Hundreds of civilian deaths, including children, linked to British military activity in Afghanistan
British forces paid compensation for the deaths of at least 289 civilians in Afghanistan, including as many as 86 children, during their combat mission, according to new analysis by Action on Armed Violence (AOAV).
Fighting by British service personnel in Afghanistan led to financial settlements for at least 289 civilian deaths across 189 incidents between 2006 to 2013, internal MOD documents show.
Ministry of Defence (MOD) payouts for Afghans killed, included as many as seven dozen children, based on reports that indicate the youth of the victim. At least 43 were female. These figures are likely just a snapshot of the reality as they are only based on UK military compensation agreements, which was not a simple process for Afghan civilians to engage with.
Overall, £688,000 was paid out by the British military for 289 deaths, meaning the average compensation for a civilian killed was £2,380. Although some of these payouts were combined with injuries and property damage, meaning this average is somewhat inflated.
Most of the deaths occurred in the Helmand province of Afghanistan, and were recorded in compensation payout data obtained under Freedom of Information requests. British military compensation payments that stated the Afghan or US military were responsible for the deaths were not included in the above figures. Little to no detail of incidents is given up to March 2009, meaning we don’t know the circumstances, age or gender of the nearly 100 civilians killed during this period.
Nonetheless, the compensation data reveals some shocking incidents.
In one December 2009 incident, the document listed four children “shot and killed by ISAF [International Security Assistance Force]”. A month later, the Ministry of Defence paid out the equivalent of £4,223.60 ($6,800) in compensation for their deaths. There is no known reportage of this incident in the English-language media.
That same month, the youngest recorded casualty was killed. He was just three years old. “During IED clearance child was killed by shock from controlled explosion (son 3 years),” the entry reads.
The amounts paid in compensation were highly inconsistent. One family received just £586.42 for the death of their ten year old son in December 2009.
In February 2008, a family was given just £104.17 for a confirmed fatality and property damage in Helmand province. No more detail was given in the military logs.
This means that, on occasion, electronics and animals were valued more than human life.
In 2009/10 FY, there were 106 instances involving property (including crops, vehicles and buildings) where the payout was greater than the death of the above unnamed ten year old boy (£586) or the unnamed 2008 casualty (£104). These included:
- £873 paid for a damaged crane;
- £662 given for the death of six donkeys “when they wandered on to the rifle range”;
- £240 was paid out for ‘Warthog damage’ in Nahr-e-Saraj, Helmand;
- £110 was given in compensation for a lost mobile phone in Camp Bastion.
There were a number of instances where lower payouts were given for a fatality, as the British military refused to accept responsibility. These were often referred to in the documents as ‘goodwill payments’. Such payouts were not included in the 289 number, but they do suggest when payouts did occur, implicit British military responsibility for the deaths was accepted.
What is comprehensively recorded is that the war in Afghanistan cost 457 British soldiers’ lives and at least £22.7 billion (2001/2-2019/20). The causes, locations and dates of these deaths were analysed in AOAV’s 2020 Report For All Was Lost.
But when it comes to Afghan civilians, estimates of overall harm or fatalities related to British activity are not recorded by the UK government.
Overall civilian harm
According to AOAV analysis of UN reports from 2007-20, at least 20,390 civilians were killed or injured by international and Afghan forces, roughly one-third of the casualties caused by the Taliban and other anti-government forces.
Overall in Afghanistan, it is estimated by the Costs of War project at Brown University, that around 47,245 civilians in Afghanistan have died violent deaths as a direct result of the conflict, between 2001-2019.
Using United Nations and The Nation data, the UK based charity Airwars compiled a minimum of 4,815 civilian deaths that were the direct result of US airstrikes.
The US also had a system of compensation. Between October 2005 and September 2014 they issued 1,630 ‘condolence payments’, which encompassed death, injury and property damage, according to a report by CIVIC. Over the bulk of America’s longest war, the US military paid a minimum of $38 million in compensation to Afghan nationals from 2003-2019.
However, few reports have delved into potential attribution to specifically British military activity.
But due to a system in which Afghans could claim compensation from the British military for harm done, ranging from a death or injury to property damage, hundreds of fatal incidents have been recorded, with varying levels of detail.
From the 6,800 compensation payments seen, between April 2006 and May 2014, there were 417 payments made for 289 civilians killed and 240 injured that the MOD accepted financial responsibility for. However, the majority (69%) of claims were denied rather than accepted, with only one-quarter of fatality claims receiving a settlement. Overall, there were 881 claims of death and 285 of injury formally denied by MOD investigators.
Of the casualties that the MOD decided to pay out for, 71 were recorded as female (43 killed and 28 injured) and at least 28 were cited as children (16 killed and 12 injured), although it could be as many as 138 minors (86 killed and 52 injured).
