Moving on from Ramadan

As October winds down expect to see numerous stories in the press regarding the death toll in Iraq.

Above chart from the Brookings Institute's Iraq Index represents U.S. fatalities in Iraq since the initial invasion. October 2006 will indeed be one of the higher months.

The nearly random nature of this chart appears senseless (and on a fundamental level it is - no argument from me there) but there is some signal in the noise. The two most notable spikes in violence represent the two battles for Fallujah. Other spikes occur with the various Iraqi elections, and some coincide with other major offensive actions on the part of the coalition.

But the chart also has a weakness. It's using western months. The Muslim world uses a lunar calendar. Applying that system to the lower axis would give some additional insight to the violence of "the past month", because the Ramadan spikes would stand out a bit better than they do when presented in a Western calendar format. Note November '03 for the first example - in 2004 Ramadan was split between October and November on our western calendar, in 2005 and 2006 it fell mostly in October.

But this year's spike also coincides with focused operations in Baghdad, and may or may not also reflect enemy awareness of pending U.S. elections. Whether they can sustain that level of violence over the next week will be one indicator if this is so - because Ramadan has drawn to a close.