The Conflict in the Moluccas: Local Youths' Perceptions Contrasted to Previous Research

(1)

Peace and Conflict Studies III Spring Semester 2013

Supervisor: Kristian Steiner

Faculty of Culture and Society

Department of Global Political Studies

The Conflict in the Moluccas:

Local Youths’ Perceptions Contrasted to Previous Research

Martin Björkhagen Words: 16 434

(2)

Abstract

The violent conflict in the Moluccas (1999-2002) has occasionally been portrayed in terms of animosities between Christians and Muslims. This study problematizes that statement by analysing several conflict drivers seen through two perspectives.

The first purpose of this study was to contrast previous research regarding conflict factors in the Moluccas to the perceptions of the local youths’. There is a research gap regarding the youths’ experiences of the conflict, which this study aims to bridge. A second purpose was to analyse discrepancy between the academic literature and the youths’ bottom-up perspective. The final purpose was to apply the theory of collective guilt to explain and analyse the youths’ memories and perceptions regarding the conflict factors in the Moluccas. The qualitative case study approach was adopted since it could include both in-depth interviews and an assessed literature review. Six in-depth interviews were conducted in Indonesia which explored the youth’s perceptions. The critically assessed literature review was used to obtain data from secondary sources regarding the same conflict factors, as was explored by the interviews.

The first part of the analysis exposed a discrepancy between the two perspectives regarding some of the conflict factors. The collective guilt analysis found that the youths only seem to experience a rather limited feeling of collective guilt. This is because all strategies to reduce collective guilt were represented in the youths’ perceptions. The most used strategy was to blame a few ‘black sheep’, or in this context a few provocateurs for the harm inflicted by the in-group, towards the out-group.

Keywords: Moluccan youths, conflict in the Moluccas, collective guilt, in-depth interview, qualitative case study

(3)

Sammanfattning

Den våldsamma konflikten på Moluckerna (1999-2002) porträtteras ibland i termer av motsättningar mellan kristna och muslimer. Denna studie problematiserar den bilden genom att analysera flera konfliktfaktorer som belyses ur två perspektiv.

Det första syftet med denna studie var att jämföra tidigare forskning gällande

konfliktfaktorer i Moluckerna med lokala ungdomars perception. Det finns en forskningslucka rörande ungdomarnas upplevelser av konflikten, vilket denna studie syftar till att överbrygga. Ett andra syfte var att analysera diskrepansen mellan den akademiska litteraturen och

ungdomarnas underifrånperspektiv.

Det avslutande syftet var att applicera teorin om kollektiv skuldkänsla för att förklara och analysera ungdomarnas minnen och uppfattningar om konfliktfaktorerna i Moluckerna. En kvalitativ fallstudieansats har valts eftersom den kan inkludera både djupintervjuer och en litteraturgenomgång. Sex djupintervjuer genomfördes i Indonesien som utforskade

ungdomarnas uppfattningar. Den källkritiskt granskade litteraturgenomgången användes för att erhålla data från sekundärkällor angående samma konfliktfaktorer, som undersöktes I samband med intervjuerna.

Den första delen av analysen visade att det finns en diskrepans mellan de två perspektiven när det gäller några av konfliktfaktorerna. Analysen av kollektiv skuldkänsla visade att ungdomarna bara upplevde en begränsad känsla av kollektiv skuld. Detta berodde på att alla strategier som minskar den kollektiva skuldkänslan var representerade i ungdomarnas uppfattningar. Den mest använda strategin var att beskylla några "svarta får", eller i detta fall, några provokatörer för skadan den egna gruppen orsakat den motsatta gruppen.

Nyckelord: Ungdomar från Moluckerna, konflikten i Moluckerna, kollektiv skuldkänsla, djupintervju, kvalitativ fallstudie,

(4)

Acknowledgements

First and foremost I want to thank my family for always being there and supporting me. I also would like to thank several other people for helping me with this study, both in Sweden, the U.S.A. and in Indonesia. I initiated the material gathering for this study when I

participated in the exchange program between Malmö University and Gadjah Mada

University, Yogyakarta. Therefore, I want to send my gratitude to these institutions for giving me this great opportunity of studying in a different environment. Further, I want to thank CSPS Research Centre in Yogyakarta for welcoming me as an intern and letting me to start with this project under their supervision. Especially Dr. Muhammad Najib Azca and Dr. Eric Hiariej at CSPS. Both have extensive research experience regarding the Moluccas and have provided me with much appreciated guidance when I started my study. I am also very grateful to Mira Andiani for the translation support she provided, including both during the interviews and transcription work. I also want to express my gratitude to my friend Dr. Jesse Rhines in Los Angeles for proofreading and for his kind encouragement. Yet, my deepest gratitude goes out to the brave youths from the Moluccas who volunteered to participate in this study. Without their contribution this study would never have been made possible.

Martin Björkhagen August 2013, Malmö, Sweden

(5)

List of Abbreviations

ALA – American Library Association

BAPPENAS – Badan Perencanaan Pembangunan Nasional (The National Development Planning Agency of Indonesia)

BKO – Bawah Kendali Operasi (Extra Non-regional Forces) BRIMOB – Brigade Mobil (Military Police Mobile Brigades) CPRU – Crisis Prevention and Recovery Unit (Jakarta) CSPS – Center for Security and Peace Studies

DDII – Dewan Dakwah Islamiyah Indonesia (Indonesian Islamic Outreach Council) FKM – Front Kedaulatan Maluku (Maluku Sovereignty Front)

HD Centre – The Centre for Humanitarian Dialogue HRW – Human Rights Watch

ICG – International Crisis Group

ICMI – Indonesian Association of Muslim Intellectuals IDPs – Internally Displaced Persons

IDR – Indonesian Rupiah

ISAI – Institut Studi Arus Infomasi

IMS – International Media Support (Copenhagen) LBH – Jakarta Legal Aid Institute

MMC – Maluku Media Centre MPC – Maluku Protestant Church NGO – Non-Governmental Organisation

OPM – Organisasi Papua Merdeka (The Free Papua Movement)

PRRI – Pemerintah Revolusioner Republik Indonesia (Revolutionary Government of the Indonesian Republic)

RMS – Republik Maluku Selatan (the Republic of South Maluku) TNI – Tentara Nasional Indonesia (Indonesian National Military) UGM – Universitas Gadjah Mada (Gadjah Mada University) UNDP – United Nations Development Programme

UNICEF – United Nations International Children’s Emergency Fund

VOC – Vereenigte Oost-Indische Compagnie (the Dutch East-Indies Company)

(6)

1. Introduction

In the recent decade Indonesia has made progress in areas such as democratisation and economic development. Yet, Indonesia struggles with poverty, widespread corruption, environmental degradation, natural disasters and civil conflicts within some of its 33 provinces (SIDA, 2013). This qualitative study explores the perceptions of local Moluccan youths’ regarding the violent conflict in the Moluccas, including its post-conflict phase. I contrast previous research regarding the conflict factors, with a bottom-up perspective gained by interviews with local youth. I then analyse and explain the results of the comparison by applying the concept of collective guilt, and subsequently discuss conclusions gained

through the literature review, and the analysis of collective guilt. In this introductory chapter I will put the research problem into context, explain the study’s purpose, and present my research questions and delimitations. In the end of this chapter I will give a more extensive outline of the study.

