Iraq War Logs: The truth is in the details

• IBC estimates that the Iraq War Logs describe hitherto unreported details of 23,000 violent incidents in which Iraqi civilians were killed or their bodies were found.
• A further 2,000 events exclusive to these logs concern the deaths of Iraqi police deaths and other security forces killed after capture.
• IBC has analysed and integrated 860 of 390,000 logs into its public database.
• From the 860 logs systematically analysed so far, we were able to add to the IBC database 61 previously unspecified incidents, 109 hitherto unreported victim names, and new demographic details for 298 victims.
• Names were most frequently (but not exclusively) found in the numerous low casualty count logs that describe one or two deaths.

Verification and integration

The number of violent deaths in a conflict is an important, and generally difficult and painstaking, fact to establish. The task requires access not just to raw numbers, but to the victim- and incident-level data which produced them: details describing who was killed, where, when, and how.

This is because there usually exist multiple originating sources of documentary evidence on casualties for any given armed conflict, with no single source likely to be fully accurate or complete for an entire war. This inevitably leads to a range of figures, some higher, some lower, with varying amounts of overlap and omissions between them, a pattern also seen in the casualty data on Iraq. A full and authoritative account of violent deaths since the country’s 2003 invasion can only be arrived at by analysing, reconciling, and combining multiple sources into an integrated whole, and this task can only be carried out reliably when it is possible to identify which incidents and which deaths are present (or missing) in each source.

From the beginning of the 2003 invasion IBC has systematically recorded not just the number of civilians killed in the Iraq conflict but also a wide range of associated details. These have included the time, location and other circumstances of each death and, whenever possible, the name and demographic (age, sex, occupation, etc.) of the dead. Consequently every detail in IBC’s public database is open to scrutiny, correction (where necessary), and verification. Most importantly in the context of the Iraq War Logs, this detailed record allows IBC’s information to be compared and reconciled with new data as it emerges.

The unprecedented level of detail in the Iraq War Logs has made this task possible in a way which previous, aggregate, releases of data by the US Department of Defense did not allow. Despite the death counts in the logs apparently being lower than the numerical totals provided in existing sources, including IBC’s count for the same period of the conflict, detailed, incident-level analysis indicates that the logs are likely to add some 15,000 deaths of civilians and police to public knowledge that were not reported elsewhere.

The fact that the Iraq War Logs contain this many previously undocumented deaths would have been impossible to discover if they were only presented as a set of death totals, as is all too often the case with official announcements or publications. Without sight of the incident- and victim-level details contained in the logs there is simply no way to know how these data in fact compare or might contribute to other sources.

Comprehension and humanisation

As well as allowing more complete numbers to be reliably derived from them, incident and victim details such as those found throughout the Iraq War Logs provide important insights in their own right. Unless these are given proper attention, we risk failing to understand not only the characteristics of violence in armed conflict but the humanity of its victims.

Each deadly incident so far reported in the public domain, and every identifiable victim, is individually listed on the IBC website, along with the details available on them. Besides date and location, such incident-level details include the weapons that were used and (whenever possible) the perpetrators involved. For victims, details collected by IBC include familial relationships as well as their age, sex, occupation, and name.

These systematically-collected data have allowed analyses such as those presented in an IBC co-authored paper in the New England Journal of Medicine,2 investigating the victim profiles of various weapons from air-launched missiles to suicide attacks, papers analysing timelines for violence at the district level,3 or in periodic IBC analyses that allow unique insights into longer-term trends, such as those following the 2007 US troop ‘surge’ in Baghdad.4

Yet many thousands of deaths in the IBC database are only listed at the level of aggregated database entries which refer only, for example, to the number of violent deaths recorded at a particular morgue within a particular month, or of multiple bodies found in a specific town on a particular day.

Our analysis of the Iraq War Logs indicates that they will provide far more detail for many of these same deaths currently listed only in aggregated form. These details will allow these deaths in the IBC database to be disaggregated into newly-defined and much more clear-cut ‘incidents’, which include accompanying details regarding their time, location, and the identity of victims.

One case vividly illustrates how the logs will add a dramatic increase to detailed knowledge of the war, even when not necessarily adding new numbers to the deaths total. IBC entry k4435 simply records 35 bodies of persons found, typically killed execution style, across Baghdad over the course of 1 November, 2006. These few data points - the number, city, date and general cause of death - were the highest level of detail that existed in the public domain about these deaths until now. The Iraq War Logs also contain 35 bodies found on the same date in Baghdad, but spread across 27 logs specifying a wide range of details, including the precise neighbourhood and time of day where particular bodies were found and, in many cases, the demographics and identities of those killed.

