Iraq Body Count: undercounting death with pro-war cash

An official funding document confirms that the Iraq Body Count (IBC), a widely-respected ‘independent’ antiwar group monitoring Iraq War casualties since its inception in 2003, has received core funding from government agencies complicit in the Iraq War, since 2009. A close review of evidence in the public record also reveals that in 2006, the Pentagon used IBC’s data to legitimise US violence under the military occupation of Iraq.

In an earlier investigation, I reported that IBC’s directors received funding from US, Swiss, German and Norwegian government foreign affairs agencies that had directly supported the war.

This funding was received through a parallel casualty-recording programme at the Oxford Research Group (ORG), run by the same directors. The funding, however, was not disclosed to the scientific journals publishing papers by IBC directors, which analysed the Iraq War in the same period.

An official document in the public record from one of IBC’s funders, however, clarifies that the ORG programme was being run by the IBC, which was therefore directly supported by this funding.

 

Response and counter-response

 

Last week, Brian Dean, a longtime IBC supporter and blogger, wrote a response dismissing this investigation as nothing more than a “smear piece.”

Leading leftwing Guardian columnist George Monbiot took to Twitter to promote Dean’s blog, describing my IBC story as “pernicious bullshit.” He accused me of promoting a “witch-hunt” against IBC based on “no evidence.”

IBC’s Josh Dougherty, a senior researcher at the organisation, similarly described my follow-up rebuttal to Dean on Twitter as “16,000(!) more words of pure bullshit,” indicating IBC’s complete denial of my reporting.

Josh Dougherty, Senior Researcher at the Iraq Body Count (2003)

Dougherty’s tweet was favourited by the Iraq Body Count’s official Twitter handle.

 

Lies, damn lies and statistics

 

IBC’s blanket denial as per Dougherty, however, contradicts information published by one of its own grant-givers.

In Brian Dean’s defence of IBC, which received a resounding endorsement from Dougherty, Dean repeatedly asserted alleged facts to the following effect:

“The funding (from sources ‘connected to US and European government foreign policy agencies and departments’) which Ahmed believes should have been ‘disclosed’ by IBC and/or its publications was not, in fact, received by IBC or any of its publications.”

Brian Dean who runs a blog called ‘News Frames,’ wrote this in his rebuttal

However, an entry in the 10th Anniversary Impact Report of one of IBC’s early funders, The Funding Network (TFN), reveals that the IBC itself was directly funded by the US, Swiss, and German governments according to its own directors.

The July 2012 impact report describes the activities of TFN, a grant-giving body, and the charitable causes TFN has funded. Among the examples of successful impacts showcased in the report is a case study authored by IBC directors, John Sloboda and Hamit Dardagan.

The entry, titled “Iraq Body Count,” describes how TFN funding enabled the purportedly ‘independent’ antiwar group to obtain financial support from the US, Swiss and German governments. The document also confirms that as far as the IBC’s own directors are concerned, for all practical purposes Iraq Body Count is indistinguishable from the casualty recording programme run through the Oxford Research Group, along with the project it seeded known as ‘Every Casualty.’

The TFN report shows that grants received by IBC financed further development of the IBC website, and also “funded IBC personnel to establish a process at the Oxford Research Group (ORG) to produce a consultation document which was circulated to 250 other stakeholders worldwide with an interest in casualty recording.”

Excerpt from The Funding Network 10th Anniversary Impact Report, authored by “Iraq Body Count — John Sloboda & Hamit Cardigan, Oxford Research Group”

The ORG process, established by IBC, led “to the establishment of the Every Casualty programme which aims to put in place a casualty recording requirement within the international system that ensures that every casualty of armed violence around the world is correctly identified and publicly acknowledged.”

Sloboda and Dardagan go on to confirm that:

“As a result of TFN’s support, we now have the support of other major funders including the Swiss and German governments and the US Congress-supported United States Institute for Peace. This has allowed us to establish partnerships with a wide range of NGOs, including virtually all the recognised casualty recording organisations around the world.”

The TFN impact report thus clarifies that Sloboda and Dardagan established the Every Casualty programme at ORG in their capacity as “IBC personnel,” and that the development of the ORG programme was led by IBC.

