The unknown Afghan body count

June was a terrible month for the war in Afghanistan. The milestone of the 300th British death was compounded by the most deadly month for the Nato-led mission since the start of the conflict.

The precise compilation of western casualties contrasts with almost criminal neglect in tracking the numbers of Afghan civilians killed since 2001. If Afghanistan is the "good war" then why are we not demanding to be accurately told how many skeletons there are in the Afghan closet?

In 2005 Donald Rumsfeld famously quipped that "death has a tendency to encourage a depressing view of war". The US defence department maintains documentation on US military personnel only, while the British ministry of defence "does not maintain records that would enable a definitive number of civilian fatalities to be recorded" – although it did confirm last month that payments to relatives of Afghan civilians killed in error by British forces have trebled over the past year. The Afghan government, characterised by massive levels of ineptitude and corruption, has failed to keep centralised records of civilian casualties which would enable it to issue annual estimates.

True to form, the International Security Assistance Force has also avoided releasing body counts – leaving it to an inconsistent patchwork group of NGOs and academics to correlate the numbers of dead Afghans. Although boosted by an occasional United Nations report, homemade body counts are largely unreliable, as they struggle to agree on a consistent methodology and are unable to keep up to date with the constant grind of killing. Perhaps the best statistics we have on the plight of the Afghans is the UN report on how for the past three decades Afghanistan has been the leading country of origin for refugees – with 2.9 million Afghans living across 71 countries.

How can any western official claim to have the best interests of the Afghans at heart when they don't even know how many they've killed? To understand the western presence in Afghanistan it is of critical importance to effectively and publicly track the lives lost as a result of both military and "insurgent" action.

In 2009 the head of the army, General Sir Richard Dannatt, told the BBC Radio 4 Today programme: "A high number of deaths inevitably makes you question what we are doing, how we are doing it. The conclusion one has to reach is, going right back to basics on this, that this mission is really important." Yet Dannatt is guilty of a moral triangulation that has typified the avoidance of a real audit into Afghan deaths.

Indeed, the constant repetition of the British death toll and fiscal expenditure is part of the "blood and treasure" argument that, in a country that supports its soldiers, places a firewall in front of any real debate on the war itself, typified by the consensus during the recent election campaign.

After sacking General Stanley McChrystal, President Obama announced that the personnel had changed, not the policy. Yet in the furore over McChrystal's attacks on the US civilian leadership, people missed how the Rolling Stone article highlighted that the former general was engaged in a battle with the military to reduce civilian casualties. McChrystal spoke of how "we've shot an amazing number of people", a reality that is most viscerally described in Sebastian Junger's account of the war.

Turning the US military into a more nuanced killing machine had been a struggle for McChrystal as soldiers saw their lives being placed at greater risk. While he was in charge, McChrystal attempted to avoid civilian casualties (known as Civ Cas) by reducing air and artillery strikes, the destroying of houses and dangerous US military driving styles. The recent Marjah offensive and upcoming Kandahar operations were highly publicised to allow civilians to leave the area. McChrystal explained to his men that "the Russians killed one million Afghans and that didn't work".

Our way out of Afghanistan, the McChrystal/Petraeus counterinsurgency strategy, emphasises protecting local populations, providing them with services such as schools and health clinics. Yet without proper tracking of the numbers of Afghans killed then the best Petraeus may be able to achieve, as he did in Iraq, is a narrative of success that ignores the far more complex and bloody reality.

James Denselow, a writer on Middle East geopolitical and security issues based at Kings College London.