members of the militia preparing to return to their homes and with their arsenal of bombs and other weapons becoming dangerously low, Christian militia leaders, pastors and other community leaders decided to take swift and devastating action against the villages. On 29 December several thousand Christian militia members travelled in trucks towards Gorua from Tobe village, according to several respondents led by Sakeus Odara. To avoid the military base, which had blocked the road north, the group drove through gardens to the north-west of Tobelo. Men from both Gorua and Popilo held out against the onslaught for approximately an hour before being overrun. According to several respondents, the scene in Gorua was chaotic. Respondents from both sides of the conflict said the Christians committed atrocities during the attack, including the extraction and eating of hearts and other body parts. Photographs taken by military personnel following the attacks confirm that corpses were disembowelled during the violence. Several Christians claimed these acts were a reversion to what they termed pre-Christian practices.60 Around 90

114 Ethno-Religious Violence in Indonesia people (almost exclusively Muslim) were killed in Gorua and the entire village was burned. As the Pasukan Merah overran Gorua, the Muslim community retreated along the road north to the village of Popilo. The Pasukan Merah advanced through Gorua destroying houses and several mosques and attacked Popilo. This village was also quickly overrun, and many Muslim men escaped into the forest. Some joined the large number of women and children hiding inside the village mosque. The Christian militia attacked the building, several men claiming they were forced into this action because the Muslims continued to fire arrows from inside. The attack against the mosque continued throughout the night until resistance from inside had been quelled. Upon entering the mosque, which was now strewn with bodies, the Christians discovered a pit underneath the pulpit where a large group of Muslims were hiding. A militia member dropped a bomb into the recess. Approximately 160 people were killed in Popilo, bringing the deaths in the two villages to around 250. It took several days for those Muslims who had managed to flee from Gorua and Popilo to travel through the mountainous forested area between Tobelo and Galela to reach Soasio.61 Christian militia also searched the forest to undertake (in the words of one militia member) a ‘cleansing operation’ of Muslims.62 At the same time a large section of the Christian militia returned south to Togoliua, led by Benny Bitjara. Several Christian leaders including Bitjara and Hersen Tinangon, who had remained in the large northern Kao village of Pediwang had organized a simultaneous attack on Togoliua from Tobelo to the north and the village of Kusuri to the south. Kusuri village had recently been built by the local government to house Tobelos relocated from the interior of North Halmahera. Local pastors had, since its resettlement, converted the community to Christianity. Villagers from Kusuri and from Kao attacked Togoliua on the morning of 29 December, reportedly led by a convicted murderer named Nyeteke. The militia from Tobelo was late in arriving at Togoliua and the Muslim village had largely been overrun by the Kusuri villagers by the time they arrived.63 Large numbers of Muslims were killed trying to prevent the Kusuri Christians from entering the village, while hundreds of others ran into the forest. As in Popilo, dozens of Muslims, mostly women and children, fled to the village mosque, Mesjid Al Islah. The Christian militia, including those who had arrived from Tobelo, attacked the mosque, throwing bombs into the building, which caused the walls to collapse in on the people cowering inside. The militia speared or shot with bows and arrows any of the terrified Muslims that attempted to escape. The Pasukan Merah killed approximately 250–300 people in Togoliua.64 Contrary to several reports in national Islamic newspapers which stated that those killed had been transmigrants from Java, all respondents in Togoliua and those involved in the attack stated that all those killed were Halmaheran people, most Tobelos like their assailants.65 The erroneous reports in the national press probably stem from Togoliua’s proximity to a transmigration settlement called Trans Suka Maju, where large numbers of Javanese reside, both Muslim and Christian. Pasukan Merah forces did attempt to attack that settlement after the destruction of Togoliua but were stopped by the local Protestant pastor, who told them there was

Dispersion – Tobelo and Galela 115 no tension in the settlement nor any risk of violence occurring.66 Atrocities reportedly occurred in Togoliua similar to those in Gorua and Popilo.67 The Christian militia searched in the forests for those villagers who had fled, killing an undetermined number. Many who escaped spent approximately two weeks in the forest attempting to reach Soasio in Galela. All buildings in the village were destroyed, including local government offices, a health clinic, a school and a new water purification plant. The army company stationed in Tobelo City requested that NHM, the company operating the mine at Gosowong in Kao, provide a backhoe to bury bodies from the village. As the surrounding sub-district was engulfed in violence, Christians in Gamhoku were no longer able to protect their Muslim neighbours. On 29 December a small number of Muslim men from Gamhoku village became involved in a clash in nearby gardens with Christians from the neighbouring village of Upa. After the Muslim men returned to Gamhoku, several bleeding from the fight, the local Christians, fearing the arrival of men from Upa, moved a large section of the Muslim community into the village church for protection. Another group remained in the village mosque. Inevitably, approximately a hundred Christian men from Upa and other adjacent villages, fresh from the violence in Tobelo City, arrived and attacked the church. During the attack, a fire broke out inside the building, and the Christian militia killed approximately 30 Muslims as they fled.

Galela erupts in violence Background The neighbouring sub-district of Galela had a population of 35,245 in 1999,68 evenly dispersed between the capital Soasio and a number of large villages located around the shores of Lake Duma (see Map 5.2).69 The majority of the Galela population are from the Galela ethnic group, closely related to the Tobelos, and many family connections exist between the two communities. The majority of Galelas and of the Galela Sub-District population are Muslim, as are the majority in the sub-district capital, Soasio, located around 20 km north of Tobelo. 70 Most villages in Galela had mixed religious populations, usually with small Christian minorities. Christians were concentrated in Duma, which was exclusively Christian, and the neighbouring villages of Dokulamo and Soatobaru. A chain of contiguous villages, either exclusively Muslim or Christian or religiously mixed, stretched along the western bank of Lake Duma (see Map 5.2). The capital Soasio and the villages located nearby such as Igobula, Togawa and Pune had Muslim majorities. The Christian community considers Duma to be the historical and spiritual centre of the Protestant Church on Halmahera. The village was the site of the first church set up by the Dutch Reformed Church and the centre of its missionary activity in North Maluku. The village of Mamuya in south Galela was also the site of the first Portuguese Catholic missionary settlement, although there are now few, if any, Catholics residing in Galela.

Map 7 Galela

Dispersion – Tobelo and Galela 117 As in Tobelo, all respondents agreed that relations between Christians and Muslims in Galela had generally been good before the Ambon conflict and the riots in Ternate and Tidore. Just as in Tobelo, close family ties existed between Muslim and Christian Galelas. Almost all respondents had a parent, uncle, grandmother or other relative from a different religion. Respondents from both religions told me they frequently celebrated religious festivals together, worked together, assisted one another in building houses and dined in each other’s homes. However, some (generally Christian) respondents stated that tension had been rising between the communities for some time. For example, in the village of Soatobaru several Christians recounted their anger at how Muslims in the village had deliberately constructed a mosque that was bigger than the church. 71 Most inhabitants of Galela are employed in the copra industry. As discussed in Chapter 2, in 1991 the Indonesian company PT Global Agronusa Indonesia (GAI) established a 2,000 ha banana plantation in the north of Galela. In 1999 GAI employed approximately 3,000 people, including large numbers of locals and people from outside north Halmahera who lived on the plantation base. In 1998, after the onset of reformasi, a group of local university students returned to the sub-district to organize protest action over the prices paid for land by GAI to local farmers. However, all respondents were adamant that this had been supported by both religious communities, and that GAI had nothing to do with the violence that broke out in 1999. 72 Rising tension The ongoing violence in Ambon created moderate tensions in Galela, particularly in the area around Lake Duma. Several Christian respondents claimed that in mid-1999 Muslim youth in Gotalamo and elsewhere were carrying out fitness and weapons training. After the destruction of the two Kao villages, Sosol and Wangeotak, in August the Protestant Church (GMIH) in Galela sent food and clothing to the Kao IDPs sheltering in Kao. This assistance subsequently caused tension with the Muslim community, members of which accused Christians of sending men and weapons to assist in the Kao retaliation against Malifut.73 As in Tobelo, Christians in Galela were angered at the riots in Tidore and Ternate. A small group of IDPs from Payahe and elsewhere in Central Halmahera arrived in Christian villages around Duma. They were few in number and their presence in Galela did not affect inter-communal relations to the extent that the much larger numbers of IDPs in Tobelo had done. However, the dissemination of rumours during November and early December appears to have increased tension to very high levels around Galela. Christian villagers said that by December it was very difficult to travel south to Tobelo, the major local city and the source of almost all supplies needed in Galela. The only road to Tobelo wound through Muslim villages such as Togawa and Igobula as well as through the sub-district capital, Soasio, which had a strong Muslim majority. Community leaders, both in the capital Soasio and in the densely populated area around the western edge of Lake Duma, attempted to lessen this rising tension and

118 Ethno-Religious Violence in Indonesia maintain peace. In Soasio, leaders from both Muslim and Christian communities met and formed the organization Galela Sariloha Family Communication Forum (Forum Kommunikasi Keluarga Galela Sariloha). The goal of the organization was to anticipate conflict and reduce any tension that arose between the communities.74 Local community and religious leaders also held a series of meetings in villages around Lake Duma to discuss ways of ensuring that violence did not occur between them. At least one Christian leader from Duma, Josafat Etha, travelled to several surrounding Muslim villages such as Ngidiho and Gotalamo. The village heads of Soatobaru and Dokulamo also met to promote peace. Leaders from both religious communities in Soakonora agreed to monitor and deflect provocation of their communities.75 The sub-district government appears to have attempted little in the way of peace-building, however. In Soasio, Christian respondents claim neither the Muslim sub-district head nor his staff did anything to ameliorate the rising tension in the capital. Members of both communities were increasingly suspicious of the intentions of the other. Just as in Tobelo, in late December the atmosphere in several villages in Galela became tense. Rumours that each community was planning an attack to coincide with either Christmas or Idul Fitri spread through all villages from Samuda to Ngidiho.76 Communities in the villages of Galela, particularly those around Lake Duma, prepared for violent conflict at least as vigorously as those in Tobelo Sub-District. Against the orders of some village heads, members of both communities made bows and arrows and spears.77 Small incidents that appeared to demonstrate the aggressive intent of the other community became more frequent. In several villages, youths from both communities threw stones at the houses of members of the other community. In the capital, Soasio, Christian businesspeople and local government staff fled the town several days before rioting broke out.78 While some community leaders were attempting to arrest the rising tension and prevent conflict, some Christian leaders from Duma and Soatobaru also appear to have planned to respond in a coordinated manner if violence did occur. Further south and somewhat isolated from the rest of Galela, the mixed village of Mamuya was divided by particularly high tensions. The population was almost evenly split between Christians and Muslims and was spatially segregated, with Muslims occupying the northern half of the village and Christians the southern half. Christians in the village felt isolated from the main Christian villages of Duma to the north and Tobelo to the south. On 16 December a stone-throwing incident led to an assault on two police officers, and on 19 December a Muslim man was badly beaten by Christian youths. Christian respondents said that for several weeks before the eventual outbreak they could not venture into the northern part of the village for fear of being attacked.79 Christians were also concerned about aggression from the Muslim village, Luari, located just across the border in Tobelo Sub-District. Muslims considered the Christians in the southern part of that village to be militant and preparing for conflict.80

Dispersion – Tobelo and Galela 119 Violence in Galela People in the capital and villages of Galela first heard of the rioting in Tobelo when several passenger vehicles that regularly travel between the two sub-districts turned back to Galela on the evening of 26 December. When further accounts reached the banana plantation, workers began to flee their compound in large numbers. The first clashes in Galela Sub-District appear to have taken place in Soasio and the adjacent village of Pune. Throughout the night of 26 December Muslim IDPs had arrived by speedboat in Soasio from Tobelo. The Christian community of Soasio fled to the GMIH church compound and spent a nervous night before a mob of Muslim men attacked the compound with homemade bombs and spears at dawn. Muslim youths from Pune, considered ‘radical’ by the Christian community, led the attacks. Christian respondents stated that the imam from the Pune mosque pleaded with the crowd of Muslim men to stop attacking the church compound, but no government officials attempted to placate the crowd. Although the sub-district police station was only 200 m away, police personnel did nothing to stop the attack.81 The Christian community was able to flee into the forest and was evacuated by villagers from Mamuya two days later.82 The church and surrounding compound was destroyed. On 27 December Muslims in the villages of Togawa and Igobula near Soasio also expelled the small Christian communities from their own villages, killing several people. After stories of rioting in Tobelo and Soasio reached the villages around Lake Duma on the morning of 27 December, crowds of armed men gathered in front of mosques and churches in the adjacent villages of Soatobaru and Dokulamo. Both sides claim the other was wearing symbols of religious identity, such as either red or white headbands. Several Muslim and Christian leaders moved between neighbouring villages.83 To members of the other religious community, these individuals appeared to be coordinating attacks with neighbouring villages. After several hours in which the opposing forces faced each other on the streets, an explosion, apparently from a homemade bazooka, ignited rioting in Soatobaru in the late afternoon of 27 December. As the strongest Christian village in Galela, Duma was always going to play a central role in any conflict in the sub-district. On 27 December, having heard of the rioting in Tobelo, the Christian villagers of Duma gathered in turns for a religious service in the local church. While small groups entered the church, the rest maintained security posts at both ends of the village. 84 Weapons were taken into the church and blessed by the pastor. The pastor instructed the residents of Duma that they should fight if attacked but not attack first.85 When villagers saw smoke rising from rioting in Soatobaru, and a group of Duma men clashed with Muslim youth from Gotalamo to the north at approximately 5 p.m., the villagers launched offensive action with devastating consequences for the area. Within three hours, the Duma militia had driven all Muslim residents of Gotalamo from their homes into the forest and destroyed the entire village. The militia destroyed the mosque in the village despite the appeals of at least one Christian leader from Duma.

120 Ethno-Religious Violence in Indonesia At dawn the following morning, 28 December, men from Duma attacked the villages of Dokulamo and Soatobaru, quickly expelling the Muslim communities. Muslims from these villages were forced to walk through the forest towards the capital Soasio, which was still controlled by Muslims. At 12.30 p.m., immediately after the attacks against Dokulamo and Soatobaru, Christians from Duma, Soatobaru, Dokulamo and Makete moved north and attacked the mainly Muslim village of Ngidiho. Prior to this attack, the Muslim village head of Ngidiho had warned Christians living in the village to leave, which they did, moving to Duma.86 One report stated that, following the departure of the Christian community, Muslims from the villages of Simau and Toweka destroyed the Christians’ vacated houses, before attacking the Christian village of Makete, where they were met and routed by the advancing Christian militia from Duma.87 However, my research suggests that Ngidiho and the villages to the east did not in fact attempt any attack against Duma before Christians from Duma attacked Ngidiho. Eight Muslims were killed in the attack. On the morning of 28 December Muslims from Igobula attacked the Christian villages of Bale and Samuda to the west. Muslims were angered that Christians from Bale had travelled to Soatobaru to help rioting Christians on the afternoon of 27 December and that all Muslims had fled the area to the west.88 Nevertheless, security concerns were paramount in this decision to attack. The village head, fearing for the safety of women and children, said that his community did not want to wait until his village was attacked by Christians from Bale and Samuda and the other larger villages to the west (Duma and Soatobaru).89 Muslim leaders decided it was better to create a buffer zone between the Christian and Muslim areas around the western edge of Lake Duma by driving the Christian community from the village. Christians from Samuda spent four days in the forest before arriving at Soatobaru and Duma. Within two days, the violence in Galela had established two exclusive Christian and Muslim areas. Christians were concentrated in the area of Duma and Soatobaru and Muslims in the area from Igobula east to Soasio. Attacks continued between Christians and Muslims from these two areas for several weeks until the Muslim community left Igobula and Togawa and moved to Soasio or to Ternate and South Sulawesi.

Insecurity, identity and violence Widespread fear developed among Christians and Muslims that they faced the danger of attack from the other community in late 1999. By December, both religious communities in Tobelo and Galela perceived aggressive intent on the part of (at least some sections of) the other community. Indeed the perceptions of both sides continue to be that the other systematically planned and initiated the violence. Both sides point to the other community’s period of intense weapons preparation leading up to the conflict, with frequent explosions and sightings of people making spears, arrows and other traditional weapons. Most stated that such preparation was either carried out exclusively by the other side, or that members of

Dispersion – Tobelo and Galela 121 their own community made weapons only in response to the obvious preparation of the other community. Most Christians from Tobelo City and the surrounding area said that, by December, large sections of the Muslim community had planned to drive all Christians from Halmahera, as demonstrated by the ‘Bloody Christmas’ rumour that circulated prior to Christmas. Incidents such as the tailoring of white robes and frequent late-night meetings between Muslim leaders (including the sub-district head) created this perception. Several incidents during the violence itself appeared to Christians to confirm the existence of such a plan. Christians claimed that upon expelling the Muslim community from the city they found in several mosques a number of implements intended for the torture of Christians. These included large piles of razors and citrus fruit, wooden implements for raping women, along with swords engraved with Arabic inscriptions. Christians argue that while Muslims were a minority across the entire sub-district, they had planned to negate this by isolating Tobelo City, where they were a strong majority. According to this account, during his late-night visits, the Muslim sub-district head of Tobelo, Agil Bachmid, had requested that on the outbreak of violence in Tobelo City Muslims in Togoliua prevent the Kao militia from passing through the village on the only road to the city. Many Christians also believe that arrangements had been made for Muslim reinforcements to arrive by speedboat from Ternate.90 One Christian man recounted how, as a friend of local Muslims, he had been standing in a Muslim guard post during the first night of the rioting when he overheard a local Muslim community leader, Musriyono Nobio, speak to somebody in Galela on a two-way radio and tell those present in the post that seven speedboats would soon arrive from Soasio carrying approximately 500 Muslim men.91 In turn, Muslims were convinced that Christians, in particular officials of GMIH and several local leaders, including some who had fled Ternate in November, pre-arranged the expulsion of Muslims from the area. Muslims claim that when GMIH evacuated IDPs from Central Halmahera to Tobelo they brought only men armed with traditional weapons. The apparent manufacture of large numbers of weapons by Christians and the arrival of thousands of Christian reinforcements after the violence had begun appeared to confirm this suspicion. There is very little evidence that Muslim leaders planned an attack or intentionally initiated conflict in Tobelo. No material evidence of such a plan exists in Tobelo. While Muslims were a majority in the city, they were a substantial minority in the surrounding sub-district. It must have been apparent to Muslims in the city that Christians in nearby villages would come to the aid of Christians in the city. In addition, they were aware that the large Christian militia that had been formed in Kao Sub-District only a few hours drive from Tobelo had close ties to Christian Tobelos and would almost certainly seek to help their co-religionists. It also seems unlikely that there was a coordinated plan to reinforce Muslims in Tobelo. Residents of Togoliua who attempted to halt the Kao militia were not acting out a conspiracy to control Tobelo, but attempting to prevent a massacre of Muslims in Tobelo City. It was highly unlikely that Muslims would travel from

122 Ethno-Religious Violence in Indonesia Ternate to Tobelo by speedboat, a journey of perhaps 24 hours. Several Muslim respondents, and also one Christian community leader, stated to me that in visiting Muslim communities Agil Bachmid was trying to prevent violence, arguing for respect for the ties of Hibua Lamo and not preparing people for violence. 92 It also appears unlikely that the Christian community or its leaders deliberately sought to initiate violence on 26 December. It is unlikely that Christians would delay for 36 hours the arrival of reinforcements from the villages to the south, thereby allowing Muslims to control Tobelo for that period, expel all Christians from the inner city and have the opportunity to destroy churches and other Christian buildings in the city. Benny Bitjara, Sakeus Odara and the Christian militia would undoubtedly have entered the city well before the morning of 28 December if such a plan had existed. While precautionary preparations for conflict were undertaken on both sides, I believe this was because of an increasing sense that a clash was inevitable rather than a desire to initiate violence. The conflict in Tobelo began as a direct result of several actions and incidents that occurred in the midst of a severe ‘security dilemma’, both physical and identity-based, between the two communities. Even before the riots in Ternate and Tidore, Muslims and Christians in Tobelo and Galela had started to become concerned about the possibility of conflict spreading to north Halmahera. However, after anti-Christian rioting began on those islands and in Central Halmahera, security concerns became acute, particularly in Tobelo City. Christians were angry at the targeting of Christians elsewhere in the region, and many Christian leaders returned to Tobelo from Ternate radicalized by the experience. More importantly, thousands of Christian IDPs were evacuated to Tobelo after violence in Central Halmahera. The situation of these people and their stories of attacks against Christian communities, the murder of pastors and destruction of churches engendered anti-Muslim feelings among the Christian community in Tobelo. This large-scale influx of Christian IDPs visibly heightened emotions among Christians and also worried local Muslims. Many believed they were at risk, particularly from the IDPs, who had no cultural or family ties with local Muslims. The combination of anger and fear caused individuals and groups on both sides to take actions that in turn increased the level of tension. Members of both communities constructed weapons, including bombs, which frequently exploded during construction. In other cases, actions by some individuals and groups appear to have been designed to intimidate the other community, if not to initiate conflict. Youths continually threw stones at houses, villagers had aggressive confrontations in the market and community leaders verbally clashed in public meetings. Several pastors, imams and political and community leaders attempted to prevent conflict, particularly by reaffirming the ethnic ties of Hibua Lamo. However, many of these attempts appear to have come with the caveat ‘We will not strike first but will fight if attacked.’93 The bravado of each side convinced members of the other community that the chance of inter-communal reconciliation was small, creating a commitment problem of the type identified by James Fearon.94 The security situation was such that even some attempts at peace building appear to have been

Dispersion – Tobelo and Galela 123 misconstrued as preparation for conflict, such as the sub-district head’s late-night visits to Muslim villages. A physical security dilemma also took hold in Galela, particularly in the month before Ramadan and Christmas. In the area around Lake Duma, concerns over the intent of neighbouring villages were particularly prevalent. As the village head of Gotalamo concluded, ‘Because people prepared, it happened.’ Acting in that grey area between offence and defence, Christians from Duma and Soatobaru appear to have agreed a coordinated response should conflict break out. Muslim leaders from the villages beside Duma also appear to have appealed to Soasio for help after violence occurred in Tobelo. The movement of Christian and Muslim leaders between the villages in this area on 27 December led to each community believing the other was finalizing plans to attack their village. Christians in Duma stated that only after they had expelled all Muslims from neighbouring villages did they feel safe. This physical security dilemma in Tobelo and Galela was exacerbated by an existential or ‘societal security dilemma’ as understood by such theorists as Buzan and Roe. The volatility of the security situation was heightened by the perception that not only lives and property were at risk, but the very existence of one’s religion in the region. The onset of Ramadan and approach of Christmas increased the sensitivity of both communities. As Hersen Tinangon, the Kao Christian leader who organized the militia that went to Tobelo and the attack on Togoliua, stated: ‘We were afraid we would never hear church bells again on Halmahera.’95 According to several Christians, if Muslims had not killed pastors and destroyed their churches, the conflict ‘did not have to happen’.96 Muslims in turn increasingly believed GMIH and local Christians were seeking to ‘Christianize’ the area. The presence of Christian Ambonese among Tobelo’s political, bureaucratic and religious elite exacerbated these concerns. This fear of Christianization explains the targeting of many of these individuals and their houses in the first 24 hours of the rioting. The perceived existential threat faced by both communities made some individuals emphasize the religious character of the dispute in confrontational incidents, in turn exacerbating the environment of tension. For example, the discovery of the Muslim tailor making white robes and headbands as symbols of Islamic solidarity seemed to confirm Christian fears. Belligerent individuals also played on religious animosity through such acts as painting provocative religious graffiti. In turn, with each religious incident, the Tobelo community polarized on the basis of religious identity and began to see it as impossible to live peacefully with members of the other community. In Galela too, much of the militancy of the Christian community can be attributed not only to concerns with physical security, but also to a desire to protect Christianity in the sub-district. Several Christian leaders stated that the reason they expelled Muslims from the villages around Duma, and subsequently did not leave the village for six months despite facing constant attack, was that Duma was the site of the first Dutch Reformed Church mission on Halmahera. This desire to defend Christianity in Galela was felt even in places where Christians constituted a

124 Ethno-Religious Violence in Indonesia very small minority, such as the capital, Soasio. The head of the Protestant Church (GMIH) in Galela advised his congregation to fight if attacked, so as to avoid being forcibly converted to Islam.97 This connection between physical security, sacred space and religious animosity was also central to the rioting in Tobelo. The concern to protect the GMIH compound in Tobelo City over the Christmas period and the transportation of a group of armed men into the city to do so, was highly instrumental in the outbreak of violence. The importance of the security dilemma to the outbreak of violence can further be illustrated by a brief examination of the one area in North Maluku in which violence did not occur, that of Weda Bay in Central Halmahera District.98 Weda Bay has been the site of a large Nickel exploration activity since 1998 by PT Weda Bay Nickel, whose main shareholder at the time was a Canadian company. In November and December 1999, as tensions reached very high levels throughout the province and at the request of local villagers, the directors of the mining operation asked the military command for approximately 12 Indonesian marines from Surabaya to be sent to the area to provide security. The marines were dispersed among the four Christian and Muslim villages that surrounded the mine and supplied with two-way radios. This provided not only comfort to the villagers regarding the possibility of external agitation or direct attacks, but also allowed them to stay in constant contact with the surrounding villages and the mine. The primary benefit of this increased communications capacity was the ability to quash the rumours that constantly reached the villages. Remarkably, no violence occurred in the area throughout the conflict elsewhere in North Maluku. This is not to argue that the violence in Tobelo and Galela was entirely spontaneous or ‘caused’ by structures of insecurity and inter-religious animosity. There were individuals on both sides of the conflict who sought to intimidate the other community. Both Christians and Muslims attempted to deter and intimidate the other side by exaggerating the strength of the forces at their disposal, leading to a breakdown in information. The neutrality of the leading government official in the area was called into question, meaning he could not act as an arbiter between the two communities. A small minority in each community was willing to initiate violence, believing they and their group would prevail and it was better to initiate violence than be attacked. In the tense atmosphere of November and December, young men from each community began to gather and adopt the attitudes of militant individuals, such as Benny Bitjara and Yahir Patty. Youths on both sides also misread the intentions of other community leaders who were attempting to prevent conflict, while appearing not to be afraid of the other community. This meant that even a small incident, such as the pelting of a house with stones, could suddenly ignite mass violence. Varying motives lay behind the actions leading up to violence in Tobelo and Galela. In many cases, individuals acted out of fear, for their own and their families’ safety, and the possibility of losing their homes and land, as well as concern over the sustainability of their religion in the area. Anger also undoubtedly played a role as each side blamed the other for past violence elsewhere and for attempting to initiate hostilities locally. In other cases, individuals deliberately aggravated

Dispersion – Tobelo and Galela 125 tension or prolonged violence in order to gain status, to ensure that payments of security money from shops continued or simply because they were drunk. Violence broke out almost simultaneously across the two sub-districts because of a combination of local security concerns, anger exacerbated by religious fervour and belligerent as well as foolish actions. One other explanation for the outbreak of violence in North Halmahera must be addressed. Several informed community leaders in Tobelo are adamant that military and police personnel exploited the tension prevailing in North Halmahera and provoked the violence. One Christian community leader claimed that in the days prior to the outbreak of violence on 26 December he witnessed military personnel warning Christians not to sleep as Muslims were planning an attack.99 He claimed the military also gave the same warning to Muslims. Military personnel certainly appear to have done little to stop the large rioting mobs on either side. Yet there is little direct evidence of instigation by the military and overwhelming evidence that societal dynamics caused the violence.100 During the rioting, most personnel were frightened, particularly by the number of Pasukan Merah and the ferocity of the violence. One young local Christian man involved in the militia recalled how military personnel would stop and ask for permission to pass through Christian guard posts, frightened that militia members would throw a bomb into their truck. In particular, the abandonment by military personnel of dozens of Muslims under their protection, and their subsequent killings, revealed the lack of professionalism of the local security forces. The lack of readiness of local security personnel for the violence appears to have also been evident in the higher ranks. One Christian man who had been present in the district military compound stated that the Deputy Commander of the company (the Commander was on training outside of Tobelo) patrolled the city with his troops. According to this respondent, the Deputy Commander fainted when he returned from the city, saying ‘he had never seen anything like it’.101 In most areas, military and police personnel were too widely dispersed to act as any deterrent to conflict or source of assurance to the population. In the one area with a large contingent of security personnel, Tobelo City, where approximately 120 soldiers were based at the district military compound, the military commanders made the decision to protect and evacuate Muslim civilians rather than forcefully oppose those rioting. In Galela, the only concentration of personnel was located in the capital, Soasio, which was overwhelmingly Muslim and therefore experienced violence for only a short period. Yet these personnel made no effort to reach the scene of most of the violence around Lake Duma in the first days of conflict. According to several respondents, officials of the Galela Sub-District government lacked any political will to prevent the conflict. Accounting for the intensity of the violence Having considered why violence broke out in Tobelo and Galela, the remaining question to be addressed in this section is: what explains the intensity of this violence? The terrible atrocities reportedly committed during the violence in

126 Ethno-Religious Violence in Indonesia Tobelo, and to a lesser extent in Galela, have been discussed throughout this chapter – beheadings, disembowelling, the consumption of body parts, the killing of women and children and those surrendering. These atrocities were accompanied by a lack of remorse on the part of those who carried them out. While it is necessary to be sceptical about these claims, given the almost universal presence of similar stories in communal violence, several factors indicate that some atrocities did occur in this case. Witnesses come from both sides of the clashes,102 and photographs too graphic to display show corpses that had been disembowelled and dismembered. Most respondents in North Maluku state that atrocities were most often carried out by orang pedalaman (literally, interior people) or Forest Tobelo who live in the more remote interior and are considered to be more warlike by those in settled coastal communities.103 Many Forest Tobelo were involved in the attacks on Togoliua, Gorua and Popilo. Christopher Duncan has argued that this blaming of the Forest Tobelo for the worst atrocities in the violence is incorrect and stems from two factors.104 First, the stereotype of the Forest Tobelo as savage and dangerous has been cultivated by local governments attempting to declare a form of terra nullius in the interior in order to open it up for development, and paradoxically also by the Forest Tobelo themselves who have sought to prevent the inward spread of coastal settlements into their gardens. Second, coastal villagers involved in the violence have used this stereotype as a convenient scapegoat to divert blame from themselves for the atrocities. They know police are unlikely to venture into the interior to investigate supposed crimes by the Forest Tobelo. My evidence confirms that the atrocities that occurred in Togoliua, Gorua and Popilo were committed not only by the Forest Tobelo but also by villagers from coastal Tobelo and from the city itself. Several factors converged to motivate the Tobelos, Kaos and other Christians to attack Gorua, Popilo and Togoliua with such intensity. Christians were angry at Muslims for apparently planning to isolate and control Tobelo City and drive Christians from North Halmahera. This perceived plan appeared to link Muslims in Tobelo with the atrocities that had occurred in Tidore and Ternate, including the murder of a pastor on Tidore and the destruction of churches. The stories of these events dehumanized Muslims as inherently violent and deserving of retribution. IDPs from Payahe, Gane and other areas that had suffered large-scale attacks by Muslims were involved in the attacks on Gorua and Popilo and were undoubtedly enraged at the destruction of their homes. The ferocity with which the Christians conducted the attacks in late December was also connected to a more strategic or ‘instrumental’ goal. As discussed, it seems likely that the Christian militia, particularly the leadership, in brutally driving Muslims from these villages sought to deter Muslims from returning to Tobelo. Another factor that played into the intensity of the violence was the sanction provided by the Christian Church. The Church response to the rising tension was not uniform. Some pastors had in previous weeks attempted to prevent pre-emptive attacks and ameliorate some of the worse excesses of the violence.105 However, several pastors provided religious service before and after such attacks, asking God to protect the militia members and blessing their weapons.106 Some

Dispersion – Tobelo and Galela 127 church officials therefore afforded legitimacy to acts of extreme violence against Muslims and the expulsion of the Muslim community from north Halmahera. It seems likely that this religious sanction for violence suppressed any remorse that Christian men may have felt, increasing both their sense of purpose and their ability to carry out further atrocities without guilt. The few leaders that departed from this belligerent interpretation of the scriptures were intimidated. Men outside the official Church structure also used religion to stir up anger and prepare men for violence. Benny Bitjara in particular, a man with no religious training, employed religious imagery. Refused Church sanction to attack Muslim communities in November, by late December Bitjara had himself assumed the mantle of Christian leader/protector. Before the entry of the Pasukan Merah into Tobelo City, Bitjara inspired his troops with a selected reading from the psalms. While the common exegesis of the psalm in question suggests that it points to God’s determination to protect His followers, Bitjara interpreted the scripture to fit the extreme context in which he and his militia members found themselves. He explained to his men that it meant that it would be God’s angels fighting against Muslims the following day. This message would have had a powerful impact on Christian men as they entered battle. The almost mythical status of Benny Bitjara and his apparent invulnerability to physical harm afforded a sense of power and righteousness to Christian militia members.107 With the assumption of Christian leadership by Benny Bitjara and other militia leaders, Christianity became synthesized with traditional, ‘pre-Christian’ war practices. Christian and Muslim militias reportedly carried out traditional ceremonies to bestow invulnerability. The revival of traditional Tobelo practices engendered pride and sense of belonging among many Tobelo men who felt disenfranchised from economic or political influence. The Tobelos take pride in their history as pirates and the Sultan of Ternate’s most feared warriors. One common story heard in Tobelo concerns a warning mothers in South Sulawesi supposedly use to frighten their children into going to sleep – ‘If you don’t sleep now, the Tobelos will come and get you.’ Some of the practices carried out in the attacks, such as eating the hearts of victims, were considered by the Tobelos and Kaos to have strong connections with the war practices of their ancestors. The legitimation bestowed by this synthesis of Christianity and pride in traditional identity led to a certitude that appears to have played a role in causing atrocities. As militia leader Hersen Tinangon stated, ‘While it was hard to kill Muslims from the same Tobelo ethnic group, there was one difference, we were right and they were wrong because we had God on our side.’ Like all other phases of the conflict, the violence in late December involved mutually reinforcing instrumental, affective, ideological and cultural influences. The violence was driven by a desire among most Christians to drive Muslims from the area and to deter their return. However, other less ‘rational’ influences – anger at past atrocities, fear of a renewed Muslim assault on Tobelo, religious sanction and traditional war practices – infused the attack with an intensity that would otherwise have been absent. For many men, the opportunity to demonstrate their bravery and ferocity in battle, particularly in what they saw as the defence of Jesus

128 Ethno-Religious Violence in Indonesia Christ and Christianity in Halmahera, provided them with an identity and status that they had never before enjoyed. These factors, combined with the excitement and frenzy of recent victory and the involvement in a large crowd under the leadership of charismatic men like Benny Bitjara, resulted in the most horrific atrocities of the North Maluku conflict.

Conclusion ‘Because people prepared, it happened.’ This statement by a village head in Galela, while not completely capturing the complexity of the Tobelo and Galela riots, goes a long to explaining the clashes. Violence spread to Tobelo and Galela because conflict elsewhere had created physical security concerns for both communities. A series of events throughout 1999, such as violence in Ambon and Malifut, caused relations between Christians and Muslims in Tobelo and Galela to deteriorate. However, most Muslims and Christians considered the Ambon conflict to be too far removed from north Halmahera, and the Malifut clashes to be too local in character, to ignite violence in their own community. It was the riots in Tidore and Ternate that dramatically altered the inter-communal relationship in Tobelo City and in villages throughout Tobelo and Galela Sub-Districts. The flight of IDPs to Tobelo, including several Christian leaders who had been members of the district parliament, along with the religious character of the violence elsewhere, created an environment in which several factors made violence more likely. First, both communities in Tobelo and Galela perceived threats to their safety, land and economic livelihoods from the other religious community. In order to circumvent this threat, each community began to prepare large supplies of weapons and undertook other preparations for conflict, in turn appearing to pose a threat to the other community. Second, each community began to consider that the other held malign intent not just toward them as individuals, but also posed a threat to the sustainability of their religion in the region. Therefore some members of each community began to emphasize the strength and solidarity of their community, in turn increasing the sense of physical and existential threat. Third, militants assumed greater status within the community and were paid protection money by local businesses. While not necessarily seeking to initiate violence, these individuals therefore had an interest in continuing volatility. In some cases community leaders more renowned for providing security for markets and economic infrastructure emphasized a selective and militaristic reading of religious texts. Indeed the detail presented in this chapter has shown how this environment of insecurity alone does not cause violence. Each religious community in Tobelo was divided into several sub-groups with different agendas and opinions regarding the use of force. The agency in any conflict situation therefore ranges from peacemaking to accidental provocation to intentional aggression. Whether the outcome is large-scale violence, uneasy peace or reconciliation depends as much on the actions and interests and identities particular to that situation as much as it does on the prevailing social structure.

Dispersion – Tobelo and Galela 129 The violence discussed in this chapter, particularly in Tobelo Sub-District, stimulated nation-wide calls for jihad against Christians in Maluku and North Maluku. In Java, radical Muslim leaders created a large militia which they dispatched to Ambon and elsewhere in Maluku Province. In North Maluku, Muslim leaders in Ternate and Tidore also created a large militia made up of IDPs from Tobelo and Galela and Muslims from elsewhere in the province. The creation of this militia and its operations on Halmahera are the focus of a later chapter. The next chapter deals with clashes that broke out on Ternate simultaneously with those in Tobelo and Galela. In this second outbreak on Ternate, however, Muslims fought against Muslims. As the next chapter demonstrates, along with ethnic, religious and economic tensions, political tension was never far from the North Maluku conflict.

6

Political exploitation – the Putih–Kuning

Introduction Towering above a busy intersection in central Ternate City is an arresting, almost shocking statue. A traditional North Malukan man stands, his machete held aloft, the other arm pointing in accusation to the north. On the statue’s base, dozens of faces scream in agony. If you follow the line of the man’s finger your eyes come to rest on the Sultan of Ternate’s palace. The inscription declares that the statue is a memorial to the ‘Bloody Ramadan Tragedy’. Yet it is not the violence in Tobelo that this statue commemorates, but instead intra-Muslim clashes that engulfed Ternate in December 1999. As the Islamic holy month drew to a close, a conflict began that would have major ramifications for political power in the new province. This new outbreak of violence erupted in Ternate while Muslims and Christians were fighting for control of the sub-districts of Tobelo and Galela. This time the violence was between Muslims from different ethnic groups and was far more political in character. The traditional guards of the Sultan of Ternate, Mudaffar Syah, along with other Ternates, fought with large numbers of Tidores, Makians and other groups from the southern areas of Ternate City. Thousands of men from neighbouring Tidore Island also joined the latter group and together they defeated Mudaffar Syah’s traditional guards. The clashes became known as the Putih–Kuning (White–Yellow) conflict, after the colours worn by the opposing groups. As discussed in Chapter 1, there are two widely held explanations for these riots, which appear so incongruous in the context of inter-religious conflict elsewhere in North Maluku. The first suggests that the violence was largely spontaneous, born out of the frustration of migrant Muslims at the arrogant behaviour of Mudaffar Syah’s guards and lingering resentment at the sultan’s protection of Christians during the November riot in Ternate. The second explanation suggests that the violence was elite-led, a consequence of Mudaffar Syah’s desperate quest to assume power in the new province. A variant of this elite competition explanation suggests the conflict marked the resurgence of the historic struggle between the Sultanates of Ternate and Tidore for dominance in the region. This chapter examines the main causes of rising tension in the city and considers whether the combatants were intentionally mobilized, and if so by whom. The chapter discusses how the violence started, what motivated those involved and,

Political exploitation – the Putih–Kuning 131 importantly, why the security forces took so little action to prevent or halt the violence once it had begun. I argue that while some non-Ternates did feel animosity towards Mudaffar Syah and his traditional guards, the violence in the city appears to have been premeditated by a wide section of the political elite and bureaucracy who felt threatened by the growing political and strategic dominance of the sultan. Before giving an account of how the Putih–Kuning conflict unfolded, I explore theoretical explanations of how political competition can play a major role in communal violence, which I believe help to explain how intra-Muslim clashes could erupt in the middle of a religious war.

