Thursdays Declared "No Math" Day At Washington Post

This is from a Washington Post story published today:

Between 2 percent and 5 percent of Iraq's 27 million people have been killed, wounded or uprooted since the Americans invaded in 2003, calculates Anthony H. Cordesman of the Center for International and Strategic Studies.

Then, five sentences later:

Since the war began, 1.6 million Iraqis have sought refuge in neighboring countries; at least 231,530 people have been displaced inside Iraq since February, when Shiite-Sunni violence exploded with the bombing of a Shiite shrine in the northern city of Samarra, according to figures from the United Nations and the U.N.-affiliated International Organization for Migration.

2-5% of 27 million is 540,000-1,350,000. 1.6 million plus 231,530 is 1,831,530.

Thus, Anthony Cordesman calculates the number of Iraqis killed, wounded or uprooted is 540,000-1,350,000. Meanwhile, the U.N. says the number of refugees alone is at least 1.83 million.

There would be nothing wrong with writing a story that contrasts these numbers. However, the Washington Post presents them as if they don't contradict each other. I think it was the whole "percent" thing that confused them, since percentages are studied around sixth grade, and Post reporters usually drop out of school by age ten.

AND: The reason I noticed this is because Anthony Cordesman is one of the best think tank denizens in Washington, yet when it comes to this subject he seemingly just makes things up. When the most recent Johns Hopkins study came out, he opined, "They're almost certainly way too high...this is not analysis, this is politics." To understand his reasoning, I read a paper (pdf) he'd just written. According to Cordesman, one reason the Johns Hopkins study can't be correct is that it would mean "every reporter actually in Iraq is radically wrong."

I'm going to go out on a limb here and guess Cordesman didn't question "every reporter actually in Iraq." In fact, I suspect the total number of reporters he asked is three or under, and the number of Arabic-speaking non-Americans among them is zero.

I can only assume the same amount of analytical rigor went into the number he "calculates" here for the Post.