Special Data Features

Journal of Peace Research

2015, Vol. 52(4) 536–550
Armed conflicts, 1946–2014 ª The Author(s) 2015
Reprints and permission:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0022343315595927
jpr.sagepub.com
Therése Pettersson & Peter Wallensteen
Uppsala Conflict Data Program, Uppsala University

Abstract
In 2014, the Uppsala Conflict Data Program (UCDP) recorded 40 armed conflicts with a minimum of 25 battle-
related deaths, up by six from 2013. This is the highest number of conflicts reported since 1999, and 11 of these con-
flicts were defined as wars, that is, conflicts generating 1,000 or more battle-related deaths in one calendar year. Further,
an escalation of several conflicts, coupled with the extreme violence in Syria, resulted in the highest number of battle-
related deaths in the post-1989 period. Yet, compared to the large-scale interstate wars of the 20th century, the number
of fatalities caused by armed conflicts in 2014 was relatively low. Additionally, seven conflicts identified in 2013 were no
longer active in 2014. However, four new conflicts erupted in 2014, all of them in Ukraine, and three previously reg-
istered conflicts were restarted by new actors. Furthermore, six conflicts reoccurred with previously registered actors. A
positive development, however, is the increase to ten of the number of peace agreements concluded and signed in 2014,
which represents a further four compared with 2013. And although this increase is part of a positive trend since 2011, it
is worth noting that several peace processes remained fragile by the end of the year.

Keywords
battle-related deaths, conflict, data, peace agreements, war

Conflicts in 2014 in the longer perspective the growing number of internationalized armed conflicts,
that is, conflicts in which one or more states contributed
Since the end of the Cold War, the number of armed
troops to one or both warring sides. Nevertheless, this ten-
conflicts in the world has decreased substantially, a trend
year period is also where we find the year with the lowest
that has been picked up by researchers arguing that wars
number of active conflicts in the post-Cold War period.
are in decline (cf. Pinker, 2011; Goldstein, 2011; Pinker
All this illustrates the fluctuations associated with the
& Mack, 2014). For the past ten years, however, the
trend in armed conflict.
Uppsala Conflict Data Program (UCDP) has recorded
Another way of looking at the trend in armed conflict is
an uneven, yet clearly visible, upward trend,1 particularly
by way of the fatalities in these conflicts. In recent years,
the conflict in Syria and the escalating violence in coun-
1
tries like Iraq, Afghanistan, Nigeria, and Ukraine, have
UCDP is one of the world-leading providers of data on organized
violence, and its dataset on armed conflict is the most widely used
resulted in the highest yearly death toll in the post-Cold
in research on civil conflicts (Dixon, 2009). An armed conflict is War period. Yet, the scale associated with the number
defined as a contested incompatibility that concerns government or of fatalities caused by armed conflicts in 2014 was still
territory or both, where the use of armed force between two parties lower than that of the large-scale wars of the 20th century.
results in at least 25 battle-related deaths in a calendar year. Of Also, one notable positive development since 2011 is the
these two parties, at least one has to be the government of a state.
increase in the number of peace agreements being signed.
For intrastate conflicts, the location is a country. For an interstate
conflict, it is two or more countries. Several countries (notably In 2014, 40 armed conflicts were active in 27 loca-
India, Myanmar, and Ukraine) have several separate conflicts going tions worldwide, representing an increase of 18% when
on at the same time, fought over different incompatibilities, which
is why the number of conflicts exceeds the number of locations.
For in-depth definitions of key concepts, see http://www.pcr.uu.se/ Corresponding author:
research/ucdp/definitions/. Therese.Pettersson@pcr.uu.se

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Pettersson & Wallensteen 537

compared to the 34 conflicts2 reported in 2013. This is The average number of active dyads, a measure of
also the highest number of conflicts reported since 1999. fragmentation in an armed conflict, however, decreased
And although the figure remains at a relatively low level in 2014, from 1.41 to 1.33, when compared to
compared to that during the immediate post-Cold War 2013.
period, the trend visible during the early 2000s, showing Since 1946, there have been 567 dyads in 259 con-
a decreasing number of armed conflicts, seems to have flicts active in 159 locations. The average number of
now been reversed. Indeed, numbers are currently up to dyads in this period is 2.19. The annual incidence of
the same level as in the second half of the 1990s. The peak conflicts and conflict dyads since 1989 are recorded in
year of 1991 saw 51 active conflicts, while the lowest Tables I and II. Figure 1 shows the trend in the number
number of active conflicts in the post-Cold War period of active armed conflicts since 1946.
was recorded in 2010, when 31 conflicts were active. What stands out in the 21st century is the lack of
Of the reported 40 active conflicts, 11 reached the large-scale interstate conflict. Only one was active in
intensity level of war – conflicts with at least 1,000 2014, the conflict between India and Pakistan, which
battle-related deaths in one calendar year – five more led to fewer than 50 fatalities. The remaining 39 con-
than in 2013.3 The number of conflict dyads4 also flicts were fought within states, but 13 of them – or
increased in 2014, going from 48 to 53.5 Eight separate 33% – were internationalized in the sense that one or
conflicts had two active dyads, the conflict in Sudan had more states contributed troops to one or both sides.
three active dyads, and in Pakistan as many as four rebel These conflicts were Afghanistan, Azerbaijan (Nagorno-
groups were fighting the government simultaneously.6 Karabakh), Iraq, Mali, Nigeria, Somalia, South Sudan,
Uganda, Ukraine (Donetsk), Ukraine (Lugansk), Ukraine
(Novorossiya), USA (the conflict with Al-Qaeda), and
2
Themnér & Wallensteen (2014) reported 33 conflicts as active in Yemen. This is an increase when compared to the previ-
2013. Since then, more detailed information regarding a previously ous year when 27% of the recorded conflicts were interna-
not included conflict in Myanmar has emerged making the conflict tionalized. The involvement of external actors in internal
active in both 2013 and 2014. The conflict between the conflicts is not a new phenomenon. However, it is note-
government of Myanmar and PSLF (Palaung State Liberation worthy that the 2014 proportion is the highest recorded
Front) concerned the status of the territory of Palaung in northern
Shan state. While the territory has been contested for more than
in the entire post-World War II period, signifying the
50 years, the conflict was first recorded as active by UCDP in continuation of a trend that has been observed in recent
2013, when it crossed the threshold for inclusion. years (Themnér & Wallensteen, 2013). The United States
3
Themnér & Wallensteen (2014) reported that seven conflicts and Russia were two of the main external warring parties
reached the level of war in 2013. Since then, however, new in 2014, involved in four and three conflicts, respectively.
information on South Sudan has resulted in lower fatality estimates
Jordan, the United Kingdom, Belgium, and France were
for the state-based conflict, making it a minor armed conflict,
instead of a war. also involved in three conflicts each. While the United
4
A dyad is defined as a pair of warring parties. In interstate conflicts, States has been one of the most frequent external warring
these warring parties are governments of states, whereas in intrastate parties since 2001, Russia entered this group based on
conflicts, one is the government and the other is a rebel group. If recent developments in Ukraine.
more than one rebel group is active in a conflict, several dyads are The presence of troops from other countries poses
recorded. For more information about the dyadic dimension of
armed conflict, see Harbom, Melander & Wallensteen (2008). The
a serious threat to conflict termination. It has been
UCDP Dyadic Dataset can be downloaded from http:// demonstrated that external military support makes
www.pcr.uu.se/research/ucdp/datasets/ucdp_dyadic_dataset/. conflicts both longer and bloodier (e.g. Elbadawi &
5
Themnér & Wallensteen (2014) reported 46 dyads as active in Sambanis, 2000; Balch-Lindsay, Enterline & Joyce,
2013. Due to a conflict in Myanmar (Palaung) (see footnote 2), 2008; Regan, 2002; Lacina, 2006). One possible expla-
and better information regarding fighting between the government
nation behind this observation is that the additional
of CAR and anti-Balaka forces, the number of active dyads in 2013
has been revised to 48. resources available to the warring parties will increase
6
A comment is warranted on the conflict in Syria, which has been their fighting power, leading to more fatalities and longer
estimated to involve more than 1,000 armed opposition groups conflicts (Record, 2006). Others argue that external
(Lund, 2013). Since it is rarely reported which group is involved in involvement makes conflicts more difficult to solve as the
a given violent incident, coding of events into dyads has been number of actors with a stake in the negotiations
impossible. The solution, used in a handful of cases, such as the
conflicts in Kashmir and Punjab, has been to simply code the
increases (Balch-Lindsay, Enterline & Joyce, 2008;
opposition side as ‘insurgents’, indicating the complexity of the Cunningham, 2010). Consequently, the involvement
situation. of external troops often means that a solution to the