The political uses of violence Many analysts have argued that competition for political power often lies behind communal violence. V. P. Gagnon, Jack Snyder and others have illustrated how politicians sometimes manipulate communal tensions in order to gain or retain power as a state goes through profound political change. Gagnon asserts that as a ‘response … to shifts in the structure of domestic political and economic power’ elites provoke violent conflict along ethnic cleavages ‘in order to create a domestic political context where ethnicity is the only politically relevant identity’.1 Snyder agrees, concluding that members of the elite will use exclusionary nationalism when several conditions apply: when economic development is low; when citizens lack the skills for political participation; and when democratic institutions, such as political parties and a professional media, are weak but the bureaucracy remains strong.2 Nationalist politicians use their access to the state bureaucracy and patronage networks to make appeals to ethnic solidarity in an attempt to survive the democratic transition.3 These individuals portray other ethnic communities as hostile or malevolent, while the lack of an independent media that would normally contradict such rhetoric precludes the community from obtaining objective information on this threat. Brass demonstrates that communal riots have long been a part of the political process in India. Riots are a ‘continuation of politics by other means’ designed to capture power by scapegoating and intimidation, which in turn consolidates the votes of one ethnic or religious group.4 In some cities, what Brass terms an ‘institutionalized riot system’ has emerged involving a range of actors from politicians to criminals.5 Central to those networks are certain individuals that ‘take as one of their purposes the protection of the status, pride, and interests of one community against presumed threats to them from another’.6 At opportune times during the political process, these individuals convert minor incidents involving Hindus and Muslims into large controversies, often involving mass mobilization and rioting in order to obtain political or other benefit for themselves and their sponsors.7 Brass concludes that communal riots are made possible by a prevailing discourse of Hindu–Muslim confrontation in India, built up over two centuries and maintained by these same interested parties.8 Anti-Muslim riots create communal solidarity among Hindus, thereby facilitating electoral victory for Hindu nationalist political parties. While Brass’s studies have mostly been focused on ‘riot prone’ areas, as

132 Ethno-Religious Violence in Indonesia will be seen in this chapter, the concept of an ‘institutionalized riot system’ can also illuminate several aspects of communal violence in areas with little history of violence or inter-communal tension. As was discussed in Chapter 4, Steven Wilkinson similarly concludes that politicians running on ethnic or religious platforms use sectarian tension as a means of unifying all members of the community behind them.9 Further, Wilkinson concludes that in many cases, government officials only order the security forces to prevent rioting if it is in their political interest.10 Violent conflict provides rare opportunities for political exploitation, opening up the potential for political outcomes that may have seemed unlikely before the violence. Just as some communal and vertical conflicts eventually become driven by profit-making or ‘rent-seeking’,11 it seems likely that some politicians undertake strategies of ‘political profiteering’ designed to exploit a conflict situation to achieve political ends quite separate from the initial precipitants of the conflict. As demonstrated by the theorists discussed above, many riots, particularly in the context of political competition, can clearly be characterized as elite-led. But a focus on the action and interests of those individuals in positions of power, while necessary, should not come at the expense of a consideration of why ordinary community members participate in violence. A fuller understanding of violent mobilization, and an ability to differentiate between events at different times and locations, comes from combining this elite focus with an account of the goals and emotions of those participating. By focusing on elite provocation and organization, participants can sometimes appear to be an amorphous mass responding to the provocative claims of their leaders. As demonstrated in Chapter 4, this can attribute too much suggestibility and inanimacy to the crowds involved. Many members of these crowds also act out of interest. In some cases this interest mirrors that of their political leaders – the undermining of a rival individual or political organization. At the same time, as many analysts of conflict have pointed out, many participants use the turmoil to seek personal gain or revenge. As Brass writes: Those screaming for blood and revenge in the crowd are making use of slogans provided to them as a justification for actions that serve either their interests or those of their political organizations, or for which they are paid, partly in cash and partly by the loot they gain – under the cover provided by the crowds so massed, by the justification given for the violence, and by the near certainty that they will escape prosecution.12

Rising political tension In 1999 Ternate was at the centre of major political change. From late 1998 to mid-1999, many of the political, bureaucratic, economic and educational elite were involved in the struggle for North Maluku’s secession from Maluku. The North Maluku community was united in this campaign, with almost all sectors of society recognizing the benefits that would come from forming their own province. While

Political exploitation – the Putih–Kuning 133 these developments have been discussed in earlier chapters, they are repeated here, as it was in this phase of the conflict that political rivalry suddenly came to dominate the ongoing conflict. After the inauguration of the new province in October, the three district governments arranged for an inaugural provincial parliament to be formed the following year based upon the results of the 1999 district elections. In these elections, the ruling Golkar Party had dominated all three district parliaments. The following year, the members of this provincial parliament would elect the first governor of North Maluku. As discussed in Chapter 2, at this time there were four candidates for the governorship of the new province: Mudaffar Syah, the Sultan of Ternate; Bahar Andili, the District Head of Central Halmahera District; Syamsir Andili, the Mayor of Ternate City; and Thaib Armain, who in late 1999 became the Regional Secretary of the provincial bureaucracy. Each of the four main candidates enjoyed substantial support in the province and therefore had good prospects of winning this election, although the first two, Bahar Andili and Mudaffar Syah, were considered the main contenders. Initially united in their campaign to separate from Maluku Province, by mid-1999 differences quickly emerged between these individuals and their followers. The sultan appeared to his rivals to be attempting to consolidate his power by pressing for recognition of the historic and cultural importance of the sultanates to the region, arguing that the province should take the name Maloko Kie Raha and assume the status of Special Region (Daerah Istimewa). However, all three other candidates and their supporters opposed both of these measures. Several students associated with Thaib Armain said they opposed Special Region status for North Maluku as it would enhance the political and cultural power of the sultan.13 But as discussed in Chapter 4, two issues with more immediate implications provided the focus of most political tension in Ternate – the location of the provincial capital and the question of who would become the province’s first governor. Housing the capital would bring immense revenue to the surrounding district through the construction of facilities, funding allocations from central government and the employment and revenue associated with local government. The Sultan of Ternate lobbied for the provincial capital to be located in Ternate. He argued that the city already possessed the necessary infrastructure for the seat of government, including government and service facilities, an airport and port. Bahar Andili and his younger brother Syamsir Andili pressed for the capital to be located in the village of Soasio on Tidore. Thaib Armain was less visible in this struggle but is said to have lobbied for the capital to be located in north Halmahera and, for that reason, received some political support from leading Christian politicians in Tobelo. Attempting to reach a compromise, but still benefit their own constituencies, Mudaffar Syah and Bahar Andili proposed locations on Halmahera within their own districts – Sidangoli in North Maluku and Sofifi in Central Halmahera respectively.14 A source of even greater tension was the issue of power in the new province. The future governor would have immense influence over the appointment of senior civil servants, which would in turn determine employment throughout the

134 Ethno-Religious Violence in Indonesia government in general. Throughout most of 1999 Mudaffar Syah was the leading contender for the first governorship, enjoying support from the Ternates and most Christians in Ternate City, as well as many Tobelos, Galelas and Kaos in northern Halmahera. The sultan also had a great deal of political influence as regional leader of the leading Golkar Party, and as chairman of the North Maluku District parliament. Likely to receive the support of most members of those parties that would dominate the provincial parliament (Golkar, PDI-P as well as representatives of TNI) when it was formed, Mudaffar Syah had a strong chance of winning a majority of votes for the governorship. Of his three rivals, Bahar Andili posed probably the strongest challenge. As the district head of Central Halmahera District, he had a strong support base in Tidore and elsewhere in that district, as well as in the southern areas of Ternate where many Tidores had settled, and would also exercise influence over all Central Halmahera members of the new parliament. Thaib Armain could not be disregarded in this campaign, either. As a Makian, and holding the highest position in the bureaucracy, he derived his support from Makians in the south of Ternate and from within the bureaucracy, which was dominated by Makians and Tidores. 15 Tensions from this elite-level struggle spilled over on to the streets as the inauguration of the province drew closer. In early September, following the initial announcement that both the temporary and permanent provincial capitals would be located in Central Halmahera District (in Soasio on Tidore and Sofifi on Halmahera), several thousand people protested outside the North Maluku District government buildings. These protesters angrily claimed that Bahar and Syamsir Andili, as well as other members of a delegation from Central Halmahera which had recently travelled to Jakarta, were using ‘money politics’ to influence central government officials. The protestors also claimed that the Andilis were trying to cause conflict in Ternate in order to undermine the city’s prospects of becoming the provincial capital.16 On 12 October 1999 the Department of Home Affairs announced National Law 46/1999 granting provincial status to North Maluku.17 The central government confirmed that Sofifi in Central Halmahera would be the provincial capital, although the established centre of Ternate, and not Soasio, would act as temporary capital until construction was finished in Sofifi.18 The new province was inaugurated in Ternate and one week later, the central government sent Surasmin, a senior official from the Department of Home Affairs, to act as caretaker governor until the provincial parliament was formed and a new governor elected. In the midst of this political struggle, the district head of Central Halmahera, Bahar Andili, announced the appointment of a new Sultan of Tidore, a position that had been vacant for several decades. While not the immediate successor to the sultanate (the son of the last sultan was living in Bogor on Java) the man, Djunus Syah, was a distant relation of the last sultan. The district head inaugurated the new sultan in October, two days after the announcement of the new province. Andili’s hurry to install a new Tidore sultan, in the context of the rivalry between himself and the Sultan of Ternate and between Ternate and Central Halmahera as discussed above, looked to some observers like a political strategy.19 The new

Political exploitation – the Putih–Kuning 135 sultan, Djunus Syah, quickly declared his support for the gubernatorial candidacy of Bahar Andili, lending the district head a great deal of traditional legitimacy within Tidore and Central Halmahera. The timing of Bahar Andili’s appointment of a new sultan suggests that it was indeed a calculated political move to counter the legitimacy that Mudaffar Syah gained from his position as a traditional, as well as political leader.

After the expulsion of Christians As previously discussed, the Malifut conflict, particularly the Kaos’ attack in October, greatly exacerbated inter-ethnic and political tension in Ternate. Many Makians, and to a lesser extent Tidores, believed Mudaffar Syah was behind the Kaos’ efforts to cancel the creation of Malifut Sub-District. Many also believed Mudaffar Syah had implicitly provoked the Kaos into retaliating against Malifut, leading to the massive attack on 25 October. They argued that in his bid to become governor, the Sultan had supported the Kaos in the Malifut dispute in order to maintain political support from them and other Christians on Halmahera who were traditionally loyal to the sultanate. This view was widespread among Makians, but also shared by some Tidores and other ethnic groups in the south of Ternate City and on Tidore Island. In addition, Mudaffar Syah had opposed the immediate evacuation of the Malifut community to Ternate, to the anger of many Makians in Ternate. He had argued instead that the Makians should stay on Halmahera until a settlement could be reached between them and the Kaos. The Mayor of Ternate, Syamsir Andili, ignored the sultan’s objections, and deployed a number of trucks to evacuate the Makians from Malifut to Sidangoli, from where they could be brought to Ternate. Andili pointed to the extremely serious humanitarian situation and the danger for Makians of remaining on Halmahera. Mudaffar Syah’s apparent support for the Kaos at the expense of the Makians united Thaib Armain with Bahar Andili and Syamsir Andili, as well as their supporters, in opposition to him. Following the influx into the city of Makian IDPs from Malifut in October, tension rose higher between Ternates and supporters of Mudaffar Syah and the other ethnic communities in southern Ternate City. When sporadic violent incidents began to occur after their arrival, Mudaffar Syah, perhaps angered that his concerns over the influx of IDPs had been ignored, deployed his traditional troops, the Pasukan Kuning, into the city. Approximately 4,000 members of the Pasukan Kuning patrolled the streets to protect the city’s infrastructure as well as some Christian property.20 As shown in Chapter 4, when rioting broke out in Ternate on 6 November the Pasukan Kuning protected Christians, evacuating many to the sultan’s palace and to the areas surrounding the palace, such as Dufa Dufa. Mudaffar Syah’s protection of Christians angered many Muslims, particularly Makians but also Muslims from ethnic groups closely connected to them, such as Tidores, Sananas and Kayoas. Graffiti appeared on the walls of the city making declarations such as ‘Bishop Syah, GMIH’, linking the sultan to the main Protestant organization in North Maluku.21 The fact that several members of the

136 Ethno-Religious Violence in Indonesia Pasukan Kuning, most visibly the militia’s leader, Yopie, were Christians from Halmahera, further angered Muslims in southern Ternate.22 This situation continued on the streets of Ternate throughout November and December. The Pasukan Kuning maintained checkpoints and bases around the city. While Mudaffar Syah’s apparent support for the Kaos and his protection of Christians during the November riots had particularly angered Makians, the armed Pasukan Kuning patrols also alienated a broader section of Ternate society. As noted in an earlier chapter, members of the Pasukan Kuning appear to have adopted a highly belligerent and arrogant attitude towards migrants, questioning their residency in the city and physically assaulting them.23 On several occasions, the guards fought with young men from the main migrant communities on Ternate (Makians and Tidores), and several of the latter group were killed. 24 Several sources stated that Mudaffar Syah and the Pasukan Kuning targeted leading Makians for assault and kidnapping. Makian respondents in Ternate claimed that during November and December a list circulated with the names of 100 Makians that the Pasukan Kuning intended to kidnap.25 The North Malukan sociologist Tamrin Tomagola reiterated this claim, naming in the national media two men who were apparently kidnapped and tortured by the Pasukan Kuning with the assistance of the police and military.26 I was not shown a copy of the list, nor did I hear of any abductions, so cannot attest the veracity or otherwise of these reports. Mudaffar Syah, through the Pasukan Kuning, thus controlled the central and northern sections of the city of Ternate. The sultan’s forces monitored movement between the south and north of the city and controlled the centre of the city. Their presence restricted access to the governor’s office, the parliamentary and bureaucratic buildings for North Maluku and Ternate City district governments, the offices of local political parties, as well as most infrastructure, such as the main telecommunications facility, the market and Khairun University. Most government departments closed down over this period, and many leading politicians and bureaucrats remained in their homes. The militia established headquarters in the provincial office for the Golkar Party (of which the sultan was chairman), located in the heart of the political district. This situation of almost martial control of Ternate City augmented the political predominance of the sultan in the new province. Despite some anger at Mudaffar Syah for protecting Christians during the November riot, the results of the June 1999 district parliamentary elections still stood. Therefore it was still widely expected that he would be elected governor by a majority of parliamentarians.27 One senior member of the Tidore sultanate’s traditional guards (Pasukan Adat) stated that Mudaffar Syah was still popular even among Tidores. Many Tidores respected the Islamic and cultural traditions associated with the Ternate sultanate, while the newly appointed Tidore sultan had not yet attained similar status. According to this respondent, despite Mudaffar Syah’s prominence in the Golkar Party long associated with President Suharto’s New Order, many people admired his perceived opposition to authoritarianism and corruption, epitomized by a

Political exploitation – the Putih–Kuning 137 Pasukan Kuning attack against the offices of the former military officer appointed district head, Ret. Lt Col. Abdullah Assagaf, in 1998. 28 In mid-December 1999 several members of the elite faction opposed to Mudaffar Syah began mobilizing their followers to counter the sultan’s political and strategic dominance in Ternate. In the village of Tomalou on the south-western side of the island of Tidore (see Map 4.2), Abu Bakar Wahid mobilized approximately a thousand men.29 A long-time advocate of the necessity of defending the Muslim community after his experiences in Ambon, Wahid referred to this militia as the Pasukan Jihad, and on 25 December 1999 he was appointed commander of the militia.30 In the southern suburbs of Ternate City, such as Mangga Dua and Bastiong, Muhammad Selang, an influential Muslim leader in southern Ternate City, also began to form a militia which he named the Pasukan Putih (White Troops).31 When these two militias merged in late December in opposition to the Pasukan Kuning, they were collectively known as the Pasukan Putih.32

Intra-Muslim conflict In late December a series of small incidents occurred in Ternate that developed into rioting. This conflict began almost simultaneously with the outbreak of violence in Tobelo City. On Monday 27 December at approximately 7.30 p.m. members of the Pasukan Kuning on guard in front of the Golkar Party building beat the driver of a car, who, by some accounts, had driven into their roadblock. The man fled to his home in the nearby suburb of Kampung Pisang. Several hours later, at approximately 11 p.m., a large group of youths, including the driver, his friends and other men from the suburbs of Kampung Pisang and Tanah Tinggi, attacked a group of the Pasukan Kuning near the Ternate Municipality government offices. The two groups fought a street brawl with stones and spears until a police unit (Brimob) arrived and attempted to separate them. After the unit commander was injured by an explosion, however, the entire unit left the area and the two groups resumed their fighting unhindered. During the clashes, the youths from Kampung Pisang attacked the ‘Maria Bintang Laut’ Catholic school, burning it to the ground. Dozens of the Pasukan Kuning had been living in the school, which, as with most Catholic buildings, was less damaged than their Protestant equivalents in the November riot. Both sides appear to have retreated, the Pasukan Kuning to the Golkar Party building, which served as their main base in the central city, and the ever-increasing numbers of Pasukan Putih to an area near the governor’s office on Ahmad Yani Street. At 5 a.m. the following day, 28 December, hundreds of the Pasukan Kuning assembled with cans of petrol. Led by Yopie, the militia attacked and set fire to houses in Kampung Pisang. The blaze spread into the neighbouring suburb of Maliaro, destroying several more houses. Two people died in the fires, and residents of the two areas fled further up the hill away from the violence.33 After attacking Kampung Pisang, the Pasukan Kuning moved south towards the areas of Tanah Tinggi and Takoma (See Map 4.1). A large number of Pasukan Putih

138 Ethno-Religious Violence in Indonesia gathered in the area around the sports stadium, striking telephone poles to attract more residents of the area on to the street. Kampung Pisang, as well as being home to the youths involved in the initial clash with the Pasukan Kuning, was also inhabited by large numbers of migrants from nearby Tidore. Many of the houses destroyed were owned by these migrants, and news of their destruction was quickly relayed there, angering their ethnic kin. In Tomalou on Tidore, Abu Bakar Wahid spoke before hundreds of young men, informing them that the Sultan of Ternate was burning the houses of Tidores and had taken control of Ternate.34 Throughout the afternoon, hundreds of men travelled from Tidore to Bastiong Port in Ternate in order to join the Makians and Tidores fighting the Pasukan Kuning. With the intervention of these men from Tidore, the Pasukan Kuning was driven back north. The two groups fought running battles, with spears, bows and arrows, homemade firearms and bombs, along the two main streets through the city. The several thousand Pasukan Kuning were outnumbered by the Pasukan Putih, which had been reinforced by thousands of men from Tidore. They were unable to maintain their positions in the central city. By late afternoon, the Pasukan Putih had forced the Pasukan Kuning back to the sultan’s palace in the north-east of the city. With the Pasukan Kuning overcome, the sultan arranged for those Christians left in Ternate to be evacuated by ship from Dufa Dufa to Halmahera.35 Having forced the sultan’s militia out of the central city, the Pasukan Putih destroyed the Golkar building which had been used as a base by the Pasukan Kuning. In explaining the destruction of the Golkar office, one leader of the Pasukan Putih, Haji Kotu, told me that many members of the militia were angry that Golkar appeared to have supported the sultan in the conflict.36 They argued that as a national political party the organization should have remained neutral. As the Pasukan Kuning retreated towards the palace, the Pasukan Putih also destroyed another small building near the palace which had formerly been the headquarters of the youth wing of the Pasukan Kuning. Many in the Pasukan Putih also prepared to raze the sultan’s palace. Haji Kotu, however, told the crowd that the palace was North Maluku’s most important cultural treasure, and should not be destroyed. The same day, 28 December, Surasmin, the interim governor of North Maluku, telephoned the Sultan of Tidore to request that he try and rein in the large number of Tidores involved in the fighting.37 In response, a delegation from the sultanate travelled to Ternate that afternoon, including the sultan and 14 others, accompanied by several hundred traditional guards.38 Upon arrival on Ternate, the Sultan of Tidore ordered the protection of the new governor’s office, stating that the building was likely to become the political centre for the entire community of North Maluku.39 At 5 p.m., as the Sultan of Tidore arrived at the Sultan of Ternate’s palace, the clashes stopped. The Sultan of Tidore and his officials instructed those members of the traditional guards already involved to withdraw.40 The sultan’s group walked through the Pasukan Putih waiting near the palace and all members of both militias sat down in the road.

Political exploitation – the Putih–Kuning 139 The leaders of the two traditional guards, Husen Alting of the Tidore traditional guards and Yopie of the Pasukan Kuning, met in the street in front of the palace. All members of the Pasukan Kuning agreed to take off their yellow hats as a symbol of their surrender, and the Sultan of Tidore and the leadership of the Pasukan Putih entered the palace grounds. The Tidore leadership forced the Sultan of Ternate to sign a document taking responsibility for the Putih–Kuning conflict and recognizing an obligation to rebuild the houses destroyed in Kampung Pisang and elsewhere.41 After Mudaffar Syah had signed the peace agreement and the Pasukan Kuning had withdrawn to their homes, the Pasukan Putih returned to southern Ternate City and Tidore.42 Soon after signing the peace agreement, Mudaffar Syah left Ternate for Jakarta. The sultan claimed he was not pressured to leave, but merely left Ternate on a routine business trip and soon returned to Ternate.43 Most people on the Pasukan Putih side of the conflict, however, claimed that the peace agreement stipulated that the sultan was obliged to effectively leave his position of authority in Ternate and North Maluku. Estimates for casualties from the Putih–Kuning conflict vary from 18 to around 200.44 While the local government report calculates that 18 people were killed in the violence, most respondents in Ternate put the figure at around 40. Many more were injured. The conflict caused a great deal of damage to housing in the city, particularly around Kampung Pisang and in areas near the sultan’s palace such as Dufa Dufa. The report on the conflict compiled by the governor’s office stated that 241 houses were destroyed. During the clashes between the Pasukan Putih and the Pasukan Kuning, which stretched from one side of the city to the other, no military or police personnel were present.45 In January the North Maluku District parliament unseated Mudaffar Syah as parliamentary chairman. Some members of parliament who had supported the sultan had already fled the island, including all Christian members. Other supporters had been convinced of the sultan’s malfeasance or had been intimidated into abandoning their support for him. Mudaffar Syah was widely accused, not only in Ternate but also elsewhere in Indonesia, of provoking the violence, and using intimidation and conflict to achieve his political ambitions. Tamrin Tomagola, a North Malukan sociologist at the University of Indonesia, called for the investigation of Mudaffar Syah for his role in the violence in North Maluku as a whole.46 Indonesia’s National Commission for Human Rights (KOMNAS-HAM) initiated an investigation into accusations of human rights abuses ordered by Mudaffar Syah, although no charges were ever brought against him. Stripped of formal political power and now thought by many Muslims in North Maluku to have been responsible for much of the violence in the province, Mudaffar Syah was no longer considered a suitable candidate for the governorship. Most respondents agree that by January 2000 Bahar Andili had become the leading candidate to assume the primary position of political power in North Maluku.47

140 Ethno-Religious Violence in Indonesia

Explaining intra-Muslim violence during religious war Most existing analyses explain the Putih–Kuning conflict as a spontaneous response to mounting animosity against the Sultan of Ternate. Many Muslims, particularly Makians and Tidores, were infuriated that Mudaffar Syah had used the Pasukan Kuning to protect Christians during the riots in November. Many were also disappointed that the province’s principal Islamic and cultural leader appeared to be more interested in ensuring political support from Christians than in defending Muslims and Islam in the region. In an article published two weeks after the Putih–Kuning conflict in the national Muslim newspaper Republika, the North Maluku academic Smith Alhadar suggested that many Muslims in North Maluku had doubts about Mudaffar Syah’s commitment to Islam.48 He pointed to the sultan’s series of (three) Christian wives, as well as his reputed penchant for drinking alcohol. Most respondents in North Maluku, including Ternates, agreed that Mudaffar Syah adhered to a more moderate form of Islam than was practised by most Makians, Tidores and other migrants in southern Ternate. Alhadar asserts that this Muslim anger came to a head when Muslims heard of the violence in Tobelo discussed in the previous chapter. Because the sultan was seen to have sided with Christians, after hearing of these riots Muslims in Ternate attacked his palace guards. Alhadar writes that after ‘Christian tribes in northern Halmahera … attacked Muslims there … on 27 December people from south Ternate in turn attacked a Catholic school housing customary guards loyal to the Sultan of Ternate.’49 The Putih–Kuning conflict is here seen as a direct result of anger about these Muslim deaths in Tobelo. There is, however, reason to doubt this explanation. As the violence began in Tobelo on the evening of 26 December, there was certainly time for people in Ternate to hear about it. Nevertheless, at the time that the Putih–Kuning clashes began on the evening of 27 December the violence in Tobelo was not particularly extensive, Muslims still controlled most of Tobelo City and most casualties had been Christian. Muslims had not yet fled the city. It seems unlikely, therefore, that the first day of conflict in Tobelo would have provoked the attack on the Catholic school ‘Maria Bintang Laut’ in Ternate as Alhadar claims. It was not until 28 December, with the entry of thousands of Christian Pasukan Merah militia into Tobelo, that Muslims suffered large numbers of casualties and all mosques in the city were destroyed. By this time the Putih–Kuning clashes in Ternate were already well under way. Many Christians in Tobelo also believe the rioting in Ternate in late December was connected to the outbreak of violence in Tobelo, although for a different reason. It is generally believed among Christians in North Halmahera that Muslims in Ternate had planned to travel to North Halmahera to come to the aid of their co-religionists. According to this account, the Putih–Kuning clashes occurred when ‘jihadis’ attempted to leave Ternate but were halted by the Pasukan Kuning and the security forces. Ahmad and Oesman, in their book written from Ternate and presenting a strong Muslim view of the violence, also stated that the Putih–Kuning conflict stopped Muslims from helping their co-religionists in Tobelo and Galela.50 However, all respondents who were involved in the

Political exploitation – the Putih–Kuning 141 Putih–Kuning conflict deny that any militia attempted to leave Ternate around this period. Therefore, although occurring simultaneously, it appears that the Putih–Kuning and Tobelo clashes were not directly connected. A far stronger case can be made, however, that the Putih–Kuning conflict was caused by the actions and attitude of the Pasukan Kuning. Several young men in Ternate who fought in the Pasukan Putih claim that it was the Pasukan Kuning’s behaviour on patrol and at checkpoints throughout the central and northern parts of the city that angered them and motivated their participation in the riot. Many people, including members of the elite, were angry that the Pasukan Kuning appeared to have replaced the military and police as the main providers of security on the streets of the city.51 It seems clear that Mudaffar Syah, through the deployment of the Pasukan Kuning, was indeed assuming control of the city. It is possible he thought his provision of security may have increased his political support among the people of Ternate. It also seems possible he was aggrieved at his failure to achieve his objectives for the new province – the name, Special Region status and the location of the capital. He may also have had legitimate concerns over the destruction of the city’s infrastructure, including the government buildings he hoped to soon occupy. Animosity towards Mudaffar Syah and the Pasukan Kuning was strongest among those ethnic groups associated with the southern areas of the city such as Makian, Tidore and Sanana. The destruction of Kampung Pisang marked the peak of this tension between ‘north’ and ‘south’ Ternate. Several leaders from the sultan’s side of the conflict, including Mudaffar Syah himself, deny the destruction of Kampung Pisang was carried out by members of the Pasukan Kuning, claiming it was the work of provocateurs intending to mobilize opposition against them.52 They suggest the men involved in the attacks on Kampung Pisang wore yellow hats and other items of clothing so as to appear to be Pasukan Kuning. Nevertheless, it seems likely that the attack was indeed carried out by some members of the Pasukan Kuning, most likely in retaliation for previous clashes. One member of the sultan’s militia admitted that the Pasukan Kuning launched the attack in revenge for insults against the sultan.53 He claimed Ternates and other supporters of the sultan were traditional people and insults against him could not go unpunished. The sight of large areas of Ternate on fire and the striking of electricity poles by people coming out of their houses (a common occurrence in the early stages of rioting in North Maluku and elsewhere in eastern Indonesia) mobilized hundreds of young men from Kampung Pisang and Tanah Tinggi and other suburbs in southern Ternate. The fact that the Pasukan Kuning had destroyed the houses of Muslims in Ternate fuelled claims that the militia was predominantly Christian and that the sultan was more interested in protecting his Christian supporters than defending the Islamic community. Many of the young men involved in the riots were also undoubtedly driven by excitement. Several interviews with participants reflected the ‘rush’ involved in large-scale battles along the main streets of Ternate and the sense of power in finally opposing the Pasukan Kuning. The burning of houses in Kampung Pisang, many of them owned by migrants from Tidore,

142 Ethno-Religious Violence in Indonesia appears to have been an important motivation for many of the Tidores who travelled to Ternate.54 This flood of reinforcements for the Pasukan Putih proved a turning point leading to the political defeat of the Sultan. However, several aspects of the conflict suggest that it was not completely spontaneous, nor stimulated solely by the behaviour of the Pasukan Kuning. As well as anger against the Pasukan Kuning among young men on the streets, by December there was a great deal of concern among Mudaffar Syah’s political rivals and their supporters at his growing political hegemony. In December 1999, despite the social unrest associated with the November riot, the political elite in Ternate and Tidore still planned to form a provincial parliament and hold gubernatorial elections in 2000. It was still widely expected that a majority of these parliamentarians would support Mudaffar Syah, as chairman of Golkar and a moderate Muslim, in the upcoming gubernatorial election.55 This eventuality threatened the aspirations of Bahar and Syamsir Andili and Thaib Armain. The likely ascension of the Sultan of Ternate to the position of governor also threatened the positions of thousands of non-elite Makians and Tidores in the district bureaucracies and future provincial bureaucracy. The governor would be responsible for all major bureaucratic appointments, including the heads of all government departments. Many existing and prospective civil servants feared that if he became governor Mudaffar Syah would give bureaucratic positions to Ternates and Halmaherans, including Christians. Makians and Tidores also resented the prospect of a return to a ‘feudalistic’ system where opportunity was based on loyalty and ethnicity rather than on education, merit and capability. The events of November and December, and the day to day necessity of dealing with Pasukan Kuning patrols on the streets of the city, had given many non-Ternate Muslims an insight into what they could expect in a North Maluku dominated by Mudaffar Syah. In line with Snyder’s argument on political transition in the former Soviet Union, it appears that Mudaffar Syah’s political rivals began mobilizing their followers for violent action to protect their political power.56 Many of these men had a great deal of influence and were able to draw upon support in the bureaucracy as well as ethnic networks. The rising resentment against Mudaffar Syah among migrant men provided an opportunity to stimulate rioting against him. The leaders of these militias on Tidore and Ternate – later to merge as the Pasukan Putih – were inspired not by Mudaffar Syah’s protection of Christians nor his supposed support for the attack on Malifut, but by their opposition to his powerful political position in Ternate. The leader of the Pasukan Putih on Ternate, Muhammad Selang, claimed people rose up because they were angry that the sultan and his troops wanted to control the city.57 Another leader stated, ‘We only wanted to show them we were the majority.’58 Other Makian leaders claimed that, by December, many people had become concerned at what they called Mudaffar Syah’s political hegemony. On Tidore, Abu Bakar Wahid mobilized young men to travel to Ternate by asserting that the sultan’s militia had not only destroyed Kampung Pisang, but also already controlled 70 per cent of Ternate City.59

Political exploitation – the Putih–Kuning 143 Pointing to the involvement of numerous Tidores in the riots, one analyst has argued that the primary dynamic in the Putih–Kuning conflict was competition between the Ternate and Tidore sultanates for political and cultural control in the new province.60 To many people in North Maluku this argument appears highly feasible, given the centuries of rivalry between the two sultanates prior to and during the Dutch colonial era. While the destruction of houses owned by Tidore migrants no doubt angered many of their co-ethnics still living on Tidore, it is unlikely that the Sultan of Tidore, Djunus Syah, played a major role in instigating the Putih–Kuning conflict, however. My interview with Djunus Syah suggested he had little interest in challenging the Sultan of Ternate, particularly militarily. Members of the Tidore sultanate’s traditional troops also denied the sultan had sent them to Ternate to join the attacks on the Pasukan Kuning. They claimed that while many of their number did join the attacks they did not do so as representatives of the sultanate. According to all officials of the sultanate and members of the traditional guards, the sultan only officially sent his troops as a means of ending the conflict in Ternate. As discussed above, the Sultan of Tidore was a recent appointee, chosen by the district head of Central Halmahera and gubernatorial candidate, Bahar Andili, and had little political ambition independent of his sponsor. Because the position of Sultan of Tidore had been vacant for around five decades, many of the members of the traditional Tidore guards had become more involved in politics than in traditional or ceremonial activities. Many members of the Pasukan Adat, particularly the leadership, were more politically minded and better educated than their counterparts in the Pasukan Kuning in Ternate. Many were bureaucrats and politicians and, hence, looked to the district head, Bahar Andili, as the main power holder in Tidore and Central Halmahera. He enjoyed widespread support in Tidore not only because of his influence as district head but also because of his campaign to have the capital of the province located in Central Halmahera District. The mobilization of men on Tidore to end the Sultan of Ternate’s growing provincial dominance was not carried out under the direct influence of Djunus Syah, but by Mudaffar Syah’s political rivals. While Ternate had only for a short period been ‘riot prone’ it is clear that some of the elements of Brass’s ‘institutionalized riot system’ existed in that city in 1999. Abu Bakar Wahid returned from Ambon after the outbreak of violence there stating his determination to prevent any repeat of the same attacks on Muslims in North Maluku. As Muslims constituted a large majority in North Maluku, particularly on Tidore (95 per cent), where Abu Bakar Wahid resided, this was not a realistic concern. It appears his main goal was similar to that of Brass’s ‘riot specialists’ – to keep a minority, in this case Christians, ‘cowed’ and Muslims ‘ready and alert for mobilization, for crowd action, and for violence if necessary should it be considered desirable for political or other reasons’.61 Mudaffar Syah was portrayed as supporting his own political ambitions and his Christian supporters at the expense of Muslims. Mudaffar Syah’s rivals used these and other issues, such as his apparent failure to live in accordance with Islam, to stimulate sectarian tension as a means of unifying Muslim parliamentarians

144 Ethno-Religious Violence in Indonesia against him before scheduled gubernatorial elections. The violence and the massive mobilization against Mudaffar Syah intimidated politicians who supported his candidacy for the governorship. The destruction of the Golkar building in particular was a clear message to the political supporters of the sultan. An important feature of the Putih–Kuning clashes was the inaction of the Indonesian military and police. It must have been clear during December that the presence and behaviour of the Pasukan Kuning, who carried spears and other sharp weapons, was creating tension and anger on the streets of the city. Yet the security forces took no action against them. Then, as rioting began, neither the military nor the police acted to stop the militia attacking Kampung Pisang and moving into the areas of Tanah Tinggi and elsewhere to fight with migrant youths. Nor did they act as the Pasukan Putih, including perhaps thousands of men from Tidore, forced the Pasukan Kuning back towards the sultan’s palace, destroying several buildings important to the provincial infrastructure, including the provincial Golkar Party office. Security personnel remained in their compounds during the conflict. There were clearly enough security personnel present in Ternate City to halt the violence. Military and police contingents were stationed in the city, each with approximately 100 men. Security personnel had an opportunity to halt the rioting in the early stages of the clashes, which took place immediately adjacent to the police compound, but did not do so. What explains this inaction? One possible explanation can be discounted. With both parties to the conflict being Muslim, religious bias did not affect the behaviour of the security forces. The sultan’s alleged protection of Christians was not sufficient to unite the city’s security forces, many of whom were Ternates, against him. There are several possible reasons for this inaction. As discussed in Chapter 2, in late 1999 there was a great deal of domestic and international pressure for the government to bring security personnel to trial for apparent human rights violations, particularly during the violence that occurred after the independence referendum in East Timor in September that year. While no military personnel explicily expressed concern to me about the possibility of accusations of human rights abuse arising from their response to the conflict in Ternate (as others involved in the violence in Tobelo and Malifut did), it is possible some may have been afraid to use the lethal force necessary to halt these clashes. Another possible explanation concerns the announcement, made only six months previously on 1 April 1999, that the Indonesian Police (Polri) were to become an autonomous institution, free of a long-standing subordinate position to the military.62 It is possible that the local military in North Maluku were seeking to demonstrate the lack of capacity of the police to deal with major internal security problems, as a means of ultimately regaining this responsibility. Confusion over which institution was tasked with responding to which disturbances may also have affected both institutions’ performance. Yet it seems likely that local political factors contributed more to this lack of response by the security forces. In 1999, after years of authoritarian and military rule, the security forces were not directly responsible to local political leaders but to their own superior officers in Jakarta, who were in turn directly responsible to

Political exploitation – the Putih–Kuning 145 the president. However, whoever became governor of North Maluku, whether Bahar Andili or Mudaffar Syah, would be likely to exercise enough influence in Jakarta to successfully seek the removal of those commanders who had opposed them. In clashes widely seen as deciding the ultimate alignment of future political power in the new province, military and police commanders appear to have felt it safer to order their personnel to remain in their barracks.

Conclusion In a conflict occurring as a new province was born, political competition was never likely to be far from the surface. This chapter has demonstrated that members of the Ternate elite exploited the existing communal conflict in the province to weaken a political rival and obtain political benefit. The large numbers of Makians and Tidores in the three district bureaucracies in Ternate and Tidore presented a sizeable constituency which shared these goals. Yet the mobilization of such a large group and the riot against the Pasukan Kuning cannot be separated from more emotional factors associated with the conflict itself, such as the Sultan of Ternate’s claimed support for Christians at the expense of Muslims. The process that his political opponents had begun with the anti-Christian riots in November came to fruition in late December. Political tension was at its height in Ternate in December 1999. Throughout November and December there had been growing tension between Mudaffar Syah and his Pasukan Kuning on the one hand and the migrant-dominated south of Ternate City on the other. Many Makians remained angry at Mudaffar Syah for protecting Christians during the November riot and for supporting the Kaos in the Malifut dispute. The Pasukan Kuning had also alienated and angered other Muslim men from Ternate by their arrogant and aggressive behaviour. However, more importantly, the Sultan’s ascension to power threatened the gubernatorial aspirations of his opponents and the dominance in local government and bureaucracy of the migrant groups which had settled over the previous decades. In late December the sultan remained a powerful political force and was the likely winner of the first gubernatorial election. In addition, his traditional troops had taken control of large swathes of Ternate City, assuming responsibility for law and order in place of the local police and thereby augmenting his electoral dominance. Around this time, several ethnic and political factions united against him because of the Malifut riots and his growing dominance. The leaders of these factions exploited the existing animosity towards him and the tension caused by the wider conflict in North Maluku to first unify the Muslim vote against him, and second, to undermine his dominance by force. Contrary to some analyses, the Sultan of Tidore played only a peripheral role in the Putih–Kuning conflict. Recently appointed by Bahar Andili, the sultan merely provided traditional legitimacy to the district head’s campaign for the governorship and played no role in instigating or exacerbating the conflict. Following a clash between the Pasukan Kuning and youths, massive force was brought to bear on the sultan and his troops. Local security forces made no effort to

146 Ethno-Religious Violence in Indonesia halt the fighting, demonstrating a lack of willingness to oppose any future power holder in the province. The sultan was defeated, and fell as a political and strategic force in Ternate, leaving his rivals to contest the governorship. 63 These events were quickly subsumed in the repercussions of the violence in Tobelo. In the following days thousands of Muslim IDPs began arriving from the violence there and from Galela in north Halmahera. Many of these IDPs related stories of the terrible events in Tobelo, angering Muslims in Ternate and Tidore. The leadership of the Pasukan Putih, which had initially been mobilized to defeat Mudaffar Syah, now formed a militia more Islamic in character which they called the Pasukan Jihad. This militia began training for the more demanding task of waging jihad against Christians on Halmahera.

7

Killing in the name of God

Introduction The intensity of the violence in Tobelo and Galela dramatically altered the North Maluku conflict. Several days after the parties to the Putih–Kuning clashes in Ternate had halted hostilities and demobilized, thousands of Muslim IDPs from north Halmahera arrived in the city. The arrival of these people, and the stories they brought with them of atrocities against Muslims, united Muslims in anger at Christians. Politics and ethnicity were forgotten as religious animosity come to the fore. Muslims from both political factions in Ternate and elsewhere in North Maluku formed a militia to retaliate against Christians in north Halmahera – the Pasukan Jihad (Jihad Force). This chapter discusses the formation and goals of the Pasukan Jihad. Those Muslims driven from their homes and seeking retribution against Christians in Tobelo and Galela found a militia already mobilized and buoyant from its victory over the sultan’s forces. The Tidores, Sananas and Makians who had made up the Pasukan Putih, along with many Ternates against whom they had recently fought, were outraged by the apparently calculated attacks against Muslim communities by Christians in Tobelo and Galela. The chapter begins with a theoretical consideration of the role of religion in the mobilization of such militias. After assessing the organization and goals of the Pasukan Jihad, the chapter then evaluates claims that several militant, national Islamic organizations were involved in the North Maluku conflict and investigates how the North Maluku militia fits into the wider environment of Islamic radicalism present in Indonesia in early 2000. The chapter then discusses the two main campaigns of the militia in Malifut and Galela. The final section brings together the conclusions of the chapter and assesses the role of religion in the violence as well as the varied response of the Indonesian security forces to the campaigns in the two areas.

Identity, strategy and the mobilization of religious militias North Maluku entered the new millennium in the grip of full-scale religious war. In Ternate thousands of Muslims mobilized to launch attacks against Christians in

148 Ethno-Religious Violence in Indonesia north Halmahera. These men prepared to spend months in the field fighting against well-fortified and -armed Christian villages, receiving no pay and obliged to forage for food in the abandoned gardens of Halmahera. While many of the men in this militia were IDPs who had lost their homes and family members in attacks by Christians, many were Muslims from elsewhere in the province who had not been directly affected by the violence in north Halmahera. This was the first time since the start of the conflict six months earlier that people had mobilized en masse and travelled to assist co-religionists of different ethnic groups. Almost all cases of violence had thus far been confined to local protagonists, combatants fighting neighbouring villages or suburbs. This case illustrates a central question in the study of conflict, often termed the free-rider dilemma. Why do people mobilize for collective action even when their interests might not be served or when they might still be served even if they remained in the safety of their homes? Several scholars have asserted that it is the obligations and loyalties of collective identity that explain mobilization in such circumstances.1 Religion is perhaps a more potent force for mobilization than other forms of collective identity because it is not only strongly linked to a sense of self, but also provides a far-reaching and uplifting ideology, powerful institutional structures and an enduring and clear-cut definition of an ‘other’. Unsurprisingly, militias fighting in religious violence often appear to be driven by religious zeal. The mobilization of the militia in North Maluku, and the violence that followed, was replete with religious symbolism. The militia formed by Muslims in Ternate and Tidore was named the Pasukan Jihad (Jihad Force), and many Muslims declared an obligation and willingness to wage jihad to defend Islam and Muslims and to prevent Christian expansionism in the region. Likewise, Christians wore large crosses into battle and declared they were fighting ‘for Jesus’ and to defend Christianity in north Halmahera. Pastors sang ‘Onward, Christian Soldiers’ during fighting. In conflicts elsewhere, militias have tended to adopt evocatively religious names such as Laskar Jihad and God’s Army, religious leaders have proclaimed their support for militants and religious sites been used as meeting places, and centres of recruitment and fundraising. Combatants have denied fearing death, proudly stating their willingness to meet God and die a martyr’s death. The apparent capacity of religion to mobilize people for violence can perhaps be explained by the profound psychological and emotional effect religious belief can have on individuals. For many people, religion provides answers to questions that cannot be satisfied by secular knowledge, such as those concerning death and ‘existential’ meaning. Religion can provide ‘individuals and groups more secure anchors for self-reference’ and hence a sense of stability.2 During crises, religion can therefore take on increased salience and provide assuredness in times of uncertainty. As Jeffrey Seul writes, during crises, loyalty to other members of your religion can become a litmus test for one’s loyalty to God.3 In conflict situations this can often lead to pressure from society to turn to acts of violence, particularly when the sustainability of the religion itself appears to be under threat. In addition, violent action can be legitimized in the eyes of believers by the use, or misuse, of the texts and teachings of the monotheistic religions.4 Some passages

Killing in the name of God 149 in the Qur’an say that the use of force is legitimate (and in some cases obligatory) for all able-bodied Muslims, particularly when they or their fellow believers have been attacked by non-Muslims.5 The Qur’an states ‘kill them wherever you find them. Drive them out of places from which they drove you … Fight against them until idolatry is no more and Allah’s religion reigns supreme’ (2:190–192).6 In order to spread the Islamic faith ‘fighting (qital) is obligatory for you, much as you dislike it’.7 Most commonly, the term jihad is used to refer to war against non-Muslims, equating in this context to the term ‘holy war’. The Hebrew Bible (Tanakh or Old Testament), central to Christianity as well as Judaism, also contains both metaphors and direct accounts of violence in God’s name. The very nature of religion with its promise of everlasting life after death also means at least some people are more willing to risk death. However, it must be noted that this legitimation of violent action requires a fair degree of selectivity in the choice of verses. The Qur’an also puts restrictions on the oppression of the weak and in some passages on the use of force other than for self-defence.8 It states ‘fight for the sake of Allah against those who fight against you but do not be violent because Allah does not love aggressors’ (2:190). While Islamic militants have often used the term jihad to legitimate the use of violence, most references in the Qur’an to the word’s root form, j-h-d, relate to striving for good and doing God’s will.9 The term has been used to mean ‘striving in the path of God’, correcting one’s internal weaknesses, the abandonment of worldly attachments, and total dedication to God.10 Similarly, pacifism has been a major tenet of Christianity. A pacifist tradition is particularly prevalent in the New Testament, according to which Jesus exhorted those who have been struck to turn the other cheek (Mt 5:39) and to ‘pray for those who persecute you’ (Mt 5:44). Despite this ambivalence in the sacred texts monotheistic religions, it is clear that during times of crisis and conflict, prescriptions of the use of violence will assume far greater prominence than passages which proscribe violence. In such cases, religious institutions are well placed to mobilize people for violence. Not only do these institutions generally exercise enormous emotional influence over members, but they also often provide social meeting places, communication networks and pools of resources.11 In crisis situations, individuals untrained in religious doctrine may also assume the mantle of religious leaders and preach an intoxicating mix of religious doctrine and militancy, as we have seen in the case of Benny Bitjara in Tobelo. As discussed in Chapter 5, Appleby has maintained that a belligerent interpretation of the scriptures is particularly effective in provoking violence when the audience does not possess adequate knowledge to challenge this interpretation.12 Yet despite the convenience of religion as a mobilizing tool, do people actually join ‘religious’ militia and participate in violence because of religious zeal and threats to their religion? Analysts often question the extent to which religion is the primary inspiration for participants in militia violence. Commentators sometimes conclude that leaders and participants in ostensibly religious conflicts are in fact fighting over economic and political spoils.13 Even in situations where members of the faiths are pitted against one another, there is often no correlation between

150 Ethno-Religious Violence in Indonesia religiosity and participation in violence. In Northern Ireland, for example, those who joined the Irish Republican Army and the Ulster Defence Association were not necessarily the most observant Catholics or Protestants, and those who could be considered the most observant were often those least likely to become involved in violence.14 In addition, those leaders using religious imagery to instigate violence may do so for very different reasons, whether out of religious fervour, political interest or because they feel pressured into it by grassroots sentiment. A consideration of the role of religion in the mobilization of the Pasukan Jihad therefore goes to the heart of a central debate in the study of conflict – how important, comparatively, are identities and interest in motivating participants in violence? The study of religion and violence suggests the relationship between religious sentiment and mobilization is often complex. It is therefore important not to exaggerate or underestimate the role of religion. Even in the face of provocation, any Muslim or Christian community will rarely have a single idea of when war is legitimate or necessary. A range of groups within each religious community will have different notions regarding the use of force which are only partly influenced by religious doctrine. Religious fervour or prejudice will motivate some people to a large extent, while others will be acting with little reference to their own religious identity or that of their opponents. Many sub-groups will be acting with clear instrumental goals in mind. Those with such goals will find utility in attaching themselves to those with identity-based fervour and vice versa. Ultimately, identity and interest interact to provide the opportunity, resources and motivation necessary for sustained mobilization and violence. The remainder of this chapter provides a detailed illustration of the mobilization of a religious militia and examines the importance of religion in this process and in the subsequent violence. It demonstrates that religion helped facilitate the mobilization and sustain the violence, but was not the primary motivation for all combatants involved. On examination of the surrounding context, it becomes apparent that the mobilization of the Pasukan Jihad was based upon a range of motivational factors. The case illustrates that we need to look beyond the highly evocative term jihad and the notions that the term conjures up to understand the true motivations involved in the mobilization and actions of ‘religious’ militias.

