Every year since we began our work in 2003 IBC has published a report summarising the main features and trends of the civilian death toll in Iraq over the previous 12 months, drawing on our systematic accumulation and analysis of civilian casualties as reported by Iraqi and international news media, NGOs and official sources.

The numbers reproduced below are derived directly from our database and may change as new data is obtained, analysed and included.

Another year of relentless violence in Iraq

IBC’s 2016 analysis of civilian deaths

1. 2016 Summary and comparison to preceding years

2016 has seen deadly violence continue to impact Iraqi civilians on a daily basis. This has been most significant this year in the northern city of Mosul and surrounding areas in Ninewa province under the control of Islamic State (IS), where it has carried out thousands of killings and executions. At the same time, the region has been under almost constant bombardment by US-Coalition and Iraqi government forces seeking to oust IS. Baghdad and its surroundings, even while not an area of direct conflict between IS and government forces, has witnessed daily attacks by roadside bombs, shootings and executions, as well as occasional mass casualty bombings. Civilians in Iraq, whether in contested war zones or not, have continued to lose their lives at levels similar to those that have characterised Iraq since mid-2013.

The annual total for civilian deaths in Iraq in 2016 was 16,361, which is within a broad range encompassing 2015 (17,578) and 2014 (20,218). These past three years are very much higher than the years 2010-2012, the least violent period since the invasion, when the annual numbers ranged from 4,167 to 4,622, and are also substantially higher than 2013 (9,852) which saw the beginning of the change from the pre-2013 levels to current levels. 1

1These figures are taken from the higher number in IBC’s range, and for December 2016 use IBC's preliminary figures. Both these and the lower end of IBC’s range tend to be exceeded over time.

2"Reductions" in this context an abstract concept, as deaths are cumulative and their number can only increase. Reported reductions are to the rate at which deaths accumulated in a given place and period. Every life lost represents an increase in their absolute number.

Table 1. Monthly civilian death toll, 2012-16
20122013201420152016
January5313571,0971,4901,374
February3563609721,6251,258
March3774031,0291,1051,459
April3925451,0372,0131,192
May3048881,1001,2951,276
June5296594,0881,3551,405
July4691,1451,5801,8451,233
August4221,0133,3401,9911,375
September4001,3061,4741,445935
October2901,1801,7381,2971,969
November2538701,4361,0211,740
December2991,1261,3271,0961,145
Totals4,6229,85220,21817,57816,361

An average of more than 1,300 civilians were killed in violence every month, the main exception being the September low of (935) which was however immediately followed by a peak of more than double that in October (1,969).

As the bar chart below visually illustrates, the 2016 total falls somewhat below the most recent peak in violent deaths in 2014, but remains far above the annual totals which preceded that peak, with the last 3 months (Oct-Dec 2016, as shown in Table 1) having been the worst quarter of the year.

Reductions in levels of violence during 2016 were very unequally distributed across the country, and were countered by sharp increases elsewhere, with Mosul in particular (and its surroundings) experiencing the worst levels ever recorded there (see section 2.4 below). 2

Bar chart showing annual civilian death tolls in Iraq, 2012-2016
Figure 1. Annual death toll in Iraq by half-year, 2012-2016, shows steep rise to 2014 peak, followed by gradual declines since. More detailed analyses can be obtained using the tools on IBC's Database page.
A mother flees fighting carrying baby and holding infant's hand
Monday, April 4, 2016: A mother and her children flee their homes during clashes between Iraqi security forces and Islamic State group in Hit, 85 miles (140 km) west of Baghdad, Iraq. Families, many with small children and elderly relatives say they walked for hours Monday through desert littered with roadside bombs to escape airstrikes and clashes. (AP Photo/Khalid Mohammed)

2. The civilians killed

As in other post-2003 years, information that was present for virtually all civilians reported killed in 2016 is the location in which they were killed or their bodies were found. This also holds true for the means by which they were killed, under five broad cause of death categories: executions; air attacks and shelling; suicide attacks; other improvised explosive devices (IEDs); 3 and gunfire not involving execution. Perpetrators, by contrast, were unknown for about one in every four civilians killed in 2016. Along with information on the time, location, cause of death and perpetrators, IBC collects demographic information on the victims themselves, including such identifying information as is available. 4

3 Non-suicide IEDs are ground-based bomb attacks such as e.g. roadside and remotely-triggerd or magnetic car bombs (etc.)

