Benefit of the Doubt

If the British government is to be believed, only one civilian has been killed by its armed forces during its air war against the Islamic State in Iraq and Syria. On 26 March 2018, according to a statement made by the then defence secretary, Gavin Williamson, a British drone was preparing to strike three IS fighters in the eastern Syrian desert when a civilian motorcyclist zoomed into the target area. It was too late for the drone’s operator to reverse course and the man was killed along with the combatants.

As part of the US-led Operation Inherent Resolve, Britain has dropped more than 4300 bombs on Iraq and Syria since 2014, many of them in densely populated urban centres, and claims to have killed more than four thousand IS fighters. The US admits that the coalition has killed at least 1437 civilians, though the likely toll is far higher. Airwars, a British civilian casualty research organisation, puts the number between eight and thirteen thousand.

In 2021, more than a thousand documents were declassified and shared with the New York Times, detailing how the US determines whether civilians have been killed or injured in its airstrikes on Iraq and Syria. If there is evidence, such as strike footage, that suggests it is more likely than not that a civilian was harmed, the report moves up the chain of command for more detailed review. Even with this relatively low standard of proof, based on a balance of probabilities, the US undercounts the deadly effects of its operations, as reporting on the ground has demonstrated. There is no reason to believe that British strikes are so much more precise than US ones, so what accounts for the dramatic difference in civilian casualties?

The explanation for the discrepancy may lie in the way the British armed forces count the dead. But so far the Ministry of Defence has refused to disclose how its processes work. Britain may use a higher standard of proof to trigger a review than the US, which would explain the carefully crafted language that British officials use (they have seen ‘no evidence’ or have ‘no reason to believe’ that more civilians had been harmed). Britain’s coalition partners told the BBC in 2020 that other possible incidents were flagged with the Royal Air Force, but the RAF concluded there was no evidence of harm to civilians. US officials said the British were ‘looking for certainty’ where there was none.

Mark Lancaster, the then armed forces minister (and now a member of the House of Lords), told the House of Commons Defence Committee in April 2019 that ‘it is not our position that there has been only a single civilian casualty as a result of our military action. What we are saying is that we have evidence of only a single, or what we believe to have been a single, civilian casualty. That is a rather different position.’

In 2021, Airwars filed a Freedom of Information request in an attempt to find out what happened on 26 March 2018 – according to the US, no coalition strikes were conducted in the area that fit the description, and none of the Syrian organisations that reliably monitor airstrikes recorded it – and to understand how the British military assesses evidence of harm to civilians from its airstrikes. The MoD rejected the request on grounds of national security, but Airwars challenged their refusal on grounds of public interest. In November 2023 Airwars took the MoD to court. The tribunal’s ruling was expected late last year but continues to be postponed.

At the tribunal, Alexander Oliver, an MoD official from the department responsible for overseeing strikes during the campaign against IS, was asked if the UK uses the same process for assessing harm to civilians as the US. He said only that they use the same process for ‘recording’ all airstrikes, and then ‘recording publicly’ if a civilian was killed. Part of the tribunal took place in private so the MoD could present classified evidence. When asked if he knew what standard of proof the military used to make decisions about civilian casualties, Oliver said he did not have that information. Asked if he meant he could not answer the question in the open, Oliver confirmed that he did not know the answer.

This is a remarkable admission. The standard of proof would determine which cases reach the MoD for review. As the civilian body responsible for exercising democratic control over the armed forces, the MoD would presumably want to know how many suspected cases never reach their desks, and the strength of the evidence used to assess the ones that do.

Then again, perhaps not. Standards of proof are not neutral features of public institutions. They are political: what counts as evidence and who must prove what to whom is always bound up in questions of power. A higher standard of proof means the MoD has effectively given the military the benefit of the doubt. The onus is then placed on civilians in Iraq and Syria to prove that they were harmed to a high degree of certainty – next to impossible unless the military were properly researching the effects of its strikes.

