I have a new article in this month's print edition on the attempt to count the number of dead in Iraq. My article focused mostly on how we process the numbers, but over the course of my research I accumulated a fair store of knowledge on the subject. Since the fifth anniversary recriminations are still in full swing, I thought it might be useful if over the next week or so, I covered the state of the debate.

In some sense, I don't think knowing the number matters1. The lower bounds of reasonable estimates are still high enough to make me think our involvement in Iraq was a bad idea, especially when considered in conjunction with the various other problems we know about, like the attacks on key infrastructure and the refugee crisis. So debating whether the number is 100,000 or an order of magnitude higher than that doesn't change my basic assessment of the situation.

But in many other ways, accuracy is tremendously important. These numbers shape the national debate; it is therefore critical that they should be as correct as possible. Also, the results from these studies have important implications for a range of policies. Knowing how bad the violence is, and what kind of violence we are dealing with, should shape many of the priorities and goals that we and the Iraqi government set for ourselves.


1Go ahead, take this out of context.

We want to hear what you think. Submit a letter to the editor or write to letters@theatlantic.com.