Muslims united against Christians Organization As the conflict between the Pasukan Kuning and the Pasukan Putih was coming to an end in Ternate, thousands of Muslim IDPs fleeing from the violence in Tobelo and Galela arrived in the city. They brought news of Christian attacks against Muslims, the destruction of mosques and terrible atrocities committed against men, women and children. These stories aroused a great deal of anger among the Muslim community in Ternate, particularly as a result of the large numbers of dead and the sense that the Muslims who died were innocent. The leaders of the Pasukan Putih, along with other community leaders and imams, began mobilizing

Killing in the name of God 151 an Islamic militia to return to Halmahera. Enthused by their recent victory over the sultan’s guards, members of the Pasukan Putih provided the basis for this militia. By this time called the Pasukan Jihad, the group drew support from every suburb and village on Ternate and Tidore. Two main centres of recruitment and mobilization were established in Ternate, the Muhajrin mosque and Toboko mosque. In Tidore, the main centre was the village of Tomalou, the home of Abu Bakar Wahid. Wahid quickly assumed leadership of the militia. Before leaving to attack Christians on Halmahera, potential recruits underwent physical examination and those considered too weak or infirm were excluded or restricted to logistical tasks. Throughout January and February, the Pasukan Jihad conducted training in martial arts, fitness and the use of weapons. Several large rallies were organized to recruit more members and publicize the intentions of the Pasukan Jihad. Leaders organized a large rally in April, symbolically held in the sports grounds in front of the exiled Sultan of Ternate’s palace. At the rally, Abu Bakar Wahid declared that reconciliation with Christian communities should be postponed until all Halmaheran IDPs could return to their homes. A wide cross-section of the North Maluku Muslim community participated in the Pasukan Jihad. Male IDPs from those ethnic groups (Makians, Tobelos and Galelas) driven out of Halmahera by Christians comprised the majority of the militia. These IDPs were joined by numerous men from the other main North Maluku ethnic groups: Ternates, Tidores, Makians, Sananas, Kayoas, and Bacans. Among the Ternates who joined the militia were many members of the Pasukan Kuning who had recently fought against their new allies. The militia assumed a far more religious character than the Pasukan Putih. The Pasukan Jihad conducted prayer sessions and heard sermons inside mosques, and during operations on Halmahera militia members were expected to pray regularly. All mujahid wore white as a symbol of their religion.15 Yet religious guidance was provided to militia members in Ternate and Tidore not by imams, but primarily by teachers and other community leaders considered to have religious knowledge. The main spiritual advisors for the Pasukan Jihad were two teachers at Islamic schools, Abdul Gane Kasuba and Albi Shamat. Some imams did assist in mobilizing the militia and several joined the Pasukan Jihad on Halmahera. Support for the jihad may not have been universal among Islamic religious leaders, however. One militia member said that certain imams faced intimidation and coercion to support the militia. Some imams and other religious leaders instructed militia members not to engage in atrocities such as the killing of women, children or people attempting to surrender. One Makian imam recounted how he forbade any Muslims to dismember or otherwise abuse corpses.16 Motivations/objectives The motivations and goals of the Pasukan Jihad in North Maluku were born out of exclusively local issues. Its formation was driven by the events in North Maluku, in particular in North Halmahera. The IDPs who had recently arrived from Tobelo and Galela had two goals in joining the Pasukan Jihad. First, they were enraged at

152 Ethno-Religious Violence in Indonesia what they saw as the premeditated nature of Christian attacks against them. In addition, they sought to defeat the Christians in their home areas so that it would be safe for them to return to their villages. For many from Tobelo and Galela this meant defeating Christians in the major villages of those sub-districts and in Tobelo City. Several respondents stated that the militia planned to drive all Christians from the area if that was necessary in order to facilitate their return to their homes. Makian IDPs in Ternate, Tidore and elsewhere had the same goal. Expelled from Malifut in October, they saw in the tide of Muslim anger an opportunity to return to their homes. Many of the leaders of the Pasukan Jihad, such as Wahda Zainal Imam and Abu Bakar Wahid, along with some politicians who supported the jihad such as Thaib Armain, were also motivated by a desire to avenge the destruction of Malifut. Makians also sought revenge for the attacks against Muslims and the strong anti-Makian sentiment that had been displayed in Tobelo. The return of Muslims to their homes was also an important goal for the Pasukan Jihad leadership. Abu Bakar Wahid claimed that the militia was motivated by a desire to protect the rights of Muslims, including the right to live wherever they chose in Indonesia.17 It is possible that leading politicians and bureaucrats supported the militia’s campaign partly because they realized that the long-term presence of Muslim IDPs in Ternate was untenable. Most, therefore, hoped to see Christians on Halmahera defeated so that the tens of thousands of Muslims could return to their villages. The displacement of additional Christians caused by attacks on Halmahera would not be a problem for the local government because it was assumed that they would mostly go to North Sulawesi, like those who had fled earlier violence. Like those from Halmahera, Muslims from the main ethnic groups in Ternate and Tidore (Ternate, Tidore, Sanana) also joined the Pasukan Jihad out of anger about what appeared to have been brutal, premeditated attacks by Christians in Tobelo and Galela. The events in Tobelo (especially Togoliua and Popilo) constituted perhaps the most terrible atrocities of the conflicts in both Maluku and North Maluku. Muslim IDPs from Tobelo brought stories of the killing of men, women and children in mosques and even those under the protection of the security forces. Stories of beheadings, forced conversions, kidnappings and rapes added to the anger. This anger against Christians on Halmahera was framed in the terms of jihad and other Islamic references. At a large rally organized by the Pasukan Jihad in Ternate in April, Abu Bakar Wahid declared to the thousands of armed men assembled there that they would fight until the day of reckoning (Hari Kiamat).18 Leaders pronounced that all able-bodied Muslim men must avenge the attacks on Muslims and protect those who remained in Halmahera. The principles of jihad and a sense of Islamic solidarity clearly motivated those Muslims (such as Ternates and Tidores) who had not been directly affected by the violence in north Halmahera. However, IDPs from Tobelo, Galela and Malifut, while no doubt inspired and energized by proclamations of Islamic solidarity, had concrete goals – retribution against Christians and the chance to return to their homes.

Killing in the name of God 153 Some Muslims, including at least one community leader who had led the Pasukan Putih against the Sultan of Ternate, refused to take part in the jihad. Several claimed they did not want jihad waged in their name, believing that the preconditions for holy war were not present. Those who did not participate and travel to Halmahera were required to pay a form of taxation to support the militia. Yet it appears little coercion was applied to men who chose not to participate and did not attend the meetings at which the militia was organized. Several male transvestites also participated in the jihad, which perhaps demonstrates the non-puritan nature of the militia. Those individuals who did join the Pasukan Jihad stood to gain a great deal of prestige within society.19 Abu Bakar Wahid claimed the primary goal of the Pasukan Jihad was to oppose and defeat a resurgence of the 1950s RMS separatist organization which he declared was operating on Halmahera.20 He proclaimed to militia members that if the Christian militia opposed the Pasukan Jihad, it was thereby opposing the Indonesian republic.21 Other Muslims in Ternate and elsewhere, including individuals not involved in Pasukan Jihad, also claimed that the RMS was present in Tobelo. Many believed that without external provocation by the RMS, the Christian and Muslim members of the Tobelo ‘family’ could not possibly have begun to kill each other.22 Assertions by the militia that they were fighting against a separatist organization were reflected in the carrying of red-and-white Indonesian national flags by the militia members.23 As discussed in Chapter 2, the North Maluku region had remained isolated from the RMS rebellion in the 1950s.24 The conflict between the forces of the Indonesian Republic and the RMS was confined to the south of Maluku Province, not affecting the northern area which was to become North Maluku. Furthermore, there was no evidence that RMS was active in North Maluku in 1999 and 2000 and no signs of RMS presence such as flags, symbols, propaganda or other documentation. The presence of Ambonese Christians in Tobelo appears to have been the only supposed evidence of the presence of this (largely defunct) organization in North Maluku. The Christian leaders most commonly accused by Muslims of ‘being RMS’, such as Jacob Soselissa, Frans Manerry and May Luhulima, deny there was any connection between Christians in North Maluku and RMS or any other separatist organization.25 The declaration by Abu Bakar Wahid and others that the Pasukan Jihad was fighting against the RMS was perhaps based more on deliberate calculation than a mistaken belief. Such statements created the impression that the militia was fighting Christians who were carrying out ethnic cleansing of Muslims in order to create a new Christian state, thereby undermining the territorial integrity of Indonesia. Through such claims, the leaders of the Pasukan Jihad calculated that the Indonesian military would not hinder their operations on Halmahera. While military commanders in North Maluku may not have believed claims about the RMS, in some cases these pronouncements by the Pasukan Jihad leadership at least gave them a pretext for avoiding a confrontation and for allowing the militia to travel to, and operate in, Halmahera. The security forces seemed reluctant to confront the Pasukan Jihad for a variety of reasons – sympathy with the militia’s goals, concern

154 Ethno-Religious Violence in Indonesia about its size and fear about the consequences of clashes, and worries about apparent connections the militia leaders had with civilian provincial power holders. Funding and supply Despite the large size of the Pasukan Jihad, the militia required little funding. The main expense was for petrol to fuel the speedboats which transported over 10,000 men from Ternate and Tidore to Halmahera, particularly those that needed to travel to Galela via Morotai (see Map 2.1). Some funding was also required for the purchase or hire of automatic weapons, obtained either from the security forces or from dealers in the southern Philippines. Christians travelled by boat from Tobelo to the southern Philippines to buy weapons in the latter stages of the conflict and claimed that members of the Pasukan Jihad did likewise.26 Once in the field of operations on Halmahera, mujahid required little funding. Most slept in the open and were fed by local communities or collected food from abandoned gardens. Most respondents said the funding raised within the Ternate and Tidore communities, largely from donations collected in local mosques, was sufficient for operations in Halmahera. According to one leader of the Pasukan Jihad who remained in Ternate to organize logistics for the militia, the local provincial and district governments also provided food and accommodation, meaning mujahid were supported as they waited for departure.27 Rank and file members of the Pasukan Jihad said they received no money for participating,28 although Christian respondents claimed they often found a large amount of money on the bodies of fallen mujahid. There is little doubt that the leadership of the Pasukan Jihad also received funding from wealthy sympathizers in other parts of Indonesia. Abu Bakar Wahid claimed he had numerous friends in Jakarta who provided him with funding.29 Several individuals in Ternate, including at least one former leader of the Pasukan Jihad, claimed that any funding donated by external actors for Pasukan Jihad operations was retained by a small number of people in the militia’s leadership. According to several respondents, the leadership of the Pasukan Jihad divided acrimoniously when two leaders, Muhammad Selang and Muhammad Albar, accused Abu Bakar Wahid of stealing donations. 30

Questions of radicalism and terrorism North Maluku Christians and external observers have claimed that the Pasukan Jihad was not entirely local in constitution and affiliation and had strong connections to external militant Islamic organizations. Christians in places such as Galela and Tobelo who fought the militia stated that many mujahid came from Ambon, Java, Sumatra and elsewhere in Indonesia.31 Christians defending villages such as Duma said they found numerous identity cards (KTP) on the bodies of deceased mujahid. According to Christian militia members, these KTP identified the men as originating from islands and provinces outside North Maluku.32 Theo Sosebeko, a

Killing in the name of God 155 leader of the Christian defence of Duma, Soatobaro and other Christian villages in central Galela, estimated that approximately 40 per cent of the members of the Pasukan Jihad were from North Maluku, the remaining 60 per cent from elsewhere in Indonesia.33 Christians from Duma also said that they had heard several languages that they could not understand, although this could be applicable to the Makian, Tidore and Ternate languages.34 They also claimed they overheard and could recognize the Ambonese language. Claims that people from outside North Maluku participated in the conflict and that the Pasukan Jihad received funding from external sources have led many observers to suggest that several extremist, national Muslim organizations were involved in the North Maluku conflict. The group most frequently said to have been involved in the conflict is Laskar Jihad, the large Java-based militia led by Ja’far Umar Thalib. For example Professor Jan Nanere, a North Maluku Christian, stated that the Laskar Jihad travelled first to Ternate and then to Halmahera in early 2000.35 The North Maluku sociologist Tamrin Tomagola also claims a fourth surge of violence began with the arrival of the Laskar Jihad.36 While not claiming the Laskar Jihad was present in North Maluku, Hasan identifies Abu Bakar Wahid (the leader of the Pasukan Putih and the Pasukan Jihad) as a field commander of Ja’far Umar Thalib.37 Laskar Jihad was formed in early 2000 by Ja’far Umar Thalib, an Indonesian of Yemeni descent and prominent preacher within the puritan salafi movement.38 Ja’far spent a number of years studying in Saudi Arabia and Yemen and engaged in jihad against the Soviet Union in Afghanistan. In 1998 he established the Forum Komunikasi Ahlussunnah Wal Jamaah (FKAWJ – The Communication Forum of the Followers of the Sunna and the Community of the Prophet).39 In early 2000, in response to major Muslim losses in the Maluku and North Maluku conflicts, in particular the Christian attack on Muslims in Tobelo in late December 1999, Ja’far established a paramilitary wing of the FKAWJ – the Laskar Jihad.40 The militia was the most prominent Islamic paramilitary organization to become involved in local communal conflict in Indonesia. The Laskar Jihad claimed that the conflicts in Maluku and North Maluku were the initial stages of an international Zionist–Christian conspiracy to Christianize Indonesia.41 In response to this threat, Ja’far Umar Thalib sought and received several fatwas from Saudi and Yemeni muftis to wage jihad against Christians in Maluku. Between May and June 2000, over 3,000 militia members left Java for Maluku, despite President Wahid’s orders that the security forces prevent their movement. Throughout 2000 and 2001, the Laskar Jihad swung the balance of power and initiative in the conflict in Ambon from Christians to Muslims, inflicting heavy losses on Christians in Maluku Province.42 The Laskar Jihad appeared to operate in Ambon and Maluku Province with the acquiescence of some elements within the military.43 Leaders and militia members were able to travel throughout Maluku and elsewhere in Indonesia without hindrance.44 In some cases, military units appeared to actually facilitate and/or assist the militia’s attacks on Christian communities. Some reports said military units guarding Christian communities withdrew moments before the Laskar Jihad

156 Ethno-Religious Violence in Indonesia and local Muslim forces attacked, for example in the suburbs of Poka (including Pattimura University) and Waai.45 Some observers explained this military assistance afforded to the Laskar Jihad as driven by a desire on the part of some military personnel to undermine the Wahid presidency. The militia’s leaders had openly opposed Wahid because of his refusal to implement sharia, his failure to protect Muslims in Maluku and elsewhere and his proposal to re-establish trade relations with Israel.46 Connections certainly did exist between the leadership of the Pasukan Jihad and the Laskar Jihad. In an interview with a reporter from the Maluku newspaper Maluku Hari Ini, Abu Bakar Wahid stated that he and Ja’far Umar Thalib were friends and shared the same ideology and goals.47 Abu Bakar Wahid was present at a large rally held by Laskar Jihad in Jakarta on 6 April 2000 and, along with Ja’far, went to the presidential palace to demand that President Wahid take firm action to protect Muslims in Maluku and North Maluku.48 The president expelled the delegation from the palace. However, all Muslims from North Maluku denied that external combatants were involved in the Pasukan Jihad. Every Muslim respondent interviewed, including the leaders of the Pasukan Jihad in Tidore and Ternate, and the militia’s leader in Galela, Samsul Bakhri, stated that no assistance was provided to the militia in North Halmahera by Laskar Jihad.49 Other community leaders in Galela and Malifut, and all mujahid interviewed, also denied that outsiders were involved in the militia.50 They claimed the only exception was a small medical team from Java which treated injured mujahid, but did not take part in attacks on Christians. None of the identity cards allegedly found on the corpses of mujahid that purportedly demonstrate a non-local origin remained in North Maluku in 2003. Several mujahid stated that Christians may have mistaken members of the Sanana ethnic group for Ambonese (Sanana Island is located close to Ambon but is part of North Maluku Province). The Pasukan Jihad leadership appears to have declined an offer of assistance from the Laskar Jihad. In early 2000, a vessel sailed into Ternate harbour carrying a Laskar Jihad delegation. The militia offered to assist the local Muslim militia to conduct military operations against Christians in North Maluku. However, the Pasukan Jihad leadership declined the offer, although the various individuals proffered different reasons for this decision.51 Abu Bakar Wahid said he declined the offer because he felt he had enough troops (18,000 by his reckoning) with which to defeat Christians on Halmahera and defend the Muslim community.52 Other leaders, such as Muhammad Albar and Muhammad Selang, said they were more concerned with preventing the involvement of outsiders in the conflict. Albar said he had been reluctant to involve the Laskar Jihad because they did not understand the problems that had led to the violence in the province and did not understand the strength of local tradition in places such as Tobelo.53 He believed the external militia were not aware of how close relations had been between Muslims and Christians in such areas before the conflict.54 Similarly, Muhammad Selang stated that he refused Laskar Jihad’s offer because he felt it would be difficult to end the conflict if people from outside the province were present. He believed

Killing in the name of God 157 Ja’far Umar Thalib was too radical and wanted to implement sharia.55 As the majority of North Maluku Muslims did not want Islamic law, he was concerned violence would arise between local and external Muslim militias.56 All of these reasons influenced the decision of the Pasukan Jihad to decline the offer of assistance. Many Muslims in Maluku Province, where the militia did play an important role, also became concerned that the militia was overly radical. Nevertheless, it was the overwhelming majority that Muslims enjoyed in North Maluku Province – approximately 80 per cent of the provincial population – that allowed Muslims there to reject the offer of assistance. In Maluku Province, Muslims did not enjoy numerical superiority, in some areas, such as Ambon City, numbering fewer than Christians. Having departed Ternate, the Laskar Jihad unit proceeded to Ambon. There are indications that other radical national and international Islamic organizations provided logistical or financial assistance to the Pasukan Jihad. Press reports from early 2000 claimed that the leaders of the militia referred to their organization as the North Maluku branch of the Islamic Defenders Front or Front Pembela Islam (FPI). FPI first emerged in Jakarta after President Suharto’s resignation as part of the Pam Swakarsa, a conglomeration of groups used by the military and police to intimidate crowds protesting outside the parliament.57 The FPI leadership also sought the addition of sharia to the Indonesian constitution, but because of the organization’s links with the Indonesian security forces claimed this must be done within the existing structure of the Indonesian state.58 The Jakarta-based FPI does not appear to have become involved in communal violence in Maluku or elsewhere. The organization focused more on the eradication of vice in Jakarta and other cities on Java, raiding and vandalizing bars, brothels and gambling centres. Those individuals in North Maluku mentioned in press reports as members of FPI include Abu Bakar Wahid, Wahda Zainal Imam, Albi Shamat and Abdul Gane Kasuba.59 Most of them now deny having been members of FPI, and one leader, Muhammad Selang, argued that journalists ‘invent things’.60 However, several respondents claimed that, in 2000, those leaders had identified themselves with FPI.61 One former mujahid in Mangga Dua, Ternate, alleged that Wahda Zainal Imam had referred to himself as the leader of FPI in North Maluku.62 Abu Bakar Wahid claimed that one of the leaders of FPI, Habib Husein al Habsyi, provided him with funding for jihad operations in North Maluku.63 Despite the adoption of the name FPI by some North Maluku Muslim leaders, the Jakarta-based organization does not appear to have been directly involved in that conflict. While some FPI funding was probably received by the leadership in North Maluku it is unlikely it was substantial. The low cost of the Pasukan Jihad meant that any funding received had little effect, neither lengthening nor worsening the conflict. Ternate moved to a more conservative observance of Islamic practices during the conflict, with alcohol consumption and prostitution taking place only in secret to avoid raids by young Muslim men. In February 2000 two nightclubs were destroyed in Ternate City. However, this development is more likely to have been a consequence of the religious violence in the province rather

158 Ethno-Religious Violence in Indonesia than of the influence of FPI or any other external organization. It seems likely that, rather than having any important links with the FPI in Jakarta, the leaders of the Pasukan Jihad adopted the (rather generic) Islamic Defenders Front name in order to obtain funding and for reasons of status. There are also some indications that the regional terrorist organization Jemaah Islamiyah (Islamic Community – JI), and through it the Al Qaeda organization, were involved in the conflict in North Maluku. JI’s main goal is to overthrow Indonesia’s current secular, democratic political system and establish a pan-Islamic state or regional caliphate based on sharia.64 An Islamic preacher formerly involved in the Darul Islam organization, Abdullah Sungkar, established JI in 1993 in Malaysia after fleeing arrest in Indonesia. The JI leadership in Malaysia maintained strong connections with small ‘cells’ of group members in Indonesia throughout their exile. Sungkar and another leader, Abu Bakar Ba’asyir, travelled to Saudi Arabia seeking funding for Jemaah Islamiyah and sent several hundred recruits to Afghanistan to assist in the mujahidin’s struggle against the Soviet Union and gain military experience. Military experience in Afghanistan provided training in bomb making and the use of weapons, and exposure to more radical international Islamic organizations (such as the al-Gama’a al-Islamiyah, an Egyptian organization thought to have links to Osama Bin Laden’s Al Qaeda). The experience infused the organization with jihadist thought.65 Some members began planning attacks on Western and other targets in South-east Asia.66 According to the International Crisis Group, after Ba’asyir assumed the leadership of Jemaah Islamiyah, the organization fractured, with one faction seeking to take a more active role in attacking Western and Christian targets.67 During 2000 JI carried out several bombing attacks against Christian churches in Medan, Sumatra, Java and elsewhere.68 As with most radical Islamic organizations in Indonesia in 2000, Jemaah Islamiyah’s leadership also saw a need to defend Muslims in Maluku and in Poso District in Central Sulawesi. It sent members and new recruits to fight in Ambon as members of an affiliated militia, Laskar Mujahidin.69 These conflict zones, along with others elsewhere in South-east Asia, became centres of recruitment and training for members of JI during 1999 and 2000. The connection between JI and the North Maluku conflict in essence boils down to one man. Fikiruddin, alias Abu Jibril, considered to have been a leading figure in the organization in 2000, and a link between JI and Al Qaeda, arrived in North Maluku shortly after the violence in December. Like the rest of the JI leadership, Abu Jibril fled to Malaysia in 1985 as the Suharto regime cracked down on Islamic radicals. He was arrested under the Internal Security Act in Malaysia in 2001 as an alleged member of JI.70 Abu Jibril was recorded on video delivering a sermon in front of a mosque in Galela in early January 2000 (not long after the initial violence in Tobelo and Galela). In the sermon, holding the Qur’an in one hand and a pistol in the other, he exhorts Muslims in North Halmahera ‘to use both the way of the book and the way of steel’ to defend Islam in the area.71 Abu Bakar Wahid confirmed that Abu Jibril

Killing in the name of God 159 was present in Ternate in early 2000 and gave spiritual guidance to members of the local militia.72 Apart from this belligerent rhetoric and ‘spiritual guidance’, the organization appears to have had little influence on the conflict. As I have argued, there was no great influx of fighters from outside North Maluku. Nor does the region appear to have been a training ground for members of JI or other radical organizations, as was the case in Ambon.73 There was also no apparent change in the level of funding for the Islamic militia. Most militia members continued to use homemade weapons and subsisted on donations from local villagers and by obtaining food from abandoned gardens. There were also no substantial changes in the strategies of the Pasukan Jihad in North Maluku. As discussed below, the Pasukan Jihad did not seek to implement sharia, a primary goal for JI and the other organizations discussed above.74 Most members of the Pasukan Jihad, including at least one leader, Samsul Bakhri, were unaware that Abu Jibril had been in North Maluku.75 There are some important similarities between the stated goals of the Pasukan Jihad and those of the militant Islamic organizations discussed earlier, particularly the Laskar Jihad. The ideology of the Pasukan Jihad was far closer to that of the Laskar Jihad than that of other organizations such as JI. Both Abu Bakar Wahid and Ja’far Umar Thalib stated that they were opposed to the RMS and were seeking to protect Indonesia’s national integrity. Both have also stated that their primary goal was to defend the Islamic community in Maluku and North Maluku which they believed was not being sufficiently protected by the national government. They also sought to win back the territory they had lost to Christians in the same way that the Laskar Jihad was attempting to win back the territory that Muslims had lost to Christians in Maluku Province. There are, however, also a number of important differences between the Pasukan Jihad and the Laskar Jihad and the other national Islamic organizations discussed in this chapter. In contrast to Laskar Jihad, FPI and Jemaah Islamiyah, neither the leadership nor members of the Pasukan Jihad advocated the creation of an Islamic state or implementation of sharia.76 Militia leaders claim that banners and graffiti proclaiming ‘This is an Islamic Province’ (Ini Provinsi Islam) were displayed by a small minority, predominantly among the youth, and were not calling for sharia. Muhammad Selang stated that he and the other leaders of the militia desired continued adherence to secular Indonesian national law in the new province. He said the Pasukan Jihad also did not advocate a wider jihad against Christians and Western interests, as did Jemaah Islamiyah.

Assaults on Christian villages The first attack by a Muslim militia in North Maluku was carried out against Christian villages on the small island of Lata Lata south of Makian Island.77 Witness accounts state that the assailants arrived in speedboats travelling from the direction of the island of Kayoa to the north. Yet it is difficult to verify the provenance of the mujahid in this case. Witness accounts state there were numerous people involved from outside Lata Lata, including from Tidore. Some accounts also claim there

160 Ethno-Religious Violence in Indonesia were a large number of Javanese in the attack.78 It is probable that this force was composed of mujahid largely from the surrounding islands, such as Kayoa and Bacan and possibly southern Halmahera. Approximately 30 people were killed in the attack, including the local Protestant pastor. From Ternate, the Pasukan Jihad also attacked Christian villages in Jailolo in early February, but was repelled by local villagers who killed several mujahid. After returning to Ternate, the Pasukan Jihad protested outside the mayor’s office, demanding that the government investigate the killing of Muslims in Jailolo. Following this demonstration, the Pasukan Jihad assembled at the port in order to depart for Halmahera. According to Nanere, military personnel prevented them from leaving Ternate and four mujahid were killed in the resulting clash.79 After these early forays, the Pasukan Jihad divided into two main groups. One group travelled to Sidangoli to attack Kao, seeking to retaliate for the destruction of Malifut and reclaim the area so that Makian IDPs could return to their homes. This group comprised primarily Makians from Malifut and elsewhere in North Maluku and was led by Abu Bakar Wahid and Haji Ishak Puasa, both Makians. The second group travelled to Morotai and Galela in the northern part of north Halmahera. This group comprised primarily IDPs from that area, including Galela and Tobelo, and was led in the field by local Galela community leader Samsul Bakhri but was organized by Muhammad Selang and Muhammad Albar in Ternate. Malifut After the Kaos’ destruction of Malifut in late October 1999, the security forces in the area were greatly reinforced. Approximately 50 army personnel were posted near Tanjung Barnabas, the port used by the mining company NHM to service the nearby Gosowong mine. A further 50 external and local army personnel were posted either at the port or at the mine itself. A police paramilitary (Brimob) unit was dispatched to the largely deserted area of Malifut. In the two months after the destruction of Malifut, Christian militias patrolled the roads between Kao and Sidangoli on the coast facing Ternate. The Kao militia, the Pasukan Merah, continued to search the public transport vans that carried people from Galela and Tobelo to Sidangoli, on at least one occasion beating a young Makian passenger. Armed militias from Muslim villages in Jailolo Sub-District also monitored the road from Malifut to Sidangoli, although no Christians travelled on this road over this period. Despite the increased security presence, the Kaos returned to Malifut in the week after the anti-Christian rioting in Ternate, and burned a small number of deserted Makian houses that remained standing. On one occasion, security personnel appear to have been intimidated by the presence of the large Kao militia in the area. In late November, army and Brimob personnel withdrew from Kao and Malifut after hearing rumours that the Kaos planned to attack them. During November and December, the Kaos demanded that former employees of NHM be allowed to return to work at the mine. On several occasions former

Killing in the name of God 161 employees, armed with machetes and other weapons, travelled to the site’s entrance and demanded a return to work, but were turned away by Brimob personnel. NHM managers informed the Kaos that no employees would be rehired until the security situation had been resolved.80 By the start of December the security situation around Kao and Malifut had improved to the extent that 30 of the army personnel stationed at Tanjung Barnabas returned to Sidangoli. Then in December and early January came the terrible events in Tobelo and the gathering storm of religious war. In January tension began to increase in Kao as stories of the mobilization of thousands of Muslims and declarations of jihad in Ternate and Tidore were heard in Kao. When stories reached Kao that Muslims were preparing to depart in order to launch an attack against them, the Kaos went on to a war footing. On 8 January the Pasukan Jihad gathered in a mosque in the village of Tomalou on Tidore, and prayed before their departure for Sidangoli. Many members of the militia were Makian IDPs, most carrying homemade weapons. According to the militia’s leader, Abu Bakar Wahid, military personnel stationed at the port of Rum on Tidore attempted to prevent the Pasukan Jihad from leaving but, upon hearing that the militia was seeking to defend the republic against separatist RMS forces on Halmahera, allowed the militia to depart unhindered. The militia travelled to Halmahera in a flotilla of approximately 40 boats, each carrying some 75 men.81 On hearing rumours that the Muslim militia had left Ternate, about a thousand Kaos, led by Benny Bitjara, gathered at the village of Dum Dum west of Malifut. By 22 January several thousand mujahid gathered in the west of Malifut Sub-District.82 That day the two sides faced each other in the Tabobo area (See Map 3.1), the mujahid outnumbering the Kaos by several thousand. The military and police units present attempted to separate them but several skirmishes took place and a few Kaos were killed. The Kao militia retreated further east, pursued by the Pasukan Jihad. The two groups again massed within sight of each other at the Tanjung Barnabas port, separated by approximately 100 army personnel, while another 100 personnel guarded the mine site at Gosowong nearby.83 The mining company hastily brought in 40 military reinforcements from Wasile Sub-District to Tanjung Barnabas port. In the late afternoon the approximately 140 military personnel present forced the Kao militia to withdraw, firing repeatedly into the ground around them. The Pasukan Jihad also withdrew west to Akelamo. Several thousand Muslims at this point abandoned the attempt to reach Kao, deterred by the strong response taken by security forces in the area, and returned home to Ternate, but approximately 2,000 militia members, mostly from Tidore but also some IDPs from Malifut, remained in Akelamo. The Kaos made several efforts to end the conflict over this period. In late January the Team of Nine Kao representatives presented a letter to the security forces stating their desire for peace, but warning that if the Pasukan Jihad entered the area they would be attacked. The Team of Nine also reassured the predominantly Javanese, and Muslim, transmigrants still living in the sub-district that they would not be harmed. The sole Kao member of the North Maluku District

162 Ethno-Religious Violence in Indonesia Parliament travelled to Tobelo to telephone the government in Ternate and request that it order the Pasukan Jihad to withdraw from the area.84 As mujahid reinforcements arrived in the Malifut area in February, bringing the total number of militia members to approximately 5,000, the security forces continued to separate the two militias and prevent any major clashes. However, the military did allow the Pasukan Jihad to destroy at least two unoccupied Kao villages in western Malifut. Because the military commander had informed the Kao militia that the villages would be protected and that they, the Kaos, should withdraw, these events served to confirm to the Kaos the need to provide for their own security. As well as taking action to prevent fighting in Malifut, the government and the security forces from this point on took a harder line on preventing further militia from leaving Ternate for Sidangoli. Their plans to attack Kao seemingly thwarted, the Makian members of the Pasukan Jihad leadership spent this period pressuring the provincial government to provide for the return of the Makian IDPs to Malifut so they could start rebuilding their homes. Abu Bakar Wahid negotiated directly with the army commander in North Maluku, Lt Col. Sutrisno, requesting that he allow the Pasukan Jihad to return to Malifut. The commander refused this request, perhaps concerned that, in reality, the militia sought to attack Kao. When several thousand mujahid did attempt to leave Ternate once more on 5 May, army troops from Infantry Battalion 732 fired on the militia, killing five mujahid. The response to this incident demonstrated that support for the jihad was not universal in Ternate. According to an Australian staff member from NHM, a petition was presented to the commander of the unit with the names of 500 Muslim Ternate citizens supporting the action taken against the Pasukan Jihad. Those groups that did manage to leave Ternate were stopped by security forces around Malifut or were deterred by the large Kao militia. On 18 May approximately 2,000 mujahid managed to leave Ternate and travelled by truck from Sidangoli to Akelamo. There, however, this group was stopped by army personnel and returned to Sidangoli,85 although perhaps 1,000 Makian IDPs who had crossed over with them remained in West Malifut waiting for permission to return to their homes. In June a small group of Kaos killed approximately ten mujahid as they were bathing in a river in the forest, in Tanjung Barnabas. The Pasukan Jihad once more retreated westward from Malifut.86 This was the last confrontation between the two forces in the Kao/Malifut area, the security forces having separated the two militias during each attempted attack by the Pasukan Jihad. Most of the several thousand mujahid who had been in Malifut returned to Ternate. It seems likely that, by early June, the Makian leaders of the Pasukan Jihad had resigned themselves to not reaching Kao, and the Makian IDPs had grown tired of the dangers in the area. After several failed attempts to militarily defeat the Kaos, Abu Bakar Wahid attempted to bring hostilities to an end and initiate reconciliation between the Makians and Kaos, meeting Kao leaders such as Jesaja Singa and Hersen Tinangon in Malifut in the presence of military personnel. Part of the motivation to start reconciliation talks was provided by a statement by the district head that the

Killing in the name of God 163 government would begin returning Makian IDPs to their homes in villages west of Malifut. Following the attempts at reconciliation by Abu Bakar Wahid, the Pasukan Jihad began to divide into a faction surrounding Wahid and one surrounding Muhammad Selang. This schism appears to have occurred for two reasons. The latter stated that he disagreed with any reconciliation with Christians on Halmahera until all Muslims could safely return to their homes. Second, animosity appears to have developed between the two groups over the disappearance of funding. Selang accused Abu Bakar Wahid and other leaders of not dispersing funding they had received from external sources and from the local North Maluku community.87 The Pasukan Jihad’s attempts to attack and destroy Kao therefore appear to have failed largely because of the efforts of army personnel in the area. The military present, including external personnel from Brawijaya 512, a company from the Surabaya-based military command (Kodam), appear to have provided a strong deterrent to large-scale attacks, although the two militias did clash on several occasions. As will be seen, this professional response did not occur in Galela. The Pasukan Jihad also appears to have been deterred by the large number of Pasukan Merah in Kao. The Kao militia had been reinforced by hundreds of Tobelos.88 Throughout these attempts by the Pasukan Jihad to attack Kao, Muslim Kaos fought alongside Christians against the Muslim militia. Galela As discussed in Chapter 5, after the conflict in Galela in December 1999, perhaps several thousand Christians had remained in the sub-district, mostly in the adjacent villages of Duma, Dokulamo, Soatobaru and Makete around the western shore of Lake Duma (See Map 5.2). These communities possessed strong village militia which prevented Muslims returning to these and neighbouring villages. Another Christian community remained in Mamuya, restricting any return to villages in the south of the sub-district, such as Luari. Several thousand Muslims remained in the sub-district capital, Soasio, and nearby villages such as Igobula, similarly restricting the movement of Christians through the area and thereby between Galela and Tobelo. From January to March, the area between Igobula and Duma in Galela Sub-District was the site of frequent clashes between Muslims and the Christians. The Christian Duma militia and the Al Istiklama Muslim militia from Igobula launched frequent tit for tat attacks on the strongholds of the other community.89 While Christians in Galela were isolated from the much larger Christian population in Tobelo, some Christian militia members attempted to go to Duma to assist their co-religionists. Small groups of Christian militia members, mostly from the village of Wari, travelled from the village of Mamuya in the south of Galela through the mountainous area to Duma.90 This group also travelled to Loloda and Ibu Sub-Districts to help Christians resist Muslim attacks. On one trip through the forest around a dozen Christians were ambushed by the Al Istiklama militia from Igobula and suffered several casualties.91 In January, Christians from Tobelo made

164 Ethno-Religious Violence in Indonesia several more large-scale attempts to reach the Christians in Galela. However, military reinforcements (Brawijaya 512) had installed a major roadblock at Pintu Batu immediately south of Soasio and forcefully repelled the Christian advance. During one attempt to pass the military guard post, military personnel shot several members of the Pasukan Merah.92 While the Pasukan Jihad was attempting to attack Kao and return the Malifut community to their homes, a much larger section of the militia was preparing in Ternate to travel to Morotai and the sub-districts of Galela and Tobelo. The majority of this militia consisted of people displaced by the violence in those areas in late 1999. For example approximately 700 men from the large Tobelo village of Gorua returned to Galela.93 In Soasio, local Muslim leaders coordinated the arrival of thousands of Muslim men from Ternate and Tidore, providing accommodation and food. A local man, Samsul Bakhri, became the leader of the Pasukan Jihad in Galela. The two remaining pockets of Christians, in Mamuya and around Lake Duma, were the initial target of the Pasukan Jihad arriving in Soasio. The large numbers of Muslim men from Tobelo Sub-District who returned to Galela with the Pasukan Jihad also hoped to attack Tobelo City.94 The first major operation by the Pasukan Jihad in Galela was a simultaneous attack against the villages of Makete and Soatobaru just after dawn on Sunday 5 March. This attack was repelled by these villages, reinforced by Christians from Duma, although one person was killed and approximately 15 injured.95 At 4 p.m. on 25 May the Pasukan Jihad attacked the more isolated Christian village of Mamuya near the Tobelo border. According to villagers, approximately 3,000 members of the Pasukan Jihad were involved, including many from Morotai.96 The militia attacked Mamuya both from the road from Soasio and from the sea. Christians in Mamuya were able to repel the attack for approximately three hours before fleeing along the road south to Tobelo or through the mountains inland from the village. Twenty-two Christians were killed in the attack, including four women.97 Estimates of the number of mujahid killed in the attack vary. According to Mamuya Christians, a large number died in the attack, although a young Tidore man tasked with collecting the corpses of mujahid during the Pasukan Jihad operations stated that only eight militia members died and three more were missing after the clash.98 After the attack on Mamuya, the Pasukan Jihad focused on the village of Duma. Christians in Duma state that their village was attacked 19 times from the first conflict in late December to the end of June 2000. Over this period, the communities of Duma, Makete and Soatobaru faced a great deal of deprivation. Isolated from Tobelo and afraid to walk to their plantations outside the village, the villagers had become very short of food. To defend itself against attack, Duma established well-organized defences surrounding the village, including hundreds of small pits filled with sharpened stakes. After the arrival of the Pasukan Jihad in Galela, the village suffered three main attacks. On 19 May approximately 2,000 members of the Pasukan Jihad attacked Duma. The Christians repelled the attack but around 40 per cent of the houses on one edge of the village were destroyed. On 29 May the Pasukan Jihad launched another major attack against Duma.

Killing in the name of God 165 According to both mujahid and Christian witnesses, a torrential downpour began as the attack started, causing the bombs prepared by the Pasukan Jihad to fail to explode. The attack was a dramatic failure. Several dozen mujahid died, many after falling on the stakes surrounding the village, others shot with firearms or arrows.99 The failure of the attack may also partly be explained by animosity between local and external members of the Pasukan Jihad. Several local Muslims stated that for this attack they had demanded that only local Galela Muslims participate, as they claimed men from Tidore, Ternate and elsewhere became frightened during the violence and undermined morale among the local members of the militia. At 4 p.m. on 18 June the commander of the military personnel guarding Duma informed a local Christian leader, Samuel Kukus, that the village would soon face another major attack and that the Pasukan Jihad had gathered in Makete.100 According to Kukus, the commander reassured him that his unit would prevent the Pasukan Jihad from reaching the village. Although the Christian militia was divided between Duma and the neighbouring village of Dokulamo, the Christians felt the number of military personnel guarding the two villages, along with their own militia, was sufficient to repel the Pasukan Jihad. After preparing the Christian militia in Duma, Samuel Kukus inspected the border of the village and discovered the corpses of several mujahid who had already fallen on the stakes placed around the village. After noting that the military were manning all guard posts around the village, the Christian militia members rested in anticipation of an attack the following day. However, when the Pasukan Jihad launched a massive attack on 19 June, the military offered no resistance.101 Christian respondents in Duma state that when the Pasukan Jihad began their attack at approximately 10 a.m., the military abandoned their posts on the edge of the village and withdrew behind the Pasukan Jihad.102 Christians and Muslims offered quite different explanations for the failure to prevent the attack. These are considered later in the chapter. Most respondents agree that approximately 10,000 Pasukan Jihad attacked Duma. Christians fought to prevent them from entering the central part of the village. Some men remained in their houses, attempting to defend them with homemade bazookas and other weapons, but after a short period of time the entire community that remained in the village, over a thousand people, was driven back towards the large village church. Women and children sheltered inside the church while men attempted to defend the surrounding grounds with homemade firearms, spears and bows and arrows. Only one man, Samuel Kukus, possessed an automatic firearm. According to Christians in Duma, three soldiers from a military post in Bale rowed across the lake to attempt to defend the Christian community, firing on the attacking mujahid. After approximately six hours, and with the village almost completely overrun, the Pasukan Jihad ceased the attack. It seems likely that they halted the attack because of the arrival of a unit of Indonesian marines from Soasio. The arrival of these additional troops appears to have motivated the Brawijaya 512 personnel stationed at Duma to begin to resist the Pasukan Jihad’s attack. The leader of the

166 Ethno-Religious Violence in Indonesia militia gave orders to halt the attack and for the militia to withdraw so that the Christians could be evacuated. The marines then used trucks to evacuate all Christians from Duma and Dokulamo to Tobelo on 20 June. The Pasukan Jihad ceased its attack in part because of the arrival of military reinforcements which were intent on halting further Christian casualties and in part because its defeat of Duma was already complete. Approximately 250 Christians died in the attacks on Duma, most during the largest and final attack on 19 June. A graveyard now occupies most of the grounds of the church, which lies in ruins. The entire village of Duma, as with the villages surrounding it, was destroyed. The Pasukan Jihad ravaged the gardens owned by the residents of Duma and the surrounding villages, chopping down coconut and other trees.103 Respondents on each side disagree on the number of mujahid who died during the conflict in Galela. According to Muhammad Selang, who led the Pasukan Jihad, approximately 200 mujahid were killed in operations on Halmahera. Christians who were in Duma estimated the number to be far higher. One Christian militia leader estimated that perhaps 3,000 mujahid died in attacks against the villages around Duma. The likely total, not calculated by local government or any other agencies, is probably closer to the smaller figure. One imam from Tidore involved with the Pasukan Jihad died in an attack on Duma.104 Respondents from both sides of the conflict said militia members killed women and children, including some who had already surrendered, often in front of military personnel. Unsurprisingly, members of both communities denied that they, or others from their group, carried out such actions. Members of the Pasukan Jihad stated that imams and religious teachers had forbidden the killing of any non-combatant. However, one mujahid acknowledged that an indeterminate number of women and children were killed during fighting or accidentally by bombs. It seems likely, however, given the large number of casualties, including women and children, that such practices were carried out by both sides. 105 Respondents from both parties to the 2000 conflict in Galela said members of the opposing militia used narcotics during the violence, both claiming to have found pills on the corpses of their enemies. These pills had allegedly been stamped with names such as ‘Mad Dog’. Muslims also recounted that when they cut the skin of Christian militia members they would bleed very slowly because of the drug. Christians claimed that Muslims acted exceedingly bravely for approximately an hour, at which point they would flee, seemingly because the effects of the narcotic had worn off.106 The veracity of these claims is difficult to establish, but given the prevalence of amphetamine and other narcotics in pill form throughout South-east Asia, their use during the violence in North Maluku is possible.

