Monday, October 20, 2008
Our Dangerous Narratives
Written by Ross Hobbes at 8:30 AMUpdate -- Matt responds: In Defense of Narratives
I have a beef with almost all political and economic commentary. It employs the Narrative Fallacy. I use the term Narrative Fallacy in the sense its presented in the writings of Nassim Nicholas Taleb, author of Fooled By Randomness and The Black Swan. It's creating a story post-hoc so that an event will seem to have an identifiable cause.
I've already made clear on this blog that I have a big problem with post-hoc thinking - and the pundit class in Washington and in the 'blogosphere' are particularly lazy in their analysis whilst creating 'narratives' to fill column inches or kill time on cable news channels. Every pundit can weave a story as to why Obama is up in the polls this week. The facts support the narrative! Of course, should Obama fall in the polls then those same facts would reconstitute themselves in a way to say the opposite - a new narrative! Here's what Taleb writes about the narrative fallacy in The Black Swan:
The narrative fallacy addresses our limited ability to look at sequences of facts without weaving an explanation into the3m, or equivalently, forcing a logical link, an arrow of relationship, upon them. Explanations bind facts together. They make them all the more easily remembered; they help them make more sense. Where this propensity can go wrong is when it increases our impression of understanding.
'Narratives' help us make sense of the world. We want a cause-and-effect - we need it, or the world doesn't 'make sense'. But I would argue, as Taleb does, that causal relationships in much of our world, particularly things like politics, economics, and social structures, are unknowable. Yet everyday we are bombarded with 'narratives' - and those narratives necessarily crowd out discussion about uncertainty and the unknowable (how boring!). The impression of understanding that is created by narratives is too dangerous to ignore. Pundits pretend to understand relationships between facts and outcomes (or more likely they fool themselves into believing so), and that lends its self to prescription. Do what happened last time and you'll be assured similar results! Narratives also have serious consequences for basic understanding. Our knowledge of political systems, economic relationships, and social constructs, become simplified.
In economics our simplifications lead businesses to rely on risk models that assume the future will look like the past - and then CEO's are surprised when those assumptions fail spectacularly! In politics we do things like remember the the 2004 Presidential campaign differently because a few thousand people in Ohio didn't go Kerry's way (just think how history would view the facts of that campaign differently had it swung the other way - and it could have - it was within any reasonable margin of error).
Narratives are an enemy of understanding. Yet we're taught to believe the opposite. Narratives do have their place - sometimes true understanding isn't required...sometimes we need simplifications. The next time you see a narrative, ask yourself, is this one of those times?
Read Matt's response here.