4 IBC's Methods pages lists other data variables collected.

5Lest we forget (30 Dec 2016) and Iraq Digital Memorial (Announcement and Facebook page).

Any serious public documentation of civilians killed will aim to record them as named individuals, as part of a record that establishes who was killed, not just how many. A recently-published companion piece to this report lists by name a sample of the individual victims in 2016 for whom further personal information has been made public, including in some cases photographs. This reflects our long-term goal to more fully humanise the victims of the war, through the forthcoming Iraq Digital Memorial project. IBC's identified victims list now spans more than 500 pages listing 25 individuals each, fewer than 10% of the deaths IBC has recorded.5

It compounds Iraq's tragic situation that detailed demographic (and particularly individualising) information is only available for so few of the victims whose deaths are reported. Information on the age and gender of victims, and how and by whom they were killed, can be of particular value in identifying the specific patterns of harm targeting these demographics. This makes it all the more important that what information is available is collected and analysed to the extent possible, even if some analyses are necessarily limited to a subset of victims.

2.1 Victim demographics

Demographic information relating to age and gender for victims killed in December 2016 is still being collected and processed, and analyses involving age and gender in this report therefore cover only the first 11 months of the year. (This data will be updated once December is processed.)

By 30 Nov 2016 15,216 deaths were recorded for 2016, and among these victims 6,162 could be identified as men, women, or children (defined as aged 17 and below). Of this number,

  • 4,755 (77%) were men;
  • 688 (11%) were women;
  • and 719 (12%) were children.

The demographic presence among civilian deaths of roughly 80% men, and 10% each of women and children, for deaths from all causes and perpetrators, has been fairly typical of the conflict since 2003. (Comparable figures for 2014 and 2015 are presented in Table 2.) These proportions do not hold across different weapons and perpetrators, however, as we show in sections 2.2.1 and 2.3.1 below.

Table 2. Annual victims identified as men, women or children, from all causes of death and perpetrators, 2014–30 Nov, 2016.
Victim records (1 Jan–30 Nov in 2016) identified as men, women, children of whom
men
of whom women of whom children
2014 20,218 9,797 (48% ) 7,977 (81%) 1,003 (10%) 817 ( 8%)
2015 17,578 7,731 (44% ) 6,061 (78%) 933 (12%) 737 (10%)
2016 15,216 6,162 (40%) 4,755 (77%) 688 (11%) 719 (12%)

2.2 How civilians were killed in 2016

6Air attacks and shelling (and other explosives that impact from the air, including artillery and mortar fire) are included in a single category because it is often difficult for witnesses on the ground to distinguish between them.

7IBC uses the term “involved” or "involving" because weapons from the different categories are sometimes used together. Thus the victim totals for these weapons categories should not be summed as this would almost never overlap.

Death by execution continues to account for by far the largest number of civilians killed in 2016 (7,170 killed, including victims of all ages) as it did in both 2014 and 2015.

Table 3. Monthly civilian death toll 2014-2016 by execution and gunfires
ExecutionGunfire (non-execution)
201420152016201420152016
January63664666416190203
February48963607345141132
March27438700357267142
April561,28346536416979
May101601413352216183
June2,373468626585264136
July402697354407161131
August1,8391,247757323167141
September474650314155141162
October6626851,33616913593
November664419731141118131
December61657720116011648
Totals7,3258,6927,1703,7742,0851,581

Deaths from air attacks and shelling also remained at a high level through 2016, as in the preceding 2 years, accounting for 2,854 deaths, and also killing children as well as men and women.6

Table 4. Monthly civilian death toll 2014-2016 by type of explosive weapon.
IEDs (non-suicide)Air attacks and shellingSuicide bombing
201420152016201420152016201420152016
January469279284108283161526569
February44925132211920513355111119
March41831826514710528211446111
April3682962751562102931723077
May33030733623248417611869297
June4163103075573501921383567
July320329254462406118149180453
August32130429765917420921116537
September4012282753993387311745100
October50530427522111115819156109
November3952563132231654859053197
December2602072942211845741141128
Totals4,6523,3893,4973,5043,0152,8541,5218661,664