(9)

Indonesia

Source: Wikimedia Commons (Map of Indonesia).

1.1 Contextualization of the Research Problem

The conflict in the Moluccas is often simplified and portrayed in terms of animosities between Christians and Muslims (HD Centre 2011: 16). However, the reality is far more complex. This study is contrasting previous research with the perceptions of local youths’1 and analyses their memories with the theory of collective guilt. To reach an understanding, I stress the

importance of addressing both the historical and systemic roots, as well as more present escalation factors, thus the youths’ memories can be explored through different angles. Beneath the calm surface of stability and economic growth of General Suharto and the New Order regime (1967-1998) was often widespread state violence to be found. The New Order roften repressed violent manifestations of tensions without dealing with the underlying causes of concern to the local population. After 1967, tensions within the archipelago often laid latent or simmered until the fall of the regime (Brown et al. 2005: 5-6). The fall of Suharto and its aftermath were bloody. In May 1998 the Indonesian capital city of Jakarta became a place of unrest, along with many other places within the archipelago. This violence2 was surprising to many Indonesians; partly because it involved shocking cruelty, but more

1 The label ’youth’ is socially constructed and historically variable (McEvoy-Levy 2006: 3). The youths in this

study were all under eighteen at the time of the conflict, but at the time of this study in the age between 17-26.

2 Schultze (2003) describes characteristics of the conflict as: “frenzied slaughter, savage mutilations, forced

(10)

importantly—it involved citizens versus citizens. According to Bertrand (2004: 133), the case of the Moluccas demonstrates the first time in Indonesia where Christians and Muslims were pitted against each other in such sectarian warfare. In the separatist conflicts it was easier to identify conflict drivers, but these new communal conflicts have a very different and more complex character (Van Klinken 2007: xviii). Across six provinces in Indonesia that experienced large-scale extended violence, at least 15,967 people were killed between the years 1998 and 2002 (Barron et al. 2012: 1). The conflict within the Moluccas was one of the most devastating (HD Centre 2011: 16). When the conflict erupted in January 1999, the Moluccas were still one province (Maluku province), but in October 1999 the government decided to split the province into North Maluku province, and Maluku province.3 They lie in eastern Indonesia, between Sulawesi and Papua. About 90 percent of the

provinces are maritime and consist of over 1,000 islands. The capital of North Maluku is Sofifi, while Ambon is the capital of Maluku Province (Brown et al. 2005: 8-9). A peace accord4 was signed in February 2002 marking the official end of the conflict. The current situation could be described as being in a post-conflict phase. Nonetheless, tensions remain high within many communities and occasional deadly violence and looting still occurs (Azca et al. 2012: 14). The fall of Suharto and the subsequent transition from autocracy to

democracy is just one of many underlying factors which have influenced the conflict dynamics.

1.2 Purpose Statement and Research Questions

This study has three purposes. The first purpose is to contrast previous research concerning conflict factors in the Moluccas, to the memories and experiences of the Moluccan youths’. There is a research gap regarding the local youths’ memories of the conflict, which this study aims to help bridge. Because the youths are university students by now, they have potential to hold important positions within the society in the future. Furthermore, their perceptions of the conflict and its factors can provide some insights regarding the future of a sustainable peace in the region. These aspects are strongly related to Peace and Conflict Studies. The inclusion of youths is often neglected when conflicting parties make peace agreements, as is the case in

3 I use the term ’the Moluccas’ to include both provinces. 4 The Malino II Agreement.

(11)

the Moluccas. In my view, youths are an important part of the population to include in any attempt to build peace in a society. It has also become a United Nations policy to promote such inclusions (McEvoy-Levy 2006: 3).

The second purpose is to analyse similarities and differences between the academic literature and the youth’s bottom-up perspective. The last purpose is to apply the theory of collective guilt to explain and analyse the youths’ memories and perceptions.

To operationalise these above stated purposes in the analysis, I use the following research questions;

 What does previous academic research tell about the conflict factors in the Moluccas?

 What are the Moluccan youths’ perceptions and experiences regarding the conflict factors mentioned in previous research?

 Is there a discrepancy between these two perspectives?

 How can the theory of collective guilt help explain the memories and perceptions of the local youths’ regarding the conflict factors in the Moluccas?

1.3 Delimitations

Several delimitations have been made to improve the quality and feasibility of this study. As mentioned above, several conflicts erupted in Indonesia in the aftermath of the Suharto era. A comparative analysis of those would be interesting, but would be too extensive to fit the frames of this study. Therefore, all light is shed on the conflict in the Moluccas, as a case study. Another delimitation is the focus on the youths’ experiences and perceptions, which means that the primary sources are limited to six in-depth interviews conducted in Indonesia. A final delimitation is the choice to use the theory of collective guilt to analyse the youths’ memories and perception of conflict factors, in relation to the secondary sources.

(12)

1.4 Outline

This study consists of seven chapters. The opening chapter starts by contextualizing and problematizing the research problem. Subsequently follows a presentation of the aims of the study and a presentation of the research questions. Then follows the delimitations I have made for the study. Chapter two presents an overview of the conflict and post-conflict scenarios. It also includes a map of the Moluccas and a demographic profile of the two provinces.

The third chapter begin with motivating the chosen methodological approach. Then the method of in-depth interviews is presented, as well as an assessment of secondary sources. Chapter four is designated to the theoretical framework and the theory of collective guilt is presented, which will be used in the analysis chapter.

In the fifth chapter the review and analysis is undertaken. Previous research is contrasted to the perspective of the local youths’ regarding conflict factors. Subsequently follow an analysis of the similarities and differences found between the two perspectives.

Chapter six explores if the theory of collective guilt can help explain the discrepancy between the youths’ perceptions and previous research. In chapter seven the conclusions of the study is presented and discussed. The chapter ends with a discussion of possible avenues for future research relating to the research topic.

(13)

2. Conflict Overview

This overview initially displays a map of the Moluccas, where the main islands are portrayed. Subsequently follow a description of the demographic profile, where the two separate

provinces are outlined. Then an overview of the conflict, the peace agreement and post- conflict phase is presented.