Harm to children
When it comes to measuring the number of children who were harmed, two figures can be produced. The MOD files refer to those harmed in relation to the individual making a claim: son, daughter, brother, sister, wife, husband etc. The lower figure, 16 child deaths and 12 child injuries, comes from cases that incontrovertibly involved a minor, either by specifying an age or describing them as ‘child’, ‘boy’ or ‘girl’.
The higher figure, 86 child deaths and 52 child injuries, included cases that mention the terms ‘son’, ‘daughter’ or ‘nephew’. Afghanistan has a young population, the median age is 18.4, so the likelihood of someone’s child (eg. son, daughter, nephew) being a minor is high.
While, of course, there are likely to have been some adult casualties who were described as ‘sons’ and ‘daughters’, the documents also contain 86 casualties over three years that have no description at all, meaning there are almost certainly more child casualties than the number this investigation was able to find.
As the violence in Helmand spiked in 2009, both civilian and British military casualties correspondingly peaked, with 100 civilians deaths that resulted in British compensation and 109 soldiers killed that year. The three commanders of Task Force Helmand during this year, General Sir Gordon Messenger (retd)., General Sir Tim Radford and Major-General James Cowan (retd)., were all approached for comment.
Fighting continued to be intense in 2010. A United Nations report cited the Taliban’s use of human shields and ISAF’s decision to “choose as a main battle ground the densely populated rural environment of Marja [Nad Ali, Helmand]” as main contributors to civilian harm in that year.
The report added: “The purpose of the Taliban’s firing from occupied houses was to use the residents as human shields in an effort to deter international forces from responding with artillery or aerial attacks. The use of human shields is prohibited by international humanitarian law.” Such a statement offers some insight into why there were so many civilian death compensation payments during that period.
The compensation process
To produce these figures, AOAV analysed a series of documents from the MOD’s Area Claims Office (ACO), from 2006-14. ACOs are established where there is a sizeable defence presence, from major operations in Helmand and Basrah to bases in Cyprus and the Falklands.
The Afghanistan unit, made up of two MOD civil servants, reduced to just one in 2011, alongside a local translator, was permanently based in the Helmand provincial capital, Lashkar Gah, but would rotate into Musa Qala and Sangin. They would work alongside the British civilian units that were supporting the local District Governor’s office: the Civil Military Co-Operation (CIMIC) group from the MOD and the Provincial Reconstruction Team (PRT) of the Foreign Office.
Serious claims, of death and injury, were passed to the UK military’s Directorate of Safety & Claims Public Liability Group (DS&C PLG) for “investigation and adjudication”.
Details on their deliberation process remain scarce. Overall, around one in three claims of fatality or wounding were deemed worthy of compensation.
The average sum paid for a death was £2,380. For an injury, it was £1,654. These averages will have been boosted as some payment totals also included compensation for property damage.
Female victims in focus
- July 2009: “Wounding; Daughter injured by propaganda air drop.” Compensation: £609.76
- October 2009: “Killed Daughter and 1 Bull.” Compensation: £602.41
- October 2009: “”Wounding; Sister shot and now paraplegic.” Compensation: £1,506.02
- November 2009: “”Daughter aged 9 shot in head. Daughter when running from one compound to another shot in head during fighting between ISAF/Ins [Insurgent]”. She survived. Compensation: £527.11
- January 2010: “Daughter hit by shrapnel from air-strike and died later of injuries.” Compensation: £666.67.
- May 2010: “Rocket attack caused fire which killed daughter.” Compensation: £607.64.
Regional disparities in MOD compensation payments
When examining compensation payments per territory, the British MOD paid out 36% more in the two decades of the War on Terror to claims originating in Cyprus than in Afghanistan – £8.44m compared to £6.18m. Over just a three year period, the MOD paid out more in Cyprus (£1.04m), where the vast majority of claims were for crop or livestock damage, than for 289 civilian deaths in Afghanistan (£688,000).
Payments in North West Europe since 2001 were more than three times greater than in Afghanistan. In 2009/10, approximately 90% of the £1.17m paid out was related to vehicle movements in and around Germany.
Such a difference between total compensation paid to the families of Afghans killed and claims in Europe that are largely based on property damage, raises questions as to the fairness of the British government’s compensation payment system based on national currency valuations, local custom, and – potentially – even nationality.
Ex-gratia: Compensation without accountability?
In response to a Freedom of Information request in 2014, the MOD stated that: “When compensation claims are received they are considered on the basis of whether or not the MOD has a legal liability to pay compensation. When there is a proven legal liability, compensation is paid.
“The amount of compensation paid is determined by common law principles which, broadly, take account, as appropriate, of an individual’s pain and suffering, degree of injury, property losses, past and future financial losses and level of care required.
