DUCKPOND

Statistical results and methods are familiarly attacked by interest groups in regard to the connection between smoking and lung cancer, or carbon emissions and climate changes. What is often forgotten, in these cases, is that the statistics, given that problems have existed, simply confirm and quantify what reliable witnesses can often verify. I am subject to correction on this matter, nevertheless the practical test of observation can usually be applied. Do reliable witnesses confirm the relation? In other words can we relate the numbers with observations. In the case of smoking and lung cancer it is relatively easy, whereas it is more difficult to record deaths in Iraq given that bodies do not all end up in morgues.

The latest estimate in The Lancet, via John Quiggin, concludes:

We estimate that, as a consequence of the coalition invasion of March 18, 2003, about 655 000 Iraqis have died above the number that would be expected in a non-conflict situation, which is equivalent to about 2·5% of the population in the study area. About 601 000 of these excess deaths were due to violent causes. Our estimate of the post-invasion crude mortality rate represents a doubling of the baseline mortality rate, which, by the Sphere standards, constitutes a humanitarian emergency.

Juan Cole explains why these figures have not been confirmed by the Iraqi Department of Health:

The Lancet study asserting that the Iraq conflict has cost the lives of between 420,000 and 780,000 Iraqis continues to generate controversy. But Dan Murphy of the CSM quotes public health officials pointing out that its methodology was sound, contrary to what Presiden Bush asserted. Murphy’s article also puts its finger on the likely source of the discrepancy between the Lancet numbers and those of the Iraqi ministry of health: The ministry employees cannot travel easily to places like Baqubah and Kut and Ramadi to collect death statistics from local officials. I can remember talking recently to a Shiite from Baghdad who said that virtually no one routinely goes to Najaf from the capital any more because the roads are too unsafe. Najaf was only an hour’s drive from Baghdad in the old days.

With any measurement, I remember this, uncertainty is to be expected. So the actual numbers could in fact be greater or less, but if the methodology is valid and reliable, the numbers of dead are large and significant. John Quiggin observes:

In particular, the lower bound estimate is now around 400 000, so that unless the survey is rejected completely, there can be no doubt about catastrophic casualties.

These deaths have followed from the invasion and the occupation and represent people who otherwise would have been expected to live. So these numbers will be significant in regard to any possible future determination as to war crimes and war crime indictments.

PLEASE NOTE:

I have deliberately not linked to John Quiggin since doing so means my post comes up on a comment box, which I would prefer did not happen. You can still connect via my blogroll and search for the post Air war in Iraq. Alternatively you could go to my other blogger duckpond and link directly.

Postscript: 15 October 2006

Brad DeLong takes up the issue of press reporting of the survey results, including by some commentators suggesting that they were released for political purposes prior to the mid term elections. I am interested to note he quotes Daniel Davies who also sees implications for war crime indictments.

Postscript: 17 October 2006

This piece by Gywnne Dyer, originally published in The Japan Times, via Common Dreams, gives a clear summary of the methodology and the results. It makes nonsense of some of the predictable criticism that I have read.