The violence ends After the fall of Duma on 19 June, all Christians had been expelled from Galela Sub-District. Yet Christians remained in Tobelo City and Kao Sub-District, continuing to prevent the return of Muslims to their homes in those sub-districts, as

Killing in the name of God 167 well as any return of the Makian community to Malifut. Controlling the main city on Halmahera, Tobelo, Christians also held the most important infrastructure in the region in the form of a port, the hospital and warehouses. Not long after the fall of Duma, the Pasukan Jihad disbanded and most members returned to Ternate, Tidore and elsewhere. For what reason(s) did the Pasukan Jihad leadership decide not to attack Tobelo City? Abu Bakar Wahid asserted that, had he so desired, the Pasukan Jihad could have controlled Tobelo in one hour, but that the caretaker governor, Surasmin, contacted him and requested that he halt jihad operations.107 However, there appear to have been several other factors that informed the decision of the leaders to disband the militia and abandon their goal of expelling Christians from Tobelo City. The Pasukan Jihad attack on Duma, the large loss of life involved and the destruction of the village received a great deal of media coverage in Indonesia. This attack, renewed violence in Maluku Province and the protracted nature of both conflicts prompted President Wahid to take stronger security measures in the region. On 26 June the president announced a state of civil emergency in the two provinces through Presidential Decision Number 88/2000. Several provincial and district government officials flew to Tobelo and met with Abu Bakar Wahid and Benny Bitjara and explained to them the implementation of the civil emergency. Civil emergency did not greatly increase the troop levels in North Maluku, where the equivalent of approximately three battalions had already been on Halmahera during the latter stages of the violence. However, the civil emergency gave security personnel increased legal authority to take more forceful action, dispelling any concerns over charges arising from over-zealous suppression of rioting. A short amnesty for militia members to hand in weapons was announced, and over the first week of July the Tobelo, Galela and Kao communities handed in thousands of homemade firearms, swords, bombs and other items.108 Three naval vessels began patrolling the northern area of North Maluku, preventing the movement of militia and the entry of automatic weapons from the southern Philippines. Increased military assertiveness may have motivated the Pasukan Jihad leadership to abandon their planned attack on Tobelo City. Two days after the government announced the civil emergency a large passenger ship left Tobelo City carrying Christian IDPs. The Cahaya Bahari was licensed to carry 500 passengers but on 28 June left Tobelo loaded with perhaps 750 people fleeing the violence in north Halmahera, mostly women, children and wounded. The vessel sank in heavy seas before reaching Sulawesi, with the loss of 492 lives. According to a Christian missionary who interviewed survivors of this tragedy in 2000, some Christian respondents stated that they believed the ship had been deliberately sunk because they had seen a smaller vessel approaching them just before the ship began to suffer difficulties.109 However, by all accounts the sea was particularly heavy and it is unlikely any small vessel would have journeyed into the open seas to sabotage the Cahaya Bahari. As the wooden vessel was severely overloaded and encountered heavy seas, it seems almost certain that the sinking was a tragic accident. This incident undoubtedly had an emotional impact on all members of the North Maluku community, including members of the

168 Ethno-Religious Violence in Indonesia Pasukan Jihad. It seems likely that the incident removed a great deal of enthusiasm for attacking Tobelo. In addition a major element of the Pasukan Jihad decision not to attack Tobelo City was the perception that it would be an enormous task. According to Muhammad Selang, after the fall of Duma almost all members of the Pasukan Jihad were exhausted. Many had experienced approximately five months in the field and been involved in several major attacks on Duma, Mamuya and other villages. Muhammad Selang stated that after Duma, members of the Pasukan Jihad were so exhausted that he was unwilling to ask them to continue the military campaign. The village of Duma, a community of several hundred, had resisted attacks by several thousand Pasukan Jihad for three months. The city of Tobelo, with a population greatly inflated by Christian IDPs to approximately 35,000, constituted a far greater challenge. In addition, the Pasukan Jihad’s operations to destroy Kao and attack Tobelo from the south had failed, meaning that the city could be attacked only from the north and by sea. Perhaps more importantly, Tobelo City enjoyed far greater military protection than villages in Galela. Unlike Duma, which was protected only by several small Brawijaya military units from outside North Maluku, including personnel who had little interest in protecting Christians, the northern road into Tobelo City was guarded by a company of approximately 100 soldiers, many of whom were of local origin. These local soldiers had a direct interest in preventing the destruction of Tobelo and the killing of its residents. Their base also abutted the ocean shoreline, allowing them to monitor the approach of vessels to Tobelo City from the north. Several mujahid told me that they had come to believe that security personnel near Tobelo were supporting Christians in the conflict. While resting after the attack on Mamuya on 25 May, the Pasukan Jihad had come under automatic weapon fire which they believed was from Christian soldiers. In addition, given their experience during December, many Muslims from Tobelo believed the local military company had done little to halt Christian attacks against Muslims. The size of the military contingent guarding the north side of Tobelo, and this belief that it would be willing to protect the Christian community, persuaded many Pasukan Jihad members not to attack Tobelo City. As the International Crisis Group points out, the violence ceased in North Maluku while it continued in Maluku, because of a ‘cantonment’ of Christian and Muslim communities, caused by displacement, rather than any process of reconciliation.110 These cantons were separated by large military contingents to the north of Tobelo City and in Malifut which had demonstrated the will to prevent attacks. In early July Muhammad Selang and Abu Bakar Wahid began ordering motor boats from Ternate to transport militia members from Galela back to Ternate.111 All mujahid from Ternate, Tidore and elsewhere in the province returned to their homes. Most IDPs from north Halmahera returned to displacement camps in Ternate. Even those from Galela, an area no longer under threat from Christian militia, returned to Ternate, choosing not to return to villages that had been destroyed during the violence.

Killing in the name of God 169

Religion and national security forces in religious war The role of religion Despite the anger caused by the violence in Tobelo and Galela and the accompanying influx of IDPs, large-scale mobilization of other Muslims would probably not have occurred in Ternate and Tidore had it not been for the work of several influential individuals in Ternate and Tidore. There was a great deal of sympathy among Muslims for the plight of their fellow Muslims from north Halmahera. Nevertheless, to launch a four-month campaign on Halmahera required coordination, funding and an ideology to stimulate and sustain participation. The first two of these have already been addressed in this chapter. The third, the role of religion, will be addressed here not only with regard to the mobilization of the Muslim militia, but also the motivations driving Christians in defending their homes on Halmahera. Undoubtedly, those Muslims from Ternate, Tidore and elsewhere not directly affected by the carnage on Halmahera were greatly motivated by anger at what had happened to their co-religionists. The atrocities committed, particularly during the attacks on Gorua and Togoliua, stimulated a great deal of solidarity among Muslims in Ternate, Tidore and elsewhere, temporarily eradicating the political rivalry that had caused the Putih–Kuning clashes. In addition, in early 2000, Christians still posed a direct threat to the several thousand Muslims remaining in and around Soasio. These circumstances appeared to legitimate calls for a holy war against Christians on north Halmahera under the principle of jihad, a fact that facilitated large-scale mobilization in Ternate, Tidore and elsewhere.112 The fact that Christian or Muslim men were fighting and dying for their religious community and for God/Allah meant they were afforded a great amount of respect by their religious community. Religious devotion and participation increased substantially during the conflict. Christians attended church in Tobelo and other areas of north Halmahera far more regularly during the violence than they did prior to the conflict.113 Involvement in the Pasukan Jihad necessitated more regular prayer than most young men were accustomed to. The importance of sacred space and buildings increased for local communities and this lengthened the violence in some cases. The decision to remain in Duma and protect the local church was taken because of the village’s status as the site of the first Protestant mission on Halmahera. Religious identity also influenced the form of mobilization of both communities, with religious ceremonies, symbolism and clothing playing a major role in each. The strongly Islamic character of the Pasukan Jihad, and in particular the framing of the campaign in the terms of jihad, helped unite Muslims from different ethnicities who had until that time been pitted against one another. This effect, of increasing religious solidarity and undermining internal division, may have been the primary function of jihad in this phase of the conflict. Religious ideology and solidarity facilitated the mobilization of a far wider group of Muslims than would otherwise have been the case.

170 Ethno-Religious Violence in Indonesia However, several factors suggest that it is important not to overstate the role of religion in the mobilization of Muslim and Christian militias and in the violence involved in this phase of the conflict. The goals of both sides to the conflict were far more strategic and physical than religious. The objective of those involved in the Pasukan Jihad was to eliminate the threat posed by Christians in their home areas, and facilitate the return of Muslims to their homes. Their actions were not driven just by hatred of Christians or a desire to eliminate Christianity from North Maluku. For Christians, too, religion cannot be considered a primary motivating factor. Christians resisted the attacks by the Pasukan Jihad because of a desire to protect their territory, homes and families. One way of assessing the influence religion had on the participants in this last phase of the conflict is to consider whether the responses of Muslims in Ternate and elsewhere would have been different if their adversaries had been Muslims of a different ethnic group. Likewise, would Christians in north Halmahera have reacted differently had they been under threat from other Christians? A partial answer to this question is provided by the violence described in the previous chapter. The Putih–Kuning conflict was fought by two almost exclusively Muslim political factions. As in those clashes, it seems likely that in the last phase of the conflict those Muslims from Halmahera would have sought the military defeat of any group that had driven them from their homes and was preventing their return. Likewise, Christians would have defended their homes against attackers of any religion. On both sides of the conflict, it was lay people rather than religious officials who most strongly espoused and utilized religious symbolism to mobilize community militia. In the Muslim community, it was leaders such as Abu Bakar Wahid, Abdul Gane Kasuba, Albi Shamat and Wahda Zainal Imam who legitimated the formation of a militia and the campaign against Christian villages on Halmahera in terms of jihad, and declared themselves the leaders of the Islamic Defenders Front (FPI) in North Maluku. Indeed, one mujahid suggested that there was some coercion of imams to support the jihad.114 Protestant pastors were more inclined to use religious symbolism in mobilizing people for violence than their Muslim counterparts, perhaps because of the more threatened position of the Christian community compared to Muslims.115 As the Pasukan Merah faced the Pasukan Jihad in Kao, Pastor Soselissa stood on the roof of a motor vehicle singing ‘Onward, Christian Soldiers’, the metaphorical nature of the hymn undoubtedly lost on militia members. However, as in the Muslim community, it was primarily militant community leaders such as Benny Bitjara who used Christian symbolism as a means of mobilizing community members. Benny Bitjara had no religious training and his religious devotion was more communal than spiritual. His high status in the community meant that no one questioned his assumption of a religious role or his use of religious doctrine. Similarly, it was secular community leaders in Duma who decided that the community should remain in the village and oppose attacks from Muslims and not withdraw to the safety of Tobelo.116 The Christian community in Galela considered Duma, where the Dutch Reformed Church was first established locally, to be the

Killing in the name of God 171 historical and spiritual centre of Protestantism on Halmahera and therefore decided to defend their village and church. However, the pastor in Duma came from outside Galela and it was local community leaders such as Theo Sosebeko and Josafat Etha who made the decision to stay. Even within the Protestant Church, leading pastors took different positions on the conflict depending on their personal perceptions and interpretations of the situation. Several prominent pastors opposed the violence, including the chairman and deputy chairman of GMIH, Pastors Aesh and Duan. As discussed in Chapter 5, Pastor Aesh refused church legitimacy to Benny Bitjara to attack Muslim communities on Halmahera in November 1999. These pastors faced intimidation and threats of murder from more militant Christians, including other GMIH pastors. The two pastors, having left Halmahera in December 1999 to seek assistance from the government in Jakarta, were unable to return to Halmahera until after the conflict. Other pastors took a very active role in legitimating and leading militia activity. Therefore even in this last, most clearly religious phase of the conflict, religion influenced different actors to varying degrees and always acted in conjunction with other motives, such as a desire for revenge, to return to one’s home, a sense of empowerment from being part of such a large group or a quest to obtain status within society. The greatest impetus for Muslim attacks against Galela and Kao was retaliation against Christians for previous violence. For Christians the primary motivation was to defend their homes and families, although religion also provided an additional inspiration. Similarities can be drawn between the Pasukan Jihad and the anti-communist violence in Indonesia in 1965 and 1966, during which some Muslims called for jihad. As Robert Cribb concluded, despite such calls ‘even for radical Muslims, however, the purpose of the killings was relatively limited’.117 The security forces A further question arising from this phase of the conflict concerns the role of the security forces. Why did the military and police stationed in Ternate and Tidore not prevent the Pasukan Jihad from departing for Halmahera? While Abu Bakar Wahid stated that military units stationed on Tidore chose not to prevent the militia’s departure from the port at Rum because they agreed with his claimed goal of opposing RMS separatists, it seems likely that the military merely used this claim as a convenient excuse. The large numbers of Pasukan Jihad members, perhaps 10,000 on Tidore and Ternate, no doubt deterred the military from taking strong action that would have led to violence. Many military commanders and personnel also undoubtedly sympathized with the goal of avenging the (apparently unprovoked) attacks on Muslims in Tobelo and elsewhere. In addition, the political faction of North Maluku politics that had gained the upper hand in the Putih–Kuning conflict now supported the Pasukan Jihad. It seems likely that military commanders would not have risked firing upon the Pasukan Jihad for fear of political repercussions.

172 Ethno-Religious Violence in Indonesia As has been shown in this chapter, the security personnel stationed in the Malifut area were far more effective at preventing violence between the Pasukan Jihad and local Christians than those in Galela. The same army battalions, Brawijaya 511 and 512, provided the majority of personnel in both areas. All units were responsible to the same provincial military commander in Ternate, Lt Col. Sutrisno. From January to June, similar troop numbers were also present in each area. Between 120 and 200 soldiers and smaller numbers of police were located in both areas throughout this period, although at most times slightly more were present in Malifut. Why was the performance of the security personnel in Malifut and Galela so different? The area between Sidangoli and Kao was certainly easier to secure than Galela. There is just one road from Sidangoli north-east to Malifut and Kao, which made it easier for small numbers of security personnel to halt and turn back advances by the Pasukan Jihad. The coast of Malifut is also located on the east side of Halmahera, preventing any easy sea access from Ternate. In contrast, the main Christian villages in Galela are located in a ring around Lake Duma and can be approached by at least three routes (along the southern and northern edges of the lake and from the coastal area in northern Galela – see Map 5.2). Nevertheless, as has been shown throughout this chapter, personnel were in positions to act forcefully against attacks on Christian villages in Galela, but they chose not to do so. It appears that the difference between the responses of the security personnel in the two areas was a matter of motivation rather than tactical considerations. One major factor in the differing responses in the two areas was the presence of the gold mine located at Gosowong in Malifut. There are several ways in which the presence of the gold mine influenced the performance of the security personnel. The mine provided a great deal of logistical support for the security personnel stationed in the area, including trucks for patrolling the road between Sidangoli and Kao and helicopters for the transportation of troops and surveillance of militias in the area. It is also likely that NHM provided funding directly to the security personnel to prevent any further conflict in the area. The presence of the mine also appears to have created greater political will in Ternate to prevent major disruptions to mining operations. Up to 140 personnel guarded the mine infrastructure, including the port at Tanjung Barnabas, the airport near Kao and most notably the front gate of the mine itself. The Australian mining staff were in constant contact with the district head, governor and military commander in Ternate requesting that adequate security measures were taken in the area. The priority given by the government and military hierarchy to securing Malifut was demonstrated when 140 personnel were transported from Galela to Kao when tension developed between Christian and Muslim Kaos in early 2000.118 While the above factors explain why the security forces responded efficiently to the threat of violence in Malifut, they do not explain why they allowed attacks against civilian populations in other locations, most notably Galela. Military and police personnel charged with defending communities in Galela failed on numerous occasions. In early 2000, the security forces played a role in preventing fighting between Duma and Igobula. However, as the Pasukan Jihad amassed in

Killing in the name of God 173 large numbers in the sub-district, the performance of the security personnel declined. Despite knowing that the Pasukan Jihad was gathering in its thousands first in Soasio and then in Makete, the security forces in the area did nothing to request they withdraw, in contrast to their demands in Dum Dum and other areas near Malifut. The main sub-district army and police contingents based in Soasio neither hindered the build up of the militia in the capital nor reinforced units around Lake Duma once the militia had moved there. When attacks did occur, army personnel quickly capitulated or withdrew from their posts. As discussed above, the military personnel based in the guard posts surrounding Duma did little to prevent attacks by the Pasukan Jihad against Duma, particularly during the major attack on 19 June. Some Christians claim that military personnel assisted attacks against their communities and otherwise treated Christians poorly during the conflict. Respondents from all Christian villages in Galela claimed TNI soldiers fired upon them as they attempted to resist the mujahid. Christian witnesses state that the military often assisted the Pasukan Jihad and a large number of Christians said they were shot by military personnel, in some cases being able to name the officer who shot them. According to one man, different military posts took different approaches to the violence. He claimed that while the military post in Bale acted neutrally, the post in Ngidiho supported only Muslims, and fired upon Christians while allowing the Pasukan Jihad to pass through the village unhindered.119 In April, even before the Pasukan Jihad attacks in Galela, Benny Bitjara and other Christian leaders in Tobelo and Kao declared that the Pasukan Merah would oppose the entry of any non-local military units into the two sub-districts because of the lack of neutrality on the part of military personnel.120 Pasukan Jihad leaders and personnel deny soldiers assisted in their attacks, however. According to Muhammad Selang, while some militia members wore military clothing they were not military personnel.121 What seems clear is that military and police units often did little to stop attacks by the Pasukan Jihad against Christian villages in Galela. Christian and Muslim respondents suggested diverging explanations for the failure of security personnel in Galela to prevent violence or assisting attacks. Some members of the Pasukan Jihad stated that the military personnel had little option but to allow them to attack Duma. One man claimed that if soldiers had fired upon the mujahid they would have been attacked by the militia, which was itself armed and far larger than most military units facing it.122 Another claimed that on the morning of 19 June most military personnel guarding Duma were resting after being awake on guard all night. During attacks on other Christian villages such as Mamuya and Makete, it is possible that the army guard posts with a contingent of perhaps 12 men were insufficient to resist the attack of, in some cases, several thousand militia members. Christians believed the military personnel intentionally allowed the Pasukan Jihad to attack for different reasons. The most common belief was that because they were Muslims most military personnel sympathized with the Pasukan Jihad and wanted to see the Christians defeated. Christians also claim that the Pasukan Jihad sometimes paid military units not to resist their attacks. Several Christians stated that military personnel would allow the Pasukan Jihad to expel Christians

174 Ethno-Religious Violence in Indonesia from a village and then extract any valuable items themselves before allowing the militia to destroy the houses. While some of these claims are difficult to verify, economic incentives certainly appear to have influenced undisciplined or provocative behaviour by military personnel throughout the conflict. Military personnel appear to have exploited the violence in north Halmahera for profit at the expense of both Muslim and Christian civilians, as the following case suggests. Despite the violence in Malifut in August and October and in Tobelo in December, several communities of Javanese transmigrants had remained in their homes in the sub-district. The majority of these communities were in west and north Kao Sub-District, while one other community was located near the village of Togoliua in Tobelo. While some of these transmigrants were Christian, the majority were Muslim. Despite religious violence in the sub-district, these communities remained in their homes throughout 1999 and early 2000, because of assurances by leaders of the Christian militia (the Pasukan Merah) that the conflict was not religious in character and they would not be harmed. One of these transmigration settlements is located at Waringinlamo, approximately 20 km north-west of Kao village. Most of the village’s inhabitants were moved from East Java in 1983, and had established adequate livelihoods as farmers and developed good relations with both the Kaos and Makians in the area. After the Kaos’ destruction of Malifut in late October, and rising religious tension following anti-Christian rioting in Tidore and Ternate, concerns among Javanese Muslims in Waringinlamo were assuaged by messages from Benny Bitjara and other leaders of the Pasukan Merah that, as Javanese, they were not involved in the conflict and were welcome to remain on Halmahera. However, this sense of relative security changed after Indonesian military personnel from Brawijaya 512 arrived to guard Waringinlamo in January 2000. According to several residents, the military personnel warned Muslim villagers that they faced an impending attack from the Pasukan Merah and were required to leave the area and return to Java. While some Muslim villagers agreed with the need to evacuate the island, others refused to leave citing the assurances of the leaders of the Pasukan Merah. In addition, they argued, they no longer had homes or livelihoods to return to in Java. Military personnel forcefully evicted those villagers that refused to leave, assaulting several men and firing bullets into gas canisters necessary for cooking.123 While it is possible the military personnel, not knowing the local situation well, considered that the village was under grave threat from Christian militia, there also appears to have been a financial incentive for the eviction of the Javanese. The military unit used trucks to transport the villagers to Sidangoli on the western side of Halmahera. Each individual was forced to pay 50,000 rupiah (approximately US$5) for this trip plus an additional 15,000 rupiah for the speedboat journey to Ternate, from where they were evacuated to Java. This was a substantial amount for small-scale rural farmers, particularly for a large family. In order to meet this cost, and because they could not take livestock, appliances and other belongings with them, villagers were forced to sell these to military personnel. The soldiers bought the items at much reduced prices and subsequently transported some items

Killing in the name of God 175 off Halmahera and sold others to Christians in north Halmahera, including IDPs from violence elsewhere.124 It therefore appears likely that there were three main reasons for the lack of action or partiality of the security forces in Galela: religious sympathy among Muslim soldiers for the Pasukan Jihad; a concern for their own safety; and financial exploitation of the conflict. In some cases, the response of security personnel in Galela and Kao varied not just from area to area but from guard post to guard post. The level of response was contingent on the surrounding circumstances and the personnel involved in any particular situation. The response would often depend on intelligence about impending attacks received by unit commanders, the degree of contact between commanders and the Pasukan Jihad before attacks and the motivations of individual soldiers.

Conclusion The Pasukan Jihad was created as a direct result of the intense violence that occurred in Tobelo and Galela in late December 1999. The events in Tobelo in particular stimulated widespread anger, and the leaders of the Pasukan Putih declared that Muslims must wage jihad to defeat the Christians who they claimed were attempting to expel all Muslims from north Halmahera. The presence of the existing militia, the Pasukan Putih, made organization relatively simple. Yet recruitment, training and the collection of adequate funding for transport and other costs took several months to complete. In addition, religious ideology was necessary to facilitate widespread recruitment of Muslims not directly affected by the violence. Religious and community leaders deliberately reiterated the principles of jihad, including an obligation to protect Muslims under attack from non-Muslims. However, while leaders framed their campaign within the principles of jihad, the militia did not espouse broader Islamic goals such as the implementation of Islamic law (sharia). The militia’s main leader also proclaimed that his primary goal was to defend the integrity of the Indonesian state from separatists. In addition, for many participants in the militia religious ideology merely provided symbolism and ideological certitude for their campaign to achieve more worldly goals. Many, if not most, of the mujahid came from Tobelo, Galela and Malifut and first and foremost sought retribution for their expulsion by Christians. Partly because of this local nature of the conflict, but also because Muslims enjoyed such numerical superiority in North Maluku, the Pasukan Jihad declined the offer of help from more extreme and militant external Muslim organizations, such as the Laskar Jihad. Religion interacted with strategic goals for Christians in north Halmahera in the same way. The religious element of the conflict added to the strength of emotions of those defending their homes and families. Many believed they were struggling for the very existence of Christianity on Halmahera. This sentiment pervaded the entire community of Tobelo and Galela, facilitating the mobilization of militias and convincing the Christians of Duma, Soatobaru and nearby villages to stay and defend themselves. Within this turmoil, many young men who were willing to risk

176 Ethno-Religious Violence in Indonesia their lives to defend their religion attained a status they would otherwise never enjoy. Despite the influence and visibility of religion in the mobilization of the militia and ensuing violence, the case demonstrates that some apparently religious militias are at heart formed as a reaction to specific socio-political contexts rather than in response to calls for jihad or other transnational ideologies. The Pasukan Jihad launched two main campaigns on Halmahera. The earliest, against the community of Kao, was designed to avenge the destruction of Malifut and return the thousands of Makian IDPs to their homes. The campaign failed largely because the security forces in the area were effective in preventing clashes. The professional and forceful response of the military and police was due to the presence of a gold mine in the area which was extremely valuable to the fledgling provincial economy. The second and larger campaign against several Christian village strongholds in Galela was far more successful. The militia destroyed several of these villages that were preventing the return of Muslim IDPs to the sub-district. In the absence of any incentive to prevent the attacks, the security forces did little to protect these villages, instead concentrating on protecting the Muslim-dominated sub-district capital, Soasio. The Pasukan Jihad halted operations on Halmahera, giving up the goal of defeating Christians in the city of Tobelo, for two major reasons. First, the conquest of that city posed too big a challenge, particularly after attacks against Kao to the south had failed, meaning that Tobelo could not be attacked from both north and south. Second, the implementation of a state of civil emergency triggered a more serious attempt by security forces to prevent violence. By July 2000 most North Malukans were finally free from the threat of violence that had tormented their region for almost a year.

Conclusion

Introduction On a day when most Indonesians were still celebrating the national Day of Independence, two ethnic communities began fighting in the area of Malifut. Two months later, an entire sub-district lay in ruins, its population displaced. By the end of the year, political and religious tensions reached their apogee as thousands of militia members fought in different areas of the province. For the next six months North Maluku was paralyzed, the security forces abandoning entire swathes of the archipelago to the violent campaigns of religious militias. The horror of nearly one year of violence came to an end in July 2000. The conflict had left over 3,000 dead, and innumerable people maimed and traumatized. Photographs which are too graphic to display in this book show bodies disembowelled, decapitated and burned. Much of the new province, particularly the largest island of Halmahera, lay in ruins. The relationship between Christians and Muslims in the region, long characterized by harmony and cooperation, was now dominated by suspicion and hatred. This study has been an attempt to understand such terrible violence. The preceding chapters have provided a comprehensive account of the initiation and trajectory of the conflict in North Maluku, and uncovered the range of structural forces and human agency involved. Each chapter focused on a major development within the conflict. After summarizing the background to the violence, this chapter reviews the findings of this study regarding each phase: initiation, escalation, dispersion, political exploitation and eventual religious war. In addition, additional consideration is given to the question of opportunity – how was violence allowed to occur in a country then facing many problems but far from being a failing state? A second section demonstrates the mutually reinforcing nature of structure, elite and mass agency, interest, emotion and identity in the conflict. By way of special illustration the section discusses the role of religion in the violence. The chapter ends by assessing the implications of this study for the methodological approach to the study of inter-communal conflict.

178 Ethno-Religious Violence in Indonesia

From harmony to bloodshed: the North Maluku conflict A nation and region in turmoil Centuries of trade and colonialism left North Malukan society with two faiths, Islam and Protestant Christianity. Despite their religious differences, the two communities had lived peacefully side by side for centuries. This stability continued throughout the New Order period of President Suharto, not based on fear of the harsh security measures occasionally employed in Indonesia during that era, but on cultural ties and a desire among the local population for development rather than confrontation. But in 1999 change hit hard in the region. While the economic crisis did not greatly affect North Maluku, the sudden collapse of national political structures that had been in place for a generation led to dramatic political change. It is clearly no coincidence that the conflict occurred during Indonesia’s transition from authoritarianism to democracy. Mostly peaceful before and after this period, between 1998 and 2001 the nation experienced several other large-scale and violent inter-communal conflicts, along with an upsurge of vertical conflict in Aceh and Papua, as well as secession in East Timor. Several studies have accounted for the seemingly close connection between democratization and civil conflict. As Snyder and others have demonstrated, conflict can often result when new opportunities, freedoms and threats come up against long-standing nepotism, patronage and other non-democratic practices.1 In North Maluku, democratization opened up political competition for the first time in decades. The eruption of severe religious violence in Ambon caused moderate concerns and animosities in the north of the province. Many North Malukans based in the provincial capital moved back to their homes relating terrible stories of the carnage in Ambon. The impending devolution of political and financial authority to the regions, along with North Maluku’s establishment as a separate province, created opportunities for political and economic gain, particularly among members of the elite. Politicians and bureaucrats spent much of 1999 repositioning themselves to assume power in the new province and its constituent districts. As the region moved towards provincial status, rivalry emerged between several political factions, each broadly associated with a different ethnic group and region. Two issues in particular caused tension between these factions: the location of the provincial capital and, more importantly, the question of who would become the province’s first governor. Each had major implications for employment and business in the region. Political and economic agendas were therefore fundamental to the rising tension in the region before the conflict. None the less, this elite-level competition did not provide the spark that ignited conflict. The genesis of the initial violence can be traced to how this competition translated into social change at the local level. As in many political transitions, members of the elite in North Maluku sought to further the interests of their own ethnic communities and patronage networks. Violence was born out of the attempt by Makians in the political elite in Ternate to resolve a

Conclusion 179 long-standing grievance among members of their ethnic group by providing the Makian community in Malifut with their own autonomous sub-district. Initiation Decades of peaceful, if not altogether friendly relations between the migrant Makians and indigenous Kaos suddenly became confrontational with the announcement that a new sub-district would be formed in Malifut. While it would incorporate several Kao villages, the sub-district was clearly designed to serve the interests of the local Makian community. The new sub-district would not only increase the territory allocated to the Makians under the terms of their original relocation, but would formally end the uncertainty associated with continuing Kao claims to the area. The location of the Gosowong gold mine within the new sub-district would also mean the Makian community would monopolize employment and other benefits associated with that resource. As an autonomous sub-district, Malifut also suddenly became a viable option to serve as a capital of one of the new districts that would make up the new North Maluku province. This eventuality would provide a major financial windfall to the community under new decentralization laws that were to come into effect in January 2001. The Kaos refused to accept the new sub-district for several reasons. They were concerned at the loss of the gold mine, believed the sub-district and its boundaries infringed their traditional land rights and communal integrity and were angered at the lack of consultation prior to the government decision. All of these factors interacted to give their objections an emotive character. But the Makian community and government officials, many of whom were Makians, paid little heed to their concerns. Long-standing political structures associated with the previous authoritarian order therefore exacerbated this dispute. When, through the ethnic nepotism that had become prevalent in local government in New Order Indonesia, one community was seen to monopolize the impending benefits of democratization and financial decentralization, the volatility of local politics increased markedly. Makians in the North Maluku District government and bureaucracy used their dominant representation in those institutions to bypass the Kaos in a decision that directly and profoundly affected them. After almost two months of attempting to gain legal recourse against the Makians for the destruction of their two villages in August, the Kaos resorted to violent action. The status and leadership of local militia leaders was crucial in this decision and its implementation. The imbalance in political influence between the two communities also emboldened some Makians to take a more ‘non-institutionalized’ and aggressive stance towards the Kaos in Malifut. With the sub-district already recognized by the central government in Jakarta, Makians in Malifut and Ternate ignored the Kaos’ objections and mobilized a group of students to pressure them into acceptance. After sustained Kao opposition, and increasing belligerence on the part of members of both communities, Makians from Malifut, led by these students from outside Malifut, attacked and destroyed two Kao villages.

180 Ethno-Religious Violence in Indonesia Had the North Maluku District government responded impartially to this first incident, it is likely the conflict would have ended at this point. However, the Kaos’ demands for compensation received little response, again owing largely to the influence of the Makians in the district government and bureaucracy. Bias within government also meant that, suffering from a lack of political representation, the Kaos were more likely to engage in ‘non-institutionalized’ strategies. Their most immediate motivation was anger at the destruction of the two Kao villages. They were also angered and frustrated by the government’s lack of impartiality, as well as their own lack of success in overturning the creation of the sub-district through institutionalized processes. These factors exacerbated their already strong sense of grievance over what they saw as the violation of their traditional ownership of Malifut, and by extension their legitimate claim to some benefit from the gold mine at Gosowong. Thus disempowered, those Kaos calling for more retributive action assumed greater influence and, in October, some 5,000 Kaos attacked and destroyed the new sub-district. This analysis demonstrates that structures of political and economic inequality between two communities can play an underlying role in the outbreak of communal violence, as asserted by Horowitz and others. In this case, the issue of territory went to the heart of the unequal relationship between the Kaos and the Makians. During the previous two decades, the Kaos had been disempowered relative to the Makians in all but their indigenous rights over the territory in and around Malifut. While land was very important to the well-being and identity of the Kaos, it was the additional importance of that land in their relative position to the Makian that explains the intensity of Kao opposition. Put differently, ownership of territory was their only source of relative superiority and pride over the Makian. In line with Horowitz’s conclusions drawn from his wide-ranging study of ethnic violence, the Kaos were reluctant to relinquish what little power they did hold relative to their more ‘advanced’ neighbours.2 Yet, demonstrating the multifaceted links between inequality and violence, in this case it was the more empowered and economically successful group which initiated violence. When attacking Sosol and Wangeotak, the Makians were to some extent acting out of a sense of disenfranchisement, a lack of legitimate rights to the land on which they had lived for two decades. But in all other respects the Makians were relatively powerful. Chapter 3 demonstrated that it was their perceived greater power vis-à-vis the Kaos, in particular their sense of immunity from prosecution arising from high-level political support, that to a large extent explained the violence. Analysis of the Malifut case also sounds a note of caution about placing too much importance on economic factors as the primary cause of violent conflict. A cursory examination of the conflagration in Malifut may suggest that the violence stemmed wholly or primarily from competition for the benefits of the mine in Gosowong. Both communities in the dispute had a material interest in this lucrative resource in the form of employment and funding for the community. Yet the emotional character of the dispute, and therefore the resort to violence, cannot be explained by reference to these interests alone. A confluence of disputed territorial

Conclusion 181 and mineral rights, threats to ethnic unity and government bias explains the emotive character of the confrontation and the decision to use violence. Escalation The Kaos’ attack on and destruction of Malifut displaced thousands of Makians, who fled to Ternate. These events, and the plight of the refugees, angered many Makians in Ternate and several Makian leaders quickly focused on retaliating against the few Kaos residing in Ternate. At the same time, Makians, both from Malifut and elsewhere, all of whom were Muslims, tried to portray the attack by the mainly Christian Kaos as religiously motivated, alleging that mosques were destroyed and the Qur’an desecrated. There was little evidence to support these assertions and they appear to have been a means of soliciting sympathy and support from Muslims of other ethnicities. However, most other ethnic groups, including Muslims, continued to see the clash in Malifut as a local dispute over territory and natural resources, and few sympathized with the Makians’ plans to seek retribution. Few non-Makian Muslims joined their demonstrations or their attempt to depart for Halmahera. This lack of support allowed the security forces to prevent the Makians both from leaving Ternate and from rioting in the city. In response to the influx of refugees and a rising number of disturbances, the Sultan of Ternate sent his traditional guards, the Pasukan Kuning, on to the streets equipped with spears and machetes. Although the sultan declared that this decision was necessary to protect the infrastructure of the city and the lives of Christian residents, the sudden appearance of a paramilitary presence on the streets, loyal only to the sultan, worried his political opponents. To them, the sultan appeared to be assuming control of the city as a deliberate strategy in the run-up to gubernatorial elections the following year, augmenting his already dominant political position. In this way, political considerations became entwined with the looming ethnic and religious tension in the province. Those Makians trying to provoke anti-Christian rioting formed an alliance with several other political rivals of the sultan’s who hoped to undermine his growing dominance. Together these individuals pressured the district police commander to release a group of Makians arrested for attacking a Christian church. In the following days, several Makian leaders stepped up their campaign to portray the Kaos’ attack as religiously motivated. The main element of this campaign was the ‘Bloody Sosol’ letter that was authored in, and disseminated from, Ternate in the final days of October and purportedly proved the existence of a Christian conspiracy to drive Muslims from Halmahera. Several days after the circulation of this letter, rioting broke out on Tidore and soon spread to Ternate. Mobs attacked and killed Christians and destroyed their homes, shops and churches. This was the first rioting explicitly based on religion. Chapter 4 examined why Christians became the target of this violence despite the Malifut dispute having had little connection to religious differences. In addition, the chapter analyzed why it was possible for violence to break out in the provincial capital, where the hundreds of armed security personnel present should

182 Ethno-Religious Violence in Indonesia have been able to halt the rioting mobs, who possessed no automatic weapons. The events in Tidore and Ternate added inter-religious tension to the greater conflict and caused violence to spread across the province. The analysis of these events therefore presents an example of how a small dispute between ethnic communities can escalate into religious war. Many studies of communal rioting conclude that particular cases of ethnic or religious rioting are spontaneous eruptions of outrage at the barbaric stories found in rumours and other forms of propaganda.3 These rumours often take on a form of agency themselves, largely denuding the actors involved of any responsibility. Most studies of the events in Tidore and Ternate concur with this understanding of riots, seeing them as an eruption of Muslim anger at the contents of the ‘Bloody Sosol’ letter.4 However, Chapter 4 demonstrated that it is important not to overstate the role of rumours and propaganda in instigating communal violence. Even in cases where propaganda appears to have ignited violence, close investigation may reveal greater intent and less suggestibility on the part of those involved. In such cases, rumours and propaganda are designed principally to make rioting that has been meticulously planned appear spontaneous, and should not be considered a cause of rioting themselves. In the attacks in Ternate and Tidore, almost all those participating in the initial violence were Makians, angered at the events in Malifut. These people sought to riot against Christians even in the absence of the ‘Bloody Sosol’ letter. They targeted Christians in general, and not just Kaos, because they believed that GMIH must have been involved in organizing the Kaos’ attack and because they had been prevented from departing for Halmahera. Once Makians had initiated violence in Tidore, the conflict accelerated as individuals from other ethnic groups began to participate. Some did so in order to loot properties, others perhaps merely for the thrill of engaging in violence and destruction. The sight of burning churches and houses also convinced some Muslims that the conflict was indeed about religion. Most ‘late arrivals’ had also identified a high level of political support that would obviate any punishment. If the rioting in the provincial capital cannot be explained purely as the spontaneous response of outraged Muslims, some other change must have occurred in the capital after late October, when the security forces prevented the crowds of Makians from burning churches and leaving for Halmahera. Chapter 4 illustrated that the difference between the situation in October and that in November was the almost complete retreat of the security forces from the city’s streets. This was facilitated by an alliance of influential figures forged by Malifut’s destruction and the sultan’s increasing dominance of the city. Those seeking to retaliate against Christians in the city allied with politicians whose main goal was to undermine the sultan. The resulting political pressure on, and immobilization of, the security forces was a more important contributor to the riots than the ‘Bloody Sosol’ letter and other propaganda. This allowed those Makians seeking to take out their anger on Christians to do so with impunity. The sultan’s rivals saw political benefit in these riots because, as a traditional protector of minorities in the region and reliant on Christian support in his gubernatorial ambitions, the sultan was forced to

Conclusion 183 defend Christians during the rioting. This protection of Christians then became a main source of mobilization for his political rivals as they sought his demise in late December. It was therefore the violence itself, particularly the killing of a pastor and the destruction of churches, facilitated by elites seeking to obtain political advantage from sectarian tension and not the ‘Bloody Sosol’ letter that acted as the catalyst for real religious violence. These acts, and the political Machiavellianism behind them, explain the escalation from local border dispute into provincial religious conflict. These riots ultimately brought the two faiths in the region into bloody confrontation. Dispersion After the anti-Christian rioting in Tidore and Ternate, violence spread to almost every area of North Maluku.5 Ethnic groups divided along religious lines in areas across Halmahera, Bacan, Morotai and other islands. In some cases, members of the same family fought against each other. Chapter 5 analyzed how violence spread throughout the region by examining in detail the cases of Tobelo and Galela Sub-Districts in north Halmahera. The killing of a pastor and the targeting of defenceless Christians in Tidore and Ternate dramatically changed the nature of the conflict throughout the region. The nature of the violence there led people elsewhere in the province to believe the region was now in the grip of religious conflict between Muslims and Christians. Christians were infuriated at the events in the two district capitals and angered and concerned that the security forces had not intervened to halt the riots. Both Muslims and Christians were increasingly afraid that their neighbours would attack them. Tobelo, as the main Christian centre in North Maluku, was additionally affected by the influx of thousands of Christian refugees from Ternate and rural areas in Central Halmahera. A spiral of insecurity ensued in Tobelo and Galela. Both sides prepared weapons and bombs and assumed a more belligerent demeanour towards followers of the other religion. Heightened tensions gave militant members of each community increased status, while at the same time any individuals or groups seeking to reconcile the two communities were marginalized and intimidated. The nature of the violence elsewhere meant that many individuals emphasized their religious identity and solidarity in a militant fashion. This ‘security dilemma’ led almost inexorably towards violence in the area. But security dilemmas do not in themselves cause communal violence. Certain interests and actions played crucial roles in converting this atmosphere of tension into violence. Perhaps because of the increased status that the confrontational situation afforded them, or for more material reasons (such as protection money), certain individuals deliberately attempted to intimidate and provoke members of the opposing community. For example, the brother of Benny Bitjara attempted on one occasion to instigate fighting in Tobelo City. The decision by unknown Muslim leaders to commission the tailoring of white robes, presumably as a means

