Organized violence,1989–2017

State-based conflict 1946–2017

Since the end of World War II, 614 dyads ² have been active in 285 conflicts in 157 locations. ³ Corresponding numbers for the post-Cold War period, that is, 1989–2017, are 371 dyads in 176 conflicts in 94 locations.

Forty-nine conflicts were active in 2017, down from 53 in the previous year, ⁴ the peak year of the entire 1946–2017 period. The high level of state-based conflict witnessed over the past few years has only been seen once before, in the early 1990s, right after the end of the Cold War. The growth of the Islamic State (IS) has driven the recent trend, as the group has tried to expand its territory resulting in a number of new conflicts (Melander, Pettersson & Themnér, 2016). During 2017, both Iraq and Syria reported the defeat of IS, and the group was indeed weakened in many of its areas of control (BBC, 2018). Nevertheless, IS was involved in 15 conflicts during 2017, a full 31% of the total number of conflicts. This is both the highest absolute number, and the highest percentage recorded since the group announced its caliphate in 2014.

OPEN IN VIEWER

Figure 2.

State-based armed conflict by type, 1946–2017

IS cells crossing the border from Iraq to Iran were involved in small skirmishes against Iranian forces already in 2016, but it was not until 2017, after a large-scale attack against the Islamic Consultative Assembly in Tehran, that the conflict in Iran reached 25 annual battle-related deaths. In Mali, the establishment of an IS territorial claim took a different form. In 2015, a faction of the al-Qaida ally al-Murabitun pledged allegiance to IS, calling itself ‘the Islamic State in the Greater Sahara’ (ISGS). However, the core IS leadership failed to acknowledge the pledge for more than a year. It was not until October 2016, when ISGS conducted a series of cross-border attacks in Burkina Faso and Niger and the core IS was under pressure in Iraq and Syria, that the group officially recognized the pledge.

As Figure 2 shows, there have been few interstate conflicts in the past decade. Of the 49 conflicts in 2017, only one was fought between states: an uptick in violence along the Line of Control (LoC) between India and Pakistan resulted in 118 deaths, the highest number recorded since 2003. Tensions had increased already in 2016, due to widespread anti-India protests in Kashmir, and remained high during 2017 amid mutual accusations of ceasefire violations.

Of the 48 intrastate conflicts, as many as 19, or 40%, were internationalized, that is, troops from external states supported one or both sides in the conflict. Both the absolute number and the percentage are high for the

OPEN IN VIEWER

Table I.

The ten most conflict-affected countries in terms of fatalities, 1989–2017

Country	Total no. of fatalities 1989–2017	Fatalities in state-based conflict	Fatalities in non-state conflict	Fatalities in one-sided violence
Rwanda	520,586	6,551	0	514,035
Syria	313,418	281,588	23,366	8,464
Afghanistan	200,316	187,470	3,294	9,552
Ethiopia	177,050	167,402	6,544	3,104
Iraq	118,902	97,976	2,943	17,983
DR Congo (Zaire)	105,482	24,640	13,520	67,322
Sudan	91,920	51,497	20,844	19,579
Sri Lanka	65,162	61,234	567	3,361
India	52,568	36,918	4,899	10,751
Nigeria	50,282	15,950	19,016	15,316
Other countries	621,430	425,724	76,224	119,482
Total	2,317,116	1,356,950	171,217	788,949

In 2017, ten state-based conflicts reached the intensity level of war, with at least 1,000 battle-related deaths. This is a decrease by two from 2016. The number of wars is still at a high level compared to the most recent decade, but not when seen in a longer perspective. As Figure 3 shows, between 1981 and 1992 the number of wars did not dip below ten.

Turning to the number of battle-related deaths in state-based conflict, Figure 1 reveals that the decrease in fatalities continued for the third year in a row. ⁵ With a fatality estimate of almost 69,000 battle-related deaths in the state-based category, 2017 witnessed a 22% decrease compared to the 88,000 battle-related deaths in 2016; this is the lowest level of fatalities since 2012. The trend is mainly due to the reduction of violence in Syria. Despite the overall decline, violence in Iraq and Afghanistan escalated in 2017, with Afghanistan experiencing its most violent year in the post-Cold War period and Iraq its second most violent year. ⁶

OPEN IN VIEWER

Figure 3.