The Moluccas

(14)

2.1 Demographic Profile

According to the census in 2000, North Maluku province had a population of approximately 670,000, with about 85 percent Muslims and 14 percent Protestants. In Maluku Province however, the demographics are evenly split (49.1 Muslims and 50.2 percent Christians), which is fairly unique in Indonesia. Included in these numbers are the migrant populations which are 9.1 percent in North Maluku, and 6.6 percent in Maluku. The majority of the immigrants are Muslims from Java and Sulawesi.

2.2 Overview of the Conflict in the Moluccas (1999-2002)

The conflict is sometimes referred to as a communal war. With nearly 5,000 lives claimed, and the displacement of nearly a third of the population of Maluku and North Maluku, I can understand and agree with that label. The violence in Maluku broke out on the last day of Ramadan, 19 January 1999. It was ignited by a dispute between an Ambonese Christian bus driver and an immigrant Bugis Muslim passenger.5 The initial violence that followed lasted for about two months, in and around Ambon, and resulted in the death of more than a thousand people (Brown et al. 2005: 15; Barron et al. 2012: 10).

In this stage, the targets of violence were predominantly Muslim migrants6 from Sulawesi Province (Bugis, Buton and Makassar). In July the same year, violence erupted again and also spread to other parts of Maluku and lasted until January 2000. The capital city of Ambon was now divided along religious lines. Later in May 2000, the conflict dynamics changed for two reasons: security forces reinforced both Muslims and Christians, but mainly the Muslim side. Further, the Muslim side was reinforced by Jihadists, predominantly from Java, joined the newly-formed Laskar Jihad militia7 (Mulyadi 2003: 80). This shifted the power-balance in the conflict to the Muslims favour. In June 2000 the government issued a civilian emergency status and thousands of army and BRIMOB were sent to the islands. Additionally the police

5 The initial violence was mainly between Muslim and Christian youth in the area called Batumerah, in Ambon

City. For further chronological details see ICG (2000), Klinken (2001, 2007), Azca (2003) and Crouch (2010).

6 This migrant group had achieved a relatively dominant power in Maluku’s informal labour market, which had

generated resentment by some in the local community (HD Centre 2011: 17).

7 Founded by the neo-Salafy fundamentalist Ja’far Umar Thalib in direct response to the conflict in Maluku. It is

the militia wing of Thalib´s Forum Komunikasi Ahl al-Sunnah Wa- al-Jama’ah, FKWAJ. Thalib preached a conspiracy theory where the United States, Israel and Christians in general are seen as the masterminds of the conflict in the Moluccas. Further readings on Laskar Jihad and Laskar Kristus see: Mulyadi (2003) and Schulze (2002).

(15)

and security forces were given increased powers, i.e., curfew (Brown et al. 2005: 16). Several sources give testimony which indicates strong partiality and co-ordination problems within many security forces (Bertrand 2004: 127). Yet, after the heavy reinforcement, violence started to decline in the Moluccas. Since that time Maluku Province as a whole did not return to major conflict, but Ambon and some other areas did experience occasional outbreaks of communal violence, and tensions remained high even long after the Malino II agreement which was signed February 11th, 2002 (HD Centre 2011: 21).

Still up until today tensions are evident in parts of the Moluccas. In North Maluku the conflict erupted later, but in just eleven months 3,257 people got killed. The initial fighting here, had a more ethnic character compared to Maluku Province (Barron et al. 2012: 11). Fighting began in August 1999 with clashed between indigenous Kao and migrant Makian. This initial violence was a result of several months of tensions related to the creating of the new sub-district of Malifut,8_{on Halmahera island. According to Wilson (2005: 89);}

Kao Muslims and Christians maintained ethnic solidarity in the face of what was ultimately ethnic antagonism, Kao Muslims and Christians joining in retaliatory attacks against Muslim Makian […] the Kao response in late October was largely free of religious symbolism.

However, Wilson’s explanation only covers the initial clashes in Kao-Malifut, where politics and struggles over resources did play a role. Yet, outside Kao-Malifut, other conflict drivers were at work. In October 1999 the fighting spread to other parts of North Maluku and took on a more religious undertone. It was influenced by the violence in the Maluku Province in

general, and by the arrival of Muslim Makian IDPs in particular. Moreover, propaganda and conspiracy theories gained ground, in the absence of coverage by the media. The area’s remote location, lack of infrastructure, and also a degree of conflict fatigue in the press can explain that (Van Klinken 2001: 5). A widespread view was that Jakarta elites were behind the violence,9 but this theory has been less and less convincing over time.10 Violence continued up until June 2000. At that time several factors combined which helped put an end to the most deadly violence in North Maluku Province e.g., imposition of civil

emergency, extra troops, conflict fatigue, and a stalemate between the two sides. North Maluku has remained relatively peaceful since, compared to Maluku province (Barron et al. 2012: 11).

8 Important factors here were issues of territory, natural resources and ethnic solidarity Wilson (2005: 89). 9 It was based on theories regarding the military being unreformed and still financed by Suharto and his Golkar

Party. This theory suggests that the military had incentives to destabilize the newly democratic government by the means of communal war, and then reinstate military rule.

(16)

2.3 The Post-Conflict Scenario

In Maluku, episodes of large-scale violence have been numerous, even after the peace agreement. In contrast, North Maluku experienced far less outbursts of violence.

After the Malino II agreement, there was a widespread feeling within some communities that the agreement lacked credibility. This became especially evident when the Muslim

negotiation leader’s house got burned down, after the signing of the agreement (HD Centre 2011: 26). In the subsequent months after Malino II, violence still occurred periodically (Bertrand 2004: 133) with the majority concentrated to Ambon city e.g., the bomb thrown into a crowd in Jan Paays street in the Christian area, which killed seven people and injuring 56 (Barron 2012: 18).

In 2004, another serious outbreak of violence happened when more than 40 people died following riots in Ambon. This happened after Alex Manuputty, a well-known Christian leader of the FKM, had the RMS flag raised at his home (Brown et al. 2005:17).

FKM was seen as a cover for a revived RMS, which was a very sensitive issue in Maluku. Even outside of Ambon, communal conflicts have escalated into violence. A ICG (2012: 5) report concludes;

[…] without more attention to the systemic roots of violence in Ambon – segregation, poor policing and lack of transparency in investigations, to name three – the eruption on 11 September [2012] can easily happen again.

(17)

3. Methodology

In this chapter I will motivate my methodological approach, explain the use of in-depth interviews, reflect on my role, explain the use of an interpreter, discuss ethical aspects of the study and finally follows a part which includes an assessment of the secondary sources. In this study I have chosen to use a qualitative design with a case study approach, and in-depth interviews as a method to explore the bottom-up perspective of the respondents’ perceptions.