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

 538
Table I. Armed conflicts, battle-related deaths,a and conflict locations, 1989–2014
Level of conflict 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Minor 30 35 39 37 34 39 32 31 33 28 29 27 30
War 10 14 12 11 9 9 8 10 7 12 11 11 8
BRD low estimate 50,114 76,156 66,765 35,320 37,564 32,280 27,466 27,625 36,981 37,158 78,605 75,199 19,927
BRD best estimate 54,224 79,580 70,520 36,459 38,671 32,836 28,419 27,879 39,727 39,315 79,597 77,354 22,347
BRD high estimate 83,095 96,986 88,465 60,593 60,808 55,256 43,583 32,963 61,185 50,331 101,984 90,138 37,858
All conflicts 40 49 51 48 43 48 40 41 40 40 40 38 38
All dyads 59 65 66 59 54 58 46 50 55 52 50 51 49
All locations 34 37 38 36 32 34 31 31 31 33 31 29 30
Level of conflict 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Minor 27 27 26 27 28 31 33 31 26 31 26 28 29
War 6 5 7 5 5 4 5 6 5 6 6 6 11
BRD low estimate 15,755 20,600 18,361 11,864 18,457 18,193 27,700 30,318 19,458 21,719 37,149 39,265 44,497
BRD best estimate 17,541 20,986 18,916 12,207 19,601 19,030 28,493 33,370 20,371 22,614 37,992 70,451 101,406
BRD high estimate 26,328 31,083 25,793 15,601 28,162 25,123 36,974 41,742 27,521 30,008 60,375 79,321 112,549
All conflicts 33 32 33 32 33 35 38 37 31 37 32 34 40
All dyads 45 44 45 40 47 44 48 47 40 51 40 48 53

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

All locations 25 25 25 23 25 25 29 27 25 30 26 25 27
a
Referred to as BRD in the table. Note that for Syria no low estimate is included for 2014, and the high estimates for both 2013 and 2014 are set to the same as the best estimates.
Pettersson & Wallensteen 539

Table II. Armed conflicts by region, 1989–2014

Region 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Europe 2 3 7 7 9 5 5 1 1 3 3 1 2
Middle East 4 7 8 6 7 6 6 7 4 3 3 3 3
Asia 13 20 14 17 13 18 15 18 19 15 16 18 15
Africa 12 13 17 14 11 15 10 12 14 17 16 15 16
Americas 9 6 5 4 3 4 4 3 2 2 2 1 2
All regions 40 49 51 48 43 48 40 41 40 40 40 38 38
Region 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Europe 1 1 2 2 1 2 3 1 1 1 2 1 6
Middle East 2 3 3 5 5 4 4 5 5 6 5 4 6
Asia 13 15 15 16 15 14 15 15 12 13 10 14 14
Africa 15 11 10 7 10 12 13 13 10 15 13 13 12
Americas 2 2 3 2 2 3 3 3 3 2 2 2 2
All regions 33 32 33 32 33 35 38 37 31 37 32 34 40

60 estimate, which makes it the most violent year in the

50 entire post-Cold War period. Compared to the second
half of the 20th century, with extremely fatal interstate
40
wars in Korea, Vietnam, Iran–Iraq, and Ethiopia–Eri-
30 trea, the number of fatalities caused by armed conflicts
20 is still lower (Lacina & Gleditsch, 2005). Although the
10 bulk of the fatalities in 2014 occurred in the conflict
in Syria, battle-related deaths increased substantially in
0
other conflicts. In fact, even when excluding Syria, more
1946
1949
1952
1955
1958
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
2012

than 47,000 deaths were recorded in 2014. This is the

Extrastate Interstate Internationalized Intrastate highest number of fatalities since the year 2000 when the
war between Ethiopia and Eritrea alone caused 50,000
Figure 1. Number of armed conflicts by type, 1946–2014 deaths. Of the ten conflicts with the most fatalities in
2013, eight became more violent in 2014. Figure 2
conflict is likely to fail if the external supporters of this shows the trend in the number of battle-related deaths
conflict do not approve of it (Pettersson, 2011). between 1989 and 2014. In 2014, Syria was by far the
most violent conflict, followed by Iraq, Afghanistan,
Battle-related deaths in 2014 Nigeria, Pakistan, Ukraine (Donetsk), South Sudan, Israel
(Palestine), Yemen, Ukraine (Novorossiya), and Somalia.
As previously noted, there were 11 conflicts reaching the The conflict in Syria went on at exceptionally high
level of war in 2014, an increase of five compared to levels of intensity throughout 2014. The year started
2013. This is the largest relative increase witnessed since with failed attempts at finding a political solution
the early 1960s although the total number of wars is still through the Geneva II talks, and the strategy of imposing
lower than during most of the 1980s and 1990s, with a sieges and starving areas under the control of Syrian
peak in 1988 when 16 wars were recorded by the insurgents continued during the year, as did the cam-
UCDP. The large number of wars witnessed in 2014 paign of air bombardments of Aleppo. Furthermore,
also translated into an increase in the number of while the government allowed its stores of chemical
battle-related deaths,7 with 101,400 fatalities as a best weapons to be removed or destroyed, there were strong
indications that chlorine gas was used in the northern
7
Please refer to page 14 for a definition of battle-related deaths as parts of the country. An important feature of the rebel
well as for further information regarding the three fatality estimates. landscape in 2014 was the severe inter-rebel fighting

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

540 journal of PEACE RESEARCH 52(4)

