Despite the end of Cold War-era civil wars, political violence rates in Africa remain remarkably high
. However, this broad statement hides an important qualification: the types of violence that have persisted in recent years have changed significantly, shifting from rebel and government activity towards violence against civilians, riots, protests, and battles by armed groups other than rebels.
The only way researchers can track this activity is through political violence data disaggregated by type, location, and time. The Armed Conflict Location and Event Dataset (ACLED) project aims to provide that service. We recently released updated information on political violence across Africa from January 1997 to December 2011 (see above for a hotspot map and timeline of violence over this period).
New to the ACLED project are real time data and trend reports on monthly conflict patterns throughout the continent. It is now perhaps the comprehensive and representative depiction of political violence as it occurs throughout Africa.
The data captures an array of actions including battles, looting, rioting, protesting, violence against civilians, and non-violent activity (such as arrests, recruitment drives, troop movements) by a variety of actors such as governments, rebels, militias, rioters and protestors.
Each event is geo-referenced by location and time-stamped by day. In our recently released version, we also included fatalities by event; distinguished conflict groups by their type (government, rebel, political militia, communal militia, rioters, protesters, civilians etc.); and specified the type of interaction between actors (e.g. government-rebel battle; rebel-civilian attack). These changes make ACLED data flexible for multiple uses.
Several new findings on conflict patterns are found using ACLED data:
- Violence against civilians accounts for approximately half of all conflict events.
- Generally, where they occur, civil wars intensely affect 19 percent of a state’s territory, yet rebel actions have drastically decreased since 2005.
- Militia activity has significantly increased since 2005 and is especially high during election periods in new democracies.
- Communal violence patterns are more widespread but affect fewer people than civil war violence.
- Civil war patterns are not strongly affected by climate changes, but communal conflict decreases during periods of local scarcity and increases on the cusp of rainy seasons.
- Political violence increases in the period from peace talk announcements to negotiations. This violence is directed towards overtaking territory and civilians are frequently the main victims.
Fine-grained data on the range of political violence in the developing world is important as it provides the opportunity to ask and potentially answer certain questions about a range of issues related to governance, economic development, and conflict dynamics themselves.
For example, ACLED data shows that violence is increasingly occurring in villages, towns, and cities across Africa. In other words, political violence may be “urbanizing.” This has important implications for how we understand the practice of politics, the geography of marginalization, and the role of trigger patterns in explaining conflict. But there are many competing theories that could explain this trend.
Using data that disaggregates by violence type allows us to probe deeper: If the environmental security framework is correct, political violence in towns may be a response to high in-migration from rural areas to cities. Violence therefore is due to competition between urban people and migrant populations.
An alternative explanation is that the poor conditions of African urban life, high rates of informal employment, and under-representation of urban communities in government might explain an increase in violence. In this case, violence is a populist response, in that civilians riot and protest in favor of government reform.
Finally, the advantages of densely populated urban areas – easier access to resources, recruits, infrastructure, and power – may attract more organized groups to contest these spaces. If this explanation were correct, we would expect higher rates of rebel activity against government forces in cities, with a clear drive to overtaking the capital.
We have not been able to address these questions before with credible and comprehensive data, but now we can. Indeed, such data is crucial for interrogating the climate-security debate.
For example, Clionadh recently co-authored an article in the Journal of Peace Research’s special issue on climate and conflict, which used ACLED data for East Africa. In that piece, she and her co-author argued that different conflict groups use their environment in accordance with their overall goals. While much of the conflict activity studied had an environmental signal, the climate signal was much weaker.
We hope that these studies and other work with disaggregated data spur more theories and explanations of conflict patterns in developing states, and that the breadth and form of the data allows for new directions within conflict research.
Clionadh Raleigh is an assistant professor in the Department of Geography at Trinity College, Dublin, external researcher at the Peace Research Institute Oslo (PRIO), and director of the ACLED project. Caitriona Dowd of Trinity College and Andrew Linke of the University of Colorado are senior researchers for ACLED.
Image Credit: ACLED.