The report shows that IBC’s founding directors, Sloboda and Dardagan, do not meaningfully distinguish between the Every Casualty programme and IBC.

In the same document, the longstanding organisational overlap between ORG and IBC, especially with regard to funding, is described by ORG founder Dr. Scilla Elworthy. She recounts her excitement at:

“….Oxford Research Group becoming the second highest funded project in TFN history with a total amount raised over the years of £58,000 for our Iraq Body Count initiative.”

Excerpt from The Funding Network 10th Anniversary Impact Report

The statement by Dr. Elworthy, ORG’s founder, confirms that IBC is equivalent to ORG’s casualty recording “initiative.”

The TFN report thus confirms that the first major grant of $100,161 from the US Institute for Peace (USIP) in 2009, was secured by Sloboda for IBC through ORG. The grant was to “define and test a generalizable framework for enumerating the casualties of armed conflict” by drawing on “a newly-created international network of casualty recording practitioners.”

Since then, the IBC-run Every Casualty programme at ORG has secured funding from the Swiss Federal Department of Foreign Affairs, the German Federal Foreign Office, and the Norwegian Ministry of Foreign Affairs. Some of this funding was used to launch Every Casualty as an independent organisation, under the directorship of IBC personnel. All four governments were supportive of the 2003 Iraq War.

While IBC’s aims would appear noble, their relationships with the very government agencies complicit in the Iraq War underscores serious conflicts of interest.

 

Systematic undercounting

 

New peer-reviewed scientific research from the Human Rights Data Analysis Group (HRDAG) out last week proves clearly that casualty counting methods advocated by IBC and Every Casualty systematically undercount death tolls on conflicts, and cannot be taken as a true reflection of the nature of violence in conflicts.

In a paper published this month by the Statistical Journal of the IAOS (the International Association for Official Statistics), HRDAG analysts Dr. Patrick Ball and Megan Price identify serious limitations to data produced by groups like IBC and Every Casualty.

“In our analyses of mass violence in Guatemala, Kosovo, Perú, Colombia, and Timor-Leste,” they write referring to their work studying conflicts over the last 20 years, “we have repeatedly found that the amount of missing data can be surprisingly large: in the chaos and fear that surrounds conflict, violence often goes unreported and consequently victims remain hidden from view.” [emphasis added]

Access to “complete data” is rare for studies of conflict violence. “What we have instead are snapshots of violence,” which in statistical terms are known as “convenience samples” that cover:

“… an unknown proportion of the total number of cases of violence. It is mathematically difficult, often impossible, to know how much is undocumented, and consequently missing from the sample.” [emphasis added]

Despite this, IBC personnel, including founding director John Sloboda, have repeatedly treated the IBC database as largely complete for analytical purposes, as illustrated by their blanket efforts to infer trends in violence — a practice that Ball and Price note is “highly misleading.”

IBC personnel have published reports and papers attempting to describe patterns of violence, and to provide casualty figures, for Syria and Iraq.

Yet according to Ball and Price, in both countries “the multiple data sources that cover these conflicts tell conflicting narratives and are not appropriate for quantitative analyses aimed at answering questions about patterns of violence.”

HRDAG draw on the work of a team led by Dr. Dustin Carpenter of Columbia University Medical Center, who note that correlations between certain types of events and media reporting of those events can generate data patterns that simply reflect the limitations of reporting, rather than actual violence on the ground.

Ball and Price warn that researchers might for instance “erroneously conclude that most victims in Iraq were killed in large events, whereas this may be an artifact of the data collection.” An even more damaging and false conclusion “might be reached if large events are centered in certain geographic regions or attributed to certain perpetrators.” Inappropriate interpretations could result in “incorrect decisions regarding security measures, intervention strategies, and ultimately, accountability.” [emphasis added]

Most significantly, they quote Dr. Carpenter et. al who write:

“Citing partial tallies as if they were scientific samples confuses the public, and opens the press and scholars to being manipulated in the interests of warring parties.”

Similarly, in a 2013 paper in The Lancet, public health experts Prof. Richard Garfield and Prof. Frederick Burkle Jr. point out that in the “passive surveillance” method used by IBC, “no special effort is made to find those deaths that go unreported.” IBC’s data is:

“… inherently biased because of scarcity or absence of independent verification, variation in original sources of information, and underestimation of mortality from violence… In research circles, random cross-sectional cluster sampling survey methods are deemed to be a more rigorous epidemiological method in conflict settings.”