120,000 year Australia, Belgium, Canada, Denmark, France, the

100,000
Netherlands, and the UK were carrying out attacks
together with the USA against IS targets in Iraq, while Jor-
80,000
dan, Bahrain, Saudi Arabia, and the United Arab Emirates
60,000 joined the USA in attacks against IS bases in Syria.
In Afghanistan, the conflict against the Taliban con-
40,000
tinued to escalate causing over 12,000 deaths, and the
20,000 violence reached its highest level in the post-1989
0 period. Further, the Afghan security forces counted an
1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 increasing number of fatalities as they took on the
Low estimate Best estimate High estimate full responsibility for security in Afghanistan. Civilian
casualties also increased and for the first time, ground
Figure 2. Battle-related deaths by type of estimate, 1989–2014 battles between the Taliban and the Afghan government
became the main cause of civilian deaths, whereas in pre-
which took place between the IS (Islamic State, previ- vious years this had been due to improvised explosive
ously known as Islamic State in Iraq and al-Sham) and devices (IEDs). The withdrawal of international forces
other rebel groups. IS also fought the regime and made continued and the ISAF (International Security Assis-
substantial advances, including gaining control of regime tance Force) mission was officially terminated in Decem-
bases in the Aleppo, Raqqah, and al-Hasakah provinces, ber 2014. But security agreements with the USA and
as well as a gas field east of Homs. NATO allowed for 12,000 soldiers to remain in Afgha-
Themnér & Wallensteen (2014) refrained from nistan in 2015, as part of the new mission ‘Operation
reporting fatality estimates for the conflict in Syria, due Resolute Support’. Yet, it appears that the absence of for-
to problems stemming from a combination of ‘issue eign troops made the insurgents able to assemble bigger
crowding’ and ‘issue fatigue’. However, in 2014 the formations and more frequently engage in ground battles
UN published a report on fatality estimates, and the against the Afghan security forces.
SOHR (Syrian Observatory for Human Rights) pro- Another conflict that escalated dramatically in 2014
vided more detailed breakdowns of their summary fig- was in Nigeria, where the security situation in the north-
ures. While it still has not been possible to carry out ern and northeastern parts of the country continued
the usual event coding procedure, the UCDP has con- to deteriorate. Jama’atu Ahlis Sunna Lidda’awati wal-
sulted different summary figures resulting in conservative Jihad (commonly called Boko Haram) made rapid terri-
estimates for the battle-related deaths, which can be torial gains in August and September, and declared the
compared to other fatality estimates provided by the establishment of an Islamic caliphate in areas under its
UCDP. For 2013 and 2014, the UCDP has added control in November. Aside from its fight against the
SOHR’s total number of rebel fatalities and total number Nigerian military, the group continued its large-scale
of fatalities on the government side (SOHR, 2015), and attacks against civilians. In April, it carried out the infa-
then excluded all rebel fatalities that were coded as part mous attack against Chibok, which resulted in the
of non-state violence by the UCDP. No civilian casualties abduction of almost 300 schoolgirls. The girls remained
have been included in the best estimate, to make sure that in captivity as the year drew to a close.
no one-sided violence was coded as part of the armed con- In Pakistan, a case of rebel fragmentation could be
flict. However, this produces a conservative estimate since observed, where the government fought four different
many civilians in fact died in fighting between the govern- Islamist groups in 2014: TTP (Tehrik-i-Taleban
ment forces and the Syrian insurgents. Pakistan: Taliban Movement of Pakistan), Lashkar-e
The developments in Syria were closely connected to Islam, IMU (Islamic Movement of Uzbekistan), and
the conflict in Iraq, where fighting between the govern- Jamaat-ul-Ahrar. During the first half of the year, peace
ment and IS escalated dramatically in 2014. At least talks were held with the biggest group, TTP, which led
12,000 people were killed as IS carried out large-scale to a reduction in hostilities. However, the decision to
attacks and seized vast areas, including Iraq’s second larg- negotiate created rifts in the organization, and after a
est city, Mosul. As IS gained ground and reports of kill- ceasefire agreement was proclaimed in March, Jamaat-
ings, abductions and torture increased, the USA and a ul-Ahrar, a splinter group from TTP, carried out a sui-
number of other countries commenced air strikes against cide attack in Islamabad. The peace talks with TTP
the group in both Iraq and Syria. By the end of the collapsed in June as TTP and IMU, a group which has

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Pettersson & Wallensteen 541

challenged the governments in both Uzbekistan and and the fighting between the government of Yemen and
Tajikistan, and fights alongside the Taliban in Afghani- Ansarallah was subsequently included in UCDP data for
stan, carried out an attack against Karachi International the first time.8 In mid-August, Ansarallah intensified its
Airport. The government responded by launching oper- calls for the government to step down and organized
ations targeting militants from TTP, IMU, and Lashkar-e demonstrations in the capital Sanaa. The peaceful protests
Islam in North Waziristan and Khyber Agency. Despite transformed into armed clashes and on 21 September
serious infighting, and a massive offensive by the Ansarallah took control of most of the capital. A peace
government, the TTP still succeeded in carrying out a agreement, which included provisions for the formation
terrorist attack against a school in Peshawar in December. of a new government and a ceasefire, was signed shortly
At least 148 people, most of them children, died in the afterwards. Despite the fulfilment of Ansarallah’s demands,
attack, which severely lowered the prospects for future the group kept its troops in Sanaa.
negotiations. By the end of the year the situation in Fighting against al-Shabaab in Somalia has resulted in
Pakistan was unusually complex. large numbers of fatalities over several years. In 2013, the
In Ukraine, the UCDP recorded four different conflicts intensity level was just below 1,000 battle-related deaths.
in 2014, two of which reached the intensity of war – Ukraine In 2014, however, the fighting once again reached the
(Donetsk) and Ukraine (Novorossiya) – causing approxi- level of war. The Somali National Army, together with
mately 2,000 and 1,500 fatalities respectively. The Ukrai- AMISOM (African Union Mission in Somalia) troops,
nian situation is further elaborated on in the next section. drove al-Shabaab from several strongholds, forcing it to
The conflict over government in South Sudan contin- relocate further south. The group increasingly relied on
ued to be active in 2014, and fighting between the govern- guerilla tactics and focused more on targeting govern-
ment and SPLM/A In Opposition (Sudanese People’s ment officials and soft, civilian targets than on conduct-
Liberation Movement/Army In Opposition), loyal to for- ing large offensives and controlling territory.
mer Vice President Riek Machar, reached the level of war.
Much of the fighting was concentrated in three key oil New conflicts: The case of Ukraine
cities, Bor, Malakal, and Bentiu, which changed hands
multiple times during the year. Both sides retaliated by Four new armed conflicts erupted in 2014, all of them
targeting civilians, often based on ethnic considerations, in Ukraine. One was fought over the control of gov-
yet attempts to negotiate continued in Addis Ababa under ernment while the remaining three concerned the sta-
the auspices of IGAD (Intergovernmental Authority on tus of territories in the eastern parts of the country:
Development) throughout the year, and several ceasefires Ukraine (Donetsk), Ukraine (Lugansk), and Ukraine
were declared, only to be breached by the parties. (Novorossiya).
The Israel–Palestine conflict reached the highest level In November 2013, Ukraine was set to sign an asso-
of intensity since the early 1980s as a result of ‘Operation ciation agreement with the EU. However, just a few days
Protective Edge’ launched in July 2014. Attempts at before the signing ceremony President Viktor Yanuko-
negotiations broke down in April and after the kidnap- vych decided to abandon the process and instead deepen
ping and murder of three Israeli youths in June, violence the country’s ties with Russia. This triggered mass
escalated to levels not seen in more than 30 years. Indi- protests in the capital, Kiev. While starting out as a
viduals connected to Hamas were suspected of the mur- demonstration, the opposition, named Maidan after the
ders, but the group officially denied involvement. Israel Independence Square in Kiev, soon became more coher-
launched aerial bombings with the expressed goal of ent, created a military force and demanded the resigna-
stopping Hamas missile fire, but also conducted a tion of the government. By the end of January 2014,
ground incursion, trying to destroy the tunnel system Maidan had occupied a large number of administrative
that Hamas used to attack Israeli targets. In late August, buildings, including the City Hall. The government
after two months of almost daily attacks, a ceasefire was resigned on 28 January, but the opposition continued
agreed to through Egyptian efforts. to call for the president to step down. In late February,
In 2014 the armed conflict between the Government of Yanukovych was dismissed by the Parliament and fled
Yemen and AQAP (Al-Qaeda in the Arabian Peninsula) to Russia.
escalated and once again reached the level of war. The USA
continued to carry out drone strikes against the group. 8
The conflict dates back more than a decade and has seen several
Moreover, in March 2014, Ansarallah (commonly referred rounds of intense fighting. However, this violence has not been
to as the Huthis) called for the government’s resignation, included in UCDP data due to the lack of a stated incompatibility.