184 Ethno-Religious Violence in Indonesia to demonstrate the solidarity and strength of local Muslims, exacerbated tension in the city. Other actions that were merely ill advised, such as the transportation of Christian men into the city to guard the GMIH compound, also led to the outbreak on 26 December. In the case of Tobelo, the security dilemma was in part facilitated by incompetence and a fear of exacerbating the situation on the part of the security forces. The police and military rarely took action against individuals clearly acting as provocateurs. Many were local personnel and felt pressure not to act against their own community. Had the security forces acted more forcefully it seems likely that violence might have been averted. Their visible presence on the streets would have assuaged the insecurity felt among the ordinary population and provided security for shops, thus taking away the interest some militia leaders had in continuing volatility. In Galela, the security forces made little effort to prevent violence in the villages around Lake Duma. The violence in Tobelo, particularly the attacks of Christians, was particularly intense because of a confluence of factors. The religious sensitivity of both communities and the involvement of Protestant pastors in the violence increased a sense among Christians that violence against their enemies was divinely sanctioned and removed any remorse they may have felt at killing non-combatants. But strategic considerations were also important. Many Christians, particularly militia leaders, sought to expel Muslims from the sub-district before they could regroup and ally with sympathetic security personnel. Through such brutal violence they also sought to deter Muslims from returning to the area. Numerous clashes broke out across North Maluku in the aftermath of the riots in Tidore and Ternate. Not all were the result of increased security concerns. Many involved more calculated attacks by one community against another, in some cases assisted by militias from neighbouring sub-districts or islands. In Tobelo and Galela, however, the most populous areas outside Ternate and those with the most evenly divided religious composition, a security dilemma provided fertile ground for violence, encouraging belligerent actions and exploitation of tensions within society. In such a situation, provocative and even unconsidered actions, which would otherwise have been ignored, had dramatic consequences. Political exploitation The violence in North Maluku had, until this point, followed a pattern common to many communal conflicts – a small dispute caused a flow of refugees and inspired propaganda and revenge attacks which subsequently escalated into a much larger sectarian conflict. Yet the Putih–Kuning clashes in Ternate in late December suggested that the conflict in its entirety possessed several strands. While Christians and Muslims were engaged in intense clashes in Tobelo, Galela and elsewhere, Ternate City again descended into violence, this time between two Muslim factions. The Muslim followers of the Sultan of Ternate fought against Muslims from the Makian, Tidore and other ethnic groups. This latter group, the Pasukan

Conclusion 185 Putih, was soon reinforced by thousands of men from Tidore Island, and the sultan’s traditional guards, the Pasukan Kuning, were quickly overrun. Most analysts assert that the clashes occurred as a result of the Sultan of Ternate and his armed militia alienating wide sections of the North Maluku community throughout late 1999. The sultan certainly appears to have engendered animosity among the migrant communities on Ternate. Claims that he supported the Kaos in their opposition to the formation of Malifut Sub-District, and his attempt to prevent Makian refugees being evacuated to Ternate, angered Makians. The belligerent behaviour of his traditional guards, who were given sanction to patrol the city’s streets, also increasingly aggrieved many Tidores and other migrants living in the city. The Pasukan Kuning’s destruction of Kampung Pisang marked the apex of this tension and caused many people to take to the streets in opposition to the sultan’s militia Nevertheless, much of the Putih–Kuning conflict suggests that it was far from spontaneous. Mobilization for action against the sultan had already been underway before the Kampung Pisang incident took place. This mobilization was coordinated by the sultan’s political opponents, who were concerned at his growing political and strategic hegemony over the city and, by extension, the new province. The sultan’s strong position in December, along with his previous support for the Kaos in the Malifut dispute, had united almost all other leading political figures and their supporters against him. In Ternate and Tidore, clients of the sultan’s main rivals mobilized militias with the goal of ousting the Pasukan Kuning from the streets of Ternate and overthrowing the sultan as a political force. While inter-religious violence raged elsewhere in the province, Muslims divided into ethnic factions and fought street battles as they determined the future balance of political power in the region. After several decades of authoritarian and non-democratic governance, intimidation remained a common means of achieving political or economic outcomes. Both leading factions in competition for political power in the new province used intimidation and militia violence to prevent a loss in power and undermine rivals. In dispatching his Pasukan Kuning on to the streets of Ternate, Mudaffar Syah undoubtedly sought to enhance his own political position. He hoped to win public support for protecting important infrastructure, while at the same time achieving strategic control over political, social and economic activity in the city. Similarly, threatened by his rising power, the sultan’s rivals mobilized militias to oppose him. Both parties also exercised political pressure on the security forces, undermining the neutrality that might have allowed them to halt rioting in its early stages. Several points of convergence exist between the building tension in Ternate and Tidore in late 1999, which culminated in the Putih–Kuning conflict, and the ‘institutionalized riot systems’ that Brass has identified as existing in some cities of India. By late 1999 the ongoing violence in North Maluku had become entwined with a political process which had implications for thousands of individuals in Ternate. During this period, several individuals, self-proclaimed defenders of the community, exploited communal tension to weaken the Sultan of Ternate as a political force. In particular, the sultan’s alleged support of Christians, ignoring the

186 Ethno-Religious Violence in Indonesia interests of Muslims provided an effective means of mobilizing wider opposition against him. As Wilkinson, Tambiah and Brass have demonstrated in the case of India, when communal tensions become caught up in political processes in this way, the security forces may become less willing to halt violence. In Ternate, military and police commanders appeared unwilling to intervene in the violence for fear of political consequences. The future power holders in the province would be capable not only of exercising influence on the careers of military and police commanders but also of hindering the economic activities of local units. This concern affected the response of military and police commanders from November 1999 to the end of the conflict in June 2000, as discussed further below. Taken in the context of the wider North Maluku conflict, the Putih–Kuning clashes illustrate the complexity and depth of issues present in a case of communal violence involving several events. First, the fighting demonstrates the pitfalls of imputing too much uniformity to ethnic or religious communities in the analysis of violence. Violence between Muslim factions showed that religion was only one of several ruptures in North Maluku society during the conflict. The clashes also demonstrate that the phases of a conflict may differ markedly from each other, depending not only on local political and social contexts, but also the use and exploitation of previous events by influential actors. Religious war Within the space of just four months, what had begun as a dispute over territory had escalated into full-scale religious war. The tensions between Muslims in Ternate dissipated as floods of refugees from Tobelo and Galela arrived in the city, recounting tales of unprovoked Christian attacks and the terrible deaths of hundreds of defenceless people in mosques. Angry refugees from Halmahera called for retaliation against Christians. Originally formed for political goals, the Pasukan Putih assumed a more religious character and was renamed the Pasukan Jihad. The militia swelled as other Muslims who had opposed the anti-Christian rioting in November joined out of fury at the events in Tobelo. With the subsequent mobilization framed in the terms of jihad, Muslim Ternates and Tidores, having only recently ceased fighting each other, became allies to exact revenge on Christians and to allow their co-religionists from north Halmahera to return to their homes. The earlier claims of several Muslim politicians that Muslims faced the threat of Christian aggression appeared to have been borne out. The violence in Tobelo achieved what the Kaos had taken a great deal of effort to avoid in their attack against Malifut – it united the North Maluku Muslim community against Christians in the province. The militia formed over several months and divided into two groups. One attempted unsuccessfully to reach Kao Sub-District, where military personnel prevented major clashes between the militia and the Kao. The other launched a series of attacks against the remaining Christian villages in Galela Sub-District. It disbanded after destroying several villages and expelling all Christians from

Conclusion 187 Galela. In Malifut, the security forces were largely effective in preventing clashes between mujahid and the Kaos in an effort to avoid too much disruption to the NHM mining operations, which provided so much revenue to the provincial coffers. In Galela, however, with little financial incentive to prevent violence, and no directive from the main civilian power holders in Ternate to do so, and with many soldiers apparently sympathizing with the Pasukan Jihad, Christians were often forced to provide their own protection. Through its campaign and rhetoric, the Pasukan Jihad appeared to fit into the rising tide of Islamic militancy in Indonesia in 2000. However, the militia was very local in both participation and goals. The militia leadership refused the assistance of the Laskar Jihad, the Java-based militia that would become prominent in the conflict in Maluku, and few men from outside the province joined the violence in North Maluku. The analysis in Chapter 7 suggests that the Pasukan Jihad had different goals from the Laskar Jihad and other more radical groups. However, also important in their refusal of assistance from Laskar Jihad was their belief that they already possessed overwhelming force relative to Christians in the region. Religion was only one of several motivations for attacks on Christians, as discussed further below. Nevertheless, the principles of jihad did provide a unifying banner under which to mobilize Muslims from all ethnic groups. The violence ended for several reasons. The implementation of civil emergency in the province by the central government increased the legality of firmer action by the security forces against combatants. Shock at the sinking of the Cahaya Bahari also diminished the motivation of many members of the Pasukan Jihad to continue fighting, and many combatants had grown weary of the violence. Yet the principal reason fighting stopped in July 2000 was the now complete separation of Christians and Muslims in the region into exclusive zones after almost a year of bloody violence. The remaining Christian populations in Tobelo and Kao Sub-Districts appeared to the Pasukan Jihad to be unassailable because of the strength of local Christian militias and the presence of military contingents to the north and south of the two areas. The security forces As outlined in Chapter 1, a crucial element in understanding the North Maluku conflict is the question of how the violence was able to happen. In a state which had a record of strongly repressing threats to internal stability how were mobs able to fight unhindered on the streets, causing such social and human destruction? As discussed in Chapter 2, in 1999 the military, and to a lesser extent the police, faced a great deal of turmoil in the form of human rights abuse allegations, declining popular legitimacy, a reduction in their political role and a lack of funding. Several commentators on reformasi-era Indonesia have concluded that the military and other individuals or groups associated with the recently ousted New Order regime of President Suharto attempted to disrupt the national reform process by provoking communal violence across the country.6 This study has demonstrated that this was not the case in North Maluku. In the initial incident in Malifut, there is no evidence

188 Ethno-Religious Violence in Indonesia of instigation by the security forces, which, indeed, appear to have been surprised by the initial attack and then frightened and overwhelmed in October by the thousands of Kao men attacking Malifut under the leadership of Benny Bitjara. However, while not in itself a cause of violence, in almost all cases the response of the security forces determined whether violence occurred, and to what extent. As demonstrated throughout, the performance of the security forces was influenced by a variety of factors. To a great extent, their response was influenced by the political transition occurring across the country and in North Maluku. In this, this study has found much common ground with Wilkinson’s argument that the actions of the security forces will in large part be determined by the interests and attitudes of the civilian political hierarchy who must order (or fail to order) the security forces to act against those causing disturbances. This concern appears particularly pertinent with regard to the urban political centres of Ternate and Tidore, where the security forces were present in greatest numbers but where they were also closest to the elite-level political rivalry occurring in the province. Military and police commanders in Ternate and Tidore were reluctant to antagonize potential future provincial power holders. This influenced their decision to refrain from becoming involved in the building tension and eventual violence in those cities. In early 2000 the Pasukan Jihad continued to enjoy free rein from the security forces partly because, having overthrown the sultan, this militia had the backing of those who had assumed power in the province. The militia’s operations on Halmahera, however, showed that the security forces took quite different approaches depending on the local context. The military forcefully prevented fighting between the Pasukan Jihad and the Kaos in the area of Malifut, largely because the civilian elite and security commanders had a common interest in preventing further violence and destruction in the areas of two of the most lucrative resource-exploitation operations in the region, in Malifut and Weda Bay. In Galela, however, the military offered scant protection to Christian villagers. With little political reason to halt the violence, other motivations influenced the security forces in Galela. The most likely explanation for their failure to deter Muslim attacks on Duma and other Christian villages was that most of the military personnel were Muslims and sympathized with the jihad. It seems possible, too, that some of the funding generated during the mobilization of the Pasukan Jihad was used to bribe military personnel to withdraw during the militia’s attacks. Other economic spoils associated with the conflict also influenced security personnel. In particular some units benefited financially by extracting payment from the displacement of thousands of villagers whom they were charged with protecting, as was shown in the forced evacuation of Javanese transmigrants in Waringinlamo in Kao. The degree of willingness of the security forces to prevent and halt violence was therefore influenced, but not determined by, the political competition occurring in the province. A range of motivations influenced the decisions of military and police commanders and personnel in North Maluku, from the provincial to the village level, on how to respond to violence. The response of the security forces

Conclusion 189 varied from location to location depending on, among other factors, the size of local militias, the prejudices of personnel, the capacity of contingents and financial considerations. The military unit placed in Malifut after the August riot was vastly inadequate to repel the Kao militia’s attack, and the failure of the security forces to dispel the growing security dilemma in Tobelo in late 1999 was at least in part because of their fear of militants. This study therefore demonstrates that no single explanation will account for the action or inaction of the security forces throughout the entire duration of a province- or state-wide conflict.

Structure, agency and motivation in violent conflict Close examination of violent conflicts, such as that contained in this study, highlights the utility of a more synthetic approach to analyzing the agency, causes and motivations involved. In isolation, an analytical focus on either structural factors, the predations of elites, the call of identity or the emotions of the masses cannot explain the onset, trajectory, duration or termination of conflict. The conflict as a whole involved a constantly shifting amalgam of overlapping political, economic and social forces, rational interests and heightened emotions. This study has demonstrated that a range of structural conditions, at the national, district and local levels, played some role in various phases of the conflict. Some political, economic and social patterns established over decades of the New Order were important in the onset and trajectory of the conflict. Norms of nepotism in local government and the use of intimidation in political processes were some of the structural conditions left by the New Order which played a role in the North Maluku conflict. Increasing religious tension across the country, along with other national developments, began to undermine decades of religious harmony in some areas of North Maluku, particularly in Tobelo. Unsurprisingly, given the uncertain and rapidly evolving situation in Indonesia in 1999, changes to many longstanding structures were also important factors in the violence. For example, as noted in Chapter 2, weakened by years of inadequate funding and suffering low morale in the face of widespread criticism in 1999, the security forces were in disarray. No one condition was determinative of any phase of the violence but interacted with other political, economic and social structures to lead to rising tension and violence. For example, political inequality played an important role in the violence in Malifut and in Ternate. Yet while structures of inequality set the scene for conflict in these areas, it was only when this inequality met other contentious issues – ties to land and economic well-being in the case of Malifut and issues surrounding the wider North Maluku conflict in general in Ternate – that they led to violence. But these conditions in themselves do not adequately explain the conflict. Similar conditions existed in numerous regions of Indonesia but few areas experienced large-scale violent conflict. Only by considering the interaction of static and changing structures with calculating, foolish and desperate human agency can the analyst account for the onset and trajectory of the North Maluku conflict. Human agency combined the effects of these different structural factors and converted the

190 Ethno-Religious Violence in Indonesia tensions inherent in such structures into overt conflict and violence. A number of small actions at crucial times and in important locations often had major consequences for the conflict as a whole. The pressure on the security forces by leading politicians, the preparation of the Kao attack in October, the killing of a pastor and the organization of the anti-Christian rioting in Ternate were all actions with great ramifications for the region as a whole. In most cases, violence was primarily organized and initiated by dominant groups concerned at potential threats to their positions of economic or political power. Many key events during the conflict were brought about in large part by the actions of elites. The mobilization of large groups of men was a crucial element of the conflict. Such coordination can often only be carried out by someone in authority. The mobilization of the Makayoa student group by leading Makians in Khairun University was crucial in this regard, as was the dispatching of the Pasukan Kuning on to the streets of Ternate, the mobilization of hundreds of Tidores to go to Ternate and the readying of the Kaos for the attack on Malifut. One of the most influential cases of elite agency was the pressure exerted on the police by leading politicians to release without charge the group of Makian men who had been arrested for attacking Christian property. Perhaps more than any other factor, this created the opportunity for rioting to occur and the conflict to escalate. Yet this study has demonstrated that the North Maluku conflict was not entirely elite-led. Indeed, in most cases, elite agency coincided with a groundswell of sentiment and interest. This is not to suggest that the violence was caused by apolitical social structures, but that we should extend consideration of the role of powerful individuals to encompass the reasons why ordinary North Malukans responded to their provocations. In numerous cases, in Malifut, Ternate and Tobelo, elite organization increased the anger, frustration or interest behind mass mobilization and generated action that may not have occurred otherwise. Many different levels of authority were observable in the conflict, confusing the distinction between leadership and mass-led spontaneity. Actions and decisions that influenced the course of the conflict were taken by individuals not generally considered members of the elite, for example the leaders of local neighbourhoods and villages, militias and local branches of religious organizations. The apparent decision of many ordinary North Malukans to respond to the machinations of politicians resulted from a coincidence of both interests and identities between masses and leaders. In the study of conflict, elite action is often seen as the result of rational calculation, designed to achieve political or other ends, while mass agency is generally seen as stemming from emotion and the ties of communal identity. While many elite actions in the conflict were driven by political interest, some powerful individuals were also simultaneously motivated by the same communal sentiment and ethnic solidarity as their constituents. Makian politicians in Ternate mobilizing crowds to avenge the destruction of Malifut while also aiming to politically weaken the Sultan of Ternate is a case in point. Likewise, some of the people on the streets during violence responded to the emotional

Conclusion 191 speeches of their leaders with one eye on the material or political advantage that might follow their participation. A range of mutually reinforcing motivations lay behind the violence. Viewing the North Maluku conflict with the benefit of hindsight, it is difficult to see how the violence could have been in anyone’s rational interest. Yet this study has demonstrated that, at numerous points during the conflict, individuals and groups made decisions with a major bearing on the onset and trajectory of the conflict, and even initiated violence, with clear instrumental goals in mind. Many actions were undertaken with objectives other than causing violence, but ultimately played an important role in doing just that, such as the decision of Makian elites in Ternate to push through PP42 with little consultation of the Kaos. In other cases, individuals or groups appear to have decided that organization of, or participation in, violence directly served their interests. Many elite actions in Ternate and Tidore were driven by political motives, in particular the mobilization of opposition to the Sultan of Ternate. Rational motives also appear to have motivated many of those ordinary North Malukans who carried out the carnage of 1999 and 2000, as they sought to protect their employment, provide security for their families or to loot shops. Yet at each stage of the conflict violent action may not have occurred, or may not have been as intense, had certain issues and actions not caused emotional responses in large proportions of some communities. Fear and anger were crucial to almost all stages of the conflict. Emotional attachments to territory, frustration at political inequality and at insecure land tenure, fear of religious and numerical domination by the opposing community and of physical attack, anger at previous violence and injustices were all important motivations driving mobilization and participation in the violence. Indeed, this study has shown that drawing a clear distinction between rational and emotional and identity-related motivations for violent action is often problematic. Throughout the North Maluku conflict, most actors’ engagement in mobilization and violence was stimulated by a combination of both instrumental calculation and emotion. Many participants undertook violence with simultaneous ethnic, religious, political and economic considerations. In Ternate, many Makians and Tidores fought the Pasukan Kuning because of their concerns over losing their dominant position in the bureaucracy, becoming politically and strategically subordinate to the indigenous Ternates and through anger at the destruction of Malifut and the sultan’s apparent support for Christians at the expense of Muslims. In Malifut, the perceived loss of employment and revenue from the gold mine accentuated the Kaos’ anger at what they saw as violation of their position as the rightful owners of Malifut. During the dispute over the new sub-district, the material implications of the Gosowong mine increased the sense of ethnic solidarity between Kaos throughout Kao Sub-District and their kin living around Malifut. Likewise, the Gosowong gold mine, and the impending creation of new districts on Halmahera, focused the attention of many Makians in Ternate on their kin in Malifut. Correspondingly, this increased solidarity

192 Ethno-Religious Violence in Indonesia heightened each community’s opposition to any loss of this material resource. This and other events during the conflict demonstrate the mutually constitutive nature of material interest and identity. North Maluku descended into violence in what can be termed an ‘identity–interest spiral’. Religion An analysis of the role of religion in the violence in North Maluku amply demonstrates the interaction of rational and affective influences in motivating those involved in violence. In general this study concurs with those analysts who conclude that religion is most often not a cause of violence, but is sometimes used as a tool by provocateurs seeking to stimulate inter-religious tension for their own more worldly goals.7 Religion played very little role in the initial outbreak in Malifut, which was essentially a result of competition over territory, made emotive by ethnic identity and the unequal treatment of communities by the district government. The crucial turning point for the conflict came in Ternate and Tidore, where influential Makians portrayed their ethnic kin in Malifut as victims of Christian aggression. Yet even in the face of this provocation claims of a campaign of Christianization in North Maluku were not accepted by most Muslims. But the actions of these rioters in murdering a pastor, burning churches and expelling and killing Christians pushed the region towards a sectarian conflict that was defined and waged almost entirely in terms of religious faith. Religion, thereafter, played a multifaceted role in determining the trajectory and intensity of the violence. It was not only used as a tool by those seeking to provoke violence, but also simultaneously heightened the emotions of combatants, facilitated the mobilization of militias, determined the form of that mobilization (the names, clothing and symbolism used by the militias) and provided the ideology that sustained those militias in the field. The use of religious doctrine by men such as Benny Bitjara and Abu Bakar Wahid to legitimate killing, along with the fears of Christians and Muslims that not just their lives but also the sustainability of their faith were under threat, made violence both more likely and more intense. Some of the Muslims involved in the Pasukan Jihad were motivated by what could be termed jihadist principles – a desire to defend Islam and defeat those who had attacked Muslims. Among the Christian community on Halmahera, too, emphasizing Christian solidarity and religious symbolism were effective means of mobilizing the unity and bravery necessary to resist the Muslim militia. Many Christians stated that their goal during the conflict was to ‘defend Jesus’ and several members of the Protestant Church evoked Christian sentiment in mobilizing combatants. Yet even in the later stages of the conflict, which appeared so religious in character, the influence of religion was always contingent on other political, strategic and economic considerations. The outbreak of violence between Christian and Muslim Tobelos in late December was caused by a rising sense of insecurity and intimidation by militants in combination with religious animosity. Similarly, the Muslim refugees from Tobelo, Galela and Malifut who participated in the Pasukan

Conclusion 193 Jihad did so to seek revenge for attacks by Christians and to return to their homes. For these people, jihad was primarily a political and strategic vehicle to achieve these goals, although for others in the militia the sense of waging holy war had real meaning. Christians also had very real motives (in the form of physical threat) to mobilize and oppose the Pasukan Jihad. During some important phases of the conflict, such as the Putih–Kuning clashes in Ternate, religion was only a peripheral issue. Throughout the conflict there was no correlation between religiosity and involvement in violence. Most militia members emphasized their religious identity and devotion during the conflict, but in times of normality were not the most religiously observant. Many devout Muslims and Christians, including religious leaders, opposed the violence and attempted to prevent it. On both sides of the conflict, it was lay people such as Benny Bitjara and Abu Bakar Wahid, far more than imams or pastors, who used religious symbolism and doctrine to motivate militia members. Yet while religion never acted autonomously as a cause of conflict, ignoring its role completely would preclude a proper understanding of much of the violence in North Maluku. Throughout the conflict religious sentiment, material interest and political motives interacted with one another, magnifying and altering the influence each would have had in isolation. As Jack Snyder, in his consideration of the causes of war, puts it ‘the effect of each element can be understood only in the context of the rest of the system.’8 Remove one factor, such as religious tension, economic inequality or political competition, and in many cases the violence would not occur. Summary and implications for research The discussion above demonstrates the wide range of factors behind the North Maluku conflict. In isolation, a focus on structural factors, human agency, identity factors, rationality, elite interest or mass sentiment would fail to explain the onset, trajectory, duration or termination of the conflict. All are surely crucial elements in any explanation of conflict – a focus on one factor at the expense of the others will invariably miss a factor crucial to the onset and development of the violence. It was the interaction of these elements which explains each important development of the conflict. A territorial dispute only triggered the North Maluku conflict because it occurred in conjunction with the Kaos’ ties to the territory and an imbalance in political representation between the two communities. These conditions were fused and made volatile by the specific decisions made by several individuals in Ternate, Malifut and Kao. In other circumstances, such a development might not have led to violence. Likewise, intentional elite provocation only led to anti-Christian rioting in Ternate, and subsequently escalated the conflict, because of its entanglement with political competition, ethnic anger and religious tension. This diversity and complexity is naturally greater when dealing with a conflict that involves a series of violent events, such as that in North Maluku. The varying dynamics involved in different locations and over time mean that, inevitably, a

194 Ethno-Religious Violence in Indonesia greater range of motivations and influences is likely. In a protracted and widespread conflict the issues central to each phase must be distinguished in order to determine how and why violence begins, changes in character, intensifies and moves beyond its initial location. Particularly important is the need to differentiate initial causal factors from later developments. This study has also illustrated the influence of local dynamics on the trajectory of a conflict. As Stathis Kalyvas has pointed out, the violence that occurs in different areas may appear as part of the whole, but may in fact be far more local in character and only tenuously connected to the primary dynamic of the wider conflict.9 In turn, these dynamics influence the path, extent and intensity of the conflict as a whole. Set against the background of religious violence in Ambon, the violence in Malifut appeared to have been the result of tension between Christians and Muslims. However, qualitative research demonstrated that religion played only a peripheral role in that dispute. Likewise, the Putih–Kuning clashes and the violence in Tobelo, which occurred simultaneously in late December 1999, were caused by wholly different dynamics which were contingent on the local context – political competition in the case of the former and insecurity and religious tension in the latter. A full explanation of each outbreak of violence requires a detailed illustration of the actions, motivations and chance events involved. The actions that caused and changed the trajectory of the North Maluku conflict – the economic opportunism and mobilization associated with the violence in Malifut, the pressure on the security forces in Ternate from politicians and the militancy of certain individuals in Tobelo – have been discussed throughout this book and need no further elaboration here. All of these conclusions suggest that the proper analysis of conflict situations requires micro-level observation and analysis. This is perhaps particularly so in the study of violent conflict because of its volatility and the misleading nature of most rhetoric that accompanies it. The identification of objective factors frequently associated with conflict situations also tells us little if we do not uncover the meaning of those objects to relevant actors, the surrounding socio-political context and the agency involved. The social complexity of each phase examined in this study, the fluidity of the issues over time and the variation from one location to another all obscure the true nature of the violence, which can easily be missed without close examination. This requires the gathering of information from participants and witnesses themselves. This is not to say that general conclusions regarding the causes or trajectories of conflict are impossible, or to cast doubt on the efficacy of well-researched comparative analyses. However, such studies should be preceded by an uncovering of the detail, temporal dynamism and geographical variation of each case. A failure to first ascertain the detail of each case may lead to erroneous conclusions and hinder the advancement of knowledge regarding communal violence.

Conclusion 195

Conclusion This study has revealed that a range of macro-structural forces and changes in those structures played a role during the conflict in North Maluku. Economic opportunism, political inequality, high-level political competition, insecurity, ethnic and religious antagonism, territory and natural resources all played not inconsequential roles during the conflict. When each main event was closely analyzed, it became possible to ascertain which causal factors played a role at which stage in the conflict. Yet while these conditions increased tension and made violence possible it was human agency that defined and caused each critical juncture. Such agency has been demonstrated throughout this study. In some cases these actions inadvertently had consequences far beyond their intention. In others, however, individuals and groups sought to exploit opportunities at the expense of other communities and in several cases deliberately provoked violence. The response of the security forces was important in allowing these tensions and motivations to reach their conclusion in violence. On several occasions, the security forces acted professionally and competently to prevent or halt violence. There were also many cases in which security personnel were outnumbered, lacked capacity, were disorganized or were otherwise unable to act forcefully against those involved in violence. However, at other critical moments within the conflict, security personnel appeared to make an intentional decision to remain disengaged, influenced by the political competition taking place in North Maluku in 1999, by personal bias or by corruption and a desire to exploit the conflict for profit. This study has gone some way to explaining why a society that had seen decades of stability and relative inter-religious harmony, a society that, in early 1999, was united in its endeavours to forge its own province, could descend into such violence. Explaining communal violence that simultaneously involves economic, political and identity issues and which is both organized and spontaneous is no easy task. The complexity of the conflict is perhaps best encapsulated in the words of the participants themselves, who were left by the spiral of events struggling to comprehend how they could have gone from relative stability to such extreme confrontation: ‘how can we explain why we killed each other … we are all one family.’

Notes

Introduction 1 This study uses English language terms for the administrative units in Indonesia. Therefore ‘sub-district’, ‘district’, ‘municipality’ and ‘province’ are used for kecamatan, kabupaten, kotamadya and provinsi respectively. When talking about districts and municipalities together, I use the term ‘district’. The term ‘region’ is used generally to refer to provinces and districts. I also anglicize the names of ethnic groups; for example when talking about members of an ethnic community, I will use ‘Kaos’ and ‘Makians’ rather than the Indonesian forms, orang Kao and orang Makian. There are no accepted English terms for these groups and I believe this usage provides uniformity and deviates from the Indonesian term as little as possible. By referring to ethnic groups in this way, I also differentiate them from the places after which they are named, i.e. Kao Sub-District and Makian Island. 2 The local sub-district officials who compiled the figures did not include members of the Pasukan Jihad who were registered as residents of other administrative areas, such as Ternate, Tidore, Sanana, etc. 3 Interview with Muhammad Selang in Ternate, 7 February 2004. Mujahid, meaning a participant in holy war, is the term used by most North Maluku Muslims to refer to participants in what was widely seen as a jihad. 4 The official statistic for the number of IDPs is 247,620 people. No breakdown of IDPs by religion was readily available. 5 I use the term Internally Displaced Person (IDP) in this study to refer to a person displaced within their own country. 6 See the book-length work by K. H. Ahmad and H. Oesman, Damai Yang Terkoyak: Catatan Kelam dari Bumi Halmahera (Shattered Peace: Dark Notes from the Land of Halmahera), Ternate: Madani Press, 2000, and J. Nanere, Halmahera Berdarah (Bloody Halmahera), Ambon: Yayasan Bina Masyarakat Sejahtera dan Pelestarian Alam (BIMASPELA), 2000. Journal articles and chapters on the conflict include: S. Alhadar, ‘The Forgotten War in North Maluku’ in Inside Indonesia, 2000; M. D. Boediman, ‘Musuhku adalah Saudarku’ (My Enemy is My Brother), unpublished dissertation, Universitas Kristen Duta Wacana, Yogyakarta, 2002; N. Bubandt, ‘Malukan Apocalypse: Themes in the Dynamics of Violence in Eastern Indonesia’, in I. Wessel. and G. Wimhofer (eds), Violence in Indonesia, Frankfurt: Abera Verlag Markus Voss, 2001; N. Bubandt, ‘The Dynamics of Reasonable Paranoia: Rumours and Riots in North Maluku, 1999–2000’, background paper for presentation at the Seminar Series of the School of Anthropology, Geography and Environmental Studies, University of Melbourne, 25 April 2002; N. Bubandt, ‘Towards a New Politics of Tradition? Decentralisation, Conflict, and Adat in Eastern Indonesia’, Antropologi Indonesia, no. 74 (May–August 2004); the final section in C. R. Duncan,

198 Notes

7

8

9 10 11 12

13 14 15 16 17 18 19 20 21 22 23 24

‘Savage Imagery: (Mis)representations of the Forest Tobelo of Indonesia’, The Asia Pacific Journal of Anthropology, vol. 2, no. 1 (2001), 45–62; C. R. Duncan, ‘The Other Maluku: Chronologies of Conflict in North Maluku’, Indonesia, 80 (October 2005), 53–80; the chapters on North Maluku in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003; T. A. Tomagola, ‘The Bleeding Halmahera of North Moluccas’, Jurnal Studi Indonesia, vol. 10, no. 2 (2000); T. A. Tomagola, ‘Krisis dan Solusi Tragedi Maluku Utara’ (The Crisis and Solution of the North Maluku Tragedy), Detikcom, 2 February 2000. The following reports primarily on the Maluku conflict also contain small sections on the North Maluku conflict: International Crisis Group, Indonesia’s Maluku Crisis: The Issues, Indonesia Briefing, 19 July 2000; International Crisis Group, Indonesia: Overcoming Murder and Chaos in Maluku, Asia Report no. 10, 19 December 2000; International Crisis Group, Indonesia: The Search for Peace in Maluku, Asia Report no. 31, 8 February 2002. The two book-length studies examining a large portion of the conflict, Nanere’s Halmahera Berdarah and Ahmad and Oesman’s Damai yang Terkoyak, present strongly Christian and Muslim accounts of the violence respectively. The different interpretations presented of the conflict, in particular those emanating from the capital city Ternate, no doubt play a role in displacing responsibility away from those currently holding power. On the struggle to interpret conflict see P. R. Brass, ‘Introduction: Discourses of Ethnicity, Communalism, and Violence’, in P. R. Brass (ed.), Riots and Pogroms, New York: New York University Press, 1996. M. Kordi and H. Ghufran, ‘Dinamika Masyarakat Makian Dan Konflik Maluku Utara’ (Makian Societal Dynamics and the North Maluku Conflict), in Hasan, Memikirkan Kembali, p. 141, Ahmad and Oesman, Damai Yang Terkoyak, p. 121, and M. Kordi and H. Ghufran, ‘Kompleksitas Konflik Maluku Utara’ (The Complexity of the North Maluku Conflict), in Hasan, Memikirkan Kembali, p. 127. Tomagola, ‘The Bleeding Halmahera of North Moluccas’. Ibid., p. 7. International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 6. Local communities would actually have exercised little control over this natural resource which, as discussed in the following chapter, was (and still is) exploited by a joint venture between an Australian and Indonesian mining company. However, the mine was still important for those communities living adjacent to the area, in terms of employment, company donations and other small amounts of revenue. Alhadar, ‘The Forgotten War in North Maluku’; and S. Alhadar, ‘Kompleksitas Konflik Di Halmahera Utara’ (The Complexity of the Conflict in North Halmahera), in Hasan, Memikirkan Kembali, p. 144. Nanere, Halmahera Berdarah, p. 62. Alhadar, ‘The Forgotten War in North Maluku’. International Crisis Group, Indonesia’s Maluku Crisis, p. 5. Tomagola, ‘The Bleeding Halmahera of North Moluccas’, p. 9. Ibid.; see note 27 for the percentage of Muslims as support for Bahar Andily’s greater political support. Nanere, Halmahera Berdarah, p. 62. The history of this rivalry is discussed in the next chapter. International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 6. Alhadar, ‘The Forgotten War in North Maluku’. This conflict is discussed in the next chapter. International Crisis Group, Indonesia’s Maluku Crisis, p. 2. Another study of the Maluku conflict that sees the North Maluku violence as an extension of the conflict in Ambon is E. Al-Jakartaty, Tragedi Bumi Seribu Pulau (The Tragedy of the Land of a Thousand Islands), Jakarta: Gubuk Kajian Mutiara Nasional, 2000, p. 13.

Notes 199 25 J. Bertrand, Nationalism and Ethnic Conflict in Indonesia, Cambridge: Cambridge University Press, 2004, pp. 129–31. 26 Ibid., p. 130. 27 Ibid., p. 131. 28 The process of pemekaran is discussed in the next chapter. Since 1999, over 100 new districts have been created in Indonesia. 29 The literature includes C. Tilly, From Mobilization to Revolution, Reading, MA: Addison Wesley, 1978; D. McAdam, Political Process and the Development of Black Insurgency 1930–1970, Chicago and London: University of Chicago Press, 1982; M. N. Zald and B. Useem, ‘Movement and Countermovement Interaction: Mobilization, Tactics, and State Involvement’, in M. N. Zald and J. D. McCarthy (eds), Social Movements in an Organizational Society, New Brunswick, NJ: Transaction Books, 1987; D. S. Meyer and S. Staggenborg, ‘Movements, Countermovements, and the Structure of Political Opportunity’, American Journal of Sociology, vol. 101, no. 6 (May 1996). 30 M. D. Toft, The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory, Princeton and Oxford: Princeton University Press, 2003, D. Horowitz, Ethnic Groups in Conflict, Berkeley: University of California Press, 1985, p. 143. 31 See for example: P. Collier and A. Hoeffler, ‘On Economic Causes of Civil War’, Oxford Economic Papers, vol. 50, no. 4 (1998), 563–73; N. Sambanis, ‘A Review of Recent Advances and Future Directions in the Quantitative Literature on Civil War’, Defence and Peace Economics, vol. 13, no. 3 (2002), 215–43; J. D. Fearon and D. D. Laitin, ‘Ethnicity, Insurgency, and Civil War’, American Political Science Review, vol. 97, no. 1 (February 2003). 32 For example Alhadar, ‘The Forgotten War in North Maluku’; Bubandt, ‘The Dynamics of Reasonable Paranoia’, p. 2. 33 Alhadar, ‘The Forgotten War in North Maluku’. 34 Ibid. 35 Bubandt, ‘The Dynamics of Reasonable Paranoia’, p. 2. 36 Ibid, p. 24. 37 See S. J. Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia, Berkeley: University of California Press, 1996, p. 81, and P. R. Brass, Theft of an Idol: Text and Context in the Representation of Collective Violence, Princeton: Princeton University Press, 1997, p. 208. 38 Tomagola, ‘The Bleeding Halmahera of North Moluccas’, p. 4. 39 Alhadar, ‘The Forgotten War in North Maluku’. 40 Ahmad and Oesman, Damai Yang Terkoyak, pp. 60–71. 41 International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 8. 42 Bubandt, ‘Towards a New Politics of Tradition?’. 43 Nanere, Halmahera Berdarah, p. 98. 44 Tomagola, ‘The Bleeding Halmahera of North Moluccas’, p. 9. 45 International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 7. 46 Alhadar, ‘The Forgotten War in North Maluku’. 47 Tomagola, ‘The Bleeding Halmahera of North Moluccas’, p. 4. 48 International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 9. 49 The free-rider concept is most commonly associated with Mancur Olson. See M. Olson, The Logic of Collective Action: Public Goods and the Theory of Goods, Cambridge, MA: Harvard University Press, 1971.

Chapter 1 1 Examples include: B. Arfi, ‘Ethnic Fear: The Social Construction of Insecurity’, Security Studies, vol. 8, no. 1 (Autumn 1998), 151–203; S. Ellis, The Mask of

200 Notes

2 3 4 5 6 7 8 9 10 11 12 13 14 15

16 17 18 19 20 21 22 23 24 25 26 27 28

Anarchy: The Destruction of Liberia and the Religious Dimension of an African Civil War, New York: New York University Press, 1999; V. P. Gagnon, ‘Ethnic Nationalism and International Conflict: The Case of Serbia’, in M. E. Brown et al. (eds), Nationalism and Ethnic Conflict (International Security Readers series), Cambridge, MA: The MIT Press, 1997; and J. Snyder, From Voting to Violence: Democratization and Nationalist Conflict, New York: W. W. Norton & Company, 2000. V. P. Gagnon, ‘Ethnic Nationalism and International Conflict: The Case of Serbia’, International Security, vol. 19, no. 3 (Winter 1994/95), p. 142. Ibid., p. 134. J. Mueller, ‘The Banality of Ethnic War’, International Security, vol. 25, no. 1 (Summer 2000), 42–70. M. Kaldor, New and Old Wars: Organized Violence in a Globalized Era, Stanford: Stanford University Press, 1999, p. 8. S. Tarrow, Power in Movement: Social Movements, Collective Action, and Mass Politics in the Modern State, Cambridge: Cambridge University Press, 1994, p. 3. J. D. McCarthy and N. Zald, ‘Resource Mobilization and Social Movements: A Partial Theory’, American Journal of Sociology, vol. 82, no. 6, p. 1215. Ibid., p. 1215. D. McAdam, Political Process and the Development of Black Insurgency 1930–1970, Chicago and London: University of Chicago Press, 1982, p. 21. B. A. Valentino, Final Solutions: Mass Killing and Genocide in the Twentieth Century, Ithaca and London: Cornell University Press, 2004, p. 61. Ibid., p. 38. Ibid., p. 64. D. Horowitz, Ethnic Groups in Conflict, Berkeley: University of California Press, 1985. A. Varshney, Ethnic Conflict and Civic Life: Hindus and Muslims in India, New Haven and London: Yale University Press, 2002. See for example P. Collier and A. Hoeffler, ‘Greed and Grievance in Civil War’, World Bank Policy Research Working Paper no. 2355, Washington, DC: World Bank, October 2001; M. L. Ross, ‘Oil, Drugs and Diamonds: The Varying Roles of Natural Resources in Civil Wars’, in K. Ballentine and J. Sherman (eds), The Political Economy of Armed Conflict: Beyond Greed and Grievance, Boulder: Lynne Reiner Publishers, 2003. B. R. Posen, ‘The Security Dilemma and Ethnic Conflict’, Survival, vol. 35, no. 1, 27–47. Horowitz, Ethnic Groups in Conflict, and T. R. Gurr, Why Men Rebel, Princeton: Princeton University Press, 1970. Posen, ‘The Security Dilemma and Ethnic Conflict’, pp. 29–31. Gurr, Why Men Rebel, p. 9. Ibid., p. 50. D. Horowitz, The Deadly Ethnic Riot, Berkeley: University of California Press, 2001, pp. 225–6. Horowitz, Ethnic Groups in Conflict, p. 143. D. Horowitz, ‘Group Loyalty and Ethnic Violence’, in C. Hermann et al. (eds), Violent Conflict in the Twenty-first Century: Causes, Instruments and Mitigation, Illinois: American Academy of Arts & Sciences, 1999, p. 101. Horowitz, The Deadly Ethnic Riot, p. 225. Ibid., pp. 226, 230. Ibid., p. 13. S. Kaufman, ‘An “International” Theory of Interethnic War’, Review of International Studies, vol. 22, no. 2 (April 1996), p. 157. Ibid., p. 158.

Notes 201 29 D. Keen, The Economic Functions of Violence in Civil Wars, Adelphi Paper 320, London: Oxford University Press, 1998. 30 S. Olzak, The Dynamics of Ethnic Competition and Conflict, Stanford: Stanford University Press, 1992. 31 Ibid., pp. 28, 209. 32 Ibid., p. 212. 33 J. D. Fearon and D. D. Laitin, ‘Explaining Interethnic Cooperation’, American Political Science Review, vol. 90, no. 4 (December 1996), p. 715. 34 J. D. Fearon, ‘Ethnic War as a Commitment Problem’, paper presented at the 1994 Annual Meeting of the American Political Science Association, New York, August 30–September 2, 1995. 35 S. N. Kalyvas, ‘The Ontology of “Political Violence”: Action and Identity in Civil Wars’, Perspectives on Politics, vol. 1, no. 3 (September 2003), p. 483. 36 S. N. Kalyvas, The Logic of Violence in Civil Wars, New York: Cambridge University Press, 2006, pp. 12–14. 37 Valentino, Final Solutions, p. 235. 38 Ibid., pp. 169–72. 39 C. Kaufman, ‘Rational Choice and Progress in the Study of Ethnic Conflict: A Review Essay’, Security Studies, vol. 14, no. 1 (January–March 2005), p. 182. 40 R. D. Petersen, Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe, Cambridge: Cambridge University Press, 2002. 41 Ibid., pp. 22–31. 42 O. Waever, B. Buzan, M. Kelstrup and P. Lemaitre, Identity, Migration and the New Security Agenda in Europe, London: Pinter Publishers, 1993, p. 23. 43 S. Kaufman, ‘Symbolic Politics or Rational Choice? Testing Theories of Extreme Ethnic Violence’, International Security, vol. 30, no. 4 (Spring 2006), pp. 52–3. 44 Collier and Hoeffler, ‘Greed and Grievance in Civil War’, p. 2. 45 Ibid., p. 17. 46 Ibid., p. 16. 47 Correspondingly, conflict study is often seen as divided into several different approaches: Essentialism (or Primordialism); Instrumentalism; Institutionalism; and Constructivism. A great deal of literature attempts to assess the progress of these main approaches to explaining conflict. See for example C. Kaufman, ‘Rational Choice and Progress in the Study of Ethnic Conflict’. See also Kaufman, ‘Symbolic Politics or Rational Choice?’. On several problems with the use of this typology, see C. King, ‘The Micropolitics of Social Violence’, World Politics, vol. 56, no. 3 (April 2004), pp. 435–7. 48 This is clearly not a new discovery – numerous analysts have recognized the need for a more synthetic and consensual approach to the study of conflict as the discussion below demonstrates. See for example: Arfi, ‘Ethnic Fear’; J. B. Rule, ‘Rationality and Non-Rationality in Militant Collective Action’, Sociological Theory, vol. 7, no. 2 (Autumn 1989), p. 158; D. Horowitz, ‘Structure and Strategy in Ethnic Conflict’, paper prepared for the Annual World Bank Conference on Development Economics, Washington, DC, 20–21 April 1998; Kaufman, ‘An “International” Theory of Interethnic War’, p. 157. 49 As stated above, Donald Horowitz claims riots and other forms of communal violence contain both organization and spontaneity: see for example ‘Structure and Strategy in Ethnic Conflict’, p. 25. See also P. R. Brass, The Production of Hindu–Muslim Violence in Contemporary India, New Delhi: Oxford University Press, 2003, p. 32. 50 A. J. Regan, ‘The Bougainville Conflict: Political and Economic Agendas’, in Ballentine and Sherman, The Political Economy of Armed Conflict. 51 Horowitz, The Deadly Ethnic Riot, p. 13. 52 Snyder, From Voting to Violence, p. 305. 53 Ibid., p. 39.

