Below I look at incidents in April 2004, during “The First Battle of Fallujah”. In some ways the account of the battle in the war logs confirms what was reported, but some of the details (such as when ceasefires were announced) are different. The war logs probably understate civilian deaths. They do, however, represent the disproportionate nature of the conflict.
A note on how it was done
I studied the First Battle of Fallujah, April 2004, using Wikileaks’s Iraq War Diary (http://wikileaks.org/iraq/diarydig). I merged the UK Guardian’s Every Death Mapped database, which contains full geographic information but no detailed text summaries of incidents, and the redacted Iraq War Diary, which contains highly redacted text summaries and highly unspecific geographic information. I did this for a geographic window around Fallujah and for the year of 2004.
I obtained background data on Iraq from the openstreetmap.org project and found the other data on mapcruzin.com, which ultimately comes from openstreetmap as well. I am still working on how to show these data with Google maps underneath and a timeline bar (I made a KML file)to see what incidents happened when and where: so in the meantime, I am including some image files I created using QGIS. Other software I used to do this study: MySQL, R, Python, OpenOffice. I got the data into shape by modifying Mike Dewar’s (of the dataists blog) scripts for accessing the Iraq War Diary.
Figure 1: Fallujah in Iraq
Figure 2: Fallujah in context. Baghdad the major (gray) urban center in the center of the Map, Fallujah is the set of points to the West of it.
Figure 3: The points are deaths in the Fallujah area for all of 2004.
The history of Fallujah 2004 in the Iraq War Diary
The incident that sparked the first battle of Fallujah (March 31-May 1, 2004), codenamed “Operation Vigilant Resolve”, was the killing of four Blackwater military contractors on March 31, 2004. The incident, and the background of the contractors, is described vividly in Jeremy Scahill’s book Blackwater. In the Iraq War Logs, which on careful examination appear highly redacted, record the incident this way:
(%%% 021900APR04) %%% SECURITY EMPLOYEES (US CIVILIANS) TRAVELING IN TWO %%% IN FALLUJAH WERE ATTACKED AND KILLED BY UNKNOWN GUNMEN. VIA , %%% AND %%% ASSISTANCE IN RECOVERING THE BODIES. 2X VEHICLES IDENTIFIED AND LOCATED AT CAMP FALLUJAH; %%% CONFIRMED VEHICLE ID. 4X BODIES RECOVERED AND LOCATED AT CAMP FALLUJAH (%%% HQ); ALL REMAINS ARE %%% DUE TO CONDITION OF CORPSES. %%% BODY REPORTED TO BE TURNED INTO TF /%%% BY MAYOR OF FALLUJAH AT %%% APR. REMAINS ARE CURRENTLY AT CAMP FALLUJAH AND HAVE BEEN EXAMINED BY MORTUARY AFFAIRS. REMAINS SCHEDULED TO DEPART CAMP FALLUJAH THIS EVENING CAMP %%% AND ONTO %%% IN %%% TOMORROW. COORDINATION BEING CONDUCTED WITH CJTF-%%% TO COORDINATE EVACUATION OF REMAINS FROM MEF AO TO BIAP.
Interestingly, these deaths are considered “civilians killed in action” in the database, the same classification, presumably, as Iraqi civilians killed in the course of the war and occupation.
According to Western news reports, what followed this incident was a siege of Fallujah, beginning around April 3, by 2000 troops, along with air strikes and raids into the city. The US declared a unilateral ceasefire on April 9, which lasted until the 13th.
For this period, April 3 to 9, in the vicinity of Fallujah, the deaths portion of the database records a number of incidents. On April 3, east of Fallujah, an Iraqi attacked Coalition Forces with an RPG. Coalition Forces returned fire and killed the attacker, seized the weapon and found some other weapons. On April 4, Marines attacked and killed one Iraqi and detained another – according to the database, the two men were “digging a mortar pit and fighting position”. On April 8, Iraqis attacked a patrol with an IED and followed up with small arms fire. They killed a CF soldier, and the CF killed 3 insurgents and detained 9. The next day, on April 9, insurgents launched rockets at CF Camp Fallujah, killing 1 CF soldier and wounding 3 soldiers and one contractor. Hours later, indirect fire wounded several CF soldiers, who stopped a truck they saw leaving the site and killed the Iraqi driver, “who attempted to run over the dismounted marines and was shot dead.”
During the “ceasefire”, there are several incidents in the database, including an “Other Defensive” incident in which an Iraqi vehicle drove towards a convoy on April 11. Marines shot the engine block, but the vehicle kept coming, so they shot the driver, who “was still alive at first medical assessment” but died shortly after. On April 11, a US tank “identified at least %% individuals moving ammunition boxes between buildings %%% and %%% in eastern Fallujah.” The tank fired on them and killed three of these individuals. On April 12, a US patrol found a weapons cache. When Iraqis fled the scene, they pursued, killed one, and detained the other.