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

542 journal of PEACE RESEARCH 52(4)

The pro-EU governmental change in Kiev, and In Egypt, the armed conflict over government restarted
Russia’s annexation of Crimea in March, led to the rise in 2014, after having been inactive since the 1990s. In
of a pro-Russian movement in the eastern parts of the the wake of the Arab Spring uprising in 2011, which
country which later escalated into a series of territorial included widespread violence against protesters by offi-
conflicts. One of the organizations formed was DPR cial security forces, the security situation in the Sinai
(Donetsk People’s Republic). It demanded sovereignty Peninsula deteriorated significantly. It worsened in
over the Donetsk region and proclaimed its independence 2013, following the military’s ouster of President
in April 2014. A military confrontation ensued, resulting in Mohammed Morsi. Several of the attacks in Sinai in
well over 1,000 battle-related deaths, and also led to events 2013 were assigned to Ansar Bayt al-Maqdis, but this
such as the downing of the civilian airliner MH17 while it violence was not included in UCDP data due to lack
was flying over the region on 17 July. Large-scale military of a stated incompatibility. In early 2014, the group
operations by both sides, including heavy shelling and announced its opposition to the government and clashes
tank offensives, were observed, along with accusations of took place throughout the year. The group changed its
Russian support for the separatists. name to Wilayat Sinai (Sinai Province) in November
Parallel to these developments, another separatist after having pledged allegiance to IS, referring to itself
group had been emerging in Lugansk. LPR (Lugansk as a province in the caliphate proclaimed by IS.
People’s Republic) was formed in April, in the same Inactive since 1990, the conflict over government in
manner as DPR, and soon demanded independence for Lebanon resumed in 2014 as a consequence of IS’s
the territory of Lugansk. DPR and LPR became strong advancements in the region. In August, IS, together with
allies and supporters of each other. On 11 May, the LPR Jabhat al-Nusra, clashed with Lebanese troops in and
had secured sufficient territory to be able to run a refer- around the border town of Arsal, following the arrest
endum on the region’s independence from Ukraine of a rebel commander. The Lebanese army managed to
which the group declared one day later. Heavy fighting retake control over the town by mid-August, but IS con-
followed in and around Lugansk city, forcing most resi- tinued to have a presence in the area.
dents to flee as much of the town was damaged. Since the ousting of President Muammar Gaddafi in
Ukraine, DPR, and LPR reached a ceasefire on 5 2011, Libya has been characterized by widespread vio-
September, under the auspices of OSCE; however, the lence between different militias. In 2014, the conflict
agreement soon proved insufficient. In order to further over government became active once again, with over
their military offensive and have a stronger voice in the 25 deaths incurred by fighting between the government
negotiations, DPR and LPR created a new, unified group and two separate rebel groups. Following the elections in
called United Forces of Novorossiya on 16 September. June 2014, two governments claimed power, one in the
The group had a new stated incompatibility, a region capital, Tripoli, and one in Tobruk.9 This led to a clash
larger than Donetsk and Lugansk combined, referred between the Zintan Brigades, supporting the parallel
to as Novorossiya. A ceasefire signed on 19 September government in Tobruk, and the Tripoli government in
with the Ukrainian government did not prevent fighting and around the capital during the second half of 2014.
from continuing at high intensity. There were also clashes between the Tripoli government
The conflicts in Ukraine have pitted the USA and and the Libya National Army, formed by retired general
the EU against Russia and created a deadlock in diplo- Khalifa Haftar. In October, the Libya National Army
matic relations, as illustrated by the sanctions imposed formally aligned itself with the Tobruk government.
on Russian individuals and organizations after the Six conflicts were restarted by previously registered
annexation of Crimea, along with further similar mea- actors: Azerbaijan (Nagorno-Karabakh), India (Garo-
sures throughout the year. And while Russia bolstered land), India–Pakistan, Israel (Palestine, see above), Mali
its military presence along the border with Ukraine, (Azawad), and Myanmar (Kokang).
Ukraine itself, in December, decided to drop its non- In Azerbaijan, skirmishes along the Line of Contact
aligned status and announced that it is applying for resulted in the conflict being active in 2014. Although
NATO membership. the violence was nowhere near as intense as in the early
1990s, the death toll was higher than in any year since
Restarted conflicts
Three previously registered armed conflicts were 9
According to UCDP definitions, the group controlling the capital is
resumed with new actors: Egypt, Lebanon, and Libya. to be considered the government.

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Pettersson & Wallensteen 543