2.2.1 Victim demographics and cause of death

Of the 15,216 victims recorded by November 30 2016, information on the weapons involved in their death (cause of death) was available for more than 99%.7 Victims who could also be demographically identified as men, women or children numbered 6,162 (40%) (see Table 2). Of these 6,162

  • Executions accounted for by far the greatest number of male adults killed (3,209, 67%), with non-execution gunfire the next most frequent cause of death (756, or 16%).
  • Female adults were, like male adults, killed most often by execution (213 , 31%) but also in more significant proportions by other weapons including air attacks/shelling, which accounted for 180 (26%) of women reported killed.
  • Air attacks and shelling accounted for the highest number of deaths among children, killing 298 (41%) of the 719 children recorded killed. The second highest cause of death among children was executions, by which means 205 ( 29%) were put to death.

Execution was by far the largest cause of death among civilians of known age and gender, as it was for all recorded civilian killings in 2016 (and the favoured means of killing by Islamic State, the year's major perpetrator, see sections 2.3 and 2.3.1 below.) Men were the principal targets of execution, women and children less often so. However it is shocking that children, in particular, should be subject to killing by execution at all.

Table 5. Weapons (cause of death) involved for civilians identified as male adults, female adults, or children, 1 Jan – 30 Nov 2016
All causes of death Execution Gunfire (non-execution) Air attacks and shelling IEDs (non-suicide) Suicide bombing

Men

3,719 3,209 (67%) 756 (16%) 183 ( 4%) 418 ( 9%) 239 ( 5%)
Women 476 213 (31% ) 134 (19%) 180 (26%) 94 (14% ) 37 ( 5%)
Children 636 205 (29%) 51 ( 7%) 298 (41%) 109 (15%) 66 ( 9%)

2.3 Who killed them

8As with weapons categories, the term “involved” is used here because deaths cannot always be attributed to a single perpetrator group in incidents involving multiple armed actors. Thus these death totals by different actors cannot be straightforwardly added due to the existence of overlaps in these categories.

The perpetrators involved in killing the greatest number of civilians (9,287) in 2016 were Islamic State (IS) and those fighting them – US-led Coalition forces, the Iraqi Government and allied militias. The means of causing death which claimed the most lives across all demographics were also those most frequently used by these forces: executions in the case of IS, and air attacks/shelling by US/Coalition and Iraqi Government and allied forces. 8

The only perpetrator for whom an absolute increase in its civilian death toll was reported between 2015 and 2016 was the US/Coalition (up from 1,466 in 2015 to 2,039 in 2016). The other groups showed a reported decrease (slight in the case of IS), but note that in 2016 as in previous years, deaths with unknown perpetrators constituted a significant proportion of the total.

Table 6. Monthly death toll 2014-2016 by major perpetrator group.
ISISAnti-Government forces (Not includeing ISIS)US/Coalition ForcesIraqi Government forces and allies
201420152016201420152016201420152016201420152016
January1680727241128320821579517636
February11,00472833924096123688021
March149978211129266432559591134
April451,27455602320010225211310088
May72791666257361446729118948761
June1,582648680353407124188173301238159
July277881723240484001227227228134
August2,2911,4058012272129901081593365921
September541743463703621492963826324068
October7217001,45219862177065132994433
November6974351,24314327241121151961213493
December52256046640189134177391141220
Totals6,7519,6209,2871,8838324393911,4662,0392,0931,852748

IBC's perpetrator information is less complete than on cause of death: in 2016, of 16,361 recorded deaths, 4,374 were by unknown perpetrators (in 2015 the equivalent number was 5,563 of 17,578 and in 2014, 9,596 of 20,218). It should not be assumed that the perpetrator breakdowns within the unknown and the known cases are similar, as some perpetrator groups may be much less likely to be identified than others. And even where a perpetrator was able to be recorded, in some incidents others might have been involved who escaped notice.

2.3.1 Victim demographics by major perpetrators

Of the victims recorded in 2016 who could be demographically identified as men, women or children, perpetrator information was available for 78% of the men, 69% of the women, and 88% of the children. As with the demographically undifferentiated numbers for 2016, one should note that this leaves a remainder involving unknown perpetrators which is highest, at nearly a third, for women, but little over 10% for children.

