It would be wonderful if in fact our Iraq casualties were decreasing but we would have to make people un-dead and uninjured to make that a reality. Perhaps you mean that the rate of new casualties is going down? If that is your position, I’d have to ask what time period we are supposed to see this decreasing casualty rate occurring?

Here are the official casualty figures (Killed and Wounded) for the past twelve months

August 536
September 544
October 704
November 606
December 371
January: 586
February: 356
March: 508
April: 562
May: 493
June: 573
July: 512

As you can see, the number of new casualties in June was DOWN from July, which was UP from May, which was DOWN from April, which was UP from March, which was UP from February…..and so on. There is no clear trend up or down over the last 12 months, or for the last 20 for that matter. You would have to go all the way back to the time period between April 2005 and November 2005 where the 1, 361 casualties in April and 1,567 in November of 2005 were higher than any months before, since or in-between.

I misspoke, I meant casualty as in DEAD US soldiers.

This was the debate with Gary Hart and proved my own point. Most people usually think death when they speak of casualties.

You new that, which is why you went specifically to the dead and wounded to make your point.

I apologize; you are much more learned than me.

It probably only matters to you that you meant DEAD soldiers. It matters not at all to the argument you were making whether you used the correct definition or the wrong one.

The point that you were unsuccessfully attempting to make was that a decrease in whatever you happended to think casualties meant at the time, showed that the NYT was incorrect in stating that a Civil War started in Iraq in January. Although I disagree with your basic premise that American casualties are an indicator of whether a CIVIL WAR is occurring in Iraq, I would point out that as an inidcator of the level of violence being experinced by U.S. troops, CASUALTIES are the most relevant.

Even if you argue that fatalties are a better indicator of violence directed toward our troops, the up and down trend in fatalities of the period in question is hardly backs up your claim of a meaningful decrease:

January 65
February 56
March 33
April 81
May 71
June 61
July 51

Throw out the two outliers (81 and 33) and the the line on the graph just wiggles a little. Leave them in and you have wide variaitons that indicate there is no overall trend.