I find that really tough. First of all, we don't collect any statistics on violence for the simple reason that it's really hard to define what violence is. If you don't know what violence is, it's not easy to get data on it. That's one part of it.
If I can accept personal failure on my account, I don't know the answer to a question that I keep asking myself. As a policy analyst, if I have developed a good policy—and you can define that good policy whatever way you want—and somebody tells me that a group of people doesn't like that policy, will engage in violence, and I have to change that policy, I ask myself how would I react to that. I don't know what the answer to the question is, but it really bothers me.