3.1 A Qualitative Case Study

With this methodological approach I am able to take advantage of the receptiveness to interdisciplinary strategies within my field of study (Wallensteen: 2011: 17; Denscombe 2000: 43). Since much of the research concerning the conflict in the Moluccas have been quantitative, it will be interesting to contrast some of the previous research with a qualitative case study approach. Moreover, I argue that this is an intrinsic case study (Creswell 2013: 98) since it involves a unique case. In my view, all conflicts are complex and include distinctive factors depending on the specific history and other contextual differences. The case study approach has many advantages i.e., different forms of qualitative data collection can be done and the researcher can identify themes or issues to study within the case (Creswell 2013: 98) e.g., the analysis of conflict factors in the Moluccas. To make use of the above stated

advantages of the approach, I have chosen to include different types of data collection

approaches in the study; both in-depth interviews and a critically assessed literature review. An ethnographic approach could also have been an interesting design for this study.

However, since extended observations on the spot in the Moluccas have not been an option for this study, I argue that the case study approach is a more feasible choice.

(18)

3.2 In-depth Interviews

This method promotes the importance of individual views which add insights into processes that cannot be fully understood through statistically documented facts and trends. I adopted this method because it is a well-suited method if one wants to explore the perspectives of the respondents, in relation to the same research problem. Therefore, it is important to cover the same themes in all the interviews, hence common themes can be analysed subsequently. Yet, it is essential to follow up on related themes that are brought up by each respondent. Further, using open-ended questions permits the interviewee to choose how extensive each answer will be (Brounéus 2011: 130). These aspects make each interview unique, even though the same interview guide is used. It is the respondent’s views and thoughts that are central to the analysis (Esaiasson et al. 2007: 258).

3.2.1 Selecting Interviewees

Six in-depth interviews were conducted in Yogyakarta, Indonesia, as well as one interview by e-mail. The in-depth interviews were conducted with students from both the Christian and the Muslim community, sharing the common denominator of growing up in the Moluccas during the violent conflict. Moreover, both male and female students are included with the purpose to gain insights from a gender perspective.

Furthermore, since they are university students now, they have the potential to hold important positions within the society in the future i.e., their experiences and outlook on the future is of interest to my study, even though the main aim is to explore their views on the conflict factors. Their ages were between 17 and 26 years old at the time of the interviews, which I see as a preferable age sample also in an ethical perspective, compared to

interviewing younger children for example. To find respondents for the interviews I put up information posters on several university campuses in Yogyakarta, where I had summarised my study, informing possible participants about anonymity and providing my contact details. After I had found a few respondents in this manner, I could perform snowball sampling to find additional respondents as well. In this selection process, I was always guided by the principle of ‘credibility’: what sources will provide best reliability and validity of the results? This principle further stresses the importance of the interviewees to be experienced and knowledgeable in the area of research (Brounéus 2011: 134). However, since my interviewees were ‘respondents’, not ‘informants’, it is their subjective experiences, thoughts and

(19)

had grown up and lived in the conflict environment, but they do not need to be ‘experts’ in regard of the conflict factors because they will not be used as sources in that sense (Esaiasson 2007: 258).

3.2.2 Conducting In-depth Interviews

I conducted the interviews during the fall semester 2012, when I was doing my internship at CSPS. I conducted two interviews at the respective university campuses of those respondents, since they suggested that and felt comfortable in that environment. The remaining interviews were conducted in a café/restaurant in Yogyakarta called Soma Yoga. I chose that café because I know from experience that it is low-key in the afternoons, and keeps traffic noise away because of the surrounding rice fields. It turned out to be a good milieu overall. The minor disadvantage of the location turned out to be the infrequent, but high-pitched, sound of grasshoppers. However, that was only a small distraction during the transcription process. I always took time to talk to the interviewees before we started the interview. I recorded all interviews and I always asked about permission to do so from each interviewee. If they wanted, they could look at the interview questions, or ask any question about the study. I had one respondent who felt more comfortable doing the interview in Indonesian, which was fine. I had stated in the information that this was an option and that I would have a translator

available, which I always had. Although I used the same interview guide, the interviews lasted from 32 minutes up to 83 minutes. This shows that each interview follows its own path, exploring follow-up questions on related themes (Brounéus 2011: 134). When conducting the interviews I paid special attention to use ‘reflective listening’. That includes empathic ability, listening-skills, and to make ethical judgements regarding when to stop asking about a certain issue (Brounéus 2011: 137).

3.2.3 Ethical Aspects, Interpretation and My Role

This study involves many aspects which might be of sensitive nature: the respondents are interviewed about their experiences and thoughts concerning an often violent conflict. Therefore, I have made several considerations in honour of the golden ethical rule; do no harm (Brounéus 2011: 141). First, I always made it clear to the participants that their participation was completely voluntary, and that they could cancel their participation at any time. That is also part of fulfilling informed consent (Kvale 1997: 107). This was explained both in English and Indonesian, so I could be assured that informed consent had not been misunderstood. Second, I had my interview guide and themes approved by my supervisor at

(20)

CSPS, Dr. Najib Azca, who have done extensive research in the Moluccas. Third, I made sure that the respondents had understood the basics of the research project before each interview and asked their approval of recording the interviews.

Furthermore, some small-talk before the actual interview is beneficial because people tend to feel more relaxed. Creating a comfortable atmosphere where the respondent feels safe and comfortable is important both in the way of conducting a fruitful interview, as well as

conducting ethical research (Brounéus 2011: 136). Anonymity for the respondents is another vital aspect to protect the respondents who volunteer to talk about sensitive subjects.

Therefore, I insisted on keeping all interviews with the Moluccan youths confidential. This includes fabricated names in the transcriptions as well.

My own identity traits i.e., gender, age, nationality, class, skin colour and how I dress may influence the information being shared by the respondents in this study (Höglund 2011: 124). My background and identity also influence my own perceptions, as well as being an outsider in the eyes of the respondents. Therefore, I stress the importance of being aware of these aspects, and especially transparent about them in the study. However, being viewed as an outsider can be positive in some instances. Kristine Höglund (2011: 124) argues,

[…] the outsider may benefit from locals’ perceptions of a ‘stranger-value’. The outsider researcher is in many cases seen as a sympathetic and interested person and is for this reason well-received […] sharing information with outsiders can be easier because they will not remain in the area or in the country.

In my view, I feel that I benefited from being an outsider in some ways; I felt I was being perceived as an interested outsider which the respondents felt they could easily talk to. Moreover, the fact that the conflict is currently in a relatively calm post-conflict phase, made it less sensitive for the respondents to reflect about the situation. Yet, one of the female Muslim interviewees only wanted to answer the questions by email, which may be linked to my identity as being, compared to her, foreign, white, slightly older and probably assumed of being a non-Muslim etcetera.