April 13 was the major day of the battle, with 125 Iraqi insurgents killed (and 3 US soldiers). It started at 4:30am on April 13, a friendly fire incident killed 1 Marine and wounded 3 others. The first major engagement was at 9am, when a US Quick Response Force was attacked at a helicopter crash site. Indirect fire wounded 6 soldiers, and two more were wounded as they tried to retreat. After the soldiers retreated, US forces hit the site with mortars and air strikes, destroying the helicopter and killing an estimated 25 Iraqis. The 12pm bombing of a building that killed 100 Iraqis is highly redacted and is quoted in full here:
/%%% WAS ENGAGED .2KM %%% OF %%%. AH-‘%%% FIRED (%%%) HELLFIRE AND (%%%) TOW %%% AT MULTIPLE BUILDING VIC OF GRID %%%. CAS ALSO DROPPED (%%%) GBU-%%%, DESTROYING (%%%) BUILDING AND DAMAGING SEVERAL OTHERS. NO REPORTED CASUALTIES. (%%%) EKIA (%%%) EWIA INITIAL REPORT. INDICATIONS THAT ENEMY CASUALTIES MAY BE HIGHER. UPDATE %%% 15APR04: FURTHER REPORTING INDICATES THAT %%% CO, /%%%, WAS INVOLVED IN INTENSE CLOSE COMBAT RESULTING FROM REPEATED AND COORDINATED ATTACKS ON THEIR POSITION AFTER SETTING UP A %%% DEGREE PERIMETER DEFENSE IVO %%%. DESPITE HEAVY CAS, COORDINATED ATTACKS BY ACF ON THE COMPANY POSITION CONTINUED THROUGHOUT THE NIGHT. INITIAL COUNT OF EKIA IS %%% (WITHIN %%% METERS OF THE COMPANY PERIMETER) AND IS LIKELY HIGHER. AFTER THE BATTLE, TOWN OFFICIALS REPORTEDLY THANKED THE MARINES BECAUSE THE ATTACKERS HAD TAKEN OVER THE TOWN AND CONTROLLED THE RESIDENTS BY FORCE. %%% EKIA, 2X EWIA.
Later that afternoon (3:45pm), US forces killed 2 insurgents in a building using mortars after reports of sniper fire. The dead men were “found with AIF paperwork”.
That evening at about 6pm, insurgents attacked (whether it was a building or a vehicle is redacted out) with RPG and small arms fire, killing 1 and wounding 1 US soldier. These “IZ” (Iraqis) were found with “AIF” (Anti-Iraqi Forces) paperwork.
At the time, US witnesses in Fallujah like Rahul Mahajan noted that there was no ceasefire. He told Amy Goodman:
“Well, they were saying that when I was there, and there was no cease-fire. I mean, I think what they did was they spent a few days without dropping the 500-pound bombs, which is what they were using, 500-pound bombs on residential areas is a terrifying thing. They were not using those, so they called it a “cease-fire… So I think the claim of a cease-fire was a joke and I see no reason to believe it now, either.”
(Rahul Mahajan, speaking to Democracy Now! April 13, 2004, cited in Iraq Body Count, see endnote #1)
Incidents recorded in the database for the next few days follow a similar pattern: insurgents attack with RPG or small arms, Marines return fire or call in air support, insurgents are killed or flee (close to midnight on April 15, the US killed 10 Iraqis this way). Or, US troops attacked insurgents while they were setting up, killing 1 or 2 and detaining several (9:30am on April 15). Similar incidents occurred on April 16 and 17. Also close to midnight on April 17, Marines killed 7 Iraqis who were “unloading weapons and material”.
While the database records no ceasefire between April 9 and 13, there are no deaths on April 18 or 19, during which there was supposedly a plan for joint Iraqi/US patrols in the city.
On April 20-21, there is a series of major incidents. A mutilated corpse was discovered and is recorded as a “murder”. At 1:13pm, a rocket attack killed 22 Iraqi detainees at the US forward operating base: “rounds hit directly on the tents which caused the high casualties.” . At 6:42pm, a US patrol killed 8 insurgents after being engaged with RPG and small arms fire.
On April 21, at 1:20pm, a major battle took place, where US aircraft and soldiers killed 36 insurgents and took 3 wounded after being engaged by insurgents “from the palm grove”:
AIF ENGAGED //%%% FROM THE PALM GROVE. /%%% RETURNED FIRE AND CALLED FOR RW CAS. A MIXED SECTION OF RW AND %%% AIRCRAFT ENGAGED THE ENEMY WITH %%% ENEMY KIA. THERE WERE THREE FWIA’: (%%%) FWIA URGENT MEDEVAC, (%%%) PRIORITY MEDEVAC AND (%%%)FWIA ROUTINE. ENGAGEMENT WAS SPREAD OVER APPROX %%%, WITH SEVERAL ENEMY ELEMENTS INVOLVED. CURRENT BDA ASSESSMENT AS FOLLOWS: TOTAL OF %%% EKIA. %%% CO’%%% REPORTED MUSIC COMING FROM THE MOSQUE TO THE SE. THE LYRICS %%% “GOD IS GOOD, GOD IS GREAT”, AND “HOLY WARRIORS COME OUT TO FIGHT”. THE MOSQUE WAS ALSO TELLING THE PEOPLE TO RISE UP AND FIGHT AND APPEARED TO BE GIVEN ORDERS.
