OF THE
TIMES
Estimating the Death Toll in Somalia
Passive sources put the violent death toll in Somalia since the U.S.-backed Ethiopian invasion in 2006 at 20,171 (Uppsala Conflict Data Program (UCDP) - through 2016) and 24,631 (Armed Conflict Location and Event Data Project (ACLED)). But an award-winning local NGO, the Elman Peace and Human Rights Centre in Mogadishu, which tracked deaths only for 2007 and 2008, counted 16,210 violent deaths in those two years alone, 4.7 times the number counted by UCDP and 5.8 times ACLED's tally for those two years.
In Libya, Libya Body Count only counted 1.45 times as many deaths as ACLED. In Somalia, Elman Peace counted 5.8 times more than ACLED - the difference between the two was 4 times as great. This suggests that Elman Peace's counting was about twice as thorough as Libya Body Count's, while ACLED seems to be about half as effective at counting war deaths in Somalia as in Libya.
UCDP logged higher numbers of deaths than ACLED from 2006 until 2012, while ACLED has published higher numbers than UCDP since 2013. The average of their two counts gives a total of 23,916 violent deaths from July 2006 to 2017. If Elman Peace had kept counting war deaths and had continued to find 5.25 (the average of 4.7 and 5.8) times the numbers found by these international monitoring groups, it would by now have counted about 125,000 violent deaths since the U.S.-backed Ethiopian invasion in July 2006.
But while Elman Peace counted many more deaths than UCDP or ACLED, this was still just a "passive" count of war deaths in Somalia. To estimate the total number of war deaths that have resulted from the U.S. decision to destroy Somalia's fledgling ICU government, we must multiply these figures by a ratio that falls somewhere between those found in other conflicts, between 5:1 and 20:1.
Applying a 5:1 ratio to my projection of what the Elman Project might have counted by now yields a total of 625,000 deaths. Applying a 20:1 ratio to the much lower counts by UCDP and ACLED would give a lower figure of 480,000.
It is very unlikely that the Elman Project was counting more than 20% of actual deaths all over Somalia. On the other hand, UCDP and ACLED were only counting reports of deaths in Somalia from their bases in Sweden and the U.K., based on published reports, so they may well have counted less than 5% of actual deaths.
If the Elman Project was only capturing 15% of total deaths instead of 20%, that would suggest that 830,000 people have been killed since 2006. If UCDP's and ACLED's counts have captured more than 5% of total deaths, the real total could be lower than 480,000. But that would imply that the Elman Project was identifying an even higher proportion of actual deaths, which would be unprecedented for such a project.
So I estimate that the true number of people killed in Somalia since 2006 must be somewhere between 500,000 and 850,000, with most likely about 650,000 violent deaths.
Nothing against the civilians, they have always been nice enough to me.