They acknowledge, though, that so far such studies have been significantly hampered in Iraq, due to worsening security conditions and other factors. The true impact of the war remains difficult to determine to this day.

 

Pentagon research uses IBC data to sanitise US violence

 

While there is no indication that IBC has deliberately undercounted violence in Iraq, its lower numbers — compared to the findings of epidemiological surveys — have been of increasing interest to the US government and the Pentagon.

In their 2013 paper in The Lancet, Garfield and Burkle Jr. point out:

“Although initially critical of the IBC findings, US and coalition governments and the political right increasingly became dismissive of the peer-reviewed mortality estimates, asserting that the deaths reported in the lower estimate sources [like IBC and others such as the Brookings Iraq Index] were accurate.”

The risk of “being manipulated in the interests of warring parties,” as Dr. Carpenter warned, was very real in the case of IBC’s inherently partial tally.

A figure at the Pentagon who displayed early interest in IBC’s low numbers is Colin Kahl, who is currently National Security Adviser to Vice President Joe Biden and Deputy Assistant to President Obama.

From 2005 to 2006, Prof. Kahl, then a political scientist at the University of Minnesota, conducted research at the Bush administration’s Department of Defense under a Council on Foreign Relations fellowship on counterinsurgency warfare in Iraq.

His Pentagon research resulted in a Foreign Affairs article published in Nov/Dec 2006, in which Kahl used the IBC database of violent deaths to prove that the US military’s conduct in Iraq was exemplary by historical standards, and continuing to improve. His Foreign Affairs piece concluded from analysis of the IBC data that:

“… a careful review of US conduct during the Iraq war reveals no broad pattern of systematic civilian victimization by US forces. US compliance with noncombatant immunity in Iraq has been higher than critics often assert, and adherence has increased over time as the US military has tried to correct its procedures in reaction to instances of noncompliance. Observed through the narrow lens of the laws of war, the US military has gone to commendable lengths to comply with the principles of distinction and proportionality in Iraq.”

The implications of this cannot be understated: The research for Kahl’s late 2006 Foreign Affairs piece, which sanitised the US role in violence in Iraq using IBC data, was conducted by Kahl under his fellowship at the Pentagon from 2005 to 2006.

It was, in other words, Colin Kahl that first saw the utility of IBC’s work to the Pentagon’s policy goals in Iraq.

After the October 2006 publication of the Lancet survey which found that 655,000 Iraqis had been killed due to the 2003 invasion and occupation, President Bush dismissed the survey during a White House news conference. “I don’t consider it a credible report. Neither does Gen. Casey,” he said, referring to the top ranking US military official in Iraq, “and neither do Iraqi officials” working with US occupying forces.

Kahl’s Foreign Affairs article, based on two-year research under his CFR fellowship at Bush’s Pentagon, came out a month after the Lancet study.

In an interview about the article at the time, Kahl specifically dismissed the Lancet findings citing his Pentagon research. Instead, he said, while IBC’s figures are likely to undercount deaths, “most estimates of the Iraqi civilian casualties put the totals somewhere between 50,000 and 150,000, but the Iraq Body Count data is the only one that breaks it down by who is doing the killing.”

Explaining his findings, he said:

“The article claims that by a historical standard, the United States is doing reasonably well. Does that mean the United States has never killed innocent civilians? No. In fact the United States has undoubtedly killed thousands of Iraqi civilians, and we shouldn’t trivialize that. On the other hand, in a single night of bombing in Tokyo 85,000 Japanese civilians were killed on purpose by the United States. Are we seeing carnage like that in Iraq by US forces? No, so I think we need to keep in mind that war is hell, and civilians are going to die and the only way to prevent that is to not fight the war in the first place.

But also keep in mind, that when we fight its incumbent upon us to take every precaution necessary, and the article argues that the United States military has done a reasonably good job by historical standards, and that that its performance has been improving over time. Could it still get better? The article concludes that it could.”

Needless to say, if the Lancet estimates — or substantively higher estimates — were anywhere near accurate, such claims would be unsustainable.

