Insights and Analysis

In Indonesia, Database Tracking Violence Provides Insights on Preventing Conflict

January 14, 2015

By Patrick Barron

From 1999 to 2008, subnational conflicts (SNCs) killed at least 100,000 people in Asia with half of the countries in South and Southeast Asia affected. Asian SNCs last on average twice as long as the global average and typically reignite after periods of calm. Policymakers and practitioners, trying to break the vicious cycle to move SNC areas from prolonged violence to peace, face two challenges. First, they need to forge peace settlements that can end large-scale violent conflict. Second, they need to craft policies and programs that ensure that the post-conflict period remains peaceful and large violence does not reemerge as it so commonly does.

A great deal of research has focused on what determines whether civil wars can end and the factors that shape the risk that civil war will resume. But there has been much less attention paid to understanding how violence declines and the ways in which some post-conflict areas manage to prevent large-scale violence from reoccurring.

Evidence from Indonesia shows the vital role that improved policing can play in preventing the return of subnational conflict.

Lessons from Indonesia

Examining success stories is one way to generate practical lessons. A new World Bank paper, co-authored by myself, Sana Jaffrey (University of Chicago), and Ashutosh Varshney (Brown), assesses Indonesia’s experience of violence and violence management.

The paper draws on a massive new dataset, the Indonesian government’s National Violence Monitoring System (NVMS), which has recorded almost 200,000 incidents of violence from 1998 to the present. Each incident is coded for where and when it occurred, the type and form of violence and the human impacts, as well as a host of dimensions such as who was involved, who intervened to try to stop the violence, and how successful such interventions were. It is the world’s largest single-country database that tracks violence-related incidents. The data allow for an assessment of how violence has evolved in Indonesia. They also allow us to identify the role that changing patterns of interventions by security forces have played.

Indonesia: a success?

Indonesia – a sprawling archipelago of more than 17,000 islands and 250 million people – saw a number of gruesome SNCs around the turn of the century. In eight provinces alone, around 21,000 people were killed in the years surrounding the fall of the New Order regime (Table 1). Other provinces experienced localized violence. As political and economic change and competition rocked the land, the risk of Indonesia fragmenting Balkans-style was real.


The existential moment passed. Peace accords, military actions, and conflict fatigue led deaths to decline precipitously. As new rules of the game were forged among political elites, Indonesia stabilized and the equilibrium of large subnational conflict was disrupted.

Whereas more than 4,500 people died in 1999, by 2003, fatalities had reduced by a half (Figure 1). While high conflict provinces (excluding Aceh) saw 1,738 people killed a year between 1998 and 2003, this declined to 365 from 2004. The drop in Aceh after the successful Helsinki peace agreement was even larger with 82 people killed per year from 2006 compared to 1,330 annually between 1998 and 2005 – a 94 percent reduction.


Enduring tensions, ongoing violence

Viewed alongside other conflicts, such as those in southern Thailand and Myanmar, Indonesia stands out as a regional star. Yet while the large SNCs ended, and deaths have remained at a stable low level, tensions and violence remain.

Old drivers of violence, such as ethnic and religious differences, continue to account for a major share of collective violence, which has increased in former high conflict areas since 2006 (Figure 2). In some post-conflict areas such as Maluku, riots and large groups clashes remain common. Other areas such as Aceh have seen an uptick in violent crime.


Recent violence is concentrated in former conflict hot spots. Incidents are now 36 percent more frequent in former high conflict provinces than elsewhere.

Improved security responses

If tensions remain, and incidents are frequent and increasing, then why has no area of Indonesia seen renewed subnational conflict?

One key reason is improved security responses. During the conflict period, only 10 percent of incidents of collective violence (involving large groups) saw any intervention from security forces (Figure 3). Following 2006, forces have intervened in half of all such incidents. While previously interventions came largely from the military, in the post-conflict years the police have been the prime responder (80% of interventions). Intervention success rates have also risen with a current average success rate of 85 percent.

Improved policing is not preventing incidents from occurring –they have been rising in recent years. Collective violence in former high conflict areas still results in more deaths than in areas that were not affected by SNCs. But policing is preventing incidents from escalating.

Indonesian policing is not without its weaknesses. Response rates in lower conflict areas are 30 percent lower than in high conflict regions. Police officers sometimes look on while mobs attack minority groups such as the Ahmadiyya. In Papua, human rights abuses continue. But in Indonesia’s former high conflict areas, intervention by the security forces in collective violence incidents has improved in quantity and quality. And this has prevented tensions and localized violence from escalating into larger-scale sustained conflict.


Practical lessons

Indonesia’s experience highlights important practical lessons for those seeking to prevent the resumption of large-scale subnational conflict.

Responses to violence in the post-conflict period matter as much as the structure of the peace settlement. Indonesia’s SNCs were brought to an end through very different means. But common across all areas has been improved policing in the post-conflict years and this has prevented violence reescalation.

State security responses can be effective even where root causes are not dealt with. Ongoing incidents and tensions suggest past grievances and incentives to use violence have not been fully addressed. But even with root causes intact, improved policing has been able to prevent large-scale conflict resuming.

Improving state security responses requires political will as much as improved capacity. More frequent and effective security interventions in Indonesia are not solely a result of improved policing capacity. Rather, it was changes to Indonesia’s political settlement that led to Indonesia’s polity stabilizing and this allowed for more effective deployment of coercive force. Strengthening state institutions in ways that satisfy the needs of competing elites may be a first necessary step to improving the capacity of the state to manage violence.

Patrick Barron is The Asia Foundation’s regional director for Conflict and Development based in Thailand. He can be reached at [email protected]. The views and opinions expressed here are those of the author and not those of The Asia Foundation.


About our blog, InAsia

InAsia is a bi-weekly in-depth, in-country resource for readers who want to stay abreast of significant events and issues shaping Asia’s development, hosted by The Asia Foundation. Drawing on the first-hand insight of renowned experts, InAsia delivers concentrated analysis on issues affecting each region of Asia, as well as Foundation-produced reports and polls.

InAsia is posted and distributed every other Wednesday evening, Pacific Time. If you have any questions, please send an email to [email protected].


For questions about InAsia, or for our cross-post and re-use policy, please send an email to [email protected].

The Asia Foundation
465 California St., 9th Floor
San Francisco, CA 94104

The Latest Across Asia


September 13, 2023

2023 Leaders on the Frontlines

Join us in San Francisco on October 12, 2023, to honor Niwat Roykaew.