Complex Systems & Certainty
I had dinner last week with the first guest on Narrative Monopoly, Martin Gurri. I won’t go into the finer details but he had one insight I believe is worth sharing — be skeptical of anyone who is certain about how complex systems will behave in the future.
Said another way, there can be no concrete predictions about the future of any complex system. By definition the system is not concrete, it is uncertain.
Complex systems have too many variables to accurately predict outputs. Sure, some predictions will be correct. But anyone who lacks the humility to caveat their predictions with the requisite confidence interval is likely selling you something.
Furthermore, the difficulty in predicting complex systems is not only that there are too many variables, but that the causes interact with each other in utterly unforeseeable ways.
Franz Ferdinand’s assassination set off a chain of events that led to WWI. On the heels of that war the world powers came together in 1928 and *outlawed war*, signing The Kellogg-Briand Pact. The complex system that is human affairs saw otherwise.
Predictions of North American air quality may come true, unless the super volcano in Yellowstone sees otherwise. Unless a meteor sees otherwise. Unless scientists working on Nuclear Fusion see otherwise.
When dealing in complex systems, walk carefully — the moment you become certain, the world may see otherwise.