During all interviews I was working together with a translator, which involves both

potential benefits and challenges. The perceived identity of the translator has the potential to influence the respondents in a similar way as my identity (described above). My translator was female, Muslim, 23 years old and was born on Sumatra. Therefore, she may also have been considered an outsider, compared to the interviewees from the Moluccas. Additionally, she did not wear any religious symbols during her work, so the respondents did not know her

(21)

religious affiliation. She had no active role in the study, apart from translating and transcribing the Indonesian part. However, I paid great attention to explain the aim of the study, the interview questions, protection of the interviewees’ anonymity, and the importance of being an empathetic listener. Since I know my translator well, there is good rapport and understanding between us. Moreover, my own intermediate knowledge of Indonesian further helped me to understand much of the context, even when the respondent spoke Indonesian. There are also benefits from working with a translator: translations make the pace slower which give time to pay closer attention to the interviewee. When communicating, people use other signs than just words i.e., how we sit and look also tell some information, between the lines (Brounéus 2011: 136). Even though I always had the translator present, all interviews but one was conducted in English, since the interviewees’ had a good level of English and were comfortable speaking it. Yet, during almost all interviews, the translator could provide help with translation of certain words and expressions.

3.3 Assessment of Secondary Sources

The secondary sources, which is the academic literature included in this study, do need to be assessed however. The previous research has been used with multiple aims. First, it was used to locate the research gap which motivates the study in the first place. Second, it was used to build an understanding of the historical context and conflict factors in the Moluccas, which is needed to be well prepared for in-depth interviewing. Finally, the previous research has been reviewed in order to contrast that research to the memory and experiences gained from the respondents. When evaluating secondary sources I have paid attention to the following factors (Dulić 2011: 38; ALA 1994: 1);

 What type of source is it (book, report, journal etc.)?

 Who is the intended audience?

 Who authored the document (i.e., his/her academic background)?

 Is the author a third party or part of the conflict in any way?

 When was the source produced?

(22)

 Does the source contain a bibliography?

 Has the material been peer-reviewed or edited?

Further, this process was gradually made easier since one builds an understanding and is able to compare different sources that are independent from one another. This process made me exclude any sources relating to news agencies from the conflict area, which is known to be controlled by the conflicting parties, and/or unprofessional and biased in their reports of the violent conflict. I found it especially useful to study the references of the

books/articles/reports to evaluate their own sources, where many became familiar to me subsequently. Furthermore, most of the journal articles, books and reports included in this study have been peer reviewed which adds to the credibility of the secondary sources.

(23)

4. Theoretical Framework

This study is not theory-driven. Instead, the theory of collective guilt has been adopted using progressive focusing. The reading through the data and continuous interpretation of the material is important in qualitative data analysis since it enables the researcher to gradually refine the focus inductively (Chambliss & Schutt 2006: 195).

I find it motivated to include this theory since it relates well to the stated purposes of the study (noted above). This theoretical framework is a useful tool for analysing similarities and differences by operationalizing the theory of collective guilt. Moreover, the insights gained from the literature review regarding the conflict factors will also be used to explain discrepancy between previous research and the perspective of the respondents’.

4.1 Collective Guilt

Our memory has long been known to not only conserve the past, but to adjust itself to current needs in certain situations. Susanne Langer (1953: 263) writes that: ‘Memory is the great organizer of consciousness’. This is a process that is done to make history conform to memory according to David Lowenthal (1975: 27). Reminders of history in the context of belonging to a certain group may influence our present-day emotional experience.

For example, both collective pride can be evoked, as well as collective guilt, depending on the aspect of history. A desire to correct the wrongs committed by the in-group may follow when associated to a feeling of collective guilt (Branscombe & Doosje 2004: 3). An

interesting aspect of collective guilt is that personal involvement is not necessary, which is one of the reasons I have integrated this theory in the study. Many of the interviewees were

(24)

not personally involved in the violence during the conflict in the Moluccas. However, the important aspect is instead the association with the ‘in-group’. Nyla Branscombe and Bertjan Doosje (2004: 3) explain that;

[…] people’s reactions to intergroup events are not usually based on having had a direct role in those events; rather such reactions will be dependent on the social identity that is contextually salient when people are reminded of their in-group’s past actions.

Therefore, a person does not need to be ‘guilty’ of anything else other than the psychological thought of associating oneself with a group that has wronged another group; it is a self- conscious emotion. That is an important difference compared with being ‘guilty’ in a legal sense. An interesting insight, in my view, about collective guilt is that it can be a rare

phenomenon. This is because there are several strategies that are group-serving, with the aim of legitimising in-group’s actions. Michael Wohl et al. (2006: 4) writes;

Once an instance of intergroup harm has been made salient, group members have a variety of means available for defending against this threat to their in-group’s identity […] Accepting responsibility for in-group harm doing, either in the past or in the present, entails the incorporation of negative elements into the group’s collective identity, and this is opposed to group members’ desire to maintain a positive group image.

There are three general techniques stating that judgements of responsibility can be deflected and collective guilt can be reduced (Wohl et al. 2006: 13);

(1) Put focus on the victimised out-group’s experiences instead of the victimising in-group’s harmful behaviour;

(2) Put the blame on the out-group for the harm they experienced;

(3) Placing blame for in-group harm on a few ‘black sheep’ who do not represent the ‘usual’ in-group member, to minimise the flaw to the in-group as a whole.

Further, the literature about collective guilt says that in ethno-political conflicts for example, collective guilt may occur without close relations between the offender and the victimised group, in contrast to an interpersonal level, where the feeling of guilt is more likely in close

(25)

relationships (Wohl et al. 2006: 4). In the violent conflict in the Moluccas, both past and present perceived experiences of injustices between the two communal groups are among the conflict drivers. According to the literature concerning collective guilt, on-going group-based inequalities, or ethnic discrimination, may also cause collective guilt (Wohl et al. 2006: 27). It could similarly be the experience of profiting from existing injustices in the society, even though that person did not have any role in creating those inequalities (Branscombe & Doosje 2004: 6). To feel collective guilt might be unpleasant, but it has the potential to lead to

something good (Wohl et al. 2006: 3);

When responsibility for harm is accepted, the guilty individual may be motivated to take action to alleviate the wrongdoing or unfair treatment. In response to feeling guilt, individuals may apologise and seek forgiveness, and they may attempt reparations to the extent that the self is perceived as morally responsible.

In that sense the acceptance for collective guilt may have positive impacts on intergroup relations, and peace and reconciliation in the long run. There are four main psychological factors that are essential for evoking collective guilt. The degree of collective guilt

experienced is influenced by the extent the following factors are true (Wohl et al. 2006: 9);

(1) If the person categorises herself or himself as a member of the harm-doing group; (2) Self-categorised group members recognise their group to be responsible for the action

considered to be harmful for another social group;

(3) Members of the group perceive the harm done to be immoral or illegitimate;

(4) The extent to which collective guilt will be felt depends on the perceived difficulty and costs to the in-group of correcting the erroneous action committed.