  • Of the 3,719 deaths of adult males with known perpetrator, the vast majority involved Islamic State (3,286 , 88%). The next highest, but much lower figures were for US/Coalition (101 , 3%) and Iraqi Government and allied forces (99 , 3%).
  • Of the 476 deaths of adult females whose perpetrators were known, the principal group involved was, as for men, Islamic State, who killed 308 (65%), with these figures (but unlike for men, more closely) by deaths involving US/Coalition forces (125, 26%).
  • Of the 636 child killings where perpetrator was known, the largest perpetrator group involved was once again Islamic State, accounting for 375, or 59% of child deaths, with US/Coalition forces (189, or 30%) and Iraqi government and allied forces, who were involved in 77 (12%) of child deaths.

All three demographic groups fell victim to Islamic State by far the most often. It is however notable that the US/Coalition caused a significant proportion of women and children reported killed.

Table 7. Male adult, female adult, and child deaths by major perpetrator groups, 1 Jan - 30 Nov 2016
Men, women, and child deaths by major perpetrators whose perpetrator was known Islamic State US/Coalition Iraqi Gov and allied forces
Men 3,719 3,286 (88%) 101 ( 3%) 99 ( 3%)
Women 476 308 (65%) 125 (26%) 44 ( 9%)
Children 636 375 (59%) 189 (30%) 77 (12%)

2.4 Where civilians were killed

Map of Iraq showing 7 most violent provinces and key cities in 2016
7,431
723
1,742
1,085
839
3,714
481
Figure 2. Map of Iraq showing 7 most violent provinces and key cities in 2016.

Figure 2 above shows the 7 Iraqi provinces (also called governorates) with the highest civilian death tolls in 2016, accounting between them for the vast majority of recorded deaths. These include areas held by IS, as well as the so-called ”Baghdad Belts” where the capital shares borders with Anbar, Salah al-Din, Diyala and Babylon, all of which led to the concentration of civilian casualties in these provinces. 9

9 Province boundaries vary across commonly used maps of Iraq, particularly for Baghdad province, which is often depicted as smaller than in Figure 2. IBC codes for both sets of boundaries and where necessary IBC province figures split the difference between these two versions in assigning deaths to bordering provinces.

Table 8. Monthly deaths by province (governorate) in 2016 (Dec 2016 data is preliminary)
NinewaBaghdadAnbarTameemDiyalaSalah al-DinBabylon
January610308179491067932
February444344206117912422
March600260245986435124
April58924816820625827
May397449130411145629
June52526431736648422
July3135986364888417
August572258114277653321
September3293273751689220
October1,33229826156525923
November1,11818874794892127
December60217218397172717
Totals7,4313,7141,7421,085839723481

Three cities with significant death toll changes between 2014, 2015, and 2016 were Fallujah, Mosul and Al-Qayyarah. Fallujah had a noticeable decline, particularly in the last half of 2016. Al-Qayyarah as well witnessed a decline in the last four months of 2016, despite being the year’s third highest city in number of civilian deaths Iraq (overtaking Fallujah). In contrast Mosul casualties have dramatically increased over the full period, with the highest monthly Mosul death toll of the entire post-invasion period being in November 2016 (1,061 in one month alone, which is significantly more than the monthly death toll for the entire country prior to June 2013).

Table 9. Monthly totals 2014-2016 for Mosul, Fallujah and Al Qayyarah (Dec 2016 data is preliminary). Figures refer only to deaths occurring within the city limits.
MosulFallujahAl Qayyarah
201420152016201420152016201420152016
January851803527115211915151
February56133244672054512114
March77594251011111224556
April9718044711218360239131
May8312326618613280131252
June215182342125183178142116
July90342136192346179139
August669652261835214115555
September2033901971476341921
October751928908024242414
November622951,06162269253015
December1062725332452255740
Totals1,2153,3135,1191,5711,420532117325764

3. Large-scale incidents in 2016 and earlier

2016 also witnessed some particularly shocking events, even by post-invasion standards. An example of that is the most deadly ground-based bombing attack in Baghdad, which was claimed by IS and hit a very crowded market in the central area of Karrada, on the 3rd of July just one day before Muslims’ Eid al-Fitr, killing 324, including women, children and members of entire families, according to the latest reports.