the ceasefire was signed in 1994. The presidents of Ethiopia (Oromiya), Malaysia (Sabah), Myanmar
Armenia and Azerbaijan met under the aegis of Russian (Karen), Myanmar (Shan), Mozambique, and Turkey
President Vladimir Putin in August, in the presence of (Kurdistan).
US Secretary of State John Kerry in September, and In Central African Republic, large-scale fighting con-
finally under the auspices of the OSCE Minsk Group tinued in 2014, with widespread atrocities committed by
in October. The meetings, which failed to reduce the different rebel groups and mobs mobilized along ethnic
tensions, involved no representatives from the Nagorno- and religious lines. However, after the resignation of the
Karabakh authorities. President, and Séléka leader, Michel Djotodia in January
In India, the conflict over Garoland, first active in 2014, violence became sectarian or stood between the
2012, again passed the threshold for inclusion. Violence anti-Balaka forces loyal to the previous President
simmered just below UCDP’s inclusion level in 2013, François Bouzizé and the ex-Séléka forces of ousted
and in 2014 a small increase in violence resulted in the President Michel Djotodia. Consequently, no conflict
conflict being recorded as active. involving the government of CAR was active in 2014.
Also in India, the interstate conflict with Pakistan was In spite of a large international presence in the form of
active for the first time since the ceasefire agreement was AU (African Union), UN, and EU troops, and a ceasefire
signed in 2003. Relations soured in August 2014 as agreement signed in July, violence among armed groups
India’s foreign secretary cancelled a meeting with her continued to create a massive humanitarian crisis.
Pakistani counterpart following reports of a meeting Fighting between the Ethiopian Government and the
between Pakistani diplomats and Kashmiri insurgents. separatist rebel group OLF (Oromo Liberation Front)
Cross-border shelling escalated in October and by the end continued, albeit on a low scale, during the year. Infor-
of the year at least 38 people had been killed in battle. mation regarding this conflict is scarce and reports failed
The conflict over Azawad in northern Mali fell below to confirm 25 battle-related deaths during 2014.
the level of inclusion in 2013 due to ongoing peace nego- The conflict in Malaysia over the territory of Sabah
tiations between the government and the Tuareg separa- emerged in 2013 when the Sulu Sultanate claimed its
tist group MNLA (Mouvement national de libération historical rights over the area. When the Sultan died in
de l’Azawad; National Movement for the Liberation of October 2013, the Sultanate announced that it would
Azawad). In the beginning of 2014, MNLA repeatedly continue to claim Sabah, but through peaceful means
accused the government of not complying with the agree- only. Even though Malaysian security forces alleged that
ment signed in 2013 and following a visit by the Malian the Sultanate planned a second intrusion, no fighting
president to the rebel-held town of Kidal, fighting esca- was recorded in this conflict in 2014.
lated. A ceasefire was proclaimed in late May but the peace In Myanmar, a broad peace process has been ongoing
process was hampered by souring relations between differ- since 2011, involving a large number of armed groups.
ent Tuareg groups. Two different rebel alliances were Although some of the groups continued to clash with the
formed, one opting for autonomy and the other for inde- government during 2014, the conflicts over Karen and Shan
pendence, and in July clashes erupted between the two. did not cross the 25-fatality threshold for inclusion. In both
By the end of the year, negotiations had restarted in conflicts, bilateral ceasefire agreements with the rebel groups
Algiers and plans were made for further talks during 2015. existed, but fighting continued on a small scale.
The conflict between the government of Myanmar Tensions remained high in Mozambique, although
and the rebel group MNDAA (Myanmar Nationalities the fighting did not result in 25 battle-related deaths dur-
Democratic Alliance Army), fighting over the territory ing the year. A peace agreement signed in August gave
of Kokang, was last active in 2009 when MNDAA was Renamo a greater say in the election committee. However,
largely crushed and lost much of its territory. In Decem- Renamo rejected the results of the October elections and
ber 2014, the MNDAA leader announced that the group threatened to establish a secessionist republic in the parts
was trying to regain some of these territories and fighting of Mozambique where the party had won a majority.
erupted in the northern parts of Shan State, narrowly Talks initiated in 2012 between the government of
making the conflict active that year. Turkey and the Kurdish group PKK (Partiya Karkeren
Kurdistan; Kurdistan Workers’ Party) have resulted in
a de-escalation of violence. Despite some small-scale
Conflicts no longer active clashes the ceasefire announced in March 2013 officially
Seven conflicts recorded for 2013 were no longer identi- continued. The close connections to the developments
fied as active in 2014: Central African Republic, in Syria remained a destabilizing factor, and hundreds

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

544 journal of PEACE RESEARCH 52(4)

of PKK fighters reportedly joined the Syrian Kurdish fourth and final annex to the Framework Agreement on the
group PYD (Partiya Yektı̂ya Demokrat; Democratic Bangsamoro signed late 2012, regulating the decommission-
Union Party) in its fight against IS for the town of ing process. The parties also agreed on a document regulat-
Kobane in July 2014. As Turkey refrained from inter- ing the borders of the Bangsamoro waters. In March 2014
vening in Kobane, massive demonstrations unfolded the Comprehensive Agreement on the Bangsamoro (CAB)
across the country, resulting in violent clashes between was signed, providing for the establishment of an autono-
police and protesters. PKK threatened to end the no- mous Bangsamoro political entity in Mindanao. The CAB
conflict period and accused Turkey of delaying the peace consolidated and affirmed the agreements signed by the par-
process by not bringing an end to the siege. In October, ties, and mediated by Malaysia, during 2012 and 2013.
Turkey shifted its position and allowed Kurdish Pesh- In Sudan, a peace process agreement was signed dur-
merga forces from Iraq to pass through the country in ing 2014. Indeed, in September, during negotiations in
order to reinforce the Kurdish forces in Kobane. By the Addis Ababa facilitated by the AU, the parties signed an
end of the year, formal negotiations including the gov- ‘Agreement on the National Dialogue and Constitu-
ernment and PKK were again being discussed. tional Process’ in which the government of Sudan and
the rebel group SRF (Sudanese Revolutionary Army)
agreed that a comprehensive political settlement was the
Peace agreements ideal option for resolving problems in Sudan, and that
A positive development, in contrast to the large increase dialogue and a constitutional process was to start after
in the number of conflicts, is an increase in the number the rules of such dialogue had been agreed upon.
of peace agreements signed during 2014. During the In 2013, the conflict in Mozambique between the
year, ten peace agreements were concluded, four more Frelimo-run government and Renamo resumed after 20
than in 2013. Of these, four were concluded in South mostly peaceful years. Fighting continued on a low scale
Sudan, three in the Philippines, one in Sudan, one in in 2014 before a peace agreement was signed by Renamo
Yemen, and one in Mozambique. leader Afonso Dhlakama and President Armand Guebuza,
In South Sudan, a number of attempts at halting the in August. The parties agreed to the integration of the
violence were made, resulting in several ceasefires and Renamo fighters into the military, in addition to changes
peace agreements. On 30 January 2014, the South Suda- in the electoral commission before the October elections.
nese government and the SSDM/A-Cobra faction In Yemen, the UN facilitated talks between President
(South Sudan Democratic Movement/Army-Cobra fac- Abd Rabbuh Mansur Hadi and Ansarallah in September
tion) signed the ‘Agreement on a Cessation of Hostilities’ 2014. The negotiations resulted in the ‘Peace and
during negotiations in Addis Ababa under the auspices of National Partnership Agreement’, stipulating the disso-
IGAD. The parties agreed to cease all hostilities and to lution of the government and giving Ansarallah broad
continue to find ways to end the conflict peacefully. powers in the new government.
On 9 May 2014, two peace agreements were concluded The increase in the number of peace agreements is
during separate negotiations in Addis Ababa. The first part of a positive trend since 2011 when only one peace
one was signed with the Cobra faction and provided for agreement was signed. However, many of the peace pro-
the establishment of a semi-autonomous administration cesses initiated in 2014 remained fragile by the end of
called Greater Pibor Administrative Area in the conflict- the year. The Yemeni rebel group Ansarallah continued
ridden Jonglei state. The second one, ‘Agreement to Resolve to fight despite its demands being met in the agreement.
the Crisis in South Sudan’, was signed by the government In Mozambique, tensions remained high and the Renamo
of South Sudan and the rebel group SPLM/A In Oppo- leader threatened to establish a secessionist republic. In
sition. Besides the immediate cessation of hostilities, the Sudan, the agreement with SRF failed to halt the vio-
agreement stated that a transitional government of lence, and in South Sudan tensions within the former
national unity was the best option for peace in the coun- SPLM/A continued in spite of several agreements
try. SPLM/A In Opposition signed another agreement concluded during the year. On a positive note, the
in Arusha on 20 October 2014. In the ‘Framework for SSDM/A-Cobra faction refrained from attacking the
Intra-SPLM Dialogue’ the warring parties agreed to government of South Sudan after its leader was
reunify the SPLM/A and work towards democracy and appointed administrator of the newly established Greater
freedom. Pibor Administrative Area. In the Philippines, the peace
In the Philippines, the government and the rebel group process with MILF has substantially reduced the violence
MILF (Moro Islamic Liberation Front) hammered out the which has been ongoing since the early 1990s.