Furthermore, if members of the group that have done harm to the other group adopt a more inclusive categorisation where the members of the victimised group are seen as fellow members, the experience of collective guilt may be more likely (Wohl et al. 2006: 11). This will be interesting to analyse further in the context of the conflict in the Moluccas, which divided the Moluccan inhabitants into religious in-groups. If this process of re-categorisation should be made possible, it usually requires a common standard of justice to be used to the now included ‘in-group’ members (Wenzel 2001: 315).

(26)

and even as a way of causing harm on the in-group. The post-conflict stage in the Moluccas thus provides a better case to study, compared with a more intense conflict. There are several obstacles that potentially hinder the experience of collective guilt (mentioned above). These strategies of trying to ‘escape’ collective guilt are of interest to my future analysis too, when I seek to explore my interviewees’ perspective of conflict factors. Besides, the potential

positive impacts on reconciliation processes are also of great interest when applying this conceptual lens in a conflict milieu. Powell et al. (2005: 510) argues that;

We showed that when collective guilt is expressed, there is also a reduction in prejudicial attitudes towards the harmed out-group. When collective guilt is coupled with a public apology, forgiveness by the out-group for the in-group’s harmful actions is more likely, making intergroup reconciliation a stronger possibility.

(27)

5. Analysis: Previous Research versus the

Youths’ Perceptions

This analysis consists of two parts. Initially I review and analyse what previous research tells in regard of the conflict factors, and contrast that picture with the perceptions and memories of the local youths’. Then follows an analysis of similarities and differences between these two perspectives.

5.1 Religion as Identity and as a Conflict Factor

The conflict in the Moluccas is often described in terms of antagonism between Muslims and Christians. The reality is far more complex, but it cannot be ignored that the conflict was almost entirely dividing the parties along religious lines (Brown et al. 2005: 10).

5.1.1 Review and Analysis of Previous Research

The division of the local population of the Moluccas is far from a new phenomenon. To understand this division, one needs to explore how Islam and Christianity have influenced the societies. Arab trading colonies had been set up in the China Sea as early as the seventh century (Wolf 1997: 234). Around 1460 the ruler of the island Ternate, in today’s North Maluku province, converted to Islam (Taylor 2003: 66). ). When the European colonial powers arrived they entered a domain of hegemonic Islam (Wolf 1997: 234). It was the

(28)

trade in cloves, nutmeg and mace, which at the time, was only to be found within the Moluccas11 (Van Gils 1994: 117). The Portuguese seized control of the important trading islands in the early 1500s and remained the dominant power until they were expelled by the increasingly powerful Dutch,12 nearly a century later (Van Gils 1994: 117). The Portuguese introduced Catholicism to the Spice Islands, and the Dutch Protestantism. The English also tried to cut in on the famed spice trade, e.g. by trying to seize the Banda islands in the 1619, but failed at that attempt13 (Wolf 1997: 240). The most important aspect is the fact that as early as 1650, complex horizontal conflicts were already apparent, splitting the local

population alongside commercial, factional and religious lines (Bertrand 2004: 115). By the end of the 17th Century, the Dutch colonial power was dominating the region, with Ambon as their main centre of operations. The Dutch colonial administration created an evident social stratification along religious lines by treating the two communities differently, e.g. local Ambonese who had converted to Christianity could get good education etcetera. This was denied to Muslims, and also people who followed native animist religions. Hence, the newly- converted Christians started to control the bureaucracy, as well as civil services (Brown et al. 2005: 10).

5.1.2 The Youths’ Perceptions

All respondents in the study confirm that an affiliation to either Islam or Christianity is part of their identity. Yet, somewhat surprising, none of the respondents talk straightforwardly about any major remaining negative affects resulting from the Dutch colonial rule.

However, religion seems very important to some of the youths, like 21 year-old Riri: “I am Catholic […] it is the strongest link to my self-identity”. Another respondent, Pat who is 24 and from the Southeast Moluccas, who also says he is Catholic, describes his own family’s integration with the other community before the conflict: “For example my sister married with a Muslim and we were all family like that. Before, Confucian, Muslim and Christian people lived together”. Furthermore, all respondents said that they had friends from the other religious community, and still do today. Yet, they also express that the schools were usually segregated depending on religion, as were many neighbourhoods even before the conflict erupted. The respondents confirm the literature as describing the segregation as worsening

11 The reason these spices was so valued was because they were extremely rare and was needed as preservatives

for meat and fish, as well valued exotic flavours of Eastern cooking.

12 Much because the power of the VOC, who gained a complete monopoly of the entirety of Dutch trade. 13 It was not until 1796 when the English finally could break the Dutch monopoly of the spice trade, by obtaining

(29)

when the conflict erupted. Dani, a 26 year old Catholic, expressed the changes in his neighbourhood and view of the other religion as follows;

In my neighbourhood [in Ambon] lived many ‘immigrants’ from Southeast Moluccas, Buton, Bugis, Manado and Java. I had so many friends from different ethnic groups and another religion […] I saw many of them for the last time when my father told them that they had to leave the neighbourhood because he [a Christian militia commander] could not protect them anymore. When the conflict began I did not know so much, but some people were yelling things about religion, […] when I was a teenager [when the conflict erupted] I started to think that I had to protect my religion, the identity.

When the respondents were asked about what role religion had as a conflict factor, they all gave answers pointing in the same direction; downplaying the role of religion as a factor causing the violence. Yet, like Dani’s quote reveals, most soon got divided from their friends with a different religion and caught up in the common rhetoric which depicted the other religious community as an enemy.

Gino, 23, who is a Muslim from North Maluku province said: “When the conflict erupted in 1999, I was a child at that time, all Muslims in my district, Sanana, saw the other religion as an enemy”. The youngest respondent Gary, 17, and a Protestant from Ambon, gave the following answer when asked if the conflict changed his relationship and view towards the other religious community;

I did not change my way of thinking about the other religion because my family has always taught me to look and treat all people the same. Even if we are different, we are all humans but with different beliefs maybe. If people have different beliefs, so what?

5.2 Different Loyalties Towards the Indonesian Nation

When Indonesia fought the Netherlands in the war of independence14 (1945-1949), many Christian Ambonese choose to side with their former colonial rulers, and even supported campaigns to pacify the republicans in other parts of the archipelago. This part will review and analyse if this question of loyalty still remain as a conflict factor.