The second largest deadly bombing of 2016 took place in Shomali village, southeast of Hilla, on the 24th of November. The suicide attack was carried out using a fuel tanker and targeted buses carrying Muslim pilgrims returning from the religious rituals of Imam Hussein Arba'een in Karbala. The powerful explosion took the lives of up to 100 civilians, with casualties not limited to Iraqis but including nationals from Iran, Afghanistan and Pakistan.

Another brutal attack shocked the world in 2016 on the 25th of March, when an IS child recruit detonated his suicide belt in the midst of a trophy award ceremony in Alexandria district, north of Hilla, killing 17 children, some as young as ten years old.

Iraqis light candles and mourn for the dead in the ruins of the Karada bombing
People light candles inside a mall at the scene of a massive truck bomb attack in Karada neighborhood, Baghdad, Iraq, Tuesday, July 5, 2016. Iraqis mourn for their dead and prepare for the Muslim holiday of Eid al-Fitr under the pall of one the worst bombings in 13 years of war. An Islamic State suicide bomber struck Baghdad's bustling commercial area of Karada on Sunday night, after people came out after breaking their daily fast. (AP Photo/Hadi Mizban)

All deadly attacks are devastating to those affected by them, no matter their scale. However for reference, the table below shows IBC’s list of all 19 suicide or IED attacks that killed 100 or more civilians in Iraq, from 2003-2016. A list for the same period that included attacks killing 50 or more would contain 97 entries (or 312 entries for those killing 25 or more, and 1,028 entries involving 10 or more deaths).

Table 10. Suicide or IED attacks killing more than 100 people, 2003-2016”
IBC pageDateIncident summary
k0211 February 2004107-109 by suicide bombers in party offices in Irbil
k0222 March 2004121 by suicide bombers in Karbala
k99128 February 2005125-136 by suicide car bomb in Hilla
k110029 August 200383-126 by car bombs at Imam Ali mosque, Najaf
k190314 September 2005111-114 by suicide bomb, Kadhimiya, Baghdad
k198129 September 200599-102 by car bombs in Balad
k471023 November 2006215 by bombs, mortars in Sadr City, Baghdad
k53563 February 2007136-137 by suicide truck bomb in market, Sadriya, Baghdad
k56596 March 2007118-120 Shiite pilgrims by suicide bombers in Hilla
k589727 March 2007152 by truck bombs in market, Tal Afar
k611218 April 2007140 by car bomb, Sadriya food market, Baghdad
k68917 July 2007159-160 by suicide truck bomb in market, Amerli, near Tuz Khurmato
k722514 August 2007516-525 by suicide fuel tanker bombs in Yazidi villages, Sinjar area
k1386619 August 2009130-132 in multiple bombings in Baghdad
k1420325 October 2009150-152 in suicide truck bombings, Haifa Street, central Baghdad
k143798 December 2009122-127 in 5 car bombings in Baghdad
k2112217 July 201595-120 by suicide car or truck bomb in Khan Bani Saad
d121343 July 2016323-324 by suicide truck bomb in Karradah district, central Baghdad
d1281924 November 201680-100 by suicide truck bomb near Hilla

4. Civilians injured

IBC does not record incidents causing injuries and no deaths, but does record the injuries, which are serious, that occur in incidents that also cause deaths. These data are systematically recorded but not individually published on the IBC website, but it is useful to analyse and understand the types of weapons and attacks that cause injuries as well as deaths.