Details present in IBC entry k4435 in its original form:

Time	Number Killed	City	Sub-city	Weapons	Victim details
	35	Baghdad		gunfire, executed

The table below shows 27 incidents in the logs recording bodies found in Baghdad on 1 November, 2006, that will likely replace the single IBC entry above. The columns in the table show the wealth of incident and victim details that will be added to the public record when the Logs are fully integrated into the IBC database:

Time	#	City	Sub-city	Weapons	Victim details
1:22	1	Baghdad	Route Al Amin, Al Sadr district	gunfire, executed
8:13	2	Baghdad	Karkh district	gunfire, executed	names, occupation, sects
10:35	1	Baghdad	Route Kamaliyah, Al Sadr district	tortured, strangled
10:45	1	Baghdad	Washshash village, Karkh district	gunfire, executed	age, gender, sect
12:00	1	Baghdad	Haifa Street, Karkh district	gunfire, executed
12:30	1	Baghdad	Al Mammon, Karkh district	gunfire, executed	gender
13:00	1	Baghdad	Diyala Bridge area, Al Resafa district	gunfire, executed	gender
13:00	4	Baghdad	Bab Al Muadam area, Al Resafa district	gunfire, executed	gender
13:45	1	Baghdad	Kadhimiya area, Kadhimiya district	gunfire, executed	name, age, gender
14:00	2	Baghdad	Al Mammon, Karkh district	gunfire, executed	names, ages, genders, sects
14:18	1	Baghdad	Al Mammon, Karkh district	gunfire, executed	age, gender
14:30	1	Baghdad	Route Al Rasheed, Hay Al Amil, Karkh district	gunfire, executed	gender
14:40	1	Baghdad	Karkh district	gunfire, executed
14:50	1	Baghdad	Al Mammon, Karkh district	gunfire, executed	name, age, gender, sect
15:00	1	Baghdad	Route Al Ghazaliya, Karkh district	gunfire, executed	age, gender, occupation
15:15	1	Baghdad	Route Al Shames Market, Hay Al Adel, Karkh district	gunfire	name, gender
15:30	1	Baghdad	Route Al Mansour, Al Ghazaliya, Karkh district	gunfire, executed	gender
16:00	3	Baghdad	Route Al Ghazaliya, Karkh district	gunfire, executed
16:20	1	Baghdad	Route New Baghdad, Al Amin, Al Sadr district	gunfire, executed, tortured	gender
16:25	1	Baghdad	Karada, Al Resafa district	gunfire, executed	gender
16:30	1	Baghdad	Route Al Thawra, Sadr City, Al Sadr district	gunfire, executed	gender
17:00	1	Baghdad	Mada’in district	gunfire, executed	gender
17:00	1	Baghdad	Route Kamaliyah, Al Sadr district	gunfire, executed	gender
17:10	1	Baghdad	Al Sadr district	gunfire, executed	gender
17:30	1	Baghdad	Route Al Rashid, Al Saydiya, Karkh district	gunfire, executed	gender
18:35	2	Baghdad	Route Al Mansour, Karkh district	assassinated	names, genders
18:40	1	Baghdad	Al Mansour area, Karkh district	assassinated	name, age, gender, occupation

Once fully analysed and coded by IBC, this one day’s account of deaths in Baghdad, rather than being presented as a single database entry of “35 bodies found,” with little accompanying details about the specific incidents or victims involved, will be split into over 20 more detailed incidents whose usefulness will extend not just to a properly-detailed historical account of the conflict but possibly also to anyone whose loved ones went missing on or near that day.

Based on our preliminary analysis of 860 logs, matching these to IBC entries for incidents in the same locations and timeframes, we estimate that the Iraq War Logs will add around 25,000 newly-specified incidents to the IBC database, of which 23,000 will define events involving civilian deaths and around 2,000 will relate to the deaths of police and Iraqi security forces killed after capture. As noted in the previous section, some of these new incidents will also add new deaths (estimated at roughly 15,000), while others will provide new event and victim details for deaths already recorded in a more aggregated, less detailed form, as in the tables shown above.5

While it is important for incidents to be as clearly defined and distinguishable from each other as possible, the same holds even more true for victims. A full list of every victim of the conflict, individual by individual, name by name, remains the ultimate long-term goal of the IBC project.

In explaining this approach we wrote, in 2008,

It is always difficult to convey the gravity and pain contained in the experiences behind the numbers. Although we may speak of dozens killed in a particular incident or thousands over a span of time, it is of course individuals who die and are mourned. And no society considers itself civilised if it fails to give the dead the final dignity of recognition not as a mass, but as individuals.6

In essence, IBC’s project is one that ultimately seeks to answer not just how many, but who has been killed as a result of the US/UK’s 2003 invasion of Iraq and the country’s subsequent collapse in civil security.

To our surprise, the Iraq War Logs contain an unprecedented level of detail regarding individual Iraqi victims. We first noticed this when matching an entry in the logs to another previously-reported case where bodies were found, this time in a mass grave containing 28 dead, recorded in the IBC database entry k4724. While the logs list the same event, they also note that the families of the dead were able to identify the victims involved, and then list all 28 names.

From our systematic evaluation of 860 of the logs (which is one five-hundredth of the total number of logs, but includes all of those listing more than 20 deaths), we have already found 109 hitherto unreported victim names, as well as new demographic data for 298 victims. However, most victim names appear to be attached to smaller incidents where one or two people were killed, and form by far the largest proportion of logs containing casualty data. Retrieving all of the victim names that may be present in these 390,000 logs will require much more work.7

But like the other casualty details in the Iraq War Logs these names too, belong in the public domain – as a memorial to the dead and public recognition of the loss suffered by their families, and, indeed, as another means of establishing the truth from its details. Only a list of named individuals, visible to all including those who knew them in life, can ever provide full verification, without omissions or duplication, of the death toll of the war.