Certainly Wrong

I was listening to CBC radio this afternoon and I heard an interview with Dan Gardner about his new book Future Babble: Why Expert Predictions Fail – and Why We Believe Them Anyway. In the book Gardner quotes a study where a range of experts in politics, economics, the social sciences and so on were asked to make what amounted to tens of thousands of predictions about the future. This was done in the 1980s and follow-ups were made at various intervals to see what actually happened. Turns out that the average expert is about as good as a coin toss or a monkey throwing darts.

Haha, silly experts, when will they learn?

Wait, there’s more to it then that. The average expert wasn’t very good, but Gardner broke it down further: some experts were actually better than average, and some were worse. The better experts were those that tended to be nuanced and doubtful about their own predictions. The experts were more likely to be wrong were those that expressed certainty about their views. This apparently matters more than political leanings, education, background, et cetera.

Are we culturally prepared to listen to nuance and self-doubt though? Not really, here’s James Fallows on the state of public discourse in the United States:

“Among the many things wrong with talking-head gab shows, which have proliferated/ metastasized in the past generation — they’re cheap to produce, they fill air time, they make journalists into celebrities, they suit the increasing political niche-ization of cable networks — is that they reward an affect of breezy confidence on all topics and penalize admissions of complexity, of ignorance on a specific topic, or of the need for time to think.”

As long as our discourse rewards glib certainty, we shouldn’t expect really high quality analysis of where we are going.