Table 11. Civilians injured per month, 2014-2016, by weapon category.
IEDs (non-suicide)Air attacks and shellingSuicide bombingGunfire (non-execution)
201420152016201420152016201420152016201420152016
January1,1468271,01133754412781139141400189109
February1,0677469563102941401422342081819230
March1,0221,0258233372471,40138515524530036124
April931949683376281298414771871956311
May724777936500302273306155610145104135
June1,0357889308783941954827818316717931
July7691,03194179859613919232344828326105
August866925846823231267385331541043271
September1,1306818646942491284021222294351117
October1,3579038773231431334541852281629656
November1,20381187134912772923112934812717227
December712769n/a321184n/a20122n/a10817n/a
Totals11,96210,2329,7386,0463,5923,8303,6751,9502,8812,2151,227916

10 For an earlier IBC analysis on the civilian wounded resulting immediately from the 2003 “Shock and Awe’” phase of the 2003 invasion, see Adding Indifference to Injury and over a longer period, see A Dossier of Civilian Casualties in Iraq 2003-2005

In 2016 (as in 2014 and 15), there were roughly the same number of civilians recorded injured as killed. However the proportion of killed to injured is highly dependent on the nature of the weapons. Our analysis shows that the great majority of injuries were associated with explosive weapons, broadly covering the weapon categories of ground bombing, air attacks and shelling, and suicide attacks: 16,449 of the total number of 17,935 were injured by weapons in these categories. This pattern holds true for the preceding three years, as indeed it has throughout the conflict, a pattern which tends to underscore the intrinsically indiscriminate nature of explosive weapons, as has been shown in many other contexts.10

5. Combatants and total deaths since 2003

Civilians, tragic as their deaths are, are not the only casualties of this war. In 2016, clashes between Iraqi military and allied militias on the one part, and insurgent groups such as IS on the other, have continued to cause thousands of deaths among these groups, particularly in Mosul and areas of Anbar.

11 For a more detailed discussion of these issues see section 4 of IBC's 2014 report

12Also see the earlier IBC announcement of this table, where the use of the Iraq War Logs and other sources is described.

Reporting on combatant casualties is often more questionable and far harder to verify than civilians, particularly since the emergence of IS in 2014.11 But even under the most conservative reading, the available data show that tens of thousands of fighters need to be added to the civilian death toll to begin to assess the full human cost of this ongoing conflict.12

All violent deaths including combatants, 2003-2016
Iraq Body Count March 2003-December 2016 188,546
Insurgents, Iraqi soldiers & other combatants killed 2010–2016 32,340
Iraq War Logs new 'Civilian' and comparable 'Host Nation' remaining 2004-2009 - central estimate (9420), ‘Host Nation’ combatant 2004-2009 - central estimate (5575), ‘Enemy’ (minus IBC overlaps) 2004-2009 - central estimate (20499) 35,494
Iraqi combatants killed March-May 2003 (4,895); Insurgents killed June-December 2003 (597); Insurgents killed May 2004 (652); Insurgents & Iraqi soldiers killed March 2009 (59) 6,203
TOTAL IRAQI DEATHS 262,583
US & Coalition military killed 2003–2016 4,832
US & Coalition foreign contractors killed 2003–2016 468
TOTAL DEATHS 267,883

6. Other sources

IBC is not the only organisation issuing casualty totals for Iraq. The United Nations in Iraq (UNAMI) issues monthly casualty totals, and their published civilian death toll for 2016 is 6,876. Such official figures have typically been lower than the figures provided by IBC, often vastly lower, as in 2016.

As we have remarked in earlier reports, when comparing differing published figures for Iraq it is important to note that on its public database IBC openly lists the violent incidents from which it derives its data, along with the original publishing sources for each entry. Progress in understanding differences between IBC and others requires looking beyond “competing totals” and examining what lies beneath them: that is, identifying which specific incidents are included in each total.

Until agencies such as UNAMI also make available the underlying data for their totals in the same disaggregated, incident-by-incident fashion that IBC does, it will remain impossible for third parties to meaningfully investigate and understand these differences (which is one reason why we and others have repeatedly called for such open publication by official sources. IBC has committed itself to abiding by recently launched international Standards for Casualty Recording which prescribe maximum possible detail and transparency in pursuit of the “ultimate objective of the recognition of every casualty” (Standard 53).

Such recognition can only be achieved where the individual incidents contained within larger totals are each published, together with whatever information is available about the people killed in them. It is this recognition which has always motivated our work, and which guides the future development of it. We are heartened that this goal is shared by increasing numbers of others, including the NGO airwars.org whose data on victims of coalition strikes, fully incorporated into our database, has greatly enriched detailed knowledge of victim identity and the incidents in which they were killed.