New York Times' Gordon Contradicts His Paper's Own Reporting In Portraying Decline In Iraqi Violence
September 8, 2007 -- 3:20 PM EST // //
As Josh notes over at TPM this morning, today's New York Times has what he calls a surprisingly "credulous" report by Michael Gordon saying that data assembled by the military shows that the surge has produced a decline in various measures of bloodshed in Iraq.
Josh aptly recommends comparing today's piece with this one in The Washington Post earlier this week showing a good deal more skepticism towards the military's assertions.
I'd like to recommend another comparison: Let's compare today's Times piece with one that appeared in The Times itself less than a week ago reporting that civilian casualties are up across Iraq.
That earlier piece, by James Glanz, relied on figures supplied by Iraqi officials. Its key conclusion was that civilian deaths had gone up in August:
Newly released statistics for Iraqi civilian deaths in August reflect the strikingly mixed security picture that has emerged from a gradual six-month increase in American troop strength here: the number of deaths across the country rose by about 20 percent since July, but in the capital itself, the number dropped sharply...So, Glanz's key conclusion was that casualties were up one-fifth in August across Iraq. Right?...figures provided to The New York Times by an Interior Ministry official who asked to remain anonymous indicated that 2,318 civilians died violently in the country in August, compared with 1,980 in July.
Yet today's report by Gordon quotes military officials explicitly making the opposite argument: That in August, "the most significant gains in reducing violence materialized not only in Baghdad, but also across Iraq."
So, which is it?
Now, slicing and dicing these numbers gets endlessly complex, and the answers often depend on which sets of numbers are being compared. But in this case, it's pretty clear what happened. Glanz compared the numbers Iraqi gave him for August with the numbers from July, and concluded that from one month to the next, civilian deaths had gone up.
Gordon, by contrast, relying on American military statistics and Iraqi ones, compared August's numbers with those of December 2006. This showed a decline, he reports. Of course, December 2006 represented an all-time high in violence, he also notes in the piece.
Which methodology do you think is a better way to assess the success of the "surge"?
Whatever your conclusion on that question, one thing seems obvious. If you're going to publish a piece showcasing the military's claim that the "most significant gains" in violence reduction occurred in August, your effort becomes a heck of a lot less credible when your own paper reached precisely the opposite conclusion less than a week ago.