Throughout his Pentagon-sanctioned article, Kahl deploys IBC’s data to justify his overlapping assumptions of supreme US benevolence in its conduct of war: that the US has always formally intended to minimise civilian casualties, and that killings of Iraqi civilians are reprehensible aberrations from that benign intent, which do not represent the general character of US military operations.

This, however, is fundamentally contradicted by authoritative testimonials from unembedded journalists in Iraq and Iraq War veterans, who confirm that media reporting failed to pick up on major and routine violent incidents against Iraqi civilians by US forces.

As I wrote in my first story on IBC, Iraq Veterans Against War (IVAW) has compiled testimony from dozens of US soldiers who served during the war, documenting that commanding officers routinely ordered the establishment of “free fire zones” in civilian areas where there were supposedly “no friendlies”, but where only civilians were actually visible. IVAW’s work provides first-hand evidence of countless cases of high-level orders to shoot at anyone, with senior officers guaranteeing no recriminations. Such incidents were never reported by the media — until IVAW published them.

In corroboration, Iraq journalist Salah Hassan told the unembedded journalist Dahr Jamail that, “In Fallujah and other cities” occupation forces committed “many crimes freely because there were no journalists there.”

Jamail himself, in his book Beyond the Greenzone: Dispatches from an Umembedded Journalist in Occupied Iraq, confirmed that he had witnessed numerous cases and evidence of violent deaths that neither Arabic nor English media ever noticed.

 

USIP, Colin Kahl, and Iraq casualty bias

 

In January 2007, Colin Kahl and John Sloboda were together in Washington DC at an off-the-record USIP panel, where Sloboda delivered a presentation about IBC’s views of the Iraq War death toll at an off-the-record USIP panel. By that time, Kahl was now an adviser to the soon-to-be president, Barack Obama, as part of the latter’s Iraq Policy Expert Group.

Previous reports at INSURGE intelligence show that USIP was directly complicit in the Bush administration’s military occupation of Iraq, having established an office in Baghdad through which USIP staff liaised with senior officials from the US military, Coalition Provisional Authority, and Iraqi government.

USIP has also demonstrated a particular bias on the question of the death toll from the war, as well as responsibility for violence.

The Bush administration’s Iraq Study Group convened by USIP concluded in its December 2006 report that the bulk of violence was due to “a Sunni Arab insurgency, Shi’ite militias and death squads, al-Qaeda, and widespread criminality.” Meanwhile, “the presence of US troops is moderating the violence.”

The report claimed that sectarian attacks that killed Iraqis, and roadside bombs targeting US personnel, were being undercounted — rather than deaths of Iraqis due to US forces. The report recommended that the “Director of National Intelligence” and the “Secretary of Defense” support changes “in the collection of data about violence and the sources of violence in Iraq.”

Since inception, federal law requires the US Secretary of Defense to be a USIP board member, among other senior government officials.

Sloboda and Kahl’s USIP panel occurred just one month after USIP published the Iraq Study Group report.

The USIP panel also included two other participants: Michael O’Hanlon, senior author of the Brookings Iraq Index; and Les Roberts, a co-author of the 2006 Lancet Iraq death toll survey.

Roberts, whose Lancet study concluded that the war had taken the lives of well over 650,000 Iraqis, was outnumbered at the USIP meeting three-to-one — Sloboda, Kahl and O’Hanlon were all staunch advocates of a far lower Iraq War death toll derived from ‘passive surveillance’ methods — that is, monitoring of media and other public record accounts of violence.

At the USIP panel, Sloboda said in criticism of the 2006 Lancet survey that “We do not think [the death toll in Iraq] could possibly be 10 times higher” than IBC’s figures.

In the summer of 2007, Kahl published a paper in the International Security journal, largely reiterating his argument developed in Bush’s Pentagon, that US military conduct in Iraq had been exemplary by historical standards.

He once again drew heavily on Iraq Body Count data and arguments by IBC personnel, in particular to argue that the 2006 Lancet findings were the by-product of “some significant sampling bias or other problem.”

By November that year, Colin Kahl was among a list of people, including Hamit Dardagan, Josh Dougherty, and Madelyn Hicks (all IBC personnel) who received thanks from future IBC adviser Michael Spagat for “sharing insights with me that were useful in the presentation of this document.”