“In Afghanistan, the UK and other ISAF members, would in certain circumstances make ex-gratia payments [meaning no liability is admitted] to individuals affected by operations involving our Armed Forces. Providing restitution for civilians who have suffered in these circumstances is the right thing to do.”
They added that within their statistics they were unable to distinguish between ex-gratia payments and those in which legal liability has been proven.
Another MOD document from 2008 explains how the compensation system would “contribute to the effort in winning the consent of the local population.” A compromise, however, was said to be needed between “expectations of the local nationals, who use local custom and religious law to settle damages, and the definition of legal liability practised by the Ministry of Defence in UK civil litigation.”
Despite this formalised system of assessment, the value of compensation payments fluctuated hugely and often with hard-to-fathom logic.
What Price A Life?
In December 2009, when a ten year old boy was killed his family received just £586.42. (The sum was first determined in US dollars).
Conversely, an incident that received media attention at the time, in which five children were injured by stray bullets from a British Apache helicopter, garnered a more substantial payout of £7,437.01, or £1,487 per wounded child.
Counterintuitively, when eight family members were killed in May 2009 by a “bombing” they received less: £7,204.97, just £900 per death.
A year before, a single wounding claim from November 2008 received £6,849.32.
And a single fatality in Kabul in November 2007 received the greatest payout of £54,347.
The level of detail given about each incident is also highly inconsistent. Some are just listed as “Fatality” with a date and compensation value.
In the detail
Other claims are far more comprehensive, sometimes giving details of the investigation and the incident. One wounding claim from July 2011, in which payment was denied, states: “Claims mother burnt when a para illum landed in their compound. There was [sic] extensive attempts to allow the claimants mother to be seen in order that her injuries could be verified. After a schura [consultation], agreement was reached where a patrol would visit with a female medic who would look at the injuries and report back.
“However, when the patrol arrived the mother refused to be seen by the female medic despite the previous agreement. An illum was fired near to the compound where the women [sic] lives on the date in question but was not reported to have been fired incorrectly. Information was received that the wounds to the mother appeared to be shrapnel type injuries and not burns. Due to the fact that permission will not be given for the mother to be examined and the lack of other supporting evidence, this claim is denied.”
Another, from February 2013, reads: “18 Days after the alleged incident the LN [Local National] finally arrives at ACO LKG [Lashkar Gah] and proceeds to tell us his family’s compound was destroyed by fire as a result of helicopter flares setting it alight. He then mentioned that the fire had killed his 2 infant brothers. ACO alerted SO2 [Name redacted] about an alleged CIVCAS [Civilian Casualty].
“LN woefully short on detail and sent back to [redacted] for further investigation. CIVCAS [civilian casualty] downgraded as no evidence of deaths has been reported, especially after 18 days. This claim will be treated with caution. Further review by MSST [Military Stabilisation Support Teams] in [redacted] resulted in LN eing [sic] sent away with no further ation [sic] required.. Claim denied and closed.”
The civilian impact of armed violence examined
Due to this sporadic level of detail given, it is only possible to ascertain the cause of harm in 208 (39%) of settled cases listed.
Gunfire was the most frequent cause of civilian harm (51%) followed by airstrikes (25%) and explosive ground fire, such as grenades and mortars (22%).
What is clear is that investigative officers would refuse to pay or downgrade compensation significantly if they disputed that British personnel were at fault.
A boy, whose age was redacted, was killed in a motorcycle accident involving a military vehicle on a roundabout. The boy’s mother was also injured. The ACO records state: “It was not ISAF fault therefore the assistance payment was abated.” They had asked for the equivalent of £902.
Similarly ex-gratia payments would be cut in certain instances. An entry for an incident in October 2011 reads: “[Redacted] Year old boy killed by Escalation Of Force (EoF) whist [sic] driving a motobike [sic] at a patrol base. This is a 75% abatement has the boy contibuted [sic] to his own demise.” This decision meant his family’s compensation was reduced by £1,840.62 to just £613.54.
One payment appears not to have been specifically claimed by an individual. This was the incident from July 2012 where three Afghan farmers were “allegedly killed in cold blood by a team of SAS soldiers”, as reported by The Sunday Times. (No paywall: Independent).
In the ACO files, the purpose of the money is described: “3xLNs SHOT BY ISAF TF (SUSPECTED INS – SEE STORYBOARD) ASSISTANCE PAYMENT TO BE MADE TO CALM LOCAL ATMOPHERICS [sic]”
The MOD denied it was compensation and thus didn’t show liability. However, Major Chris Green, who was a British Army captain in the area at the time, told the Sunday Times in 2017: “To say it’s not compensation to the family is nonsense, really. This is what everybody called it on base, including the officers directly involved in paying the money. I am unaware of any money paid to the families of insurgents.”
