Writing in the Wall Street Journal recently, Paul Campos offered a version of John Mueller’s argument that statistically speaking, the risks posed by terrorism are very small. Kevin Drum says “this line of argument — that terrorism is statistically harmless compared to lots of other activities — will never work. For better or worse, it just won’t. So we should knock it off.”
I don’t know if they’re right or not. What I know Kevin is wrong about is that “never” is far too long a time frame on which to make this kind of claim. Some time over the summer I was on a panel with Tyler Cowen and he convinced me that a lot of people interested in politics spend much too much time thinking about in-the-now partisan political controversies. If you think back to the year 1602 and who was making real contributions 400 years ago, it’s not the people who played a key role in winning some early 17th century political controversies. Rather, the people who made a difference are the people who were promoting better, different ways of thinking about things. Promoting science and rational inquiry, tolerance, whatever.
It’s quite true that human beings do not have a great intuitive grasp of statistical arguments or a great love for them. But the world would be a better place if people thought of these things in a more statistically informed way. Likewise it’s true as Jon Chait says that people generally think differently about intentional murders than thinks like car crashes. But this, though it’s definitely a fact of life, is also a problem that it would be good to ameliorate over the long run. People tend to view threats stemming from identifiable, individual villains as more problematic than impersonal ones. But while this is a fact of life, it’s also a mistake. If we do something to very slightly reduce the risk of a terrorist attack that has the inadvertent consequence of causing a large number of additional highway deaths then that would be a mistake.