State-based armed conflict by intensity level, 1946–2017

The past five years have witnessed high levels of fatalities compared to most years since the end of the Cold War. Nevertheless, the wars of the 21st century have been nowhere near as lethal as the large-scale wars of the 20th century, such as the Chinese Civil War, Korea, Vietnam, and the two World Wars (Lacina & Gleditsch, 2005). ⁷ Together with the substantial decrease in fatalities witnessed over the past three years, this lends support to the claim that battle-deaths are in decline and that the world today is increasingly peaceful (e.g. Goldstein, 2011; Pinker, 2011, 2018).

The state-based conflicts in Syria caused just over 20,000 deaths during the year, around 29% of the total number of fatalities recorded in state-based conflict in 2017. Down by 20,000 from 2016, this was the lowest number of fatalities in Syria since 2011. The decline was partly due to the concentration of the fighting to fewer areas and clearer delineation of the battle fronts. In part, it can also be attributed to several ceasefire agreements and de-escalation zones agreed upon during the year. In contrast to earlier years, most of the fatalities occurred in the conflict with IS over the territory called the Islamic State, rather than over governmental power.

At its height, IS in Syria and Iraq controlled an area the size of Portugal. By the end of 2017 the group had lost control of almost all of its territory and was confined to a few pockets of land along the Iraq–Syria border. Since IS was also engaged in a non-state conflict with the Kurdish-dominated Syrian Democratic Forces (SDF), which led to the fall of IS’s de facto capital Raqqa in October (more on this below), Syrian government forces were able to push the group out from areas on the western side of the Euphrates river. At the same time, SDF continued to take territory from IS on the eastern side of the river. By the end of October, the new de facto capital of the Islamic State – Mayadin, in Deir Ezzor province – fell to the Syrian army, severely reducing IS’s access to the Iraqi border (Toumaj, 2017).

Non-state conflict 1989–2017

UCDP has recorded a total of 670 non-state conflicts since 1989, with a yearly average of 37 active conflicts. While the increase in non-state conflict since 2010 was reversed in 2016, it again rose substantially in 2017, reaching a new peak of 82 active conflicts. As in previous years, Syria and Mexico dominated this picture, together responsible for 28% of the total number of non-state conflicts. However, the number of active conflicts was also on the rise in the Democratic Republic of Congo (DRC) and the Central African Republic (CAR). In CAR, a new pattern of intra-Muslim violence emerged, in addition to the widespread violence pitting Muslims against Christians. After a failed attempt to unite the mainly Muslim groups that used to be part of the ex-Séléka alliance, these actors intensified attacks on groups based among the mainly Muslim Fulani people, claiming that they were foreigners. This led to new alliances and enmities, resulting in an increasing number of non-state conflicts as well as higher levels of one-sided

OPEN IN VIEWER

Figure 4.

Fatalities in non-state conflict, by region, 1989–2017

Africa saw a large increase in the number of non-state conflicts in 2017 compared to 2016, from 35 to 50. In the Middle East, 19 non-state conflicts were recorded in 2017, up by two from 2016. As was the case in 2016, the vast majority of the conflicts in the Middle East took place in Syria. The Americas had 11 active non-state conflicts in 2017, an increase from 8 in 2016. For the past five years, non-state conflicts in the Americas have been concentrated in Mexico. In 2017, however, three new conflicts between criminal gangs were recorded in Brazil. At the end of 2016, the longstanding alliance between two of the country’s biggest criminal organizations, Comando Vermelho and Primeiro Comando da Capital (PCC), broke down. As a result, fighting between these two groups and their allies escalated in 2017. Finally, just as in 2016, two non-state conflicts were recorded in Asia in 2017, both of them in Afghanistan.