202 Notes 54 55 56 57 58

59

60 61 62

63

64

65 66 67 68 69 70

Arfi, ‘Ethnic Fear’. Ibid., p.156. Ibid., p. 168. Ibid. D. A. Lake, ‘International Relations Theory and Internal Conflict: Insights from the Interstices’, International Studies Review, vol. 5, no. 4 (December 2003), p. 86. Surveying several detailed case studies of violence, James Fearon and David Laitin provide a good overview of some possible answers to this dilemma. Possible explanations include: a psychological bias towards believing the leaders of one’s group; information imbalances between leaders and masses; a long-standing discourse of ethnic animosity; and the possibility that participants are actually taking advantage of ethnic provocation to pursue their own agendas: J. D. Fearon and D. D. Laitin, ‘Violence and the Social Construction of Ethnic Identity’, International Organization, vol. 54, no. 4 (Autumn 2000), pp. 845–55. Many internal conflicts are also complicated by the intervention of external parties, at national or international levels. While the actions (or lack thereof) on the part of Jakarta had some influence on the trajectory of the violence, the North Maluku conflict remained almost entirely local in scope so this consideration does not apply here. R. Brubaker and D. Laitin, ‘Ethnic and Nationalist Violence’, Annual Review of Sociology, vol. 24 (1998), p. 446. S. J. Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia, Berkeley: University of California Press, 1996, p. 214. For example, the Ambon conflict in eastern Indonesia quickly evolved from clashes between local Christians and Muslim migrants from South Sulawesi into religious violence. See International Crisis Group, Indonesia’s Maluku Crisis: The Issues, Indonesia Briefing, 19 July 2000, and Overcoming Murder and Chaos in Maluku, Asia Report no. 10, 19 December 2000. See for example A. A. Engineer, ‘Introduction’, in A. A. Engineer (ed.), The Gujarat Carnage, New Delhi: Orient Longman, 2003; P. R. Brass, Theft of an Idol: Text and Context in the Representation of Collective Violence, Princeton: Princeton University Press, 1997; Tambiah, Leveling Crowds. Kalyvas, ‘The Ontology of “Political Violence”’, pp. 475–6. Of rioting during the Tragic Week in Barcelona in 1909, the governor is reported to have said: ‘On each street they shouted different things and fought for different purposes’, H. Thomas, The Spanish Civil War, rev. edn, New York: The Modern Library, 2001, p. 18. Kalyvas, ‘The Ontology of “Political Violence”’, p. 479. K. R. Young, ‘Local and National Influences in the Violence of 1965’, in R. Cribb (ed.), The Indonesian Killings 1965–1966: Studies from Java and Bali, Victoria, Australia: Aristoc Press, 1990. Kalyvas, ‘The Ontology of “Political Violence”’, p. 481. Individuals interviewed included members of the Ternate, Tidore, Makian, Tobelo, Galela and Kao ethnic communities. H. Blumer, Symbolic Interactionism: Perspective and Method, Berkeley: Prentice Hall, 1969. Ibid., p. 2.

Chapter 2 1 Kabupaten Maluku Utara and Kabupaten Halmahera Tengah. In April 1999 Ternate City was made a municipality (kotamadya), giving it the same status as a district with its own parliament. I use the full term North Maluku District when referring to the district. I use the term North Maluku when referring to the entire region.

Notes 203 2 Interview with Yusuf Abdurrahman, former rector of Khairun University, in Ternate, 12 January 2004. In reality, after several centuries of movement between islands, the differences between ethnic communities are perhaps not clearly demarcated, although most people strongly identify with their ethnic community and differentiate themselves from other communities. Linguistically, the region is divided between Papuan language speakers, located mostly in northern Halmahera, Ternate and Tidore, and Austronesian speakers, located mostly on Makian, Bacan and in southern Halmahera: L. Y. Andaya, The World of Maluku: Eastern Indonesia in the Early Modern Period, Honolulu: University of Hawaii Press, 1993, p. 104. 3 Although many Makians believe their community constitutes perhaps 50 per cent of North Maluku’s population. 4 Under transmigrasi the central government resettled families from overcrowded areas of the country, such as Java and Bali, to less crowded areas in eastern Indonesia and elsewhere. 5 Official statistics for religious affiliation state that Muslims constitute 80 per cent of the North Maluku population. However, many Christians, including the Protestant Evangelical Church on Halmahera (GMIH), dispute this, suggesting that the true figure is more likely to be approximately 65 per cent, with Christians comprising the remaining 35 per cent of the population. 6 C. G. Kiem, Growing up in Indonesia: Youth and Social Change in a Moluccan Town, Saarbrücken and Fort Lauderdale: Verlag Breitenbach Publishers, 1993, p. 139. 7 More complete accounts of North Maluku’s history can be found in: Andaya, The World of Maluku; J. Haire, The Character and Theological Struggle of the Church in Halmahera, 1941–1979, Frankfurt: Verlag Peter D. Lang, 1981; W. A. Hanna and D. Alwi, Turbulent Times Past in Ternate and Tidore, Banda Naira: Yayasan Warisan dan Budaya Banda Naira, 1990. 8 Andaya, The World of Maluku, p. 1. Archaeological records from the Middle East suggest that the clove, native only to Maluku, had been traded along long-distance networks since at least 1700 BC. 9 From the early sixteenth century, Ternate came to exercise power over Jailolo, which faced it across the strait on Halmahera. Likewise, Tidore eventually incorporated Bacan into its own kingdom: Haire, The Character and Theological Struggle, p. 9. 10 Ambon recognized the authority of Ternate until the Dutch colonial administration assumed control of the island in the late seventeenth century: Andaya, The World of Maluku, p. 84. 11 Ibid., p. 97. This relationship can still be observed in present-day linguistic terminology. Tobelo and other communities (both Muslim and Christian) use the terms ke atas (above) to signify any direction toward Ternate and ke bawah (below) for any direction away from Ternate. These terms are used rather than north and south and can make for a confusing bus ride from Tobelo southward towards Ternate (ke atas). 12 For example, Galela was required to send the sultan 20 men for service each year along with large amounts of rice: Hanna and Alwi, Turbulent Times Past, p. 259. 13 Andaya, The World of Maluku, p. 96. 14 Ibid. 15 Ibid., p. 55. Intermarriage between the two kingdoms was common, and in 1999 the two sultans were (albeit distant) relations. 16 Haire, The Character and Theological Struggle, p. 92. 17 R. Z. Leirissa, ‘The Idea of a Fourth Kingdom in Nineteenth Century Tidorese Maluku’, in E. K. M. Masinambow (ed.), Maluku dan Irian Jaya (special issue of Buletin Leknas, vol. III, no. 1), Jakarta: Lembaga Ilmu Pengetahuan Indonesia, 1994, p. 181. 18 Andaya, The World of Maluku, p. 157, and Hanna and Alwi, Turbulent Times Past, p. 135. 19 Ibid., p. 252.

204 Notes 20 Andaya, The World of Maluku, p. 57. 21 Haire, The Character and Theological Struggle, p. 91. 22 J. Villiers, ‘Las Yslas de Esperar en Dios: The Jesuit Mission in Moro 1546–1571’, Modern Asian Studies, vol. 22, no. 3 (1988), p. 595. Christian conversion was apparently spurred by a series of volcanoes and earthquakes. In 1533, 35,000 people were baptized in 29 villages around ‘Moro’. Remains of the Portuguese settlement can be seen in the forest on a mountainside behind the village of Mamuya. 23 Hanna and Alwi, Turbulent Times Past, pp. 57–8, 75. 24 Andaya, The World of Maluku, p. 146. 25 The Portuguese were particularly hostile towards Muslims, a result of the centuries-long occupation of Portugal by Moors from North Africa: ibid., p. 123. 26 Ibid., p. 131. 27 Ibid., p. 132. 28 Haire, The Character and Theological Struggle, p. 107. 29 N. Ishige, ‘Limau Village and its Setting’, in N. Ishige (ed.), The Galela of Halmahera: A Preliminary Survey, Osaka: National Museum of Ethnology, 1980, p. 8. 30 Haire, The Character and Theological Struggle, p. 144. 31 Ibid., p. 67. 32 Ibid., p. 182. The Ambonese, who constituted the majority of the pastors in the region at that time, also practised a form of Christianity syncretized with local Ambonese adat. 33 After spice production and the accompanying colonial administration had been relocated to Ambon, Christians on that island were provided with superior education and employment opportunities in the bureaucracy and Dutch military compared to Muslims: R. Chauvel, Nationalists, Soldiers and Separatists: The Ambonese Islands from Colonialism to Revolt 1880–1950, Leiden: KITLV Press, 1990, pp. 20, 27. 34 Interview with Galela Muslim leader Haji Umar T Baendi, former village head of Igobula, Galela, in Igobula, 30 October 2003. The Japanese occupied Halmahera from 1942 to 1945. Japanese forces occupied most areas of the region, including Ternate, but were stationed in greatest concentration in Kao, where 50,000 troops were based and built an airport. After the deployment of large numbers of American troops led by General MacArthur on the large northernmost island of Morotai in the latter part of World War II, North Maluku, particularly north Halmahera, became the scene of intense conflict. The wrecks of several Japanese warships lie off the coast of Kao: J. Leith, ‘Resettlement History, Resources and Resistance in North Halmahera’, in S. Pannell and F. von Benda-Beckmann (eds), Old World Places, New World Problems: Exploring Issues of Resource Management in Eastern Indonesia, Canberra: Centre for Resource and Environmental Studies, Australian National University, 1998, p. 113. 35 Interview with local historian Adnan Amal in Ternate, 15 January 2004. In 1958, several military officers in North Sulawesi launched the Permesta rebellion. Supporters of the rebellion in North Maluku fought against government military units on Halmahera and Ternate: Hanna and Alwi, Turbulent Times Past, p. 271. Interview with Ternate historian Herry Nachrawy in Ternate, 14 January 2004. 36 Interview with Ternate historian Herry Nachrawy in Ternate, 14 January 2004. 37 R. Chauvel, ‘Ambon: Not a Revolution but a Counterrevolution’, in A. R. Kahin (ed.), Regional Dynamics of the Indonesian Revolution: Unity from Diversity, Honolulu: University of Hawaii Press, 1985, p. 257. On the RMS rebellion, see also Chauvel, Nationalists, Soldiers and Separatists. 38 Chauvel, ‘Ambon: Not a Revolution but a Counterrevolution’, p. 259. 39 Ibid. While the largely Christian RMS separatist movement was defeated, the Christian community in Ambon continued to enjoy dominance in gubernatorial and civil service positions until the 1990s.

Notes 205 40 See for example Kiem, Growing up in Indonesia, p. 54. 41 Interview with Herry Nachrawy, in Ternate, 14 January 2004. 42 Many Muslims in North Maluku refer to the movement as Republik Maluku Sarani – the local term for Christian. 43 Haire, The Character and Theological Struggle, p. 54. 44 Many of the Christian Ambonese suspected of being involved in RMS were pastors or others associated with GMIH. 45 For a discussion of Darul Islam, see C. van Dijk, Rebellion under the Banner of Islam, The Hague: Martinus Nijhoff, 1981. 46 Interview with Christian community leader Urubungus Djawa in Tobelo Sub-District, 24 January 2004. 47 Interview with Urubungus Djawa and several other Christians in Tobelo Sub-District, 2003–4. 48 Kiem, Growing up in Indonesia, p. 55. 49 Ibid. 50 Ibid. 51 P. A. Meyer and M. Hardjodimedjo, ‘Maluku: The Modernization of the Spice Islands’, in H. Hill (ed.), Unity and Diversity: Regional Economic Development in Indonesia since 1970, Oxford: Oxford University Press, 1989, p. 558. 52 It is estimated that almost three-quarters of Maluku Province (before the separation of the North Maluku region) was forested and two-thirds of the logs extracted in Maluku came from the north: ibid., pp. 550, 564. 53 Ibid., p. 557. 54 Newcrest Mining Ltd website: www.newcrest.com.au PT Nusa Halmahera Mineral is 82.5 per cent owned by Newcrest and 17.5 per cent owned by Aneka Tambang. 55 A good discussion of the migration of young people to Ternate from rural areas on other islands to attend school can be found in Kiem, Growing up in Indonesia, pp. 73–7. 56 Makians officially make up 9 per cent of the North Maluku District population. Statistic from ‘Maluku Utara Dalam Angka, 1999’ (North Maluku in Figures, 1999), Badan Pusat Statistik (Central Bureau of Statistics), Ternate. Their proportion in Central Halmahera District is greater, however. There are also, according to several sources, many others in North Maluku who have one Makian parent but do not identify themselves as Makians in censuses. There has traditionally been much intermarriage between Makians and several other ethnic groups such as Tidores, Sananas and Kayoas. 57 Interview with Christian community leader Urubungus Djawa in Tobelo Sub-District, 24 January 2004. 58 The perception of this dominance held by other ethnic groups, especially those from Halmahera, is probably even greater than the reality. Makians represent most political parties including Golkar, PDI-P (Democratic Party of Indonesia: Partai Demokrasi Indonesia – Perjuangan) and PPP (the United Development Party: Partai Persatuan Pembangunan). 59 Thaib Armain was also the head of the bureaucracy for North Maluku District from 1989 to 1997. In 2002 he was elected governor of North Maluku Province. 60 Interview with Yusuf Abdurrahman, Khairun University campus, Ternate, 12 January 2004. 61 For example, in 1998, following a dispute with the district head, Abdullah Assagaf, the sultan’s traditional guards (Pasukan Adat), also sometimes known as the Young Generation of Sultan Baabullah (Gemusba), attacked Abdullah Assagaf’s office. The guards destroyed a great deal of equipment in the office, including computers and furniture. 62 This was despite societal and familial pressures against converting, as required by Indonesian law when the partners are of different religions.

206 Notes 63 Kiem, Growing up in Indonesia, pp. 107, 111–12. 64 Ibid., p. 116. To some extent these divisions in Ternate society were also associated with political divisions between Golkar (of which the sultan was a representative) and the party with a strong Islamic character, PPP. 65 Badan Pusat Statistik, Bappenas and UNDP, Indonesia Human Development Report 2001. Towards a New Consensus: Democracy and Human Development in Indonesia, Jakarta, 2001, p. 2, and H. Hill, ‘Indonesia: The Strange and Sudden Death of a Tiger Economy’, Oxford Development Studies, vol. 28, no. 2 (2000), p. 118. 66 Badan Pusat Statistik, Bappenas and UNDP, Indonesia Human Development Report 2001, pp. 2–8. 67 The World Bank estimated that from 1990 to 1996 highly mobile short-term debt trebled. Cited in Hill, ‘Indonesia: The Strange and Sudden Death of a Tiger Economy’, p. 124. 68 Badan Pusat Statistik, Bappenas and UNDP, Indonesia Human Development Report 2001, p. 1. 69 For the sake of uniformity this study uses the more recent term TNI for the military and Polri (Polisi Indonesia) for the national police. 70 In the 1950s newly independent Indonesia had a brief period of democracy. 71 Some of the more resource-rich provinces such as Aceh and Papua accused Jakarta (or more generally Java) of having exploited local resources with scant compensation. 72 The impact of this policy is discussed further in the next chapter. 73 H. Crouch, ‘Wiranto and Habibie: Military–Civilian Relations since May 1998’, in A. Budiman et al. (eds), Reformasi: Crisis and Change in Indonesia, Melbourne: Monash Asia Institute, 1999, pp. 140–1. In 1997 this number was reduced to 75. 74 Ibid., p. 145. 75 J. Kristiadi, ‘The Armed Forces’, in R. W. Baker et al. (eds), Indonesia: The Challenge of Change, Singapore: Institute of Southeast Asian Studies, 1999, p. 101. 76 The monthly salary of ordinary soldiers is between Rp550,000 and Rp900,000 (US$55–90) and for high-ranking officers it is between Rp2 million and Rp6 million (US$200–600): ‘Cash Strapped Military Recipe for Corruption’, Asia Times, 15 March 2003. 77 Editorial, ‘Current Data on the Indonesian Military Elite: January 1 1999–January 31 2001’, Indonesia, vol. 71 (April 2001), p. 1. The editors estimate that, in 1999, the military had control of over 300 companies through foundations and cooperatives (ibid., footnote 3). 78 D. Bourchier, ‘Skeletons, Vigilantes, and the Armed Forces’ Fall from Grace’, p. 152. 79 H. Crouch, ‘Political Update: Megawati’s Holding Operation’, in E Aspinall and G. Fealy (eds), Local Power and Politics in Indonesia: Decentralization and Democratization. Singapore: Institute of Southeast Asian Studies, 2003. See also G. Robinson, ‘Indonesia: On a New Course?’, in M. Alagappa (ed.), Coercion and Governance: The Declining Political Role of the Military in Asia, Stanford: Stanford University Press, 2001, p. 242. 80 This lack of resources was felt most strongly in the regions, because the government’s funding allocation to the security forces almost certainly remains in Jakarta. Therefore while the security forces as a whole are forced to obtain 70 per cent of their required income from ‘off-balance-sheet’ activities, in the more remote regions the percentage is closer to 100 per cent. 81 Asian Development Bank, ‘Country Governance Assessment Report Republic Indonesia’, draft report, September 2002. 82 See for example the anti-Madurese riots by Malays in West Kalimantan in 1999: J. S. Davidson, ‘The Politics of Violence on an Indonesian Periphery’, South East Asia Research, vol. 11, no. 1, 59–89. 83 B. Singh, ‘The Indonesian Military Business Complex: Origins, Course and Future’, Working Paper no. 354, Canberra: Strategic and Defence Studies Centre, 2001, p. 10.

Notes 207 84 Polls taken in Jakarta, Medan and elsewhere found that 46 per cent of the population believed that the TNI did not work in the interests of the nation: Bourchier, ‘Skeletons, Vigilantes, and the Armed Forces’ Fall from Grace’, p. 155. 85 International Crisis Group, National Police Reform, Asia Report no. 13, February 2001. 86 Eighty-seven per cent of Indonesians consider themselves Muslim. 87 M. Mietzner, ‘Godly Men in Green’, Inside Indonesia, 53, January–March 1998. 88 See A. Azra, ‘The Islamic Factor in Post-Soeharto Indonesia’, in C. Manning and P. Van Dierman (eds), Indonesia in Transition: Social Aspects of Reformasi and Crisis’, Singapore: Institute of Southeast Asian Studies, 2000, p. 313. 89 J. Bertrand, Nationalism and Ethnic Conflict in Indonesia, Cambridge: Cambridge University Press, 2004, p. 80. 90 On the connection between more contemporary radical groups and Darul Islam, see M. van Bruinessen, ‘Genealogies of Islamic Radicalism in Post-Suharto Indonesia’, South East Asian Research, vol. 10, no. 2, (2002), and on the origins of the Darul Islam movement, see J. C. Santosa, ‘Modernization, Utopia and the Rise of Islamic Radicalism in Indonesia’, unpublished PhD dissertation, Boston University, 1996, p. 248. 91 Pancasila was first articulated by Sukarno in 1945 and has since been enshrined as the national ideology. The five principles of pancasila are: belief in one God, humanitarianism, national unity, democracy and justice. 92 Santosa, ‘Modernization, Utopia and the Rise of Islamic Radicalism in Indonesia’, p. 326. 93 These incidents were attributed to a radical Islamic group under the leadership of Husein Ali al Habsyi, a teacher at an Islamic religious school (pesantren) in Malang. 94 Bertrand, Nationalism and Ethnic Conflict in Indonesia, p. 99. 95 K. Steenbrink, ‘Muslim–Christian Relations in the Pancasila State of Indonesia’, Muslim World, vol. 88, no. 3 (1998), 32–52, and Bertrand, Nationalism and Ethnic Conflict in Indonesia, pp. 100–2. 96 ‘Kami tidak akan mundur’: interview in Tobelo, North Maluku in 2003. 97 On the violence in Poso, see L. V. Aragon, ‘Communal Violence in Poso, Central Sulawesi: Where People Eat Fish and Fish Eat People’, Indonesia, vol. 72 (October 2001), 45–79; G. J. Aditjondro, ‘Kerusuhan Poso dan Morowali, Akar Permasalahan dan Jalan Keluarnya’ (The Poso and Morowali Riots, the Roots of the Problem and the Way Out), paper presented at seminar on Application of Emergency Status in Aceh, Papua and Poso?, Hotel Santika, Jakarta, 7 January 2004; and D. McRae, ‘Criminal Justice and Communal Conflict: A Case Study of the Trial of Fabianus Tibo, Dominggus Da Silva, and Marinus Riwu’, Indonesia, vol. 83 (April 2007), 79–117. 98 There were violent incidents in at least seven locations throughout Indonesia on the same day, including Cirebon, Tegal, Pemalang, Sugihwaras, West Kalimantan and Bolang Mongondow district in North Sulawesi, according to ‘Mass Violence Mars Indonesia’s Eid al-Fitr Celebrations’, Agence France Presse, 21 January 1999. The nature of the incidents, i.e. anger over recent elections, retribution on a suspected thief, violence against police because of brutality, would suggest that the religious character of the date was not determinative. 99 ‘Panic in Ambon – National Human Rights Group Ambushed’, Kompas, 3 February 1999. 100 ‘Death Toll in Indonesian Religious Riot Reaches 24: Official’, Agence France Presse, 21 January 1999, ‘Warning Shot Marks Tension in Riot Hit Indonesian Town’, Agence France Presse, 23 January 1999. 101 See the Ambon Information Website, where figures obtained from the Regional Office of the Ministry of Transmigration are displayed: www.websitesrcg.com/ambon/ Transmig.htm. Accessed May 2002.

208 Notes 102 ‘Unrest in Ambon on Lebaran Day’, Kompas, 21 January 1999. 103 Ambonese reportedly resented the influx of Bugis because of a belief they created slums and caused rising crime: Human Rights Watch, ‘Indonesia: The Violence in Ambon’, Human Rights Watch Report, 1999. Prior to the violence, both Christian and Muslim Ambonese and Bugis had also complained of the violence of the migrant Butonese population: D. Mearns, ‘Class, Status and Habitus in Ambon’, in D. Mearns and C. Healey (eds), Remaking Maluku: Social Transformation in Eastern Indonesia, Special Monograph no. 1, Centre for Southeast Asian Studies, Northern Territory University, 1996, p. 102. 104 See report by Richard Rowat, the MSF Field Coordinator in Ambon, online at: www.websitesrcg.com/ambon/documents/Advocat1.htm. 105 J. Bertrand, ‘Legacies of the Authoritarian Past: Religious Violence in Indonesia’s Moluccan Islands’, Pacific Affairs, vol. 75, no. 1 (2002), p. 73. 106 See J. D. Goss and T. R. Leinbach, ‘Development and Differentiation: A Case Study of a Transmigration Settlement in West Seram’, in Mearns and Healey, Remaking Maluku, p. 92, and F. von Benda-Beckmann and T. Taale, ‘Land, Trees, and Houses: Changing (Un)Certainties in Property Relationships on Ambon’, in Mearns and Healey, Remaking Maluku, p. 52. 107 International Crisis Group, The Search for Peace in Maluku, p. 2. Human Rights Watch, ‘Indonesia: The Violence in Ambon’. 108 C. J. Bohm, ‘Brief Chronicle of the Unrest in the Moluccas 1999–2001’, Crisis Centre, Diocese of Amboina, p. 4. 109 International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 3. 110 Bertrand, ‘Legacies of the Authoritarian Past’. The RMS rebellion of the 1950s was also preceded by Christian concerns at being incorporated into a predominately Muslim state: Chauvel, ‘Ambon: Not a Revolution but a Counterrevolution’, p. 260. 111 G. van Klinken, ‘The Maluku Wars: Bringing Society Back In’, Indonesia, 71 (April 2001), p. 11. 112 G. J. Aditjondro, ‘Guns, Pamphlets and Handy-Talkies: How the Military Exploited Local Ethno-religious Tensions in Maluku to Preserve their Political and Economic Privileges’, in I. Wessel and G. Wimhofer (eds), Violence in Indonesia, Hamburg: Abera Verlag Markus Voss, 2001. 113 Between 165 and 600 Ambonese sailed to Ambon from Java at the end of 1998, either expelled for having inadequate identification papers (KTP) or voluntarily for the religious holidays of Christmas and Ramadan. These groups included the leaders and members of Ambonese gangs in Jakarta 114 International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 5. 115 The distance between Ternate and the Maluku provincial capital of Ambon was approximately 500 km, which is approximately the same as the distance between Jakarta and Surabaya: Meyer and Hardjodimedjo, ‘Maluku: The Modernization of the Spice Islands’, p. 549. 116 Interview with North Maluku historian Herry Nachrawy in Ternate, 14 January 2004. 117 Ibid. As in 1999, during the Sukarno and Suharto eras, the creation of a new province required the agreement of the government of the province from which the new province would be created. 118 The violence in Ambon cannot be considered a factor in the mobilization to create the province, as efforts to do so preceded the start of that conflict. 119 Many North Malukans also perceived that the new president, B. J. Habibie, would be more sympathetic to the region, coming as he did from Gorontalo in Sulawesi: F. Ammari and J. W. Siokona (eds) Ternate: Kelahiran dan Sejarah sebuah Kota (Ternate: The Birth and History of a City), Ternate: Ternate City Government, 2003, p. 135. 120 Interview with Hasbi Yusuf, currently a lecturer at Khairun University and in 1998 a member of the FPPMU, in Ternate, 5 January 2004.

Notes 209 121 Bahar and Syamsir Andili are originally from Tidore, although of Gorontalo heritage. 122 Numerous interviews in Ternate and elsewhere in North Maluku, 2003–4. 123 President Sukarno granted Yogyakarta Daerah Istimewa status because of the city’s involvement in the independence struggle against the Dutch. 124 Several interviews in North Maluku.

Chapter 3 1 Sidney Tarrow defines a social movement as ‘collective challenges by people with common purposes and solidarity in sustained interaction with elites, opponents, and authorities’, S. Tarrow, Power in Movement: Social Movements, Collective Action, and Mass Politics in the Modern State, Cambridge: Cambridge University Press, 1994, p. 3. 2 J. D. McCarthy and N. Zald, ‘Resource Mobilization and Social Movements: A Partial Theory’, American Journal of Sociology, vol. 82, no. 6, p. 1216. 3 Perhaps because the focus of most Social Movement theorists has been rebellion and insurgency, most theorists have focused overwhelmingly on the responses of different factions of the state’s elite to social movements. 4 On counter-mobilization, see M. N. Zald and B. Useem, ‘Movement and Countermovement Interaction’, in M. N. Zald and J. D. McCarthy (eds), Social Movements in an Organizational Society, Brunswick, NJ: Transaction Books, 1987, and D. S. Meyer and S. Staggenborg, ‘Movements, Countermovements, and the Structure of Political Opportunity’, American Journal of Sociology, vol. 101, no. 6 (May 1996), p. 1635. 5 See for example D. McAdam, Political Process and the Development of Black Insurgency 1930–1970, Chicago and London: University of Chicago Press, 1982, p. 56, and Zald and Useem, ‘Movement and Countermovement Interaction’, p. 248. 6 Meyer and Staggenborg ‘Movements, Countermovements, and the Structure of Political Opportunity’, p. 1650. 7 D. McAdam, ‘Tactical Innovation and the Pace of Insurgency’, American Sociological Review, vol. 48 (December 1983), p. 735. 8 Recent scholarship in the Resource Mobilization tradition has also emphasized the importance of emotion to social movements. See R. R. Aminzade and D. McAdam, ‘Emotions and Contentious Politics’, in R. R Aminzade et al. (eds), Silence and Voice in the Study of Contentious Politics, Cambridge: Cambridge University Press, 2001, p. 15. 9 D. Horowitz, Ethnic Groups in Conflict, Berkeley: University of California Press, 1985, p. 143. 10 Ibid., pp. 147–75. 11 For example, Anthony Smith defines an ethnie as a ‘named human population with a shared ancestry, myths, histories and cultures, having an association with a specific territory, and a sense of solidarity’: A. D. Smith, The Ethnic Origins of Nations, Oxford: Blackwell Publishers, 1986, p. 32. 12 M. D. Toft, The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory, Princeton and Oxford: Princeton University Press, 2003, pp. 1–6. 13 There are obviously many other factors relevant to these conflicts, but ties to territory are central to both. 14 Kao is now located within North Halmahera District. Until October 1999, the area of North Maluku was not yet a province but comprised two districts (North Maluku and Central Halmahera) and one city (Ternate). The area that is now the district of North Halmahera was, at that time, several sub-districts within North Maluku District. 15 Badan Pusat Statistik Kabupaten Maluku Utara, Ternate (Central Bureau for Statistics for North Maluku District), ‘Maluku Utara Dalam Angka, 1999’.

210 Notes 16 There are no official statistics for the number of Muslims in Kao. Abdurrahman Hongi, a Kao Muslim community leader, stated that over a thousand Muslims live in Kao (interview in Kao, February 2004). 17 These settlements are largely in West Kao but are also situated at Waringinlamo, 15 km from the capital, and near Pediwang in the north-east of the sub-district. The inhabitants of these settlements are generally Muslim Javanese. However, the composition of many of these villages has changed substantially following the conflict, as almost all Muslim Javanese returned to Java, and their houses and land were often bought (cheaply) by Christian IDPs from elsewhere on Halmahera (see Chapter 7). 18 For a discussion of this resettlement, see R. Lucardie, ‘Spontaneous and Planned Movement among the Makianese of Eastern Indonesia,’ Pacific Viewpoint, vol. 26, no. 1 (April 1985), 63–78. 19 Many Makians state that the government forced them to depart from their villages. 20 Interviews in North Maluku, 2003–4. 21 Makians living in Malifut and Kaos testify to the welcoming attitude of the Kaos in 1975 and the existence of close relations between the two communities. For example, Fahri Yamin, a Makian staff member for the Malifut office of PDI-P, interviewed in Malifut, 3 November 2003, and Nathanial Bitjara, a Kao community leader interviewed in Kao, 29 December 2003. 22 In 1999, approximately 75 per cent of the Malifut population was successfully engaged in farming clove, cocoa and coconut (for copra), most of which was sold to Kao villages and Tobelo for export (interview with the Malifut Sub-District Head and head of agriculture, Malifut, 20 January 2004). The Makians are widely considered by most ethnic groups in North Maluku to be highly successful farmers. 23 Interview with Ahdan Abdul Gani, Kepala Desa Samsuma village, Malifut, 2 February 2003. 24 Interview with Suratman in Ternate, 13 January 2004. Suratman was involved in the push for legalized status for Malifut. North Maluku District Head Instruction (Instruksi Bupati KDH Tk. II Maluku Utara) no. 09/8/MU/1995 closed the island to reoccupation. See H. H. Sitohang et al., Menuju Rekonsiliasi Di Halmahera (Towards Reconciliation on Halmahera), Jakarta: Pusat Pemberdayaan untuk Rekonsiliasi dan Perdamaian, 2003, p. 76. 25 Lucardie, ‘Spontaneous and Planned Movement’, p. 76. 26 Members of the Kao and Tobelo communities regularly expressed this stereotype to me. They often related a story in which the Makians were described as an elephant which first asks if it may rest its trunk inside a house, then his head and finally his whole body until there is no room left for the original inhabitants. 27 Interviews with large numbers of Kaos. It is difficult to obtain statistics indicating inequality in funding, although the perception is widespread and deeply held among Kao. 28 Several Makians, including members of the Makian elite in Ternate, expressed this stereotype to me. 29 Interview with Pastor Rein Salakparang. 30 I heard this from several sources, including Mohtar Adam, a Makian on the academic staff at Khairun University, interviewed in Ternate, 17 February 2004. 31 Confidential interview in North Maluku, 2003. By 1999 Kao ethnic solidarity became far stronger in the face of ethnic antagonism with the Makian. 32 Newcrest Mining Ltd website: www.newcrest.com.au. PT Nusa Halmahera Mineral is 82.5 per cent owned by Newcrest and 17.5 per cent owned by Aneka Tambang. 33 Interview with an Australian employee of Newcrest in Gosowong, 2003. 34 Interview with several Australian employees of Newcrest, Gosowong mine, 2003. 35 Muksin H. Abdullah, ‘Kerusuhan Maluku Utara Dalam Perspektif Sosial Budaya’ (The North Maluku Riots in a Socio-cultural Perspective), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North

Notes 211

36 37 38 39

40 41 42

43 44

45 46

47

48

49 50

Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003, p. 132. Interviews with several Makians, including Suratman, 13 January 2004. Interview with Suratman, 13 January 2004. Interview with several former students involved in lobbying the provincial and central governments. The names of these informants are confidential because of the sensitivity of the information. Most members of the Makayoa group, or individuals associated with it, now hold positions in the local bureaucracy or universities. Yusuf Abdurrahman told me it was ‘stupid’ that the Kaos in villages such as Sosol wanted to remain part of Kao Sub-District – the location of infrastructure clearly dictated that they should be within Malifut. Interview on Khairun campus, Ternate, 12 January 2004. Interview with Lt Col. (Ret.) Abdullah Assagaf, Bandung, 28 February 2004. Ibid. Interviews in North Maluku 2003–4. Several respondents also stated that some Makian elite held the goal of making Malifut the provincial capital. In the early 1980s there were plans among the North Maluku elite to create an independent North Maluku Province. At that time there was discussion of making Malifut the capital of the new province. The vice-president, Adam Malik, visited Malifut in 1982. According to one respondent, this visit was widely seen as intended to assess the area as a potential provincial capital. Interview with Tobelo community leader, Urubungus Djawa, in Tomahalu village, 24 January 2004. Interview with Abdullah Assagaf, district head of North Maluku July 1994–July 1999, in Bandung, 28 February 2004. The eventual North Halmahera District now covers only the eastern side of the peninsula. Interviews in North Maluku, 2003–4, including members of the North Maluku District parliament. Even if the structure of two districts and one municipality had remained after the new province had been established, it was likely that North Maluku District would require a new capital. If, as seemed likely in early 1999, Ternate was to become the provincial capital, it was likely that it would be considered unwieldy for that city to fulfil a role as the capital of both the province and a district. Interview with Hein Namotemo, in Tobelo, 2 December 2003. In May 1999, one month prior to the start of production at Gosowong, Law 22/1999 and Law 25/1999 were signed by President Habibie and thereby passed into law. Thanks to Professor Harold Crouch for information regarding the timing of these laws. The decentralization process began in January 2001. This proportion is specified in one of the implementing regulations following that law. According to Regulation (PP) 104/2001, Section 10, this royalty is divided into Land Rent (Iuran Tetap) and (the much smaller) Exploration and Exploitation Rent (Iuran Eksplorasi dan Eksploitasi), 80 per cent of each going to the region. Of the 80 per cent of Land Rent, 16 per cent is paid to the province and 64 per cent to the producing district or municipality (Kabupaten/Kota Penghasil). Of the 80 per cent of Exploitation Rent, 16 per cent is paid to the province, 32 per cent to the producing districts and 32 per cent to the remaining districts in the province. Information obtained from the provincial mining department in Ternate in the form of statements of royalty payments by Nusa Halmahera Mineral to the Department of Finance in Jakarta. I was unable to obtain figures for 1999 or 2000, but I assume the mine at that early stage of exploitation, despite the conflict, was producing at a rate closer to the 2001 rate. Land Rent is obviously far smaller than Exploitation Rent. In the first quarter of 2002, Nusa Halmahera Mineral paid US$5,354.47 in Land Rent as opposed to US$465,407 in Exploitation Rent. As was the case with North Maluku District, a large proportion of the staff of both the bureaucracy and government were likely to be Makian. Law 25/1999 provided no guidelines as to what portion of mining revenue must be

212 Notes

51 52

53 54 55 56 57 58 59 60 61 62 63 64

65 66 67 68 69 70

71 72 73 74 75

passed down to the sub-district government, leaving a ‘producing’ sub-district dependent on the goodwill of the district government. In 2004, ex-district head Abdullah Assagaf stated to me there was a consensus in the district parliament in the early part of 1999 to form the sub-district. Interview in Bandung, 28 February 2004. The full name of this law is Peraturan Pemerintah No. 42/1999 tentang Pembentukan dan Penataan Beberapa Kecamatan di Wilayah Kabupaten Daerah Tingkat II Maluku Utara (Government Law No. 42/1999 on the Formation and Organization of Several Sub-Districts in North Maluku District). The use of student groups to lobby the central government may have been intended to have this effect. It is possible that the wider Kao community influenced the communities of those villages set to be included in Malifut, making them subsequently reject the sub-district. Interview with Suratman, a member of the Makayoa student group, in Ternate, 13 January 2004. Interview in Gosowong, October 2003. For example, Arnol Nanlohy, a Kao Christian, interviewed 10 January 2003, and Mr Moumou, a Kao Christian and village head of Sosol in 1999, both argued that the loss of the mine was not a primary source of opposition to PP42. Interviews with numerous Kao. Sitohang et al., Menuju Rekonsiliasi Di Halmahera, pp. 78–9. Ibid. Interviews with Kao community leaders in Kao, 2003–4. Interview with a member of the Makayoa student group. Several Jailolo villages were also included in the new sub-district. I was told of the symbolism of the grave by, among others, Haji Ahmad and Abdurrahman Hongi, two Kao Muslim community leaders, in an interview in Kao, 19 January 2004, and by Imam Langar, a Kao Muslim religious leader, in Kao, 1 October 2003. A Christian Kao community leader stated to me the principle of ‘hidup bersama mati sekubur’ (Ternate, confidential source). Interviews in Kao 2003–4. Interviews in Sosol and Wangeotak in 2003. Interview with Abdullah Assagaf in Bandung, 28 February 2004. Interviews with members of the Makayoa group. Confidential interview with a Kao community leader, North Maluku, 19 January 2004. Interview with Ahdan Abdul Gani, village head Samsuma village, Malifut, 2 October 2003. Confidential interviews with several Kaos in North Maluku 2003–4. Kaos state that in August, the new sub-district head of Makian di Malifut, Husen Kuda, gave an ultimatum to the Kaos, stating that if they did not want to join ‘please raise your feet and get out of this new sub district’ and that ‘the houses of those who do not wish to join will need to be burned’. The date for this speech is uncertain, but some Kaos state that Kuda made the speech on 18 August, the day the violence occurred. As only Kaos related this story, and no Makians, I cannot confirm it. Several respondents stated that around this time the district head Abdullah Assagaf recognized the volatility of the situation in Malifut and initially urged the district parliament to implement PP42. Confidential interview with a Makian community leader, Ternate, January 2004. The Imam of Tahane mosque confirmed to me that ‘Allahu Akbar’ was shouted from the mosque, although he said he was not present for the riot. Confidential interview with a Makian community leader in Ternate. Wangeotak is almost adjacent to Sosol, although the two villages are separated by 100–200 m of Makian houses.

Notes 213 76 Interview with Commander Franciscus Arisusetio, in Tobelo, 29 September 2003. Arisusetio is now the commander of the military company (Kompi Senayan C) for North Halmahera District. 77 The Kao community historically served as important members of the Ternate Sultanate’s military forces. 78 Confidential interviews in Kao in 2003. This was also stated in an unpublished report by GMIH, released in December 1999. 79 Interview with the sultan in Ternate, February 2004. 80 Interviews with members of the Team of Nine such as Pastor Salamena, Mr Bitjara and Haji Muksin. 81 Confidential interview in Kao with a member of the Team of Nine. 82 Interview with the Sultan of Ternate, Mudaffar Syah, in Ternate, 17 February 2004. Members of the Team of Nine also told me the sultan was sympathetic to their demands, but that he told them it was difficult to circumvent a government law. This is an important point, as the sultan is considered by many Makians to have provoked the Kaos. 83 Confidential interview in Kao with a member of the Team of Nine. 84 Ibid. 85 Before the initial incident, Benny Bitjara was living in Kupa Kupa village, Tobelo Sub-district, but subsequently he spent a great deal of time in Kao, overseeing Kao military preparations. Benny Bitjara would also come to be known as Benny Doro, ‘Doro’ being the name of his village in Kao. Prior to the conflict, Benny Bitjara was in charge of security at the Pertamina installation at Kupa Kupa. Some military preparations had been carried out simultaneously with the Kaos’ diplomatic efforts, but were accelerated over time as the Kaos received no response from the government. 86 Unsurprisingly, each side believes the other intended to attack that day. 87 Makians I interviewed suggested that there were between 40 and 100 Makians present. Kaos whom I interviewed, however, stated that there were as many as 600 Makian, although one Kao man who had been present guarding the border suggested there were 200. Interview with Kao Muslim, Mohtar Ismael, in Kao, 19 January 2004. 88 Interviews with Makians in Malifut and Ternate, including Ipor, a man from the Malifut village of Tahane, in Ternate, 15 January 2004. 89 Separate interviews with Mohtar Ismael and Darwan, Muslim Makians guarding Kalijodo on Sunday 24 October 1999, in Kao,19 January 2004. The Kaos suggest that the Makians timed the attack to occur during the church service (ibadah) on Sunday morning, when most Christians would be ill prepared, and the small numbers of Muslim Kaos on guard could be easily overrun or persuaded to join the Makians. 90 Large numbers of Makians interviewed stated the community was short of food; interviews in Malifut and Ternate, 2003–4. Interview with Ipor, Ternate, 15 January 2004. 91 Confidential interviews, North Maluku, 2003–4. This statement does not by any means prove that this happened, but does add to the list of factors discussed above. 92 Most Kaos (and Tobelos) consider the communities of West Kao, often called orang pedalaman (interior people) to be more ‘warlike’ and animist than other Kao communities: numerous interviews on Halmahera, 2003–4. 93 Interview with Benny Bitjara, Kupa Kupa village, 18 September 2003. As is almost always the case in such events, the exact number of Kao troops is difficult to obtain. Makian have suggested there were as many as 20,000. Benny Bitjara stated he led 15,000. However, the fact that the total population of Kao Sub-District was only 27,000 suggests that both these figures are exaggerated. The more likely figure is probably approximately 5,000. At least some women and children were involved. No members of the Tobelo ethnic group assisted in the attack. 94 Interview with Satta Sabar, Ternate, 4 January 2004. Satta Sabar was a community leader in Malifut at the time of the conflict and now is a staff member in the North Maluku District Department of Agriculture (Kantor Petanian) in Ternate. Apparently

214 Notes

95 96 97 98 99

100

101

102 103

104

Makian from Ternate and elsewhere had attempted to reach Malifut from Ternate but had been blocked by Indonesian security forces at Sidangoli, the port village on the western side of Halmahera. Interview with Satta Sabar, 4 January 2004. Interviews with Benny Bitjara, Kupa Kupa village, 18 September 2003 and Satta Sabar, Ternate, 4 January 2004. It seems likely that the schools were not destroyed because they were built with funds from Nusa Halmahera Mineral. The mosques in Malifut were subsequently ransacked and then destroyed during the wider conflict on Halmahera. Interview with Benny Bitjara. As Christians are a 20 per cent minority in North Maluku, it is obvious why the largely Christian Kao would not want the wider population to see the conflict as religious. The issue of the destruction of mosques in Malifut would nevertheless become a central issue in the development of the conflict in North Maluku. A series of national laws, from the Basic Agrarian Law No. 5 1960 to the Basic Forestry Law 1967 and the Mining Law 1968, state that customary title is recognized only so long as it does not interfere with national and state interests. See International Crisis Group, Communal Violence in Indonesia: Lessons from Kalimantan, Asia Report no. 19, 27 June 2001, pp. 15–16. The Imam of Mesjid Raya, Ngofagita village, in Malifut told me there had been Christian Ambonese in Kao who had been provoking local people. One Makian told a story of how an Ambonese IDP in Wangeotak had shouted ‘Yesus menang!’ (Jesus wins!) following a ‘spike’ during a volleyball game. This required a great amount of effort as most Muslims in Ternate and elsewhere in the region continued to view the Malifut conflict as ethnic in character (see Chapter 4). The Kaos maintained civil relations with Javanese transmigrants even as conflict broke out between Muslims and Christians in the neighbouring sub-district of Tobelo in December 1999. Many Javanese transmigrants fled North Maluku during the conflict, not because of attacks by Christian militia but pressure from Indonesian military personnel, as discussed in Chapter 7. There was no violence between Muslim and Christian Kaos throughout the conflict. The two communities stayed at peace even during the religious conflict in late December between Muslim and Christian Tobelos in Tobelo City, 100 km to the north.

