Measuring conflict is hard. It is easy to speak to victims and conclude that something bad is happening, or to conclude from major events – a specific battle, a grievous atrocity – that violence is present. But proving the intensity of a conflict, and how far it has spread, or assessing it relative to other conflicts, is a challenge of an entirely different magnitude.

This is where the Armed Conflict Location and Event Data Project (ACLED) comes in. Run out of the University of Sussex, the project monitors and catalogues all reported instances of political violence and conflict across Africa, south Asia and south-east Asia. It is an enormous, laborious process, involving more than a dozen researchers and a dizzyingly complicated spreadsheet. But it delivers a unique collection of numbers: big data on where and how conflict is happening.

The project has just released its overview of political violence and protest in Africa in 2016. Its findings make for sobering – albeit illuminating – reading.

The headlines won’t surprise anybody. Libya, Nigeria, Somalia and South Sudan are the continent’s major crisis areas, with significantly more recorded instances of violence and fatalities than anywhere else, accounting for 33% of all violent conflict in Africa last year.

Of these, however, Somalia is in considerably worse shape than the other three. While there were approximately 740 armed, organised events in Libya, Nigeria and South Sudan, there were three times that number in Somalia. “In effect, Somalia’s violence is equal to the combined violence of Libya, South Sudan and Nigeria,” the report found. This should raise some tough questions for the African Union Mission in Somalia (AMISON): after 10 years and billions of dollars, the peace is far from being kept.

Naturally, Somalia also has the highest fatality count. But the raw figure masks a more interesting statistic: counterintuitively, individual incidents in Somalia are less violent than in the other three major conflict countries (there are just a lot more of them). On average, each incident in Somalia results in 2.5 fatalities. That figure is four for Libya, 4.5 for South Sudan, and a massive six in Nigeria. Nigeria is also extreme in that fatalities “are overwhelmingly against civilians compared to battles between armed agents, and fatalities when the state retakes territory are also markedly higher than in other crisis contexts”. On the other hand, actors in the Libyan conflict appear to avoid directly targeting civilians.

Elsewhere, the absence of full-blown civil war or large-scale insurgencies masks disturbing levels of violence. “Despite lesser media coverage, a number of countries across the continent witnessed lower yet sustained rates of armed conflict, as state and non-state actors continue to use violence to influence political dynamics or consolidate their position vis-à-vis other competitors. The political nature of these low-level conflicts is such that, unless a political solution to the crises is found, violence is likely to persist or to escalate in the near future. This situation is common in several African states, but particularly intense in Burundi and Mozambique,” said the Armed Conflict Location and Event Data Project.

In 2016, Mozambique went from 19 violent incidents in 2015 to 92. In Burundi, although conflict between armed groups declined markedly last year, this simply reflects the changing dynamics of the violence. “Rather than seeking direct confrontation, government forces and armed militias widely resorted to violence against unarmed civilians and targeted political assassinations to either consolidate the grip on power or manifest their strength,” said the report.

When it comes to protests, there are some signs that Africans are beginning to find their voices, with a 4.8% increase last year in the number of events involving rioters and protesters. The increase is mostly attributed to Chad, Tunisia and Ethiopia, but South Africa remains the continent’s undisputed protest capital – and, as ACLED notes, the police seem to be doing their best to keep it that way: “Police often resorted to violent means in the attempt of curbing protests, but this repression ended up feeding more disorder. With new general elections scheduled in 2019 and growing in-fighting within the ruling party, violence is likely to feature prominently over the coming months in South Africa.”

ACLED’s data is perhaps most useful when it is used to examine continent-wide trends. For example, some 34% of incidents in 2016 involved state forces, which is high compared to recent years, suggesting that governments are adopting a more forceful approach to maintaining power. This is complemented by another trend, which is the increase in violence committed by political militias, which accounted for 30% of incidents. These are defined as groups that seek to shape and influence the existing political system, but do not seek to overthrow national regimes (the best example of these is the Imbonerakure in Burundi, a “youth organisation” which functions as enforcers for the ruling party).

The bottom line is that while conflict in Africa is not going away any time soon, the nature of that conflict is changing: “The overall patterns are clear: battles and large-scale wars are on the decline, as they have been for quite some time. In their place are multiple, co-existing agents who engage in a variety of strategies to make their place within the political landscape: local militias, pro-government militias, political militias working at the behest of politicians and political parties, civil society organisations forming protest movements, external groups seeking local partners (such as ISIS), and more occasionally, rebel groups seeking to overtake the government. These groups may use similar forms of violence — including attacking civilians, bombing, clashing with security forces, rioting — but they are distinct in their goals,” concluded ACLED. DM

ACLED’s data portal is available here

Photo: South Sudanese soldiers on their vehicle patrol a street in Juba, South Sudan, 20 December 2013. EPA/PHILLIP DHIL