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Pettersson & Wallensteen 545

Replication data Pettersson, Therése (2011) Pillars of strength: External sup-

port to warring parties. In: Therése Pettersson & Lotta
The complete datasets (UCDP/PRIO Armed Conflict
Themnér (eds) States in Armed Conflict 2010. Research
Dataset, UCDP Dyadic Dataset, and UCDP Battle- report 94. Uppsala: Universitetstryckeriet.
Related Deaths Dataset) updated to 2014 are found Pinker, Steven (2011) The Better Angels of our Nature. New
at http://www.pcr.uu.se/research/ucdp/datasets/. Older York: Viking.
versions of these datasets can also be found at this address Pinker, Steven & Andrew Mack (2014) The world is not
(all datasets) and http://www.prio.org/Data/Armed-Con- falling apart. Slate 22 December.
flict/UCDP-PRIO/ (the UCDP/PRIO Armed Conflict Record, Jeffrey (2006) External assistance: Enabler of insur-
Dataset). The tables and figures in this article were created gent success. Parameters 36(3): 36–49.
directly from the Excel sheets at the UCDP web page. Regan, Patrick M (2002) Third-party interventions and the
Detailed descriptions of the individual conflicts are found duration of intrastate conflicts. Journal of Conflict Resolution
in the UCDP Conflict Encyclopedia at http://www.ucdp. 46(1): 55–73.
uu.se/database. Replication data for this article can be found SOHR (2015) 76021 people killed in Syria in 2014. Syrian
Observatory for Human Rights, 1 January (http://syriahr.
both at http://www.pcr.uu.se/research/ucdp/datasets/
com/en/2015/01/76021-people-killed-in-20014/).
replication_datasets/ and www.prio.no/jpr/datasets. Themnér, Lotta & Peter Wallensteen (2013) Armed con-
flicts, 1946–2012. Journal of Peace Research 50(4):
Acknowledgements 509–521.
Themnér, Lotta & Peter Wallensteen (2014) Armed con-
Numerous colleagues in Uppsala have contributed to the flicts, 1946–2013. Journal of Peace Research 51(4):
data collection, notably Marie Allansson, Mihai Croicu, 541–554.
Emma Elfversson, Garoun Engström, Maria Greek,
Helena Grusell, Stina Högbladh, Gabrielle Lövquist,
Ralph Sundberg, Samuel Taub, and Lotta Themnér. A Appendix 1. Armed conflicts active in 2014
special thanks for comments on the first draft by Maria
Greek and Margareta Sollenberg. This list includes all conflicts that exceeded the mini-
mum threshold of 25 battle-related deaths in 2014 and
fulfilled the other criteria for inclusion.1 The column
References Year(s) shows the latest range of years in which the con-
Balch-Lindsay, Dylan; Andrew J Enterline & Kyle A Joyce flict has been active without interruption. The start year
(2008) Third-party intervention and the civil war process. is found in parenthesis in the Incompatibility column,
Journal of Peace Research 45(3): 345–363. which indicates when the armed conflict reached 25
Cunningham, David E (2010) Blocking resolution: How battle-related deaths for the first time. If a conflict has
external states can prolong civil wars. Journal of Peace been inactive for more than ten years or if there has been
Research 47(2): 115–127. a complete change in the opposition side, the start year
Dixon, Jeffrey (2009) What causes civil wars? Integrating refers to the onset of the latest phase of the conflict. For
quantitative research findings. International Studies Review
more complete information on the conflict- and dyad
11(4): 707–735
history, see (a) the UCDP/PRIO Armed Conflict Dataset
Elbadawi, Ibrahim A & Nicholas Sambanis (2000) External
interventions and the duration of civil wars. World Bank and the UCDP Dyadic Dataset at http://www.pcr.uu.se/
Policy Research Working Paper 2433. research/ucdp/datasets/ and (b) the Uppsala Conflict Data
Goldstein, Joshua S (2011) Winning the War on War: The Program’s online conflict encyclopedia at http://
Decline of Armed Conflict Worldwide. New York: Dut- www.ucdp.uu.se/database. The column ‘Intensity in
ton/Penguin. 2014’ displays the aggregated conflict intensity in terms
Harbom, Lotta; Erik Melander & Peter Wallensteen (2008) of the number of battle-related deaths. Thus, if more
Dyadic dimensions of armed conflict, 1946–2007. Journal than one dyad is active in the conflict, the intensity col-
of Peace Research 45(5): 697–719. umn records their aggregated intensity. Three fatality
Lacina, Bethany (2006) Explaining the severity of civil wars. estimates are given in the table: low, best, and high.2
Journal of Conflict Resolution 50(2): 276–289.
Lacina, Bethany & Nils Petter Gleditsch (2005) Monitoring
trends in global combat: A new dataset of battle deaths.
European Journal of Population 21(2–3): 145–166. 1
See p. 549 for further information regarding definitions.
Lund, Aron (2013) The non-state militant landscape in Syria. 2
See p. 549 for a definition of battle-related deaths as well as for
CTC Sentinel 6(8): 23–28. further information regarding the three fatality estimates.

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

 546
Intensity in 2014

Location Incompatibility Opposition organization(s) in 2014 Year(s) Low Best High

EUROPE
Azerbaijan Territory (Nagorno- Republic of Nagorno-Karabakh3 2014 45 46 54
Karabakh) (1991)
Russia Territory (Caucasus Emirate) (2007) Forces of the Caucasus Emirate 2007–14 167 167 295
Ukraine Government (2014) Maidan 2014 87 87 106
Territory (Donetsk) Donetsk People’s Republic4(2014) 2014 1,456 1,996 2,225
Territory (Lugansk) Lugansk People’s Republic5(2014) 2014 712 712 1,027
Territory (Novorossiya) (2014) United Armed Forces of 2014 1,558 1,558 1,635
Novorossiya6
MIDDLE EAST
Egypt Government (2014) Wilayat Sinai (Sinai Province) 2014 208 208 208
Iraq Government7 (2004) IS (Islamic State) 2004–14 12,537 12,598 14,126
Israel Territory (Palestine) (1949) Hamas (Harakat al-Muqawarna al-Islamiyya: Islamic Resistance 2014 1,651 1,665 1,750
Movement),PIJ (Palestinian Islamic Jihad)
Lebanon Government (2014) IS (Islamic State) 2014 126 126 126
Syria Government (2011) Syrian insurgents8 2011–14 – 53,948 53,9489
Yemen Government10 (2009) AQAP (Al-Qaeda in the Arabian Peninsula), Ansarallah (Supporters of 2009–14 1,660 1,660 2,046
God)
(continued)
3
Nagorno-Karabakh supported by troops from Armenia.
4

