If truth is the first casualty of war, one of the ways in which this is true is in relation to the casualty statistics themselves. As part of the time-tested war propaganda, each side minimises its own casualties and exaggerates estimates of the damage inflicted on the enemy forces, strategic-industrial targets and public morale. The estimates of costs and timelines for victory are similarly downsized.

All of this has been true of the Iraq War. On the economic costs, people like Paul Krugman in his New York Times column and the Nobel Laureate Joseph Stiglitz have done much to highlight the magnitude of the true figures.

With respect to the numbers of Iraqi civilian casualties after the 2003 war and the ensuing insurgency, however, the Bush Administration has largely got away with little or no accountability. The American public has been left dazed and confused with a maze of claims, counter-claims and disinformation campaigns where often, if the statistics are damning, then the methodology is criticised and the motives of the scientists are questioned. Some of the tactics to discredit the studies' findings and their authors are lifted straight from the old (and enduringly relevant) Yes Minister and Yes Prime Minister television series.

If a diplomat is someone who will persuade you to die for his country, then Iraq will indeed be an apt metaphor for the George W.Bush brand of international diplomacy. The first point to note is the moral bankruptcy of an administration and a coalition that would wage a war of aggression retroactively in response to humanitarian atrocities committed by the Saddam Hussein regime, yet, as a matter of deliberate policy, refuse to collect statistics on how many civilians were being killed as a consequence and in the wake of the war. Where is the outrage in the US press and public at this gross immorality?

Others stepped in in an effort to fill the statistical breach. For the 18-month period after the war, a US medical team calculated the civilian casualty based on a scientific household survey and came up with the stunning figure of 98,000 deaths.

The results were published in the highly regarded medical journal The Lancet in October 2004. Coalition governments disputed the findings, but failed to provide numbers of civilian casualties themselves whose accuracy could be assessed against the article's.

The methodology they employed is called clustered sampling, which is the rule in public health studies, for example of epidemics. The alternative technique, called passive-surveillance systems, relies on waiting for reports of deaths to come in, rather than reporters going out randomly into the field to see if anyone has been killed in a violent attack. For this reason, it tends seriously to undercount mortality, in epidemics as in violence.

The Johns Hopkins University study team interviewed a total of 7868 people in 988 households about births and deaths that had occurred since January 1, 2002. Based on these interviews, the team calculated the number of deaths caused by the war by comparing the aggregate death rates before and after March 18, 2003, and attributing some 60 per cent of the excess deaths directly to the violence (from both sides), with the remaining being due to accidents, disease and infant mortality.

The rather large range of possible death numbers, from 8000 to 194,000, reflects the small sample size for a study of this type. Nevertheless, the figure of 98,000 is the most likely number in that huge range. This does not mean, therefore, that any number in that range is just as probable as any other number. The further away we move from 98,000, in either direction, the lower the probability of that number, so that the lowest estimate of 8000 is just as (un)likely as the highest estimate of 194,000.

Experts consulted by The Economist hardly a left-wing, anti-war propaganda tract confirmed that the study had been carried out to the standard professional level. Australian epidemiologists and public health experts I spoke to confirmed that the methodology used for the study is standard practice in the profession and was correctly followed by the Johns Hopkins team.

Because the study was published on the eve of the last US presidential election, it became an easy target for suspect political motives rather than the quest for scientific truth. While the public database Iraqi Body Count estimated the Iraqi civilian toll at around 25,000 deaths, the Graduate Institute of International Studies in Geneva re-examined the Lancet study to conclude that the more accurate estimate should be around 40,000 deaths.

A second Johns Hopkins team went back and replicated the study, using a 12,801-person sample drawn from 1849 households, and published their second round of findings in The Lancet as well (October 2006). The repeat exercise broadly confirmed the large numbers of excess deaths compared to pre-2003 levels. Their revised total of the ''excess'' number of people killed was 655,000 in the three-year period from March 2003 to March 2006. Of these, they estimated that 601,000 were due to violent causes.

Since then, the widely watched website Iraq Body Count has revised its figures to between 80,000 and 88,000 killed by 2008. But a survey conducted by the British-based, non-government-funded Opinion Research Business, published early this year (Reuters, January 30), concluded that about 1.03 million Iraqis have died as a result of the conflict caused by the 2003 US-led invasion. They gave their margin of error as 1.7 per cent, meaning that the likely range of the true figure was 946,000 to 1.2 million.

Readers, like governments, will no doubt tend towards the casualty figures that best suit their views and opinions on the war. The basic sad fact remains that very large numbers of Iraqi civilians have been killed through direct violence of sectarian and revenge killings and many through the structural violence of disruptions to critical health services, medical supplies and nutritional requirements.

The still sadder fact is that neither the Iraqi Government nor the coalition forces have deemed Iraqi lives lost worthy enough to be counted accurately. Dignity in death is clearly not a human right for Iraqis. No one will be called to account in national or international criminal justice forums.

The opinions expressed in this article/multimedia are those of the author(s) and do not necessarily reflect the views of CIGI or its Board of Directors.