Global reports have routinely underestimated the true number of people killed in armed conflict, including during the Iraq conflict, according to a report released Monday.
The annual Small Arms Survey said it's possible as many as 39,000 Iraqis have died as a result of conflict in the Iraq from May 2003 to October last year, more than twice media estimates of between 10,000 and 15,000.
It derived its figures from a study published in the Lancet medical journal last October, which estimated 98,000 more civilians had died in Iraq since March 2003 than would otherwise have been expected to have perished. The British government has rejected those findings, which were based in part on projections.
Monday's report, from the Geneva-based Graduate Institute of International Studies, said that between 80,000 and 108,000 people were killed as a direct result of conflict around the world in 2003 -- two to four times higher than current estimates.
It said death tolls are usually based on misleading official estimates or media reports gathered despite intense efforts to keep reporters away from the fighting.
"Examples of politically motivated misdirection about conflict casualties are numerous," the report said. "Traditionally, this was called propaganda; the modern word is spin."
Other recent examples of such distortion include Bosnia, Kosovo and Afghanistan, the report said. Advocacy groups and non-governmental organizations are also to blame, sometimes over-reporting or playing down casualties to further their own ends.
The report said that between 60 and 90 percent of all deaths during conflict are caused by small arms and light weapons -- everything from pistols to rocket-propelled grenades to rifles.
That number highlights just how grave a concern small arms are to conflict. And when small arms are not used directly, they can still play a crucial role -- as during the Rwanda genocide, when Hutu extremists rounded up Tutsis at gunpoint and then often massacred them with machetes.



