Two Perspectives on Civilian Fatalities in the 2003 Iraq War U R LEE Division of Economics, University of Bradford, 26 June 2015
Civilians at the Forefront of War Civilians account for 5 percent of all casualties in WWI 50 percent of all casualties in WWII 90 percent or more after WWII through 1980s (Chesterman, 2001) 2
Military and Civilian Fatalities in the 2003 Iraq War 3500 3000 US military hostile deaths Civilian violent deaths (IBC minimum) 2500 2000 1500 Mission accomplished (1 May 2003) Surge (Jan-May 2007) 1000 500 0 Mar-2003 Mar-2004 Mar-2005 Mar-2006 Mar-2007 Mar-2008 Mar-2009 Mar-2010 3
Military and Civilian Fatalities in the 2003 Iraq War Mil. 160 140 120 US military hostile deaths Civilian violent deaths (IBC minimum) 3500 3000 2500 Civ. 100 2000 80 60 1500 40 1000 20 500 0 Mar-2003 Mar-2004 Mar-2005 Mar-2006 Mar-2007 Mar-2008 Mar-2009 Mar-2010 0 4
IBC vs. Pentagon s Iraq War Logs IBC UK NGO Media, Iraqi government, morgues Violent civilian deaths from beginning up to now Iraq War Logs Pentagon s archives Leaked by Wikileaks in 2010 (Manning and Assange): A scoop of the century Violent deaths between 2004 and 2009 (except May 2004 and March 2009) Similarities Persistency Incident-based records (see picture) Behavioural differences between US military authorities and media
6
IBC vs. Pentagon s Iraq War Logs cont. IBC Only civilians (non-combatants) Minimum and maximum Police in non-paramilitary roles (i.e., traffic police) Iraq War Logs Civilians : Civilians Host Nation: Iraqi security forces (police) Enemy: Insurgents and anti-coalition forces Friendly: Coalition forces
IBC vs. Pentagon s Iraq War Logs cont. Data Pentagon s War Logs IBC Dataset Categories Civilian Host nation Enemy Number of deaths (Baghdad) Number of incidents Deaths per incidents 66,081 (56%) 15,196 (33%) 23,984 (27%) Civilian + host nation 81,277 (52%) Minimum 85,132 (54%) Maximum 92,027 (54%) 34,009 8,064 9,417 42,073 21,865 22,108 1.94 1.88 2.55 1.93 3.89 4.16 Duration 70 months during 2004-2009 (except May 2004 and Mar 2009)
IBC vs. Pentagon s Iraq War Logs cont. 9
Monthly Civilian Fatalities Monthly death toll 4,000 3,500 3,000 2,500 Pentagon IBC 2,000 1,500 1,000 Kolmogorov- Smirnov equality test p value =.000 Without 2004 and 2005 =. 280 500 0 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
Monthly Civilian Fatalities: Baghdad 3,000 Pentagon Baghdad IBC Baghdad 2,000 1,000 0 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
Monthly Civilian Fatalities: Baghdad cont. IBC 4000 IBC Baghdad Baghdad Morgue Release 3000 2000 1000 0 Mar-03 Mar-04 Mar-05 Mar-06 Mar-07 Mar-08 Mar-09 Mar-10 Mar-11
Monthly Civilian Fatalities: Outside Baghdad 1,500 Pentagon excluding Baghdad IBC excluding Baghdad 1,000 500 0 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
Discrepancies in Anbar Najaf Salah al-din Kerbala Basrah Baghdad
Civilian Fatalities across Governorates
Major Discrepancies: Fallujah 700 Anbar Governorate Falluja District First Battle of Falluja (April 2004) Second Battle of Falluja (November 2004) 350 0 2004 2005 2006 2007 2008 2009 2004 2005 2006 2007 2008 2009 Civilian Minimum Enemy * Battle in April 2004 * Battle in November 2004 IBC: 577 civilians 668 civilians Pentagon:1 civilian, 386 combatants 2 civilians, 479 combatants
Wilcoxon Signed Rank-sum Test (Equality Test of Distribution)
Wilcoxon Signed Rank-sum Test (Equality Test of Distribution) cont. Consistent figures to some degree at monthly, governorate and district levels if excluding the first two years Consistent figures if added by Host Nation
Conclusion Both are incomplete lists Both are consistent with each other if excluding 2004 and 2005 Comparative analysis could provide a clue of significantly different standards between US military authorities and media in counting violent civilian deaths Media may have inflated the number of violent deaths US military authorities might have been Less attentive in counting civilian deaths at the initial period Less mindful in distinguishing civilian losses out of insurgent deaths in major military offensives
Monthly Civilian Fatalities Monthly death toll 4,000 3,500 Pentagon IBC 3,000 2,500 2,000 1,500 1,000 500 Mission accomplished (1 May 2003 ) Surge (Jan-May 2007 ) 0 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09
Conclusion cont. We don t do body counts -Gen Tommy Franks, 2002- Political implications Precise counting is requisite To enhance civilian protection To understand evolutionary intensity of war If IBC is closer to true number, the US seems to have failed to grasp reality in Iraq at the initial period Great transparency and consistency will help military authorities To understand spread of violence To prevent unnecessary escalation
22 Thank you!