(30)

According to the literature, different loyalties to Indonesia was a sign of the very beneficial treatment Indonesian Christians had received under the Dutch. It also reflected a fear of domination of Muslim Javanese also in the Moluccas. Under the Dutch colonial system, the Moluccas were part of the East Indonesian state. When it became clear that the Dutch were losing the war, Dr. Soumokil, the East Indonesian Minister of Justice, declared the

independent RMS, in April 1950 (Brown et al. 2005: 11).

This declaration of independence was not accepted by the new Republic of Indonesia and was eventually defeated with military force. Yet, the RMS is still a symbol and factor of political and religious division in Ambon. The integration of Maluku into the state was also met with long resistance and Malukan leaders kept in contact with other rebellious

movements in the archipelago, i.e., PRRI in Sumatra and Permesta movement in Sulawesi, which opposed Sukarno´s regime until 1961.

Therefore, Sukarno sought to encourage Maluku to integration by allocation of resources intended for different development projects. Further, not all Christians in RMS had supported the Dutch and those who had instead supported the Republic of Indonesia, were awarded by Sukarno, typically by different roles in the central government (Brown et al. 2005: 11).

None of the respondents expressed any loyalty or support for the Moluccas to be independent from the Indonesian nation. Most had little to say about this topic, but some identified the RMS movement to still be a source contributing to the continuing violence in the region. Gino argues;

[…] the riots in Ambon are sometimes identified with the movement of RMS. There are a couple of RMS leaders who want the independence of Maluku, this is screwing up the situation […] every year they always make terror like bombing or putting their flag up when they claim the independent RMS […] they are like OPM in Papua who wants independence, and a lot of their leaders now stay in the Netherlands.

(31)

5.3 Autocracy, Islamisation and Transmigration

This part will review and analyse General Suharto’s rule and his Islamisation policy in the Moluccas, as well as transmigration effects.

When Suharto came into power the Moluccas lost many of the few privileges it had received under Sukarno. Changes were also made politically as Suharto appointed governorships to external military officials. Furthermore, any actors within the civil society who bothered the regime were arrested and often accused of being an RMS sympathizer (Brown et al. 2005: 12). Moreover, the heavy-trafficked shipping routes to the Moluccas were redirected away and the intra-island trade of the eastern part of the archipelago was re-arranged. In the northern part of the province, the centralisation of power in Jakarta made the impact of the rivalry between the sultanates of Tidore and Ternate weaker (Brown et al. 2005: 12). The acceleration of religion as identity has its roots in the New Order’s purge against the Indonesian communist party (PKI). Indonesians without a stated religion were automatically suspected as being communists. Hence, every Indonesian citizen was required to carry identity cards stating their religion. And not any religion of choice, but only the government approved religions: Catholicism, Protestant Christianity, Buddhism, Hinduism and Islam. Therefore, thousands of people, communists as well as other people who were atheists, converted to any of these religions, in fear of being punished for being an atheist (alleged communist)

(Bertrand 2004: 74). In Maluku province, where the population was almost split evenly between the two communities, these policies became a source of growing tensions. Further, the once privileged Christian community in Indonesia in general, and in Maluku in particular, became increasingly marginalised compared to before. This was an effect of the late

Islamisation of Suharto’s New Order. One of the most significant changes during Suharto’s Islamisation was the creation of ICMI, which signalled the departure from the regime’s previous policy of denying political platforms to Muslims. According to Bertrand (2004: 114), the conflict in Maluku “was a consequence of New Order policies that disrupted the balance of power between the two communities”. Further, patrimonial networks also reinforced group identities, as influential and profitable positions within the civil service became a source of competition between the religious communities. Therefore, even before 1999, religious identities in Maluku were highly politicised and the groups were poised for conflict (Bertrand 2004: 114). The official transmigration policy under Suharto, which

(32)

encouraged people to move from overpopulated islands (i.e. Java), to less populated ones, had impacts for the Moluccas. Together with spontaneous migration, it changed the fragile ethnic and religious balances. The majority of the migrants and transmigrants were Muslims from Java, but also Muslims from Sulawesi. It was not just a demographic change, but also a social and economic aspect: the transmigration policies entitled them to communal land which had traditionally been owned by local communities. In some places infrastructure projects

intended for the transmigrants displaced or negatively affected the local communities (Brown et al. 2005: 12).

The respondents did not state the long autocratic rule under Suharto as a conflict factor. Similar for all the youths’ answers were likewise that they do not describe any tensions towards the migrant communities before the conflict erupted, unlike the literature. The same trend is mostly accurate for what the literature describes as Islamisation in the region. However, both Pat, and Mimmi a 23 year old Muslim from Ambon, argued that inequalities between the religious communities did have some influence. Mimmi said;

At first in the conflict yes [effect of inequalities]. Maybe it [the conflict] was about why some people got important positions in the government, […] but I think it is equal between Muslims and Non-muslims now.

When I asked Pat about if Suharto’s Islamisation policies had affected, as a conflict factor, he replied: “Yes, I think it did, even if I think it was not a big problem”. Dani had a different view compared to the other respondents regarding Suharto’s Islamisation: “Rather it should be called Javanisation because so many Javanese people come to take jobs in Ambon. In the Suharto era just one Governor was from the Moluccas”.

5.4 The Peace Agreement and Role of the Government

This part includes a review and analysis of the Malino II agreement and further analyses the Indonesian government’s role according to the different perspectives.

The national government had long failed in its reconciliation attempts and was even accused of neglecting the conflict by many. A more resolute attempt was initiated under the backing of

(33)

the Coordinating Minister for People´s Welfare, Jusuf Kalla, to repeat the successful Malino peace talks.15 On Kalla´s initiative the Governor of Maluku got to select one representative from each side in the conflict, and then that representative selected a larger group of

representatives from each community (HD Centre 2011: 20). After three days of negotiations in Malino, the Malino II peace agreement was signed on February 11th, 2002.16 The

agreement looks ambitious on paper and contains eleven specific points, i.e. end all disputes, disarm militias, establish an independent investigation team, call for voluntary refugees to return and to rehabilitate mental health, as well as social, economic and public infrastructure (ReliefWebsite 2002). Graham Brown et al. (2005) writes:

Stakeholders consulted as part of BAPPENAS/UNDP process, confirm that the Malino II agreement retains broad legitimacy in the province and it is still seen as important in bringing an end to the violence by providing a ‘platform for future peace action’. However, as later violent events testify the agreement did not manage to achieve sustained absence of violence.

The weaknesses of the agreement lie not in its content, but rather in how the process went about. Firstly, the process was top-down oriented since the government had already made a ready draft of the agreement. Secondly, more time for the representatives of the different communities to engage in this important agreement would have been beneficial in a long-term perspective. Further, religious leaders were not part of the process. This aspect gave the agreement less credibility with many within the religious communities (HD Centre 2011: 26). In Maluku, episodes of large-scale violence have been numerous, even after the peace

agreement. In contrast, North Maluku did experience far less outbursts of violence. After the Malino II agreement, there was a widespread feeling within some communities that the agreement lacked credibility. This became especially evident when the Muslim negotiation leader´s house got burned down, after the signing of the agreement (ibid.).