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Donetsk People’s Republic supported by troops from Russia.
5
Lugansk People’s Republic supported by troops from Russia.
6
United Armed Forces of Novorossiya supported by troops from Russia.
7
The government was supported by troops from Australia, Bahrain, Belgium, Canada, Denmark, France, Jordan, Netherlands, Saudi Arabia, United Arab Emirates, United Kingdom, and
United States of America.
8
A large number of groups have been active. Some of the larger groups in 2014 were Ahrar al-Sham, Jaysh al-mujhahideen, Jund al-Aqsa, IS, Jabhat al-Nusra li al-Sham, Jaysh al-Islam, Liwa al-
Haqq, Liwa al-Tawhid, and Suqour al-Sham.
9
The high estimate for the conflict in Syria in 2014 is set to the same as the best estimate, although the UCDP acknowledges that this is a conservative estimate.
10
The government was supported by troops from the USA.
 (continued)

Intensity in 2014

Location Incompatibility Opposition organization(s) in 2014 Year(s) Low Best High

ASIA
Afghanistan Government11 (1979) Taliban 1978–2014 11,520 12,311 15,675
India Territory (Bodoland) (2013) NDFB-S (National Democratic Front of Bodoland-Songbijit faction) 2013–14 30 34 47
Territory (Garoland) (2012) GNLA (Garo National Liberation Army) 2014 23 26 32
Territory (Kashmir) (1990) Kashmir insurgents12 1990–2014 176 177 205
Government (1991) CPI–Maoist (Communist Party of India-Maoist) 1996–2014 163 177 228
India, Pakistan Territory (Kashmir) (2014) 2014 34 37 52
Myanmar Territory (Kachin) (2011) KIO (Kachin Independence Organization) 2011–14 83 83 84
Myanmar Territory (Kokang) (2009) MNDAA (Myanmar National Democratic Alliance Army) 2014 7 67 107
Myanmar Territory (Palaung) (2013) PSLF (Palaung State Liberation Front) 2013–14 37 37 203
Pakistan Territory (Baluchistan) (2004) BLA (Baluchistan Liberation Army), BRA (Baluchistan Republican Army) 2011–14 90 90 90
Government (2007) TTP (Tehrik-i-Taliban Pakistan: Movement of the Taliban in Pakistan), 2007–14 2,948 2,951 3,019
Lashkar-e-Islam (Army of Islam), IMU (Islamic Movement of
Uzbekistan), Jamaat-ul-Ahrar (The Freedom Fighters Group)
Philippines Territory (Mindanao) (1972) ASG (Abu Sayyaf Group), BIFM (Bangsamoro Islamic Freedom 1993–2014 276 280 288
Movement)
Government (1969) CPP (Communist Party of the Philippines) 1999–2014 187 190 199
Thailand Territory (Patani) (2003) Patani insurgents13 2003–14 61 68 72
AFRICA

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

Algeria Government (1991) AQIM (Al-Qaeda Organization in the Islamic Maghreb)14 1991–2014 83 107 147
DR Congo Territory (Katanga) (2013) Kata Katanga 2013–14 28 58 58
(continued)
11
Supported by the NATO-led International Security Assistance Force (ISAF) that in 2014 included troops from: Albania, Armenia, Australia, Austria, Azerbaijan, Belgium, Bosnia
and Herzegovina, Bulgaria, Canada, Croatia, Czech Republic, Denmark, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Hungary, Iceland, Ireland, Italy, Jordan,
Latvia, Lithuania, Luxembourg, Macedonia FYR, Malaysia, Mongolia, Montenegro, Netherlands, New Zealand, Norway, Pakistan, Poland, Portugal, Romania, Slovakia, Slovenia,
South Korea, Spain, Sweden, Tonga, Turkey, Ukraine, United Arab Emirates, United Kingdom, and United States of America.
12
A large number of groups have been active. Some of the larger groups in 2014 were Lashkar-e-Taiba, Jaish-e-Mohammed, and Hizb-ul-Mujahideen.
13
E.g. BRN-C (Barisan Nasional Revolusi-Coordinate), PULO (Patani United Liberation Organization), and GMIP (Gerekan Mujahideen Islam Pattani).
14
Until January 2007, AQIM was known as GSPC (al-Jama’ah al-Salafiyah lil-Da’wah wa’l-Qital: Groupe Salafiste pour la prédication et le combat: Salafist Group for Preaching and Combat).

547
 (continued)

Intensity in 2014

548
Location Incompatibility Opposition organization(s) in 2014 Year(s) Low Best High
Government (2012) APCLS (Alliance des patriotes pour un Congo libre et souverain: Alliance 2012–14 34 56 60
of the People for a Free and Sovereign Congo), PARC-FAAL (Parti pour
l’action et la reconstruction du Congo-Forces armées alleluia: Party for
Action and the Reconstruction of the Congo-Allelujah Armed Forces)
Ethiopia Territory (Ogaden) (1993) ONLF (Ogaden National Liberation Front) 1998–2014 25 25 220
Libya Government (2014) Libyan National Army, Zintan Brigades 2014 322 322 322
Mali Territory (Azawad) (2007) MNLA (Mouvement national pour la libération de l’Azawad: National 2014 80 80 80
Movement for the Liberation of Azawad)
Government15 (2009) AQIM (Al-Qaeda Organization in the Islamic Maghreb), al-Murabitun 2012–14 95 95 129
(The Sentinels)16
Nigeria Government17 (2009) Jama’atu Ahlis Sunna Lidda’awati wal-Jihad (Group Committed to 2011–14 4,480 4,621 5,006
Propagating the Prophet’s Teachings and Jihad)18
Somalia Government19 (2006) al-Shabaab (The Youth) 2006–14 894 1,140 2,271
South Sudan Government20 (2011) SPLM/A In Opposition (Sudan Liberation Army/Movement In 2011–14 719 1,667 2,993
Opposition)
Sudan Government (1983) SARC (Sudanese Awakening Revolutionary Council), SRF (Sudanese 1983–2014 815 856 2,163
Revolutionary Army),21 Darfur Joint Resistance Force
Uganda Government22 (1980) ADF (Alliance of Democratic Forces), LRA (Lord’s Resistance Army) 2013–14 864 864 1,036
AMERICAS
Colombia Government (1964) FARC (Fuerzas armadas revolucionarias colombianas: Revolutionary 1964–2014 113 113 113
Armed Forces of Colombia)
USA Government23 (2001) Al-Qaeda (The Base) 2001–14 103 103 104
Total number of battle-related deaths in 201424 44,497 101,406 112,549

15
The government was supported by MINUSMA involving troops from Bangladesh, Belgium, Benin, Burkina Faso, Burundi, Cambodia, Chad, China, Egypt, Estonia, Finland,