Chapter 4 1 The only location in North Maluku that did not experience violence was a small area around the Weda Bay nickel mine in (then) Central Halmahera District. The reasons for the absence of violence in this area are discussed in the following chapter. 2 C. Tilly, The Politics of Collective Violence, Cambridge: Cambridge University Press, 2003, p. 119. 3 D. Horowitz, The Deadly Ethnic Riot, Berkeley: University of California Press, 2001, pp. 74–5. 4 This propaganda is a form of what the Social Movements literature calls ‘framing processes’. See for example D. McAdam et al., Comparative Perspectives on Social Movements, Cambridge: Cambridge University Press, 1996, p. 8. Social Movement theorists have argued that even dramatic changes in objective conditions will not stimulate collective action without a process of making people aware of the nature of their condition, the source of their grievance and their collective capacity to act against it.

Notes 215 5 S. J. Tambiah, Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia, Berkeley: University of California Press, 1996, p. 81. 6 Ibid. 7 P. R. Brass, Theft of an Idol: Text and Context in the Representation of Collective Violence, Princeton: Princeton University Press, 1997, pp. 96, 177. 8 J. D. Fearon and D. D. Laitin, ‘Violence and the Social Construction of Ethnic Identity’, International Organization, vol. 54, no. 4 (Autumn 2000), p. 874. 9 Horowitz, The Deadly Ethnic Riot, p. 73. 10 Tambiah, Leveling Crowds, p. 124. 11 S. I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India, Cambridge: Cambridge University Press, 2004, p. 4. 12 Ibid., p. 6. 13 Ibid., p. 5. 14 By the time of writing in 2005, the new sultan has assumed some of the traditional authority of the Sultan of Ternate. 15 The permanent capital of North Maluku Province will be Sofifi on Halmahera. Since the division (pemekaran) of North Maluku District in 2001 and district elections in April 2004, Ternate has become the capital of the district of West Halmahera. Following the district elections held in April 2004, North Maluku District was officially divided into North Halmahera (Halmahera Utara) and West Halmahera (Halmahera Barat). 16 Of the other major parties, PPP had seven. 17 In 2004 the city also obtained municipal status. 18 Interview with a member of the Central Halmahera District parliament. 19 Most export commodities are forestry products such as plywood, agricultural products such as copra, clove and nutmeg, fish products (live and frozen) and minerals, particularly nickel and gold. 20 Interviews with both Muslim and Christian respondents. 21 C. G. Kiem, Growing up in Indonesia: Youth and Social Change in a Moluccan Town, Saarbrücken and Fort Lauderdale: Verlag Breitenbach Publishers, 1993, p. 107. 22 Most respondents agreed that Islamic norms had become more strictly enforced following the conflict. 23 Confidential interview with a Christian refugee from Tidore in Tobelo. 24 Abu Bakar Wahid eventually led the local Islamic militia in North Maluku, the Pasukan Jihad discussed in Chapter 7. 25 Confidential interview with a member of the Central Halmahera District parliament. 26 This information is widely acknowledged and was given to me by various respondents, including a member of the North Halmahera District parliament. 27 ‘Demo Berebut Ibukota’ and ‘Para Penghianat Rakyat’, Ternate Pos, 7–13 September 1999. 28 Ternate Pos, 7–13 September 1999. 29 ‘Misteri Angka Sembilan’, Ternate Pos, 7–13 September 1999. 30 GPM also operated several churches in North Maluku along with the local Protestant Church, GMIH. 31 See A. H. Jati, Dan Bundapun Menangis (And Mothers Weep), Manado, 2001, p. 76. 32 Ibid., p. 64. 33 Numerous interviews in Ternate, Tobelo and elsewhere. As made clear in the previous chapter, mosques had not in fact been destroyed, largely because of the involvement of Muslim Kaos in the attack and because Kao leaders wanted to avoid creating the impression that the clashes were caused by religious animosity. 34 This organization collected money, food and clothing for the thousands of Makian refugees. Interview with Wahda Zainal Imam, a candidate for the North Maluku provincial parliament for Partai Bintang Reformasi, in Ternate, 11 February 2004. In 1999 Imam was a PPP member of the Ternate Municipality parliament.

216 Notes 35 Confidential interview with a staff member at Khairun University. 36 Several confidential interviews in Ternate, January 2004. 37 Several confidential interviews, including people who owned houses attacked by these crowds and who were forced to run to the Central Halmahera District police and military compounds and elsewhere. 38 The sources of this information are confidential, but include the Lurah of Mangga Dua, a police officer, a member of Palang Merah Indonesia (the Indonesian Red Cross), members of the district parliament and young men resident in those suburbs. 39 Confidential interview with a police officer, February 2002. 40 Confidential interview with a Tidore man living in Mangga Dua, Ternate, in Ternate, 5 February 2004. The interviewee stated that the young Makian said ‘We have to return to Malifut because they killed us and destroyed our homes, how can you be quiet?’ In line with the mounting claims that the Malifut incident was religious in character, he used the term ‘we’ to refer to Muslims. 41 Confidential interview with a Makian staff member of Khairun University, in Ternate. 42 The fact the shop owners were Chinese appears to have played little role in causing them to be targeted for any reason other than they were Christian. 43 Confidential interview with a member of North Maluku District Police Resort (Polres). 44 Ibid. 45 Andili is of mixed Tidore and Gorontalo heritage. 46 Interview with Wahda Zainal Imam, Ternate, 11 February 2004. 47 Interview with Ibrahim Fabanyo, Ternate, 11 February 2004. 48 Interview with Pastor Aesh, Tobelo. 49 Respondents who stated that these people produced the ‘Bloody Sosol’ letter included a police officer from the North Maluku District Police Resort (Polres) who was involved in the investigation into the riot, confidential interview, February 2004. 50 Confidential interview with a member of the public service in Ternate, in a refugee camp in Manado, February 2004. 51 Confidential interview with a member of the Central Halmahera Parliament (DPRD) in 1999. I have not provided the names of the men the respondent gave as they were not corroborated by any other source and the information could not be verified for publication. 52 This speech, given in the closed meeting in the district parliament, was recounted to me by two members of that parliament. 53 Confidential interview with a member of the North Maluku District Parliament, in Ternate, 4 January 2004. This was also told to me by other respondents in Ternate who had not been present in the parliament. 54 Interview with senior parliamentarian, Ternate, 11 February 2004. 55 Indonesiana is so called because it was very diverse, home to a large minority of Christians and numerous ethnic groups including Makians, Chinese, Papuans, Makassarese and Bugis. One Makian resident of Indonesiana told me that the meeting was announced as between Muslim and Christian community leaders, which he felt was ‘stupid and strange’. 56 Confidential interview with a member of the Tidore ethnic group and member of the Central Halmahera District parliament. 57 Report produced by the North Maluku Governor’s office and interviews with local government officials in Tidore. 58 Interviews with several people in Indonesiana, Soasio, Tidore, including a local government official, 9 February 2004. 59 Witnesses who stated that there were large numbers of Makians involved in the riot included one Makian who lived in Indonesiana, Soasio. Confidential source.

Notes 217 60 As well as being told to me by Christian residents of Ternate, this claim is also made in Jati, Dan Bundapun Menangis, p. 94. 61 Interview with a member of the North Maluku District parliament. 62 See for example Jati’s interview with a members of the security forces: Jati, Dan Bundapun Menangis, p. 87. 63 Confidential interview with a Christian resident of Ternate who was sheltering in the church at the time, now living in Kao. 64 North Maluku Provincial Government, ‘Kronologis Kerusuhan Bernuansa Sara di Propinsi Maluku Utara’ (A Chronology of the Sectarian Riots in North Maluku Province), April 2000. The rioters did far less damage to the Catholic church than they had to the GPM and GMIH Protestant churches. 65 I did not undertake fieldwork in this sub-district, and therefore rely upon published and unpublished analyses, as well as accounts given to me by people in other sub-districts. 66 According to Nanere, Muslims attacked Christians attending prayers at church in the village of Lola, killing all of the approximately 100 people present. The crowd involved in the attack burned the church and Christian homes: J. Nanere, Halmahera Berdarah (Bloody Halmahera), Ambon: Yayasan Bina Masyarakat Sejahtera dan Pelestarian Alam (BIMASPELA), 2000, p. 92. 67 N. Bubandt, ‘The Dynamics of Reasonable Paranoia: Rumours and Riots in North Maluku, 1999–2000’, background paper for presentation at the Seminar Series of the School of Anthropology, Geography and Environmental Studies, University of Melbourne, 25 April 2002, p. 24. 68 Horowitz, The Deadly Ethnic Riot, p. 75. See also D. Horowitz, ‘Group Loyalty and Ethnic Violence’, in Charles Hermann et al. (eds), Violent Conflict in the Twenty-first Century: Causes, Instruments and Mitigation, Illinois: American Academy of Arts & Sciences, 1999, p. 105. 69 Ten per cent of the Ternate population is Christian and perhaps 2 per cent of the Tidore population. The outbreak of anti-Christian violence therefore undermines arguments made by some analysts that a security dilemma is necessary for conflict to occur. Kaufman for example argues that a mutual fear of extinction must be present for conflict to occur: S. J. Kaufman, ‘Spiraling to Ethnic War: Elites, Masses and Moscow in Moldova’s Civil War’, International Security, vol. 21, no. 2 (Fall 1996), p. 109. 70 For these reasons, a coalition was formed between the sultan’s rivals for the governorship, Bahar and Syamsir Andili and Thaib Armain. Until late October, in addition to being rival candidates for the governorship, these individuals had taken different positions on several issues, including the location of the provincial capital. 71 While in November 1999 the new North Maluku provincial parliament had not yet been formed, it was likely to be formed on the basis of the results of the mid-year district elections. Therefore, as in the three district parliaments, the provincial parliament would be dominated by Mudaffar Syah’s own party, Golkar, and the other major secular, nationalist party, PDI-P. Once this parliament had been formed, its members would then elect North Maluku’s first governor.

Chapter 5 1 Interview with a Muslim community leader in Tobelo Sub-District in 2003. 2 The realist theory in international relations argues that in an international system of anarchy, where there is no overarching power, each state will naturally strive for its own security, thereby threatening other states in the system. 3 B. R. Posen, ‘The Security Dilemma and Ethnic Conflict’, Survival, vol. 35, no. 1, pp. 29–31.

218 Notes 4 Ibid., p. 28. See also D. A. Lake and D. Rothchild, ‘Spreading Fear: The Genesis of Transnational Ethnic Conflict’, in D. A. Lake and D. Rothchild (eds), The International Spread of Ethnic Conflict: Fear, Diffusion and Escalation, Princeton: Princeton University Press, 1998. 5 J. D. Fearon, ‘Ethnic War as a Commitment Problem’, paper presented at the 1994 Annual Meeting of the American Political Science Association, New York, 30 August–2 September 1995, and D. A. Lake and D. Rothchild, ‘Containing Fear: The Origins and Management of Ethnic Conflict’, International Security, vol. 21, no. 2 (Fall 1996), 41–75. 6 Lake and Rothchild, ‘Containing Fear’, p. 46. 7 Ibid., p. 47. 8 Posen, ‘The Security Dilemma and Ethnic Conflict’, p. 32. 9 O. Waever et al., Identity, Migration and the New Security Agenda in Europe, London: Pinter Publishers, 1993, p. 23. 10 For an application of this concept to ethnic conflict, see P. Roe, ‘Misperception and Ethnic Conflict: Transylvania’s Societal Security Dilemma’, Review of International Studies, vol. 28 (2002), 57–74. 11 ‘Momentous’ events (such as religious riots elsewhere) can eradicate an individual’s normal indifference towards their religious identity: S. Kakar, ‘The Time of Kali: Violence between Religious Groups in India’, Social Research, vol. 67, no. 3 (Fall 2000). 12 Horowitz points to numerous riots that were preceded by rumours of impending aggression by the group that actually becomes the eventual target: D. Horowitz, The Deadly Ethnic Riot, Berkeley: University of California Press, 2001, p. 79. 13 B. R. Weingast, ‘Constructing Trust: The Political and Economic Roots of Ethnic and Regional Conflict’, in K. Soltan et al. (eds), Institutions and Social Order, Ann Arbor: University of Michigan Press, 1998, p. 165. See also Horowitz, The Deadly Ethnic Riot, p. 319. 14 Ibid., pp. 74, 529. 15 R. S. Appleby, The Ambivalence of the Sacred: Religion, Violence, and Reconciliation, New York: Rowman & Littlefield, 2000, p. 69. 16 M. Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence, Berkeley: University of California Press, 2000, p. 146. 17 A. Hasenclever and V. Rittberger, ‘Does Religion Make a Difference? Theoretical Approaches to the Impact of Faith on Political Conflict’, Millennium, vol. 29, no. 3 (2000), p. 656. 18 N. Z. Davis, ‘The Rites of Violence: Religious Riot in Sixteenth-Century France’, Past and Present, vol. 59 (May 1973), pp. 81–2. 19 Ibid., p. 83. 20 Ibid., p. 85. 21 Posen, ‘The Security Dilemma and Ethnic Conflict’, p. 33, and Lake and Rothchild, ‘Containing Fear’, p. 55. 22 That such situations do not always lead to violence is pointed out in J. D. Fearon and D. D. Laitin, ‘Explaining Interethnic Cooperation’, American Political Science Review, vol. 90, no. 4 (December 1996), p. 715. 23 ‘North Maluku in Figures, 1999’, Central Statistics Agency (Badan Pusat Statistik), North Maluku District, Ternate. 24 Van Fraassen considers the Kao to actually be of the ‘South Tobelo’ ethnic group: C. F. van Fraassen, ‘Types of Socio-Political Structure in North-Halmaheran History’, in E. K. M. Masinambow (ed.), Halmahera dan Raja Ampat: Konsep dan Strategi Penelitian (Halmahera and Raja Ampat: Research Strategies and Concepts), Jakarta: LIPI, 1980, p. 122. 25 N. Ishige, ‘Limau Village and its Setting’, in N. Ishige (ed.), The Galela of Halmahera: A Preliminary Survey, Osaka: National Museum of Ethnology, 1980, p. 8.

Notes 219 26 Togoliua is officially a sub-village (dusun) of the village of Tobe. 27 On Pela Gandong, see D. Bartels, ‘Guarding the Invisible Mountain: Intervillage Alliances, Religious Syncretism and Ethnic Identity among Ambonese Christians and Moslems in the Moluccas’, unpublished PhD thesis, Cornell University, 1977. 28 Interview with Ar Sidiumar, Tobelo Muslim, Tobelo City, 2003. 29 This was also the case for Christian migrants from Ambon, East Nusa Tengarra and other areas of Indonesia. Migration similarly affected Pela Gandong in Ambon: D. Bartels, ‘Your God is No Longer Mine: Moslem–Christian Fratricide in the Central Moluccas (Indonesia) after a Half-Millennium of Tolerant Co-Existence and Ethnic Unity’, unpublished article, 2000. 30 Such graffiti were very common during the conflict in 1999–2000. During my fieldwork in 2003, new symbols had been added, such as ‘Taliban’ and ‘CIA’. 31 There are somewhat different versions of this story. Christians state the dispute arose because of a linguistic misunderstanding since Bouwens’s skills in the Indonesian language were not of a high standard. Muslims, however, and several Christians, stated that Bouwens told the Muslim men that ‘Islam is dirty’. Several Christian respondents also stated that during the conflict in 2000, Bouwens told them that there was no need to be concerned because an American warship was not far from the coast of the island of Morotai, ready to protect them against the ‘jihadist’ militia (the Pasukan Jihad). Pastor Bouwens declined my request for an interview, citing ill health. 32 Interview with several Christian men in the neighbourhood of Gura, Tobelo, 2003. 33 Interview with the former Sub-District Head of Tobelo, Agil Bachmid, in Ternate, 7 February 2004. 34 Several confidential sources from the Christian community in Tobelo. 35 By the time of the outbreak of violence, some Muslims in Tobelo as well as in Ternate and elsewhere believed RMS was present in North Maluku, particularly in Tobelo. 36 Interview with a young Muslim man from Talaga who was involved in the clash, 6 January 2004. Several Christians in Tobelo also mentioned the Ibu incident. 37 Confidential interview with a Makian community leader who was asked to assist with tents for the militia, interviewed in Ternate, January 2004. This informant refused to assist the militia and was threatened. 38 Possible connections between this militia and the large Java-based militia of the same name are discussed in Chapter 7. 39 I did not do fieldwork in these areas so the above information is based upon J. Nanere, Halmahera Berdarah (Bloody Halmahera), Ambon: Yayasan Bina Masyarakat Sejahtera dan Pelestarian Alam (BIMASPELA), 2000, pp. 91–3. 40 It is possible the government relied upon GMIH to evacuate the IDPs, but it seems more likely that the government was unwilling to offer support to Christians given the atmosphere at the time. 41 Interviews with several Christian respondents in Tobelo, 2003–4. 42 Interviews with several respondents in Tobelo, 2003, including Agil Bachmid, members of FKPKHU and Muslim and Christian community leaders. 43 ‘Orang Islam adalah saudara kami, tapi etnis Makian dan Tidore adalah oknum perusuh yang harus angkat kaki dari Kecamatan Tobelo.’ This information is given in the Muslim account of the conflict: K. H. Ahmad and H. Oesman, Damai Yang Terkoyak: Catatan Kelam dari Bumi Halmahera (Shattered Peace: Dark Notes from the Land of Halmahera), Ternate: Madani Press, 2000, p. 59, n. 4. Christian respondents also agreed this message was conveyed to the political delegation. 44 Several members of the Muslim community gave me this information. 45 Interview with Pastor Charles Kaya in Tobelo. 46 Interviews in Gosoma, 2003. 47 This area stretched from the port and market area around Kampung Cina, west to Gosoma, north to Gura and south to Kaliseratus.

220 Notes 48 A senior politician and a pastor told me in confidential interviews that they notified militia leaders of rioting in the city and requested they send Christian reinforcements, Tobelo 2003. 49 Interview with Benny Bitjara and other leaders of the militia, as well as several Muslim Kaos. 50 Interviews in Tobelo, Kao and Togoliua. Several men in Togoliua stated they laid the coconut trees across the road. 51 Interviews with three Christian and three Muslim community leaders in Gamhoku, 29 January 2004. Relations between the Christian and Muslim communities in this village had been deteriorating after the Tidore and Ternate violence and the spread of the ‘Bloody Christmas’ rumour. Christian IDPs from Weda and Payahe were housed in the village but no violent incidents had occurred during November or December. 52 Interview with Sakeus Odara, Tobelo, 3 September 2003. However, no Javanese were killed in the violence. 53 Interview with Benny Bitjara in Kupa Kupa village, Tobelo Sub-District, 18 September 2003. Thanks to Father Tom Lawn for the excerpt from the psalm. 54 Interview with Benny Bitjara in Kupa Kupa village, Tobelo Sub-District, 18 September 2003. 55 Interview with a Muslim resident of Kampung Baru, 2003. 56 The claim that the soldiers did not have firearms is doubtful. Contradicting this claim, Christians recounted how a Muslim Koramil officer had shot a Christian man guarding a church in Gosoma. 57 Several Christian shop owners said that Christian militia had looted and destroyed their shops. 58 Many shop owners have not yet managed to reconstruct their businesses. 59 The scarred asphalt remains as a grisly reminder of the violence. 60 Several Tobelo men told me this was common in ‘black magic’, using the English-language term for traditional animist spiritual beliefs. The eating of hearts was said to make one’s body feel hot (panas), and increase bravery and strength. 61 Interview with the village head of Gorua, 26 September 2003. 62 Interview with a Christian militia leader in Wari Village, Tobelo Sub-District, 17 October 2003. 63 The Kao Christian leader Hersen Tinangon told me the Kao were late returning because they delivered the body of a Kao man to his girlfriend. 64 Interviews in Togoliua with the head of the village (Ketua Dusun), and two community leaders, Hasan Andi Makulau and Irwan Sangaji, 23 September 2004. 65 On reports that those killed were Javanese transmigrants, see ‘Kepolisian Benarkan 216 Transmigrasi Dibantai di Mesjid Maluku Utara’ (Police Confirm 216 Transmigrants Butchered in North Maluku Mosque), Republika Online Edition, 17 January 2000; available online at: http://media.isnet.org/ambon/Republika05.html. Accessed May 2002. 66 Interview with the pastor from the Trans Suka Maju transmigration settlement, in Gamhoku village, Tobelo Sub-District. 67 Residents of Togoliua state that several dozen children were taken hostage during the attack and are still being kept against their will by Christians in villages such as Kusuri, Kupa Kupa, Efi Efi and Upa. According to residents of Togoliua these children are currently forced to live as Christians, made to eat pork and attend Christian services. I was not able to obtain verification of this claim nor evidence to disprove it. 68 ‘North Maluku in Figures, 1999’, Central Statistics Agency (Badan Pusat Statistik), North Maluku District, Ternate. In 1999 the sub-district had a slightly larger male (18,634) than female population (16,611). 69 Soasio (2,464), Dokulamo (2,166), Soatobaru (1,638), Togawa (1,558), Soakonora (1,498), Mamuya (1,498), Duma (1,147). These figures are taken from a survey of

Notes 221

70

71 72

73 74 75 76 77 78 79 80 81 82 83

84 85 86 87 88 89 90 91 92 93 94

communities in the area of a volcano near Mamuya and should be taken as estimates only. There are no official statistics for the size of the religious communities in Galela. However, one source states that the Muslim population was 20,062 and the Christian population 11,131. See F. Adeney-Risakotta, ‘Mobilising Conflict through Media in the North Moluccas’, seminar paper presented at Indonesian Conflict Studies Network, Bandung, Indonesia, 7–12 April 2003, p. 16. Confidential interview with Christian community leaders, in Soatobaru, 27 October 2003. Numerous interviews in Galela and Tobelo, 2003 and 2004. Aditjondro mentions the Sinar Mas interest in GAI and the banana plantation in Galela in his discussion of the violence in Maluku: G. J. Aditjondro, ‘The Political Economy of Violence in Maluku’, Green Left Weekly, vol. 397 (2003). However, as I have stated, I found no evidence to suggest external actors involved in this venture had reason to provoke unrest around the plantation, and all local respondents stated there was no connection between the business and the conflict. GMIH officials denied this. Interview with Samsul Bakhri, Islamic community leader, in Soasio, 20 October 2003. Interview with Syamin Basyir, Soakonora Muslim community leader, in Soakonora, 30 January 2005. Confidential interview with two Christian community leaders in Soatobaru, October 2003. The village head of Dokulamo, Abu Bakar Dabidabi, told me he ordered villagers in Dokulamo not to make weapons, interview in Dokulamo, 24 October 2003. Interview with the head of GMIH for Galela, Pastor Kaleb Kakale, in Tobelo, 30 January 2004. Interview with several Christian men, Mamuya, 2003. Interview with a Muslim community leader in Mamuya. The sub-district military compound (Koramil) is located outside Soasio, perhaps 10 km from the site of the first attack. The fact the Christian community was able to flee with few casualties suggests the imam persuaded the crowd to allow them to escape. Alternatively, the size of the attacking crowd was at this stage still small. Several Muslim community leaders travelled from Gotalamo and Dokulamo to the capital, Soasio, telling Christians manning roadblocks in Duma and Soatobaru that they intended to inform the government that there were individuals attempting to provoke each community in their villages. Interview with Josafat Etha, Christian community leader, Duma, 16 October 2003. Ibid. Interview with the village head of Ngidiho, Haji Basiron Ayub, in Ngidiho, 31 January 2004. Adeney-Risakotta, ‘Mobilising Conflict through Media in the North Moluccas’. Interview with former Igobula village head, Haji Umar Baendi, in Igobula, 30 October 2003. Interview with Yamin Galela, village head of Igobula, in Igobula, 30 October 2003. Almost all Christian respondents interviewed in Tobelo believe this to have been the case. Interview with Robert Tunggal, in Tobelo, 22 January 2004. Confidential interviews in Tobelo, 2003–4. In some cases, community and religious leaders attempted conflict prevention initiatives so as not to be seen as provocateurs. J. D. Fearon, ‘Rationalist Explanations for War’, International Organization, vol. 49,

222 Notes

95 96 97 98 99 100

101 102 103 104 105 106 107

no. 3 (Summer 1995), p. 381. See also Fearon, ‘Ethnic War as a Commitment Problem’. Interview in Pediwang, Kao Sub-District, 5 December 2003. Interviews in Tobelo, 2003 and 2004. Interview with the head of GMIH for Galela, Pastor Kaleb Kakale, in Tobelo, 30 January 2004. This information is based on a telephone interview with Malcolm Baillie of PT Weda Bay Nickel. Confidential interview Tobelo, 2003. It is not beyond the realms of possibility, however. During my fieldwork in late 2003, a series of bombs exploded in Tobelo, almost reigniting conflict between the two communities. After suspicions arose that the Brimob unit from Makassar stationed in the city was responsible, popular protests drove the district head to have the unit replaced by another, from Manado. Following this rotation, the bombing stopped. Possible reasons for instigation of conflict by security personnel include increased payment for operations during a civil emergency. However, it is worth pointing out that most Christian leaders believe the Muslim military commander was ‘disturbed’ because he had believed Muslims would easily control the city. Interviews with Muslim men from Gorua and Popilo and Christian men involved in the attacks on these villages. The term Forest Tobelo is used by Christopher Duncan: C. R. Duncan, ‘Savage Imagery: (Mis)representations of the Forest Tobelo of Indonesia’, The Asia Pacific Journal of Anthropology, vol. 2, no. 1 (2001), 45–62. Ibid. Two senior pastors who attempted to prevent violence were the chairman and vice-chairman of GMIH, Pastors Aesh and S. S. Duan. One pastor who provided such a service before clashes was Pastor Yacob Soselisa. Christian respondents recounted several stories about Bitjara which demonstrated his magical powers. As the Pasukan Merah attacked the south of Tobelo City on 28 December a bomb landed in front of Bitjara but failed to explode. Bitjara strode to the bomb and kicked it back towards the Muslim guard post, where it exploded and killed several men. The story of Bitjara catching a bullet with only a slight injury to his hand was also widely known and believed among Christians in north Halmahera.

Chapter 6 1 V. P. Gagnon, ‘Ethnic Nationalism and International Conflict: The Case of Serbia’, International Security, vol. 19, no. 3 (Winter 1994/95), p. 132. For example, in Yugoslavia, following threats to their positions and power after the fall of the Soviet Union, a wide coalition of ‘conservatives in the Serbian party leadership, local and regional party elites … orthodox Marxist intellectuals … and parts of the nationalist army provoked a conflict along ethnic lines’, ibid., p. 142. 2 J. Snyder, From Voting to Violence: Democratization and Nationalist Conflict, New York: W. W. Norton & Company, 2000, p .39. 3 For example, in his discussion of the transition from Soviet authoritarian rule to democracy in the Caucasus, Snyder argues that the prevalence of kin-based patronage networks were irreconcilable with democratic processes: ibid., p. 204. 4 P. R. Brass, The Production of Hindu–Muslim Violence in Contemporary India, New Delhi: Oxford University Press, 2003, p. 231. 5 Brass also used this term in Theft of an Idol: Text and Context in the Representation of Collective Violence, Princeton: Princeton University Press, 1997, p. 208. 6 P. R. Brass, ‘Introduction: Discourses of Ethnicity, Communalism, and Violence’, in

Notes 223

7 8 9 10 11 12 13 14 15 16 17

18 19 20

21 22

23 24 25 26 27 28 29 30 31 32 33

P. R. Brass (ed.), Riots and Pogroms, New York: New York University Press, 1996, p. 13. Brass, The Production of Hindu–Muslim Violence, pp. 117, 242. Brass, ‘Introduction: Discourses of Ethnicity, Communalism, and Violence’, p. 15. S. I. Wilkinson, Votes and Violence: Electoral Competition and Ethnic Riots in India, Cambridge: Cambridge University Press, 2004, p. 4. Ibid., p. 5. D. Keen, The Economic Functions of Violence in Civil Wars, Adelphi Paper 320, London: Oxford University Press, 1998, p. 12. Brass, The Production of Hindu–Muslim Violence, p. 362. Interviews in Ternate, 2003–4. This information is widely acknowledged and was given to me by various respondents including a member of the North Halmahera District parliament. Sofifi eventually became the permanent provincial capital. Numerous interviews in Ternate and elsewhere in North Maluku, 2003–4. ‘Demo Berebut Ibukota’ and ‘Para Penghianat Rakyat’, Ternate Pos, 7–13 September 1999. The demonstrators held banners claiming ‘Bahar and Syamsir Andili are provocateurs’ (‘Bahar – Syamsir Andili provokator’). The full title of the law is Undang Undang Nomor 46/1999 Tentang Pembentukan Propinsi Maluku Utara, Kabupaten Buru, Dan Kabupaten Maluku Tenggara Barat (Law 46/1999 on the Formation of North Maluku Province, Buru District and West Southeast Maluku District). As at the time of writing, construction of government facilities is still taking place in the village of Sofifi. Confidential interviews in Ternate, 2003–4. The figure of 4,000 is cited in S. Alhadar, ‘Anatomi Kerusuhan Sosial di Maluku Utara’ (The Anatomy of the Social Riots in North Maluku), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003, p. 121. Other pieces of graffiti stated ‘Jesus is an animal’ and ‘Jesus is a drunk’. Several residents of Ternate claimed the Pasukan Kuning included many Christians from Halmahera. Respondents included Haji Kotu, a prominent member of the militia that would eventually oppose the sultan, the Pasukan Putih, interviewed in Ternate, 3 January 2004. Some respondents stated they were outraged at this behaviour as the members of Pasukan Kuning often could not read their identity cards Confidential interview with a member of the Pasukan Kuning. Interviews with several Makians in Ternate, 2003–4. T. A. Tomagola, ‘Krisis dan Solusi Tragedi Maluku Utara’, Detikcom, 2 February 2000. This was stated to me both by members of the Pasukan Putih as well as more traditional supporters of the sultan. Confidential interview in Ternate, 2003. Interview with Abu Bakar Wahid in Tomalou, Tidore, 7 January 2004. Abu Bakar Wahid is of mixed Makian/Tidore descent, and, according to several respondents, prior to the conflict in North Maluku had ties to Thaib Armain. Interview with Abu Bakar Wahid in Tomalou, Tidore, 7 January 2004. As will be seen in the next chapter, after leading the Pasukan Putih during the Putih–Kuning conflict, Selang would also become the leader of the Laskar Mujahidin, the Ternate branch of the regional Islamic militia, the Pasukan Jihad. Following the violence in Tobelo and Galela the Pasukan Putih in Ternate and Tidore became commonly known as the Pasukan Jihad. Interviews with several residents of Kampung Pisang and Maliaro in 2004.

224 Notes 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63

Interview with a young man in Tomalou, Tidore. Interview with man from Dufa Dufa, Sucipto, in Ternate, 15 January 2004. Interview in Ternate, 3 January 2003. Interview with the Sultan of Tidore, Djafar Junus Sjah, in Soasio, Tidore, 9 January 2004. Confidential interviews in Soasio, Tidore, 2004. Interview with the Sultan of Tidore, in Soasio, Tidore, 9 January 2004. Confidential interview with a member of the sultanate government, in Soasio, Tidore, 2004. Interview with a senior member of the Tidore traditional guards, in Soasio, Tidore, 2004. After the Pasukan Kuning had dispersed, several houses owned by Christians were destroyed in northern Ternate City. Interview with the Sultan of Ternate, in Ternate, 17 February 2004. The higher estimate is given in Yurnaldi, ‘Dendam yang tak Berkesudahan’ (Revenge without End) in Hasan, Memikirkan Kembali, and ‘Dendam yang tak Berkesudahan’, Kompas, 2 January 2000. Interview with the Sultan of Tidore. See for example Kompas, 22 January 2000. Bahar Andili would eventually die of natural causes during his campaign to become governor. S. Alhadar, ‘Tragedi Maluku Utara’, Republika, 18 January 2000. S. Alhadar, ‘The Forgotten War in North Maluku’, Inside Indonesia, no. 63 (July–September 2000). K. H. Ahmad and H. Oesman, Damai Yang Terkoyak: Catatan Kelam dari Bumi Halmahera (Shattered Peace: Dark Notes from the Land of Halmahera), Ternate: Madani Press, 2000, p. xxi. Interview with Ibrahim Fabanyo, deputy chairman of North Maluku District parliament, in Ternate, 11 February 2004. Interview with Mudaffar Syah, Sultan of Ternate, and several other respondents in Ternate, 2003–4. Confidential interview with a member of the Pasukan Kuning in Dufa Dufa, Ternate City. Several interviews in North Maluku 2003–4. Interview respondents included members of the Tidore ethnic group residing in central and south Ternate, and on Tidore. This was stated to me by members of the Pasukan Putih as well as more traditional supporters of the sultan. Snyder, From Voting to Violence, p. 39. Interview with Muhammad Selang, in Ternate, 7 February 2004. Interview with Haji Kotu, in Ternate, 3 January 2004. Interview with a young Muslim man who joined the Pasukan Putih, in Tomalou, Tidore, 15 February 2004. Alhadar, ‘The Forgotten War in North Maluku’. Brass, ‘Introduction: Discourses of Ethnicity, Communalism, and Violence’, p. 13. The police assumed primary responsibility for internal security and law and order, including responding to both communal conflict and violent insurgency. Thaib Armain would eventually be elected North Maluku’s first governor after Bahar Andili’s untimely death from natural causes during campaigning. The North Maluku provincial parliament’s first choice for governor, a former Suharto minister, Abdul Gafur, was overturned following accusations of corruption.

Notes 225

Chapter 7 1 F. Polletta and J. M. Jasper, ‘Collective Identity and Social Movements’, Annual Review of Sociology, vol. 27 (2001), p. 284. 2 J. R. Seul, ‘Ours Is the Way of God: Religion, Identity, and Intergroup Conflict’, Journal of Peace Research, vol. 36, no. 5 (1999), p. 558. 3 Ibid., p. 559. 4 Other studies have discussed the violence inherent in sacred texts and I will not attempt to do so here; see, for example, B. Moore Jr, Moral Purity and Persecution in History, Princeton: Princeton University Press, 2000. 5 F. M. Donner, ‘The Sources of Islamic Conceptions of War’, in J. Kelsay and J. T. Johnson (eds), Just War and Jihad: Historical and Theoretical Perspectives on War and Peace in Western and Islamic Traditions, Westport: Greenwood Press, 1991, p. 47. 6 Taken from B. Tibi, ‘War and Peace in Islam’, in S. H. Hashmi (ed.), Islamic Political Ethics: Civil Society, Pluralism, and Conflict, Princeton: Princeton University Press, 2002, p. 179. 7 Ibid. 8 Donner, ‘The Sources of Islamic Conceptions of War’, p. 47. 9 P. L. Heck, ‘Jihad Revisited’, Journal of Religious Ethics, vol. 32, no. 1, p. 97. 10 Ibid., pp. 97–8. 11 J. Fox, ‘Do Religious Institutions Support Violence or the Status Quo?’, Studies in Conflict and Terrorism, vol. 22 (1999), p. 120. 12 R. S. Appleby, The Ambivalence of the Sacred: Religion, Violence, and Reconciliation, New York: Rowman & Littlefield, 2000, p. 69. 13 See, for example, A. Hasenclever and V. Rittberger, ‘Does Religion Make a Difference? Theoretical Approaches to the Impact of Faith on Political Conflict’, Millennium, vol. 29, no. 3 (2000), 641–74. 14 S. Bruce, Politics and Religion, Cambridge: Polity Press, 2003, p. 84; S. Bruce, ‘Religion and Violence: What Can Sociology Offer?’, Numen, vol. 52 (2005), p. 16. 15 Mujahid, meaning a participant in holy war, is the term used by most North Maluku Muslims to refer to participants in the jihad. 16 Interview in Malifut in 2003. 17 Several interviews in Ternate, 2003–4, and interview with Abu Bakar Wahid in Tomalou, Tidore, 7 January 2004. 18 Ternate Pos, 25 April–1 May 2000. 19 Several interviews in Ternate, 2003–4. 20 Interview in Tomalou, Tidore, 7 January 2004. 21 Ibid. 22 For example interview with North Maluku historian Herry Nachrawy, in Ternate, 14 January 2004. 23 Christian militia members in Tobelo told me they were angry to see Muslims carrying red-and-white Indonesian flags in clashes with fellow Indonesians. 24 See, for example, C. G. Kiem, Growing up in Indonesia: Youth and Social Change in a Moluccan Town, Saarbrücken and Fort Lauderdale: Verlag Breitenbach Publishers, 1993, p. 54. 25 Interviews with all those mentioned, in Tobelo 2003–4. No separatist sentiment was conveyed to me during fieldwork in 2003 and 2004. 26 Interviews in Tobelo Sub-District, 2003. 27 Interview with Muhammad Albar, in Ternate, 10 February 2004. 28 For example interview with one young Muslim man from the Tidore ethnic group, in Ternate, 5 February 2004. 29 Interview in Tomalou, Tidore, 7 January 2004. 30 Several interviews in Ternate in 2003–4. It is impossible to verify this claim but this tension between the former leadership of the militia was still manifest in early 2004.