Smaller incidents (patrols and failed ambushes) continued until another major incident on April 24:
%%% RECON BN PATROL OBSERVED SEVERAL SUSPICIOUS IZ MALES WITH SMALL ARMS AND MORTAR SYSTEM IVO FALLUJAH. CAS ON STATION FURTHER OBSERVED A GROUP OF IZ MALES IN THE HOUSE COURTYARD. PATROL ENGAGED GROUP, WHICH RAN TO RETRIEVE WEAPONS AND RETURNED FIRE. CAS ENGAGED HOUSE AND VISIBLE TARGETS %%% IN AN ESTIMATED 25X EKIA. GROUND QRF RESPONDED. CORDON AND SEARCH OF AREA WAS CONDUCTED. THE ATTK ALSO RESULTED IN 1X CF (US) WIA (NON LIFE THREATENING). AS OF 240510D: %%% REPORTED THAT ALL SMALL ARMS FIRE HAD CEASED. %%% CO, %%% RECON IS SET IN AN %%% SHAPE AROUND THE HOUSE OBSERVING IT. AT FIRST LIGHT MARINES ENTERED AND CLEARED THE HOUSE WITH NSTR. ACTIONS CONCLUDED.
Many of the incidents show the Marines engaging with Iraqis aggressively, with incidents (as one on April 24 where 5 Iraqis were killed) starting with US forces “observed 6 IZ (Iraqi) males congregating around a mosque, carrying RPGs and small arms. Enemy assumed firing positions and fired at position…”, or (on April 25, in which 1 Iraqi was killed) observing “5 IZ (Iraqi) males moving to a weapons cache, assume firing positions, and begin firing”. A battle at a mosque on April 26 killed 6 insurgents and 1 Marine.
In the last few days of April, the database records several major (failed) attacks by large groups of insurgents who were basically wiped out by US air support. On April 28, 24 insurgents were killed by bombing when they engaged Marine patrols from buildings. On April 29, first 7, then 12 insurgents were killed in similar incidents.
On May 1, the US announced that it was withdrawing from Fallujah, turning patrols over to Iraqi forces with their support. This tactic failed by September, and the US came back in November and re-took Fallujah.
The total numbers for April 2004 in Fallujah:
Total Killed in Action 340
Coalition Forces Wounded 42
Coalition Forces Killed 7
Host Nation Wounded 0
Civilians Wounded 93
Civilians Killed 24
Enemy Wounded 14
Enemy Killed 309
Many of the 93 civilian wounded and 24 civilian dead in the database are actually military contractors working for the US. The database also doesn’t seem to have any civilian casualties or “collateral damage” incidents for this battle. For those figures, the Iraq Body Count estimate of 600, compiled from all the major media reports, may be the best available.
While some of the dynamics of the military conflict were known and understood from media reports, there are various disconnects between what the public would get from the media and from studying the US’s own war logs. These include:
1. The disproportion of the military conflict (the battle of Fallujah had 309 Iraqis killed and between 7-33 US forces killed, depending on how many of the civilian deaths in the database were contractors).
2. The difference between reported ceasefire incidents and activity reported in the database.
As with the Afghan war diary, it is also important to note that those entering the data would have every incentive to classify all those Iraqis killed, civilian or otherwise, as insurgents. Enemy military casualty figures during wars of occupation should always be viewed with extreme suspicion.
Things did calm down in Fallujah for May (33 deaths), but gradually increased again, with 47 deaths in June, 86 in July, 154 in August, 234 in September and 123 in October, culminating in the second battle of Fallujah in November-December 2004 (310 deaths and 193 deaths). Most of the deaths each month were classified as “enemy killed in action”, and there were 1339 “ekia” out of a total of 1723 killed in 2004 in the Fallujah region.
Figure 4. The Google Fusion Tables for Fallujah 2004. This seems to be the best way to display the information and includes summary information. It’s a little too small here, so see the table on google.
Figure 5. The map of deaths in Fallujah for all of 2004 on Google Maps – a work in progress. I will be working on trying to get it to display more information than the Reportkey field [ADDENDUM: SEE FIGURE 4, as google fusion tables seems to solve the problem]
1) Iraq Body Count’s analysis of Fallujah :http://www.iraqbodycount.org/analysis/reference/falluja-april/
2) Glossary: Al Jazeera, October 24, 2010. “The Secret Iraq Files: US military jargon.” http://english.aljazeera.net/secretiraqfiles/2010/10/20101022172727420698.html