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

France, Germany, Ghana, Guinea, Guinea-Bissau, Italy, Ivory Coast, Jordan, Liberia, Mauritania, Nepal, Niger, Nigeria, Norway, Rwanda, Senegal, Sierra Leone, Sweden,
Tajikistan, Togo, United Kingdom, and United States of America.
16
Al-Murabitun was created by a merger of the Signed-in-Blood Battalion and a faction of MUJAO in 2013.
17
The government was supported by troops from Cameroon, Chad, and Niger.
18
Previously coded as Boko Haram, which is the name commonly used in news media. However, Jama’atu Ahlis Sunna Lidda’awati wal-Jihad is the group’s official name.
19
The government was supported by troops from Burundi, Djibouti, Ethiopia, Ghana, Kenya, Nigeria, Sierra Leone, and Uganda.
20
The government was supported by troops from Uganda.
21
SRF is a coalition consisting of SPLM/A-North (Sudan People’s Liberation Movement/Army-North), SLM/A (Sudan Liberation Movement/Army), SLM/A-MM (Sudan Liberation
Movement/Army-Minni Minawi), and JEM (Justice and Equality Movement), formed in November 2011.
22
The government was supported by troops from DR Congo and South Sudan.
23
In 2014, the USA was supported by Afghanistan and Pakistan.
24
No low estimate for Syria included. The high estimate for the conflict in Syria is set to the same as the best estimate.
Pettersson & Wallensteen 549

Definitions of civilians killed in crossfire, indiscriminate bombings,

An armed conflict25 is defined by the Uppsala Conflict etc. All deaths – military as well as civilian – incurred in
Data Program (UCDP) as a contested incompatibility such situations are counted as battle-related deaths.
that concerns government or territory or both where the UCDP codes three different fatality estimates – low,
use of armed force between two parties results in at least best, and high – based on the reliability of reports and
25 battle-related deaths in a year. Of these two parties, at the conflicting number of deaths that can be reported for
least one has to be the government of a state. any violent event.
The incompatibility is the stated (in writing or
verbally) generally incompatible positions. A more  Low estimate: The UCDP Low estimate consists
detailed definition can be found on UCDP’s webpage, of the aggregated low estimates for all battle-
at http://www.ucdp.uu.se. related incidents during a year. If different reports
The conflicts are divided according to type of provide different estimates and a higher estimate
conflict:26 is considered more reliable, the low estimate is
also reported if deemed reasonable.
 Interstate armed conflict occurs between two or  Best estimate: The UCDP Best estimate consist
more states. of the aggregated most reliable numbers for all
 Internationalized internal armed conflict occurs battle-related incidents during a year. If different
between the government of a state and internal reports provide different estimates, an examina-
opposition groups, with intervention from other tion is made as to what source is most reliable.
states in the form of troops. If no such distinction can be made, UCDP as a
 Internal armed conflict occurs between the govern- rule includes the lower figure given.
ment of a state and internal opposition groups.  High estimate: The UCDP High estimate consists
of the aggregated high estimates for all battle-
The conflicts are also divided according to their inten- related incidents during a year. If different reports
sity into two categories: provide different estimates and a lower estimate is
considered more or equally reliable, the high esti-
 Minor armed conflicts: at least 25 battle-related mate is also reported if deemed reasonable. If
deaths in a year but fewer than 1,000. there are incidents when there is some uncertainty
 War: at least 1,000 battle-related deaths in a year. about what parties have been involved, these are
also included in the high estimate.
Battle-related deaths are those fatalities that can be
related to combat in a conflict dyad. Typically, battle- It is the best estimate of battle-related deaths that deter-
related deaths occur in what can be described as ‘normal’ mines both whether or not a dyad will be included in the
warfare involving the armed forces of the warring parties. UCDP data (i.e. the best estimate needs to be 25 or higher)
This includes traditional battlefield fighting, guerrilla and whether the intensity is recorded as minor or war.
activities (e.g. hit-and-run attacks/ambushes) and all kinds
of bombardments of military units, cities, and villages, etc.
The targets are usually the military itself and its installa- Appendix 2. Unclear cases in 2014
tions, or state institutions and state representatives, but Cases that have been completely rejected on the grounds
there is often substantial collateral damage in the form that they definitely do not meet the criteria of armed
conflict are not included in the list below. For the con-
25
flicts listed here, the available information suggests the
UCDP also codes two other categories of organized violence: possibility of the cases meeting the criteria of armed con-
non-state conflict and one-sided violence. Datasets on these
can be downloaded from http://www.pcr.uu.se/research/ucdp/
flicts, but there is insufficient information concerning at
datasets/ucdp_non-state_conflict_dataset_/ and http://www.pcr.uu. least one of the three components of the definition: (a)
se/research/ucdp/datasets/ucdp_one-sided_violence_dataset/ respectively. the number of deaths, (b) the identity or level of organi-
Furthermore, narratives on the cases are available at www. zation of a party or (c) the type of incompatibility. The
ucdp.uu.se/database. list of unclear cases for the entire 1946–2014 period is
26
UCDP has also coded a fourth type, extrasystemic armed conflict,
currently under review, but will be published at http://
a conflict that occurs between a state and a non-state group outside its
own territory. These conflicts are by definition territorial. The last www.pcr.uu.se/research/ucdp/datasets/. The unclear aspect
such conflict ended in 1974, so this category is not applicable in may concern an entire conflict or a dyad in a conflict that
Appendix 1. is included in Appendix 1.

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015

550 journal of PEACE RESEARCH 52(4)

Location/Government Opposition organization Unclear aspect

Burundi Unclear Identity of organization
Burundi FNL-Ubugabo-Burihabwa (Forces for the National Number of deaths
Liberation-Ubugabo-Burihabwa)
China ETIM (East Turkestan Islamic Movement) Identity of organization
Pakistan Forces of Hafiz Gul Bahadur Incompatibility
South Sudan, Sudan Number of deaths
Sudan SLMJ (Sudan Liberation Movement for Justice) Incompatibility
Sudan JEM-Bashar (Justice and Equality Movement – Bashar) Incompatibility
Tunisia Ansar al-Sharia Identity of organization
Yemen Southern Movement Level of organization

THERESE PETTERSSON, b. 1982, MA in Peace and Professor of Peace Studies, University of Notre Dame
Conflict Research (Uppsala University, 2008); project (2006– ); director of the Uppsala Conflict Data
leader, Uppsala Conflict Data Program, Department of Program (UCDP) and the Special Program on the
Peace and Conflict Research; editor of States in Armed International Targeted Sanctions (SPITS); author of
Conflict since 2010. Understanding Conflict Resolution: War, Peace and the
Global System, 4th updated edition (Sage, 2015) and
PETER WALLENSTEEN, b. 1945, PhD (Uppsala Peace Research: Theory and Practice (Routledge, 2011)
University, 1973); Senior Professor of Peace and Conflict where Chapter 9 gives the history of UCDP; co-author
Research, Uppsala University since 2012, previously Dag (with Isak Svensson) of The Go-Between: Jan Eliasson
Hammarskjöld Professor of Peace and Conflict Research and the Styles of Mediation (US Institute of Peace Press,
(1985–2012), and Richard G Starrman Sr Research 2010).

Downloaded from jpr.sagepub.com at Uppsala Universitetsbibliotek on September 29, 2015