The presentation concerned a talk delivered by Spagat, a professor of economics at Royal Holloway University, titled ‘How Many Dead (Really?): Fact and Fiction Regarding Civilian Casualties in Iraq.’ The presentation at George Mason University in November 2007 attempted to critique the larger Iraq death toll estimates from the 2004 and 2006 Lancet surveys.

Spagat summarily dismissed the latter respectively as containing “little, if any, useable information” and possessing “serious quality problems and a number of unexplained anomalies,” before concluding that “IBC is the best source on violent killings of civilians in Iraq.”

By 2008, Sloboda and Dardagan had incorporated Conflict Casualties Monitor as the nonprofit company running IBC.

That year, Colin Kahl, as director of Center for a New American Security’s (CNAS) Middle East Security Programme, lead authored a CNAS report on US policy in Iraq. The report, co-authored with CNAS founder and then CEO Michele Flournoy, also drew on IBC data to argue that the “security situation in Iraq had significantly improved due to the influx of additional US troops utilizing improved counterinsurgency techniques.”

Kahl was still on the Obama Campaign’s Iraq team at this time. As for Flournoy, she had previously been a senior Pentagon official under the Clinton administration, serving as Principal Deputy Assistant Secretary of Defense for Strategy and Threat Reduction and Deputy Assistant Secretary of Defense for Strategy. In late 2008, she was a Review Team Lead at the Pentagon for Obama’s transition, before being appointed as Under-Secretary of Defense for Policy under Bush administration hangover Defense Secretary Robert Gates.

In September 2008, Kahl presented his and Flournoy’s findings at another USIP panel. Once again, the panelist who was critical of the US role in Iraq, Charles Knight — co-director of the Project for Defense Alternatives — was outnumbered three-to-one.

While Knight had argued that the US intervention in Iraq had been an unmitigated disaster, and that the US military presence was the main factor in galvanising sectarian violence, Colin Kahl and two other panelists — including USIP senior fellow Rend al-Rahim Francke, who was appointed by Bush administration as Iraqi ambassador to the US from 2003 to 2005 — largely agreed that the surge strategy was generating lasting success, implying that “some US military presence will be necessary for the near- to medium-term.”

The following year, Colin Kahl would also join Flournoy in the Pentagon. Obama appointed Kahl to the post of Deputy Assistant Secretary of Defense for the Middle East, a position he held until December 2011.

Also in 2009, USIP authorised its first major grant to John Sloboda for IBC to run what would become ORG’s Every Casualty programme, which has been followed by further core funding since then.

At that time and for the duration of much of USIP’s funding to ORG-IBC, USIP’s Executive Vice President was Tara Sonenshine, a colleague of Flournoy in the Clinton administration, who had served as Director of Foreign Policy Planning and Deputy Director of Communications for the National Security Council, as well as Special Assistant to the President. Like Flournoy and Kahl, Sonenshine also went on to join the Obama administration from 2012 to 2013 as Under Secretary of State for Public Affairs and Diplomacy.

In 2010, Michael Spagat formally joined the IBC-run Every Casualty programme, receiving a two-year grant from ORG for casualty-recording research.

 

Michael Spagat, the Pentagon and NATO

 

Spagat was already well-known to defence circles. From 2004 to 2007, he received grants totaling $334,135 from a Pentagon contractor, Radiance Technologies, to develop databases on armed violence in Colombia and Mexico. Radiance’s raison d’etre is to provide “operational support for the Department of Defense (DoD), armed services, intelligence agencies, and other Government organizations.”

Radiance’s Pentagon work in this period on Latin America was on behalf of the US Army’s Foreign Military Studies Office (FMSO), but the project also covered countries in the Middle East, including Iraq.

In 2009, Spagat collaborated with IBC director Madelyn Hicks to create the ‘Dirty War Index’, (DWI) a tool to measure human rights abuses in armed conflicts. The DWI has, however, been heavily criticized by the statistics experts at HRDAG, who question its “basic utility and applicability.” The DWI could, due to underreporting of extreme violence, artificially “inflate the proportion of ‘clean’ outcomes, obscuring what they are intended to measure” — namely, illegal violence against civilians.