The sharp increase in the number of non-state conflicts was accompanied by a rise in the number of fatalities from this type of violence. Figure 4 shows a new peak in fatalities in 2017, with more than 13,500 people killed in non-state conflict. Much of this is due to the conflict between IS and SDF in Syria, which was the most lethal non-state conflict during the year, just as in 2016. This conflict alone caused almost 5,000 deaths as the Kurdish-dominated SDF launched the ‘Great Battle’ to expel IS from its de facto capital Raqqa in June (Rudaw, 2017). After more than four months of intense fighting between the two non-state groups, amid heavy airstrikes carried out by the US-led coalition, SDF captured the city in October. Following the Raqqa offensive, SDF continued to push back IS from its former strongholds in large parts of Deir Ezzor province. As a consequence, the Middle East was the region where most fatalities in non-state conflicts were incurred. However, the overall increase in deaths was also caused by escalating violence in Africa. UCDP recorded an almost 50% increase in the number of deaths in the region in 2017, mainly due to the fighting in CAR discussed above.

While fatalities in non-state conflict peaked in 2017, they remain but 15% of the total number of fatalities in organized violence. As Figure 1 shows, the number of people killed in state-based violence dwarfs that of non-state conflict, suggesting that it would be inadvisable to infer a substitution effect between the two forms of violence. While some non-state conflicts can be seen as de facto proxy wars between rival states, most non-state conflicts are connected to weak state apparatuses and the lack of territorial control by the government (Fjelde & Nilsson, 2012). The four countries in which the majority of non-state conflicts took place in 2017 – Syria, Mexico, Nigeria, and CAR – are examples of such weak states.

One-sided violence 1989–2017

UCDP has recorded a total of 266 actors engaged in one-sided violence since 1989, with a yearly average of 33 active actors. In 2017, the number of actors carrying out one-sided violence increased substantially, from 27 ⁸ to 33. An escalation of violence by several groups in DRC following President Kabila’s failure to hold elections partly caused this increase. This is the highest number of actors active since 2005, a return to the previous higher levels seen in the 1995–2005 decade. The overall downward trend witnessed during the last decade continued, with an uneven yet clearly visible decrease since 2002–03 when 45 actors were recorded in both years.

UCDP estimates show that a state or a formally organized group targeted and killed almost 7,000 civilians in 2017. Figure 5 reveals that while the number of one-sided actors increased compared to 2016, the number of fatalities remained at the same level. Although that

OPEN IN VIEWER

Figure 5.

Fatalities in one-sided violence, by type of actor, 1989–2017

One much-publicized case of one-sided violence during 2017 took place in Myanmar (Burma) where the government targeted and executed civilians mainly belonging to the Muslim Rohingya ethnic minority. Médecins Sans Frontières (MSF, 2018) reports that more than 6,000 Rohingya were killed in violence in August and September 2017. However, UCDP records only around 750 civilian fatalities from one-sided violence in its best estimate. Information on the circumstances of killings, including actors involved and how many people were actually killed, is extremely scarce due to restricted media access to the area. The number reported by MSF is based on surveys carried out in refugee camps and then extrapolated to estimate the total number of victims. The UCDP, however, only includes fatalities from events where the death itself and the context in which it occurred, including actors involved, can be identified. Moreover, while MSF’s figure presumably includes considerable one-sided violence carried out by the government forces, it also contains other types of violent deaths, including one-sided violence carried out by other organized groups as well as state violence targeting rebel groups, which in UCDP data is recorded to the state-based armed conflict category.

Although the killings of civilians by the Burmese government constituted one of the most lethal cases of one-sided violence in 2017, non-state actors caused the vast majority of the global fatalities from one-sided violence. This has been the case for most years since 1989; in only seven out of 29 years were governments responsible for more deaths than non-state groups. Non-state groups are generally weaker than their state opponents and terrorist attacks against civilians can be used as a tactic to overcome this asymmetry (e.g. Hultman, 2007; Kalyvas, 2006; Pape, 2005). Governments, on the other hand, face more pressure from the international community, which may influence their decision to refrain from targeting civilians. They may also choose to outsource this type of repression to pro-government militias (Carey, Colaresi & Mitchell, 2015).

IS continued to be the actor responsible for most one-sided violence – around 35% of the total number of deaths – although the killings carried out by the group decreased substantially from 2016. As the group has been militarily weakened and pushed back in many of its formal strongholds, some experts have warned that attacks against civilians may increase (e.g. IHS Markit, 2017; Winter, 2017). So far, however, the data do not seem to support this forecast although the group has continued to carry out suicide attacks in numerous countries around the world, with the majority taking place in Syria, Iraq, Egypt, and Afghanistan.