There are always trade-offs. A lower standard of proof might mean more false positives, while a higher standard might mean more cases slip through the cracks, and bad actors and harmful practices persist undetected. The approach that an oversight body chooses depends on what it wants to know. This is why standards of proof tend to fall when an institution loses credibility. Following recent high-profile cases of police violence, for example, the requirement for bringing cases of police misconduct fell from ‘beyond reasonable doubt’ to ‘balance of probabilities’.

This is not the first time that Britain’s defence establishment has had trouble with counting. The tribunal’s decision could also shed light on what (if anything) has changed since the Chilcot Inquiry into the UK’s role in the Iraq war. John Chilcot’s report revealed a host of problems with the way the British defence establishment handled information, the weight it gave to evidence and the inferences it passed off as fact.

The inquiry also found that the government ‘has a responsibility to make every reasonable effort to understand the likely and actual effects of its military actions on civilians’ and should have done so during Iraq. But Adam Ingram, Tony Blair’s armed forces minister, told the inquiry that establishing these facts would not have changed what happened on the ground. ‘It would have confirmed what everyone knew, but it wouldn’t have led to a solution.’

But what did everyone know? On the eve of the 2003 invasion of Iraq, Tony Blair told Parliament that Saddam Hussein’s regime would be responsible for ‘many, many more deaths even in one year than we will be in any conflict’. In the face of mounting evidence that the coalition’s military activities were killing thousands of Iraqi civilians, Blair publicly stuck to his line.

Behind the scenes, however, things seemed less certain. In 2004, Blair wrote to cabinet colleagues asking how many civilians had been killed in Iraq. ‘The figure of 15,000 is out there as a fact – is it accurate?’ he asked. ‘Accuracy’ became the basis on which Blair’s government rejected estimates produced by media reports, civil society, and academic researchers. Senior officials from civilian and military branches of government repeatedly told Blair that producing ‘accurate’ figures was simply not possible and should be avoided. ‘This is not merely our public line but our genuine judgment,’ an MoD official wrote in October 2004.

But as the extent of violent civilian deaths across Iraq became impossible to deny, the government adopted a new argument. Civilians were dying in Iraq, Blair admitted in 2006, but they were being killed by ‘terrorists’, not by coalition soldiers. In private, officials looked for data that would support this claim while also countering research that suggested excess deaths may have reached 100,000.

The politics of counting those killed in the war became so distasteful to the government that officials resorted to making dubious legal arguments against collecting data at all. An official from the FCO wrote that Britain had no obligation under international humanitarian law to understand how many were actually killed so long as risks to civilians were assessed before an operation. In effect, refusing to accept any civilian casualty figures became yet another form of unjustified certainty (an ‘unknown known’, as Stephen Colbert put it in a conversation with Donald Rumsfeld – something we know, but choose to ignore). If reliable figures did not exist, then everyone’s counts could be wrong, and there was still a world in which Britain could plausibly claim that things were not so bad. Or later – as the extent of civilian suffering became impossible to deny – that Britain was not to blame.

This sort of thinking was supposed to have been overturned after the Iraq war. Yet two decades on, we still know next to nothing about how Britain makes its assessments of civilian casualties. Investigations by Airwars found that many more civilians, including children, were killed and disabled at sites hit by British airstrikes in the war against IS. The families were not contacted by the British armed forces or the British government. As Airwars says, this suggests they have not made ‘every reasonable effort’ to understand the impact of their military operations or learn from past mistakes. Despite the trust that was lost during the Iraq war, and the lengthy, expensive public inquiry justified as necessary to rebuild public trust, the British military still expects to be given the benefit of the doubt when it comes to civilian casualties.

Near the end of the hearing in November 2023, the MoD’s witness was asked if he had confidence in the procedures for assessing harm to civilians from British airstrikes. Oliver said he did. When the appellant’s lawyer asked him why, he said only that he had been involved in the MoD’s military operations overseas for more than a decade, and there were many experienced people who worked in the department. When asked why the public should share his confidence, Oliver repeated that there are many experienced people who work in the department. Asking, in other words, for the British public – and indeed the families of Iraqis and Syrians killed by British airstrikes – to give him and his colleagues the benefit of the doubt.