Spagat and Hicks worked with Lt. Col. Ewan Cameron, then Head of Health Services Development at NATO Headquarters Regional Southern Command in Kandahar, Afghanistan, to adapt the DWI tool to create “the Civilian Battle Damage Assessment Ratio” (CBDAR) — a new metric that would purportedly help minimise civilian casualties from NATO operations.

Although the CBDAR metric was claimed by NATO to have reduced civilian casualties in southern Afghanistan, this claim has been contested by another IBC advisor, Prof. Marc W. Herold, of the Whittemore School of Business and Economics, University of New Hampshire. Herold documented an increase in Afghan civilian casualties in the same period, rather than a decline.

In April 2010, the same year Spagat was consulting for the USIP-funded IBC programme at ORG, the journal Defence, Peace and Economics, published Spagat’s major critical review of the 2006 Lancet survey of Iraqi deaths.

In a footnote, Spagat once again acknowledged the input of Colin Kahl, who at the time of publication was a senior Pentagon official for Middle East affairs. Spagat also acknowledged IBC personnel, namely John Sloboda, Hamit Dardagan, Josh Dougherty, and Madelyn Hicks.

 

Whose fraud?

 

In his response to my original investigation, Brian Dean makes much of my allegations of statistical fraud on the part of IBC scholars, especially Spagat. Yet it was Spagat and other IBC personnel who first laid charges of fraud at the door of the Lancet teams.

Spagat’s paper drew on similar arguments he had tested earlier in his 2007 presentation at George Mason University, where he had also acknowledged Kahl’s “insights.” As with that presentation, Spagat’s 2010 paper leveled serious charges of fraud, negligence, and ethical breaches at the 2006 Lancet Iraq death toll survey.

However, in my previous investigation I undertook an in-depth analysis of each of Spagat’s core criticisms of the Lancet survey, finding that, to the contrary, almost all of his arguments were spurious, and in several cases I found that Spagat had created statistical artifacts based on entirely false assumptions about Iraqi society to prove his accusations of fraud.

My investigation even led statistics expert Prof. Andrew Gelman of Columbia University, a critic of the Lancet survey based on Spagat’s work who felt that my original report had not sufficiently represented his negative opinion of the survey, to concede in the Washington Post that:

“Ahmed may be correct that the Burnham study was performed well and that Spagat’s criticisms are baseless and that my acceptance of Spagat’s criticisms were misinformed — as I wrote above, I know nothing about Iraq.”

As is widely acknowledged, the IBC provides a valuable service in collating data from media reports and other sources about violent deaths in Iraq. There is no evidence that IBC deliberately undercounts deaths.

But this investigation confirms that the US government and the Pentagon has sought to use IBC’s work for its own ends, namely to draw attention to lower counts of civilian deaths, and to legitimise US military policies in Iraq.

In total, four pro-war governments are currently involved in funding IBC and IBC personnel since 2009, none of which has been declared in the scientific journal articles related to the Iraq War by IBC authors.

In particular, Colin Kahl, a senior Pentagon official in the Obama administration, has consistently used IBC data to sanitise US violence in Iraq, and to dismiss the higher estimates of epidemiological studies in the Lancet.

IBC personnel, especially Spagat, Hicks, Sloboda and Dougherty, have played the lead role in casting doubt on the Lancet studies, and higher death toll estimates in general. The result of their concerted campaign has been a widespread perception that IBC’s lower count of violent deaths is the most accurate, while higher estimates are completely bogus.

But if the higher Iraq death toll estimates of the Lancet, other surveys and studies are correct — some of which imply total deaths of over 1 million Iraqis to date since 2003 — this would corroborate the point made by HRDAG and others, that IBC’s data is really nothing more than a useful but inherently limited snapshot of the true scale of violence and its impact in Iraq.

In which case, the allegedly profound patterns of violence IBC scholars are inferring from their media-based conflict data — not just for Iraq, but for Syria and other conflicts — are simply statistical artifacts reflective of the serious limitations of embedded corporate media reporting on wars in which the US and Western Europe have strategic interests.

This article was amended on 11th June 2015 to ensure that Brian Dean’s statement about IBC’s funding on his ‘News Frames’ blog was conveyed using his own words.