‘Just a fraction of reality’
It is likely that the deaths recorded in the ACO files are an underestimation of the deaths caused by UK forces. These 289 deaths are only those that were brought as complaints by the affected people. Not all those who suffered due to British actions will have been aware of such options. And many who were would’ve been threatened by the Taliban who looked dimly on anyone who entered foreign bases.
In his 2013 book ‘Investment in Blood: The True Cost of Britain’s Afghan War’, former military intelligence officer in Iraq and ‘Justice Advisor’ in Afghanistan, Frank Ledwidge estimated, based on journalist and witness accounts, that, in 2007, there were at least 220 civilians killed or injured by British forces. The ACO documents list just 36 for that year.
Comparing the compensation given to members of the British military under the Armed Forces Compensation Scheme, Ledwidge calculated: “It would be fair to say that the lives and limbs of Afghans are assessed in financial terms at between one-twentieth and one-hundredth of those of a British soldier.”
Ledwidge told AOAV: “There were perverse incentives in place to keep the civilian casualty number down, and these were in place until the end of the war. That’s why these casualty figures are just a fraction of the reality.”
Prosecutions of British Military Personnel
Overall, from 529 deaths and injuries paid out for by the MOD, 17 British military personnel have been charged in relation to civilian casualties and 15 prosecuted, according to documents obtained by AOAV via FOI.
This would either take the form of court martial or in private summary hearings. In 2011, Grenadier Guardsman Daniel Crook was dismissed from the army and sentenced to 18 months in prison after stabbing a 10-year old boy in the kidney with a bayonet. The morning after drinking a large quantity of vodka, the soldier was on patrol and encountered Ghulam Nabi.
Any compensation paid to this boy’s family was not listed in the compensation documents available to the public – possibly owing to this incident being escalated beyond the Helmand-based claims office.
His father, Shah Zada, told The Guardian in 2012 that Crook ordered his son to stop but, due to his age, the boy didn’t understand.
Crook “took hold of the boy’s shoulder and stabbed him in the region of his kidneys with his bayonet”, prosecutors from the Service Prosecuting Authority told the court martial.
AOAV excluded a number of deaths and injuries from the total figures due to various reasons and did not include incidents that were denied by the MOD in our headline figures. Our methodology is detailed below.
A MOD spokesperson told AOAV: “Every civilian death is a tragedy and the UK always seeks to minimise the risk of civilian casualties through our rigorous targeting processes, but that risk can never be removed entirely.
“The amount of compensation paid is determined by legal principles which consider the degree of injury and both past and future losses, settlements also reflect local customs and practice.
AOAV started with multiple files containing thousands of claims, the vast majority concerning small payouts for property damage. We trimmed the dataset down into just fatalities and woundings and divided them into ‘Settled’ [received compensation] and ‘Denied’ [received no compensation].
Names and precise dates were included in the original documents but we have discounted these in our reporting in order to protect the identity of those who received any compensation from the British military.
Within the fatality and wounding claims that were listed as ‘settled’, a handful of entries contained details in which the MOD investigators stated there was no British or ISAF blame. Three fatality listings (1% of total) detail that investigators have found no British or International Security Assistance Force (ISAF) fault but have offered a small sum as a “goodwill payment”. These were excluded from our total civilian death figure of 289.
Occasionally, it appears that a complaint would be lodged by an Afghan civilian for an incident that was linked to a non-British foreign force. The ACO documents feature a fatal road traffic accident between a military vehicle and a motorcycle in 2011. It’s listed as “actually a US incident for which they have refused to take responsibilty [sic]. Flagged up to [redacted]. $ is offered as a good will payment.”
This was the only fatality or wounding in the ACO settled files that listed the US as being culpable. However a 2008 MOD report on liability claims explains that just under 9% of claims were “transferred to other Troop Contributing Nations within Regional Command (South)”. It seems likely that aside from the above incident, cases that were the liability of other nations were not included in the ACO files seen by AOAV.
Overall, these outlier cases, as well as the cases where payments were abated or claims were outright denied, suggest that when payouts did occur, implicit British military responsibility for the deaths was accepted.
When it came to measuring children, we arrived at two figures.The language of the ACO file refers to victims in their relation to the person who has lodged the complaint. Often, beside the terms ‘son’, ‘daughter’ and ‘nephew’ is followed by ‘age ()’ with the number redacted.
The first figure of 28 child casualties (16 deaths and 12 injuries), involved cases that explicitly mention a child, either by specifying an age or describing them as ‘child’, ‘boy’ or ‘girl’.
The second figure of 138 child casualties (86 deaths and 52 injuries), included cases that mention the terms ‘son’, ‘daughter’ or ‘nephew’. Afghanistan has a young population, the median age is 18.4, so the likelihood of someone’s child (eg. son, daughter, nephew) being under 18 is high.
Research support provided by: Riya Kamboj, Ana-Marija Apostoloska and Ryan Cunnington
Did you find this story interesting? Please support AOAV's work and donate.