226 Notes 31 Numerous interviews in Tobelo, Galela and Kao 2003–4. 32 Interviews with Christians in Duma, Tobelo and elsewhere in North Maluku, 2003–4. See J. Nanere, Halmahera Berdarah (Bloody Halmahera), Ambon: Yayasan Bina Masyarakat Sejahtera dan Pelestarian Alam (BIMASPELA), 2000, p. 156 for a list of the identification cards allegedly found on mujahid. 33 Interview in Tobelo, 11 January 2003. Several Christians also claim that one Afghan was identified among the dead, although this seems unlikely. 34 Interview with Christopher Selong, Duma, 17 October 2003. 35 Nanere, Halmahera Berdarah, p. 154. 36 T. A. Tomagola, ‘The Bleeding Halmahera of North Moluccas’, Jurnal Studi Indonesia, vol. 10, no. 2 (2000), p. 7. 37 N. Hasan, ‘Faith and Politics: The Rise of the Laskar Jihad in the Era of Transition in Indonesia’, Indonesia, vol. 73 (April 2002), p. 159. 38 Salafism, a very puritan stream of the Saudi Wahhabist doctrine, demands a literal reading of the Qu’ran and application of sharia. Indonesian salafi leaders often seek guidance from sheikhs in Saudi Arabia and Yemen. 39 For a good discussion of FKAWJ, see Hasan, ‘Faith and Politics’. 40 Ibid., p. 159. 41 Ibid., p. 165. 42 For a good discussion of the operations of the Laskar Jihad, see International Crisis Group, Indonesia: Overcoming Murder and Chaos in Maluku, Asia Report no. 10, 19 December 2000. 43 While the attitude of military personnel to Laskar Jihad varied from unit to unit, in general the militia was not initially opposed by the military, although clashes between the two did occur later in 2001. 44 International Crisis Group, Indonesia: The Search for Peace in Maluku, Asia Report no. 31, 8 February 2002, p. 14. Military support for Laskar Jihad seems to have begun with the training of militia members by military personnel at Bogor near Jakarta. Reports claimed that when members of Laskar Jihad arrived in Ambon, military personnel provided them with weapons: G. Fealy, ‘Inside the Laskar Jihad – An Interview with the Leader of a New, Radical and Militant Sect’, Inside Indonesia, 31 March 2001, p. 29 45 See the various June 2000 reports of the Crisis Centre, Diocese of Amboina. 46 R. W. Hefner, ‘Civic Pluralism Denied? The New Media and Jihadi Violence in Indonesia’, in D. F. Eickelman and J. W. Anderson (eds), New Media in the Muslim World: The Emerging Public Sphere, Bloomington: Indiana University Press, 1999, p. 3. Hefner writes that Western intelligence reports suggest army officers transferred US$9.3 million to Laskar Jihad, ibid., p. 10. 47 Maluku Hari Ini, 18 July 2000. 48 Jakarta Post, 7 April 2000. 49 Interviews with Abu Bakar Wahid, Muhammad Selang, Muhammad Albar in Tidore and Ternate, and with Samsul Bakhri in Soasio, Galela, 20 October 2003. 50 For example interview with Ahmad Pilo, a sub-district Muslim community leader in Ngidiho, Galela, 21 October 2003. 51 This information is consistent with interviews conducted by Professor Harold Crouch with members of Laskar Jihad: personal communication. 52 Interview with Abu Bakar Wahid, in Tomalou, Tidore, 7 January 2004. 53 Interview with Muhammad Albar in Ternate, 10 February 2004. 54 Ibid. 55 Interview with Muhammad Selang in Ternate, 7 February 2004. 56 Ibid. 57 The group’s leader was another Islamic preacher of Yemeni descent, Habib Rizieq Shihab, a sayyid or descendant of the prophet Muhammad. 58 FPI appeared to have contacts with military and police officers, including senior

Notes 227

59

60 61 62 63 64 65 66 67 68

69 70 71 72 73 74 75 76

77

78

personnel such as former General Wiranto. It is unlikely that senior officers shared the goals of FPI, but may have utilized the organization as an expedient way of mobilizing large numbers of people at certain times. ‘500,000 Jihad Legion Supporters Gather in North Maluku Province’, The Indonesian Observer, 24 April 2000. As North Maluku has a population of 800,000, it is clear 500,000 was a substantial overestimate. See also ‘Islamic Defenders Reject Reconciliation over Maluku Conflict’, Antara, 26 April 2000. Interview with Muhammad Selang in Ternate, 7 February 2004. Respondents include Makian and Tidore men in Ternate and Dan Murphy, a journalist with the Christian Science Monitor: personal communication in Ternate, February 2004. Confidential interview in Ternate, 12 February 2004. Interview with Abu Bakar Wahid, in Tomalou, Tidore, 7 January 2004. A Bubalo and G. Fealy, Joining the Caravan? The Middle East, Islamism and Indonesia, Lowy Institute Paper 5, New South Wales: Lowy Institute for International Policy, 2005, p. 79. Ibid., p. 81. International Crisis Group, Al-Qaeda in Southeast Asia: The Case of the ‘Ngruki Network’ in Indonesia, Asia Briefing no. 20, 8 August 2002, p. 19. International Crisis Group, Indonesia Backgrounder: How the Jemaah Islamiyah Terrorist Network Operates, Asia Report no. 43, 11 December 2002, pp. 3–4. Subsequently, members of Jemaah Islamiyah have been responsible for the worst terrorist attacks carried out in Indonesia. On 12 October 2002 a Jemaah Islamiyah cell detonated two large bombs targeted at two nightclubs in Kuta, Bali, attempting to kill Westerners. The bombings killed 202 people, both Westerners and Indonesians, and injured a further 300. In August 2003 the organization exploded a bomb in front of the American-owned J. W. Marriot hotel killing 12 people, and in September 2004 detonated a car bomb in front of the Australian Embassy killing nine. International Crisis Group, How the Jemaah Islamiyah Terrorist Network Operates, p. 22. Abu Jibril was deported from Malaysia to Indonesia and has not been charged with being a member of JI in Indonesia. This recording is discussed in D. Murphy, ‘Al Qaeda’s New Frontier: Indonesia’, Christian Science Monitor, 1 May 2002. Interview with Abu Bakar Wahid, in Tomalou, Tidore, 7 January 2004. When other members of Jemaah Islamiyah arrived in North Maluku after the conflict, they appear to have been disappointed that violence had ended. See International Crisis Group, How the Jemaah Islamiyah Terrorist Network Operates, p. 24. In 2001 in Ambon, Laskar Jihad members, under the leadership of Ja’far, executed a militia member who had been sentenced to death under sharia. Interview with Samsul Bakhri in Soasio, Galela, 20 October 2003. Interviews with Muhammad Selang and Muhammad Albar in Ternate, 7 February 2004 and 10 February 2004. Some analysts have also reported that some Ambonese Muslims, while at first welcoming the protection afforded by Laskar Jihad, soon resented the imposition of a stricter form of Islam, as well as the continuing violence. This was demonstrated in the July 2000 statement by the Ambon Secretary of Majelis Ulama Indonesia, Malik Selang, that all external actors should leave Maluku: Crisis Centre, Diocese of Amboina, report no. 25, 25 July 2000. I did not undertake fieldwork on this island or interview anybody involved. This information was taken from interviews with IDPs from Lata Lata in Ambon posted on the internet by the Masariku Network: www.malra.org/posko/malra.php4?oid= 151531. It must be noted that all these accounts are given by Christians and I cannot confirm their accuracy. According to this account, by early February the entire village had been forcibly

228 Notes

79 80 81

82 83

84 85 86

87 88 89 90 91

92 93 94

95

96 97 98 99 100 101 102

converted to Islam and all men, women and children circumcised. According to an individual who had interviewed the Lata Lata IDPs, Brawijaya 511 Battalion military personnel supervised the procedures and remained on the island supervising mosque attendance and other Islamic practices: interview in North Sulawesi, 2004. Nanere, Halmahera Berdarah, p. 154. Information received from a former NHM employee, 2005. Interview with Abu Bakar Wahid, in Tomalou, Tidore, 7 January 2004. One well-informed attendee at a seminar where I presented a photograph of the mujahid leaving Tidore believed it was very likely that the vessels in question were Indonesian Navy gunships. Interview in Pediwang, Kao, 5 December 2003. NHM began evacuating further personnel to a ship berthed at Tanjung Barnabas port by helicopter. On 24 January NHM provided a helicopter to fly the national Human Rights Commissioner, Bambang Suharto, in to the area to talk to both parties, although this visit appears to have achieved little. It was impossible for a Kao parliamentarian to return to Ternate during the conflict and any requests to the government had to be made by telephone. Some reports apparently stated that the militia was targeting several Ambonese staff members of PT NHM. Interview with leader of the Pasukan Merah in Kao, 2003. According to the militia leader, the Kaos were able to convince the military unit commander, and through him the mujahid, that the Pasukan Jihad was surrounded by Kaos. According to him it was for this reason the Muslims left the area. Several interviews in Ternate 2003–4. Interview with Pasukan Merah leader Sakeus Odara in Tobelo, 3 September 2003. Interview with Nasrun Alih, the leader of Al Istiklama militia in Igobula, 31 October 2003. Interview with Eddie Tobelo, Pasukan Merah militia member, in Wari, Tobelo, 3 September 2003. Interviews with Eddie Tobelo and with Nasrun Alih, leader of the Al Istiklama militia, in Igobula, 31 October 2003, both of whom were present on opposite sides of this ambush. A large number of Christians believe that a Christian leader from Duma gave information about their movements to the Muslim militia in exchange for money. For example, Sakeus Odara and Cornelis Hohakay were shot and wounded while attempting to reach Galela. Interview with the village head of Gorua, Tobelo, 26 September 2003. One Muslim man from Tobelo who returned to Galela stated he was ready to attack Tobelo City ‘if he had to’: interview with a Muslim community leader in Tobelo City, 19 September 2003. As discussed below, the Pasukan Jihad did not attack Tobelo for several reasons. Nanere writes that 31 members of the Pasukan Jihad died in the attacks and Christians found Indonesian name cards revealing that victims were not just from Ternate but also Surabaya. Also found, according to Nanere, were two name cards of military personnel from a Brawijaya battalion near Surabaya: Nanere, Halmahera Berdarah, pp. 155–6. Interview with approximately 15 Christians in Mamuya, 29 October 2003. Interview with the village head, Mamuya, Interview in Ternate, 5 February 2004. Interview with Yamin Galela, village head of Igobula, in Igobula, 30 October 2003. Interview with Samuel Kukus, leader of the Duma Christian militia, in Tobelo, 26 January 2004. Interviews with both Christian and Muslim respondents present on 19 June, in Galela and Tobelo, 2003–4. Interview with Samuel Kukus, leader of the Duma Christian militia, in Tobelo, 26

Notes 229

103 104 105

106 107 108 109 110 111 112 113 114 115

116 117 118 119 120 121 122 123 124

January 2004. According to Kukus, the military personnel guarding the northern edge of the village laughed at the Christian militia as they withdrew. In 2003 as Duma Christians were preparing to return to their homes, their gardens were destroyed once more, leaving many local Christians still unable to establish economic sustainability. Interview with Nasrun Alih, kapita of Al Istikama militia in Igobula, 31 October 2003. The kinds of atrocities that had been committed in Tobelo Sub-District are also said to have occurred in the fighting in Duma and elsewhere in Galela. In Duma some Christian men were said to have engaged in the practice of eating the hearts of Muslim victims. Interview with a Christian community leader in Duma, October 2003. This may explain why several Galela men told me they demanded that external mujahid should not join a certain attack against Duma, as their fear during battle allegedly lowered the morale of the local men. Interview with Abu Bakar Wahid, in Tomalou, Tidore, 7 January 2004. The Ternate Pos quotes a police officer as stating that 19,000 homemade firearms, machetes, spears and other weapons were surrendered or taken from citizens in this amnesty: Ternate Pos, 15–21 July 2000. Confidential interview with a Christian missionary in Manado, February 2004. International Crisis Group, Overcoming Murder and Chaos in Maluku, p. 12. Interview with Muhammad Selang in Ternate, 7 February 2004. Under the concept of jihad, any war against non-believers who had previously attacked Muslims is generally permissible: Tibi, ‘War and Peace in Islam’, p. 178. Confidential interview with a Protestant pastor in Tobelo City, 2003. Certain Muslim community leaders certainly opposed the jihad, arguing the conditions for jihad were not present. From a quantitative study of ethno-religious conflicts Fox has concluded that ‘whether religious institutions promote protests is highly dependent upon whether the religion itself is threatened’: Fox, ‘Do Religious Institutions Support Violence of the Status Quo?’, p. 131. Clearly this decision was also taken in the context of believing that the military personnel stationed in the area would assist in protecting the village. R. Cribb, ‘Introduction: Problems in the Historiography of the Killings in Indonesia’, in R. Cribb (ed.), The Indonesian Killings 1965–1966, Victoria: Aristoc Press, 1990. Despite some tension, no violence erupted between Muslims and Christians in Kao village. Interview with Theo Sosebeko, in Tobelo, 1 November 2003. See Ternate Pos, 2–8 May 2000, which reports GMIH Pastor Rudi Tindage explaining why Christians refused the entry of non-local military units. Interview with Muhammad Selang in Ternate, 7 February 2004. Interview with a member of the Pasukan Jihad, in Ternate, February 2004. Several interviews in Waringinlamo, 3 November 2003. Ibid.

Conclusion 1 J. Snyder, From Voting to Violence: Democratization and Nationalist Conflict, New York: W. W. Norton & Company, 2000, p. 39 2 D. Horowitz, Ethnic Groups in Conflict, Berkeley: University of California Press, 1985, p. 166. 3 See for example D. Horowitz, The Deadly Ethnic Riot, Berkeley: University of California Press, 2001, p. 75. 4 See for example N. Bubandt, ‘The Dynamics of Reasonable Paranoia: Rumours and Riots in North Maluku, 1999–2000’, background paper for presentation at the

230 Notes

5 6

7 8 9

Seminar Series of the School of Anthropology, Geography and Environmental Studies, University of Melbourne, 25 April 2002, p. 24, and S. Alhadar, ‘The Forgotten War in North Maluku’, Inside Indonesia, no. 63 (July–September 2000). As discussed in Chapter 5, the only area not to experience violence was the immediate vicinity of the Weda Bay nickel mine in Central Halmahera District. For claims that a great deal of the violence in reformasi-period Indonesia was instigated in a covert fashion by actors associated with the New Order regime, see for example G. J. Aditjondro, ‘Guns, Pamphlets and Handy-Talkies: How the Military Exploited Local Ethno-religious Tensions in Maluku to Preserve Their Political and Economic Privileges”, in I. Wessel and G. Wimhofer (eds), Violence in Indonesia, Hamburg: Abera Verlag Markus Voss, 2001, and K. O’Rourke, Reformasi: The Struggle for Power in Post-Soeharto Indonesia, St Leonards: Allen & Unwin, 2002. See for example, A. Hasenclever and V. Rittberger, ‘Does Religion Make a Difference? Theoretical Approaches to the Impact of Faith on Political Conflict’, Millennium, vol. 29, no. 3 (2000), 641–74 J. Snyder, ‘Anarchy and Culture: Insights from the Anthropology of War,’ International Organization, vol. 56, no. 1 (Winter 2002), p. 34. S. N. Kalyvas, ‘The Ontology of “Political Violence”: Action and Identity in Civil Wars’, Perspectives on Politics, vol. 1, no. 3 (September 2003), pp. 475–6.

Bibliography

Abdullah, Muksin H., ‘Kerusuhan Maluku Utara Dalam Perspektif Sosial Budaya’ (The North Maluku Riots in a Socio-cultural Perspective), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. Adeney-Risakotta, F., ‘Mobilising Conflict through Media in the North Moluccas’, seminar paper presented at Indonesian Conflict Studies Network, Bandung, Indonesia, 7–12 April 2003. Aditjondro, G. J., ‘Guns, Pamphlets and Handy-Talkies: How the Military Exploited Local Ethno-religious Tensions in Maluku to Preserve Their Political and Economic Privileges’, in I. Wessel and G. Wimhofer (eds), Violence in Indonesia, Hamburg: Abera Verlag Markus Voss, 2001. —— ‘The Political Economy of Violence in Maluku’, Green Left Weekly, vol. 397 (2003). —— ‘Kerusuhan Poso dan Morowali, Akar Permasalahan dan Jalan Keluarnya’ (The Poso and Morowali Riots, the Roots of the Problem and the Way Out), paper presented at seminar on Application of Emergency Status in Aceh, Papua and Poso?, Hotel Santika, Jakarta, 7 January 2004. Ahmad, K. H. and Oesman, H., Damai Yang Terkoyak: Catatan Kelam dari Bumi Halmahera (Shattered Peace: Dark Notes from the Land of Halmahera), Ternate: Madani Press, 2000. Alhadar, S., ‘The Forgotten War in North Maluku’, Inside Indonesia, no. 63 (July–September 2000). Available online at: www.insideindonesia.org/edit63/alhadar.htm. —— ‘Anatomi Kerusuhan Sosial di Maluku Utara’ (The Anatomy of the Social Riots in North Maluku), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. —— ‘Kompleksitas Konflik Di Halmahera Utara’ (The Complexity of the Conflict in North Halmahera), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. Al-Jakartaty, E., Tragedi Bumi Seribu Pulau (The Tragedy of the Land of a Thousand Islands), Yakarta: Gubuk Kajian Mutiara Nasional, 2000. Aminzade, R. R and McAdam, D., ‘Emotions and Contentious Politics’, in R. R. Aminzade, J. A. Goldstone, D. McAdam, E. J. Perry, W. H. Sewell Jr, S. Tarrow and C. Tilly (eds), Silence and Voice in the Study of Contentious Politics, Cambridge: Cambridge University Press, 2001, pp. 14–50.

232 Bibliography Ammari, F. and Siokona, J. W. (eds), Ternate: Kelahiran dan Sejarah sebuah Kota (Ternate: The Birth and History of a City), Ternate: Ternate City Government, 2003. Andaya, L. Y., The World of Maluku: Eastern Indonesia in the Early Modern Period, Honolulu: University of Hawaii Press, 1993. Appleby, R. S., The Ambivalence of the Sacred: Religion, Violence, and Reconciliation, New York: Rowman & Littlefield, 2000. Aragon, L. V., ‘Communal Violence in Poso, Central Sulawesi: Where People Eat Fish and Fish Eat People’, Indonesia, vol. 72 (October 2001), 45–79. Arfi, B., ‘Ethnic Fear: The Social Construction of Insecurity, Security Studies, vol. 8, no. 1 (Autumn 1998), 151–203. Azra, A., ‘The Islamic Factor in Post-Soeharto Indonesia’, in C. Manning and P. van Dierman (eds), Indonesia in Transition: Social Aspects of Reformasi and Crisis, Singapore: Institute of Southeast Asian Studies, 2000. Badan Pusat Statistik, Bappenas and UNDP, Indonesia Human Development Report 2001. Towards a New Consensus: Democracy and Human Development in Indonesia, Jakarta, 2001. Bartels, D., ‘Guarding the Invisible Mountain: Intervillage Alliances, Religious Syncretism and Ethnic Identity among Ambonese Christians and Moslems in the Moluccas’, unpublished PhD thesis, Cornell University, 1977. —— ‘Your God is No Longer Mine: Moslem–Christian Fratricide in the Central Moluccas (Indonesia) after a Half-Millennium of Tolerant Co-Existence and Ethnic Unity’, unpublished article, 2000. Bertrand, J., ‘Legacies of the Authoritarian Past: Religious Violence in Indonesia’s Moluccan Islands’, Pacific Affairs, vol. 75, no. 1 (2002), 57–85. —— Nationalism and Ethnic Conflict in Indonesia, Cambridge: Cambridge University Press, 2004. Blumer, H., Symbolic Interactionism: Perspective and Method, Berkeley: Prentice Hall, 1969. Boediman, M. D., ‘Musuhku adalah Saudarku’ (My Enemy is My Brother), Unpublished thesis, Universitas Kristen Duta Wacana, Yogyakarta, 2002. Bohm, C. J., ‘Brief Chronicle of the Unrest in the Moluccas 1999–2001’, Crisis Centre, Diocese of Amboina. Bourchier, D., ‘Skeletons, Vigilantes, and the Armed Forces’ Fall from Grace’, in A. Budiman, B. Hatley and D. Kingsbury (eds), Reformasi: Crisis and Change in Indonesia, Clayton: Monash Asia Institute, 1999. Brass, P. R. (ed.), Ethnic Groups and the State, London: Croom Helm, 1985. —— ‘Introduction: Discourses of Ethnicity, Communalism, and Violence’, in P. R. Brass (ed.), Riots and Pogroms, New York: New York University Press, 1996. —— Theft of an Idol: Text and Context in the Representation of Collective Violence, Princeton: Princeton University Press, 1997. —— The Production of Hindu–Muslim Violence in Contemporary India, New Delhi: Oxford University Press, 2003. Brubaker, R. and Laitin, D. D., ‘Ethnic and Nationalist Violence’, Annual Review of Sociology, vol. 24 (1998), 423–52. Bruce, S., Politics and Religion, Cambridge: Polity Press, 2003. —— ‘Religion and Violence: What Can Sociology Offer?’, Numen, vol. 52 (2005), 5–28. Bubalo, A. and Fealy, G., Joining the Caravan? The Middle East, Islamism and Indonesia, Lowy Institute Paper 5, New South Wales: Lowy Institute for International Policy, 2005.

Bibliography 233 Bubandt, N., ‘Malukan Apocalypse: Themes in the Dynamics of Violence in Eastern Indonesia’, in I. Wessel and G. Wimhofer (eds), Violence in Indonesia, Hamburg: Abera Verlag Markus Voss, 2001. —— ‘The Dynamics of Reasonable Paranoia: Rumours and Riots in North Maluku, 1999–2000’, background paper for presentation at the Seminar Series of the School of Anthropology, Geography and Environmental Studies, University of Melbourne, 25 April 2002. —— ‘Towards a New Politics of Tradition? Decentralisation, Conflict, and Adat in Eastern Indonesia’, Antropologi Indonesia, no. 74 (May–August 2004), 11–30. Chauvel, R., ‘Ambon: Not a Revolution but a Counterrevolution’, in A. R. Kahin (ed.), Regional Dynamics of the Indonesian Revolution: Unity from Diversity, Honolulu: University of Hawaii Press, 1985. —— Nationalists, Soldiers and Separatists: The Ambonese Islands from Colonialism to Revolt 1880–1950, Leiden: KITLV Press, 1990. Collier, P. and Hoeffler, A., ‘On Economic Causes of Civil War’, Oxford Economic Papers, vol. 50, no. 4 (1998), 563–73. —— ‘Greed and Grievance in Civil War’, World Bank Policy Research Working Paper no. 2355, Washington, DC: World Bank, October 2001. Cribb, R., ‘Introduction: Problems in the Historiography of the Killings in Indonesia’, in R. Cribb (ed.), The Indonesian Killings 1965–1966: Studies from Java and Bali, Centre of Southeast Asian Studies, Monash University, 1990. Crouch, H., ‘Wiranto and Habibie: Military–Civilian Rrelations since May 1998’, in A. Budiman, B. Hatley and D. Kingsbury (eds), Reformasi: Crisis and Change in Indonesia, Clayton: Monash Asia Institute, 1999. —— ‘Political Update: Megawati’s Holding Operation’, in E Aspinall and G. Fealy (eds), Local Power and Politics in Indonesia: Decentralization and Democratization. Singapore: Institute of Southeast Asian Studies, 2003. Davidson, J. S., ‘The Politics of Violence on an Indonesian Periphery’, South East Asia Research, vol. 11, no. 1, 59–89. Davis, Natalie Zemon, ‘The Rites of Violence: Religious Riot in Sixteenth-Century France’, Past and Present, vol. 59 (May 1973), 51–91. Donner, F. M., ‘The Sources of Islamic Conceptions of War’, in J. Kelsay and J. T. Johnson (eds), Just War and Jihad: Historical and Theoretical Perspectives on War and Peace in Western and Islamic Traditions, Westport: Greenwood Press, 1991. Duncan, C. R., ‘Savage Imagery: (Mis)representations of the Forest Tobelo of Indonesia’, The Asia Pacific Journal of Anthropology, vol. 2, no. 1 (2001), 45–62. —— ‘The Other Maluku: Chronologies of Conflict in North Maluku’, Indonesia, vol. 80 (October 2005), 53–80. Editorial, ‘Current Data on the Indonesian Military Elite: January 1 1999–January 31 2001’, Indonesia, vol. 71 (April 2001). Ellis, S., The Mask of Anarchy: The Destruction of Liberia and the Religious Dimension of an African Civil War, New York: New York University Press, 1999. Engineer, A. A., ‘Introduction’, in A. A. Engineer (ed.), The Gujarat Carnage, New Delhi: Orient Longman, 2003. Fealy, G., ‘Inside the Laskar Jihad – An Interview with the Leader of a New, Radical and Militant Sect’, Inside Indonesia, 31 March 2001. Fearon, J. D., ‘Rationalist Explanations for War’, International Organization, vol. 49, no. 3 (Summer 1995), 379–414. —— ‘Ethnic War as a Commitment Problem’, paper presented at the 1994 Annual Meeting

234 Bibliography of the American Political Science Association, New York, 30 August–2 September 1995. Fearon, J. D. and Laitin, D. D., ‘Explaining Interethnic Cooperation’, American Political Science Review, vol. 90, no. 4 (December 1996), 715–35. —— ‘Violence and the Social Construction of Ethnic Identity’, International Organization, vol. 54, no. 4 (Autumn 2000), 845–77. —— ‘Ethnicity, Insurgency, and Civil War’, American Political Science Review, vol. 97, no. 1 (February 2003), 75–90. Fox, J., ‘Do Religious Institutions Support Violence or the Status Quo?’, Studies in Conflict and Terrorism, vol. 22 (1999), 119–39. Gagnon, V. P., ‘Ethnic Nationalism and International Conflict: The Case of Serbia’, International Security, vol. 19, no. 3 (Winter 1994/95), 130–66. —— ‘Ethnic Nationalism and International Conflict: The Case of Serbia’, in M. E. Brown, Owen R Cote Jr, Sean M Lynn-Jones and Steven E Miller (eds), Nationalism and Ethnic Conflict (International Security Readers series), Cambridge, MA: The MIT Press, 1997. Goss, J. D. and Leinbach, T. R., ‘Development and Differentiation: A Case Study of a Transmigration Settlement in West Seram’, in D. Mearns and C. Healey (eds), Remaking Maluku: Social Transformation in Eastern Indonesia, Special Monograph no. 1, Centre for Southeast Asian Studies, Northern Territory University, 1996. Gurr, T. R., Why Men Rebel, Princeton: Princeton University Press, 1970. Haire, J., The Character and Theological Struggle of the Church in Halmahera, 1941–1979, Frankfurt: Verlag Peter D. Lang, 1981. Hanna, W. A. and Alwi, D., Turbulent Times Past in Ternate and Tidore, Banda Naira: Yayasan Warisan dan Budaya Banda Naira, 1990. Hasan, N., ‘Faith and Politics: The Rise of the Laskar Jihad in the Era of Transition in Indonesia’, Indonesia, vol. 73 (April 2002), 145–69. Hasenclever, A. and Rittberger, V., ‘Does Religion Make a Difference? Theoretical Approaches to the Impact of Faith on Political Conflict’, Millennium, vol. 29, no. 3 (2000), 641–74. Heck, P. L., ‘Jihad Revisited’, Journal of Religious Ethics, vol. 32, no. 1, 95–128. Hefner, R. W., ‘Civic Pluralism Denied? The New Media and Jihadi Violence in Indonesia’, in D. F. Eickelman and J. W. Anderson (eds), New Media in the Muslim World: The Emerging Public Sphere, Bloomington: Indiana University Press, 1999. Hill, H. ‘Indonesia: The Strange and Sudden Death of a Tiger Economy’, Oxford Development Studies, vol. 28, no. 2 (2000), 117–39. Horowitz, D., Ethnic Groups in Conflict, Berkeley: University of California Press, 1985. —— ‘Structure and Strategy in Ethnic Conflict’, paper prepared for the Annual World Bank Conference on Development Economics, Washington, DC, 20–21 April 1998. —— ‘Group Loyalty and Ethnic Violence’, in Charles Hermann, Harold K. Jacobson and Anne S. Moffat (eds), Violent Conflict in the Twenty-first Century: Causes, Instruments and Mitigation, Illinois: American Academy of Arts & Sciences, 1999. —— The Deadly Ethnic Riot, Berkeley: University of California Press, 2001. Human Rights Watch, ‘Indonesia: The Violence in Ambon’, Human Rights Watch Report, 1999. International Crisis Group, Indonesia’s Maluku Crisis: The Issues, Indonesia Briefing, 19 July 2000. —— Indonesia: Overcoming Murder and Chaos in Maluku, Asia Report no. 10, 19 December 2000. International Crisis Group, National Police Reform, Asia Report no. 13, February 2001.

Bibliography 235 International Crisis Group, Communal Violence in Indonesia: Lessons from Kalimantan, Asia Report no. 19, 27 June 2001. —— Indonesia: The Search for Peace in Maluku, Asia Report no. 31, 8 February 2002. —— Al-Qaeda in Southeast Asia: The Case of the ‘Ngruki Network’ in Indonesia, Asia Briefing no. 20, 8 August 2002. —— Indonesia Backgrounder: How the Jemaah Islamiyah Terrorist Network Operates, Asia Report no. 43, 11 December 2002. Ishige, N., ‘Limau Village and Its Setting’, in N. Ishige (ed.), The Galela of Halmahera: A Preliminary Survey, Osaka: National Museum of Ethnology, 1980. Jati, A. H., Dan Bundapun Menangis (And Mothers Weep), Manado, 2001. Juergensmeyer, M., Terror in the Mind of God: The Global Rise of Religious Violence, Berkeley: University of California Press, 2000. Kakar, S., The Colors of Violence: Cultural Identities, Religion and Conflict, Chicago: University of Chicago Press, 1996. —— ‘The Time of Kali: Violence between Religious Groups in India’, Social Research, vol. 67, no. 3 (Fall 2000), 877–99. Kaldor, M., New and Old Wars: Organized Violence in a Globalized Era, Stanford: Stanford University Press, 1999. Kalyvas, S. N., ‘The Ontology of “Political Violence”: Action and Identity in Civil Wars’, Perspectives on Politics, vol. 1, no. 3 (September 2003), 475–94. —— The Logic of Violence in Civil Wars, New York: Cambridge University Press, 2006. Kaufman, C., ‘Rational Choice and Progress in the Study of Ethnic Conflict: A Review Essay’, Security Studies, vol. 14, no. 1 (January–March 2005), 178–207. Kaufman, S. J., ‘An “International” Theory of Interethnic War’, Review of International Studies, vol. 22, no. 2 (April 1996), 149–72. —— ‘Spiraling to Ethnic War: Elites, Masses and Moscow in Moldova’s Civil War’, International Security, vol. 21, no. 2 (Fall 1996), 108–38. —— ‘Symbolic Politics or Rational Choice? Testing Theories of Extreme Ethnic Violence’, International Security, vol. 30, no. 4 (Spring 2006), 45–86. Keen, D., The Economic Functions of Violence in Civil Wars, Adelphi Paper 320, London: Oxford University Press, 1998. Kiem, C. G., Growing up in Indonesia: Youth and Social Change in a Moluccan Town, Saarbrücken and Fort Lauderdale: Verlag Breitenbach Publishers, 1993. King, C., ‘The Micropolitics of Social Violence’, World Politics, vol. 56, no. 3 (April 2004), 431–55. Kordi, M. and Ghufran, H., ‘Kompleksitas Konflik Maluku Utara’ (The Complexity of the North MaluluConflict), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. —— ‘Dinamika Masyarakat Makian Dan Konflik Maluku Utara’ (The Dynamism of the Makian People and the North Maluku Conflict), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. Kristiadi, J., ‘The Armed Forces’, in R. W. Baker, M. H. Soesastro, J. Kristiadi and D. E. Ramage (eds), Indonesia: The Challenge of Change, Singapore: Institute of Southeast Asian Studies, 1999. Lake, D. A., ‘International Relations Theory and Internal Conflict: Insights from the Interstices’, International Studies Review, vol. 5, no. 4 (2003), 81–9.

236 Bibliography Lake, D. A. and Rothchild, D., ‘Containing Fear: The Origins and Management of Ethnic Conflict’, International Security, vol. 21, no. 2 (Fall 1996), 41–75. —— ‘Spreading Fear: The Genesis of Transnational Ethnic Conflict’, in D. A. Lake and Rothchild, D. (eds), The International Spread of Ethnic Conflict: Fear, Diffusion and Escalation, Princeton: Princeton University Press, 1998. Leirissa, R. Z., ‘The Idea of a Fourth Kingdom in Nineteenth Century Tidorese Maluku’, in E. K. M. Masinambow (ed.), Maluku dan Irian Jaya (special issue of Buletin Leknas, vol. III, no. 1), Jakarta: Lembaga Ilmu Pengetahuan Indonesia, 1994. Leith, J., ‘Resettlement History, Resources and Resistance in North Halmahera’, in S. Pannell and F. von Benda-Beckmann (eds), Old World Places, New World Problems: Exploring Issues of Resource Management in Eastern Indonesia, Canberra: Centre for Resource and Environmental Studies, Australian National University, 1998. Lucardie, R., ‘Spontaneous and Planned Movement among the Makianese of Eastern Indonesia’, Pacific Viewpoint, vol. 26, no. 1 (April 1985), 63–78. McAdam, D., Political Process and the Development of Black Insurgency 1930–1970, Chicago and London: University of Chicago Press, 1982. —— ‘Tactical Innovation and the Pace of Insurgency’, American Sociological Review, vol. 48 (December 1983), 735–54. McAdam, D., McCarthy, J. D. and Zald, N., Comparative Perspectives on Social Movements, Cambridge: Cambridge University Press, 1996. McAdam, D., Tarrow, S. and Tilly, C., Dynamics of Contention, Cambridge: Cambridge University Press, 2001. McCarthy, J. D. and Zald, N., ‘Resource Mobilization and Social Movements: A Partial Theory’, American Journal of Sociology, vol. 82, no. 6, 1212–41. McRae, D., ‘Criminal Justice and Communal Conflict: A Case Study of the Trial of Fabianus Tibo, Dominggus Da Silva, and Marinus Riwu’, Indonesia, vol. 83 (April 2007), 79–117. Mearns, D., ‘Class, Status and Habitus in Ambon’, in D. Mearns and C. Healey (eds), Remaking Maluku: Social Transformation in Eastern Indonesia, Special Monograph no. 1, Centre for Southeast Asian Studies, Northern Territory University, 1996. Meyer, D. S. and Staggenborg, S., ‘Movements, Countermovements, and the Structure of Political Opportunity’, American Journal of Sociology, vol. 101, no. 6 (May 1996), 1628–60. Meyer, P. A. and Hardjodimedjo, M., ‘Maluku: The Modernization of the Spice Islands’, in H. Hill (ed.), Unity and Diversity: Regional Economic Development in Indonesia since 1970, Oxford: Oxford University Press, 1989. Mietzner, M., ‘Godly Men in Green’, Inside Indonesia, vol. 53 (January–March 1998), 8–9. Moore, Barrington, Jr, Moral Purity and Persecution in History, Princeton: Princeton University Press, 2000. Mueller, J., ‘The Banality of Ethnic War’, International Security, vol. 25, no. 1 (Summer 2000), 42–70. Murphy, D., ‘Al Qaeda’s New Frontier: Indonesia’, Christian Science Monitor, 1 May 2002. Nanere, J., Halmahera Berdarah (Bloody Halmahera), Ambon: Yayasan Bina Masyarakat Sejahtera dan Pelestarian Alam (BIMASPELA), 2000. Olson, M., The Logic of Collective Action: Public Goods and the Theory of Goods, Cambridge, MA: Harvard University Press, 1971. Olzak, S., The Dynamics of Ethnic Competition and Conflict, Stanford: Stanford University Press, 1992.

Bibliography 237 O’Rourke, K., Reformasi: The Struggle for Power in Post-Soeharto Indonesia, St Leonards: Allen & Unwin, 2002. Petersen, R. D., Understanding Ethnic Violence: Fear, Hatred, and Resentment in Twentieth-Century Eastern Europe, Cambridge: Cambridge University Press, 2002. Polletta, F. and Jasper, J. M., ‘Collective Identity and Social Movements’, Annual Review of Sociology, vol. 27 (2001), 283–305. Posen, B. R., ‘The Security Dilemma and Ethnic Conflict’, Survival, vol. 35, no. 1, 27–47. Regan, A. J., ‘The Bougainville Conflict: Political and Economic Agendas’, in K. Ballentine and J. Sherman (eds), The Political Economy of Armed Conflict: Beyond Greed and Grievance, Boulder: Lynne Reiner Publishers, 2003. Robinson, G., ‘Indonesia: On a New Course?’, in M. Alagappa (ed.), Coercion and Governance: The Declining Political Role of the Military in Asia, Stanford: Stanford University Press, 2001. Roe, P., ‘Misperception and Ethnic Conflict: Transylvania’s Societal Security Dilemma’, Review of International Studies, vol. 28 (2002), 57–74. Ross, M. L., ‘Oil, Drugs and Diamonds: The Varying Roles of Natural Resources in Civil Wars’, in K. Ballentine and J. Sherman (eds), The Political Economy of Armed Conflict: Beyond Greed and Grievance, Boulder: Lynne Reiner Publishers, 2003. Rule, J. B., ‘Rationality and Non-Rationality in Militant Collective Action’, Sociological Theory, vol. 7, no. 2 (Autumn 1989), 145–60. Sambanis, N., ‘Review of Recent Advances and Future Directions in the Quantitative Literature on Civil War’, Defence and Peace Economics, vol. 13, no. 3 (2002), 215–43. Santosa, J. C., ‘Modernization, Utopia and the Rise of Islamic Radicalism in Indonesia’, unpublished PhD dissertation, Boston University, 1996. Seul, J. R., ‘Ours Is the Way of God: Religion, Identity, and Intergroup Conflict’, Journal of Peace Research, vol. 36, no. 5 (1999), 553–69. Singh, B., ‘The Indonesian Military Business Complex: Origins, Course and Future’, Working Paper no. 354, Canberra: Strategic and Defence Studies Centre, 2001. Sitohang, H. H., Hidayaturohman, Nafi, M. D., Ramdhan, M. and Subekti, S., Menuju Rekonsiliasi Di Halmahera (Towards Reconciliation on Halmahera), Jakarta: Pusat Pemberdayaan untuk Rekonsiliasi dan Perdamaian, 2003. Smith, A. D., The Ethnic Origins of Nations, Oxford: Blackwell Publishers, 1986. Snyder, J. From Voting to Violence: Democratization and Nationalist Conflict, New York: W. W. Norton & Company, 2000. —— ‘Anarchy and Culture: Insights from the Anthropology of War’, International Organization, vol. 56, no. 1 (Winter 2002), 7–45. Steenbrink, K. ‘Muslim–Christian Relations in the Pancasila State of Indonesia’, Muslim World, vol. 88, no. 3 (1998), 32–52. Tambiah, S. J., Leveling Crowds: Ethnonationalist Conflicts and Collective Violence in South Asia, Berkeley: University of California Press, 1996. —— ‘The Nation-State in Crisis and the Rise of Ethnonationalism’, in E. N. Wilmsen and P. McAllister (eds), The Politics of Difference: Ethnic Premises in a World of Power, Chicago: University of Chicago Press, 1996. Tarrow, S., Power in Movement: Social Movements, Collective Action, and Mass Politics in the Modern State, Cambridge: Cambridge University Press, 1994. Thomas, H., The Spanish Civil War, rev. edn, New York: The Modern Library, 2001. Tibi, B., ‘War and Peace in Islam’, in S. H. Hashmi (ed.), Islamic Political Ethics: Civil Society, Pluralism, and Conflict, Princeton: Princeton University Press, 2002, pp. 175–93.

238 Bibliography Tilly, C. From Mobilization to Revolution, Reading, MA: Addison Wesley, 1978. —— The Politics of Collective Violence, Cambridge: Cambridge University Press, 2003. Toft, M. D., The Geography of Ethnic Violence: Identity, Interests, and the Indivisibility of Territory, Princeton and Oxford: Princeton University Press, 2003. Tomagola, T. A., ‘The Bleeding Halmahera of North Moluccas’, Jurnal Studi Indonesia, vol. 10, no. 2 (2000). —— ‘Krisis dan Solusi Tragedi Maluku Utara’, Detikcom, 2 February 2000. Valentino, B. A., Final Solutions: Mass Killing and Genocide in the Twentieth Century, Ithaca and London: Cornell University Press, 2004. van Bruinessen, M., ‘Genealogies of Islamic Radicalism in Post-Suharto Indonesia’, South East Asian Research, vol. 10, no. 2 (2002), 117–54. van Dijk, C., Rebellion under the Banner of Islam, The Hague: Martinus Nijhoff, 1981. van Fraassen, C. F., ‘Types of Socio-Political Structure in North-Halmaheran History’, in E. K. M. Masinambow (ed.), Halmahera dan Raja Ampat: Konsep dan Strategi Penelitian (Halmahera and Raja Ampat: Research Strategies and Concepts), Jakarta: LIPI, 1980. van Klinken, G., ‘The Maluku Wars: Bringing Society Back In’, Indonesia, vol. 71 (April 2001), 1–26. Varshney, A., Ethnic Conflict and Civic Life: Hindus and Muslims in India, New Haven and London: Yale University Press, 2002. Villiers, J., ‘Las Yslas de Esperar en Dios: The Jesuit Mission in Moro 1546–1571’, Modern Asian Studies, vol. 22, no. 3 (1988), 593–606. von Benda-Beckmann, F. and Taale, T., ‘Land, Trees, and Houses: Changing (Un)Certainties in Property Relationships on Ambon’, in D. Mearns and C. Healey (eds), Remaking Maluku: Social Transformation in Eastern Indonesia, Special Monograph no. 1, Centre for Southeast Asian Studies, Northern Territory University, 1996. Waever, O., Buzan, B., Kelstrup, M. and Lemaitre, P., Identity, Migration and the New Security Agenda in Europe, London: Pinter Publishers, 1993. Weingast, B. R., ‘Constructing Trust: The Political and Economic Roots of Ethnic and Regional Conflict’, in K. Soltan, E. M. Uslaner and V. Haufler (eds), Institutions and Social Order, Ann Arbor: University of Michigan Press, 1998. Wilkinson, S. I., Votes and Violence: Electoral Competition and Ethnic Riots in India, Cambridge: Cambridge University Press, 2004. Young, K. R., ‘Local and National Influences in the Violence of 1965’, in R.Cribb (ed.), The Indonesian Killings 1965–1966: Studies from Java and Bali, Centre of Southeast Asian Studies, Monash University, 1990. Yurnaldi, ‘Dendam yang tak Berkesudahan’ (Revenge without End), in I. Hasan (ed.), Memikirkan Kembali: Maluku dan Maluku Utara (Rethinking: Maluku and North Maluku), Makassar: Lembaga Penerbitan Universitas Hasanuddin (LEPHAS), 2003. Zald, M. N. and Useem, B., ‘Movement and Countermovement Interaction: Mobilization, Tactics, and State Involvement’, in M. N. Zald and J. D. McCarthy (eds), Social Movements in an Organizational Society, New Brunswick, NJ: Transaction Books, 1987.

Index

Ambon 31, 36; and Maluku conflict (1999 to 2002) 2, 6–8, 28–9, 31, 43–5, 48, 69, 77–80, 89, 103, 117, 128, 155, 157–9, 178; history 32, 34–5, 45–6 Andili, Bahar 6, 46–7, 74, 78–9, 93–4, 133–5, 138, 142–3, 145 Andili, Syamsir 46–7, 78–80, 83, 92–4, 133–5, 142 Armain, Thaib 37, 46–8, 78, 80–1, 83, 92–94, 133–135, 142, 152 Assagaf, Abdullah 57–59, 61–62, 137 Bitjara, Benny 64–6, 70, 105, 110–12, 114, 122, 124, 127–8, 149, 161, 167, 170, 173–4, 183, 188, 192–3 ‘Bloody Sosol’ letter 8–9, 26, 84–6, 91–2, 94, 181–3 Brass, Paul 9, 12, 21, 72–3, 94, 131–2, 143, 185–6 Cahaya Bahari 3, 167, 187 Civil Emergency 3, 167, 176, 187 decentralization (also regional autonomy) 29, 39–40, 48; impact on conflict 4, 58, 67, 179 Duma 115; during conflict December 1999 to February 2000 2, 3, 117–20, 123, 125, 184; during attacks by Pasukan Jihad 154–5, 163–70, 172–3, 175, 188; history 34, 101 FPI (Front Pembela Islam) 104, 157–9, 170 Galela Sub-District 26, 31–2, 34–6, 96, 99, 101, 115; conflict in 2–3, 10–13, 92, 96–7, 99, 106, 113–15, 117–19, 121–126, 128–30, 140, 146–7, 150–2,

154–6, 158, 160, 163–76, 183–4, 186–8, 192 GMIH (Gereja Maseh Injili Halmahera) 34, 52, 77, 84–5, 101, 104–5, 107–8, 117, 119, 121, 123–4, 135, 171, 182, 184 Gorua 101, 107–9, 113–15, 126, 164, 169 Horowitz, Donald 8, 17, 21–2, 51, 68, 73, 91, 180 Imam, Wahda Zainal 80–3, 88, 93–4, 104, 152, 157, 170 Internally Displaced Persons (IDPs) 3, 10, 64, 69–71, 80–3, 88–90, 104–6, 114, 117–9, 121–2, 126, 128–9, 135, 146–8, 150–2, 160–3, 167–9, 175–6 Jemaah Islamiyah 158–9 Jibril, Abu (aka Fikiruddin) 158–9 Kalyvas, Stathis 19, 25, 194 Kao Sub-District 1, 26, 31–2, 52, 54–6; conflict in 2, 5–8, 13, 49–50, 56–71, 161–4 Laskar Jihad 4, 12, 148, 155–7, 159, 175, 187 Makayoa (student group) 57, 61–3, 68–9, 190 Makian Island 5, 32, 49, 52, 55, 57, 59, 92, 104 Malifut 2, 4–9, 11–2, 26, 36, 48–50, 54–71, 79–96, 103–4, 110, 117, 128, 135, 142, 144–5, 147, 152, 156, 160–4, 167–8, 172–7, 179–82, 185–94 Maluku conflict (1999 to 2002) see Ambon, TNI

240 Index Mamuya 3, 33, 115, 118–9, 163–4, 168, 173

Syah, Djunus – the Sultan of Tidore 12, 134, 138–9, 143–5

NHM (Nusa Halmahera Mineral) 36, 56–60, 115, 160–2, 187

Tambiah, Stanley 9, 24, 72–3, 186 Ternate 29–31, 36–8, 45, 47, 55, 76–8: history 30, 32–5, 74; riot – November 1999 2, 8–10, 26–7, 71, 88–95, 181–3, 192; see also Putih–Kuning conflict Tidore 29, 31–2, 47, 76–8; History 30, 32–4, 37–8, 48, 74; riot, November 1999 2, 8–9, 71, 86–8, 91; see also Putih–Kuning conflict TNI – the Indonesian Military: nationally 39–42, 44, 47, 155–7; role in North Maluku conflict 2–3, 7, 13, 26, 62, 64–6, 71, 79, 82–3, 86, 88–9, 93, 95, 103–4, 106–9, 112–13, 124–5, 134, 136, 139, 141, 144–5, 153, 160–6, 168, 171–174, 176, 184, 186–9 Tobelo City and Sub-District 3, 12, 31–2, 35–8, 48, 58, 67, 84, 96, 99–102; conflict in December 1999 2–4, 10–11, 89, 103–115, 120–9, 183–4, 192 Togoliua 101, 107, 110, 113–5, 121, 123, 126, 152, 169, 174

Pasukan Jihad 3, 13, 137, 146–76, 186–8, 193 Pasukan Kuning 74, 83–4, 88–9, 93, 135–45, 150–1, 181, 185, 190–1 Pemekaran 7, 40, 60, 67; see also Malifut Popilo 109, 113–15, 126, 152 PP42 (Government Regulation 42/1999) 59–66, 70, 191 Putih–Kuning Conflict 11–12, 130–1, 137–47, 184–6 RMS (Republik Maluku Selatan) 35, 91, 103, 109, 153, 171 Security Dilemma 11, 17, 19, 97, 99, 122–4, 183–4, 189 Sulawesi, North – refugees in 3, 89, 104, 152 Syah, Mudaffar – the Sultan of Ternate 6, 11–12, 37–8, 46–8, 63–4, 74, 78, 80, 83, 90, 92–5, 106, 127, 130–1, 133–46, 153, 181, 184–5, 190–1

Wahid, Abu Bakar 77, 137–8, 142–3, 151–63, 167–8, 170–1, 192–3