A Layman’s Evaluation of Risk: Likelihood vs. Magnitude

I woke up this morning to a Twitter feed that got the juices flowing regarding ubiquity. Nassim Nicholas Taleb quoted a tweet containing a link to a Washington Post editorial titled “You’re more likely to be fatally crushed by furniture than killed by a terrorist“. The text of Taleb’s tweet suggested he wasn’t too impressed with the article’s content.

More from Taleb…

A bit later, Glenn Greenwald posted a tweet referencing the same article, with what I interpret as a more favorable description.

I suggested to Greenwald he read Taleb’s tweets about the article.

I don’t know whether or not it was my suggestion that prompted what happened next, (though I’m totally taking credit for it 🙂 ) but Greenwald responded to Taleb’s original tweet on the subject.

Taleb, his contempt for “journos” being rather famous by now, quoted Greenwald’s tweet back with a bit of sting.

Granting that I admire both Taleb and Greenwald, I hope they don’t end up in some sort of net.feud resembling the one between Greenwald and Sam Harris. But that’s not the real point of this post, and now that the preamble is over what will follow is a crude layman’s description of what I think Taleb’s point is about the article’s deficiency (with a bit of additional opinion from the same crude layman).

I don’t find the article especially disagreeable, but the author seems myopic insofar as he gives consideration primarily to the likelihood of a harmful event and not its magnitude. The title’s reference to furniture is indicative of this narrowness of thought. While I’m perfectly willing to grant that I am much more likely to be injured by an unfortunate encounter with a footstool or chair than I am by a terrorist attack, a basic feature of the furniture-related mishap is that it can only happen to me. But if one of the survivors of the Oklahoma City bombing were injured by a flying desk, we don’t describe their injury as resulting from furniture. We recognize that a) they survived a terrorist attack, and b) many other people were far less fortunate than the survivor struck by a desk. Yes, being injured or killed by a truck bomb is far less likely, but the magnitude of the event is much greater. (The Oklahoma City bombing killed 168 people and injured several times more.) Taleb expresses the concern over magnitude when he mentions, “naive comparisons across statistical classes, fat tails vs thin tails” and the more blunt, “Doesn’t take math to get that terrorists can kill millions while furniture can’t.”

Taleb advocates vigilance. While there’s certainly a great deal of poorly-founded fear and security theater surrounding both recent events such as the attacks in Paris and past attacks like 9/11, there’s little sense in failing to acknowledge that vigilance can further reduce the likelihood of terrorist attacks. Though the author of the editorial does not suggest forgoing vigilance, it doesn’t seem too great a leap to that view given the article’s narrow content.

Perhaps more important is the moral component to why concern over terrorism differs from concern over injuries from furniture. If I’m worried about suffering an injury from slipping in the shower, I’m worried about it happening to me or to people I’m close to. Maybe I put some of those non-slip appliqués on the floor of the shower. If I’m concerned about terrorism, it’s true I will almost certainly go my whole life without being caught in an attack myself. But my concern isn’t for myself, or even those near me, so much as it is for the as-yet-unspecified number of complete strangers, whether nearby or in some part of the world remote from me, who will suffer if a terror event happens to them. And a lot more of them will suffer, as did recently in Beirut and Paris, than can suffer from any individual instance of shower slippage.

Having said that, a particular point in the article with which I agree is that of the misspent trillions fighting the War on Terror. So much money is spent on immoral and ill-conceived war that could go to benefit things like cancer research, covering the cost of care and treatment for a variety of ills. A lot sickness and death may be preventable but for the lack of funds.

I hope that Taleb and Greenwald get along. I hope that my attempt at explaining likelihood and magnitude as features of risk wasn’t entirely incompetent. Mostly, I hope we can stand down from the War on Terror and recognize that vigilance need not be bought through fear, killing and destruction.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s