In the subsequent months after Malino II, violence still occurred periodically (Bertrand 2004: 133) with the majority concentrated to Ambon city e.g., the bomb thrown into a crowd in Jan Paays street in the Christian area, which killed seven people and injuring 56 (Barron 2012: 18). In 2004, another serious outbreak of violence happened when more than 40 people died following riots in Ambon. This happened after Alex Manuputty, a well-known Christian leader of the FKM, had the RMS flag raised at his home (Brown et al. 2005:17). FKM was

15 The Malino I peace agreement had helped bringing an end to the conflict in Central Sulawesi; see Brown et al.

(2005: 51).

(34)

seen as a cover for a revived RMS, which was a very sensitive issue in Maluku, and still is. Even outside of Ambon, communal conflicts have escalated into violence.

The views of the legitimacy of the Malino II peace agreement differ between the respondents, and some state that they do not know so much about it i.e., Gary. Both Gino and Dani were in favour of the initiative. Gino argues: “[the agreement] was a breakthrough in solving this problem”. However, they both shared the view, which is shared by all respondents, that it took too long for the government to act. Dani said:

I still remember in December in year 2000, President Gus Dur [Abdurrahman Wahid] said that the Ambonese should take care of their situation themselves […] it was kind of like ‘wow’, the President of our state is just letting us like that…

Riri and Mimmi both claim that the [Moluccan] people did more to support peace, compared with the government. Mimmi said: “When the conflict started they [the Indonesian

government] was so slow […] for me the conflict in Ambon did not go back to normal because of the government, but because of the people themselves”. Overall the respondents confirm the picture of a peace agreement which was quite in-effective, as described in most literature as well.

5.5 The Role of the Elites and Provocateurs

Here follows a review and analysis of the different aspects relating to both national and local elites. Provocateurs were often blamed for being behind much of the conflict and this aspect will also be covered here.

The violence in the Moluccas has many connections with local elite’s struggles for resources and important positions and should be viewed in a context of decentralisation and

democratisation, which created great political uncertainty across the nation (Brown et al. 2005: 30-31). Most large incidents of violence need to be organized and the coordination and mobilisation was usually done by elite leaders. So, who is included into this elite? Previous

(35)

research mentions regional politicians, traditional, and religious leaders, as well as commanders of criminal gangs and militias. These elites often exist in many levels e.g. village, district, or province (Barron et al. 2012: 26). One factor that contributed to why elites used violent methods is that there was a lack of accurate information. Both sides therefore continued with violence to show that they were strong, or because they under- estimated the other side. Another reason is that episodic violence is sometimes used to increase the size of the resources being sent for peace-building and peacekeeping operations,17 in the post-conflict period. This is relevant to the conflict in the Moluccas, since episodes of violence continued to erupt many years after the Malino II peace agreement was signed in 2002. Literature regarding war economies has revealed that even when there are various other reasons behind extended violence than materialistic, logics of predation often sees an increase in significance over time. For example it can create opportunities for criminal rent-making activities i.e. ‘protection service’, trade in drugs, minerals, and monopolisation of different economic sectors. In this conflict, local elites have had incentives to use violence since it has paid off in the past, and also because the risk of engaging in violence is still low. This was evident after the peace agreement too; potential trouble-makers have many times been rewarded with important jobs and resources, with the intent of more or less bribing them to become peaceful. Violence has even been used by elites to deter corruption investigations of funds received by the elite, which evidently has worked since some investigations suddenly have been called off, and few big shots have been convicted (Barron et al. 2012: 122). While elite motivations are an important part behind the violence in the Moluccas, it is needed to state that the

motivation on a grass-root level also is an important piece of the puzzle. Without tensions and some acceptance of violence as means for solving conflict, the incentives for elites to use violence would decrease (Barron et al. 2012: 30). In the Moluccas and many other provinces of Indonesia, the elites often use ‘premans’18 which may include gang members, ex-

combatants, thugs or other ‘violence specialist’, to pursue political goals19_{(Wilson 2010:} 200). The process of fast decentralization, that saw important means of power and control get transferred to the district level, further increased the incentives for elites and religious groups

17 More than 630 million USD was spent on international programs to disarm, demobilise, and reintegrate former

combatants in the year of 2007(Barron et al. 2012: 28).

18 Meaning gangster or thug.

19 ‘…these groups [preman] are often involved in racketeering and violent entrepreneurship, both as a service to

supply a ready market for forms of intimidation and in the form of a coercive relationship imposed on their imidiate communities’ (Wilson 2010: 200).

(36)

to gain access to key bureaucratic and political jobs.20 This eventually led to the use of violence as politicians sometimes used militia or thugs to pursue their goals (Barron et al. 2012: 41).

Elites emphasised religious difference and the (potential) threat of other confessional groups to mobilise masses in violent actions, often for self-serving purposes (ibid.).

When some of these underlying motivations became more evident as the violent conflict continued, peace efforts started to include means of pleasing the needs of the elites, which often were divided along religious lines. Both bureaucrats with significant power and local politicians are included into this group, as well as leaders of militia groups. Many of these persons were present in more than one of these categories (Barron et al. 2012: 41-42). The democratisation process that was on-going in all of Indonesia also meant that important positions now had to be obtained through rallying popular support. In the past this has many times been achieved by currying favour with the national elite in the capital city Jakarta. Thus, the motivations to mobilize along religious and ethnic lines increased (Brown et al. 2005: 30). The national elite include members of the army and also economic and political elites linked to the Suharto family. It is clear that the national elite lost much of its former power and access to resources from the decentralisation process. Therefore, it has been widely suspected, but never overwhelmingly proved, that some within the national elite tried to provoke violent conflict in the Moluccas, and other provinces too. By doing so, they would try to reassert the need for a strong central leadership and resources to the military (Brown et al. 2005: 31). A (HRW: March 1999) report mention three categories of theories with relation to immediate causes, which ignited the communal war:

 Provocations by people within the Christian community aiming to revive local domination through bolstering support for the RMS and attacking Muslim migrants;

 Provocation by local military officers through mobilisation of RMS support but with the intent of benefitting from the anticipated Muslim backlash;

20 The law 22/1999 on Regional Government and Law 25/1999 on Fiscal Balance between the Regions and the

Central Government led to a significant transfer of authority, as well as resources, to districts (Barron et al. (2012: 41). Regional government spending went up 15 percent between 2000 to 2002 (ibid.).

The Conflict in the Moluccas: Local Youths&apos; Perceptions Contrasted to Previous Research