by Brian Awehali
When Edward Lorenz gave a talk in 1972 entitled “Does the Flap of a Butterfly’s Wings in Brazil Set Off a Tornado in Texas?,” he distilled the main essence of his thoughts on predictability, interdependence and “chaos theory” in one pithy question.
Lorenz was a mathemetician and a meteorologist who, in the early 1960s, discovered that weather simulation models he was developing were exhibiting chaotic, non-predictive behavior, despite a fixed set of variables and no apparent equipment malfunction. Two identical weather simulation machines, side-by-side, given the same variables to process. Wildly different results. How?
Lorenz eventually concluded that it was a “dependence on initial conditions” — in this case, the fact of computers rounding variables to decimal points: 3.12879 expressed as 3.13, etc. Even extending the number of decimal points in the simulators did not produce matching results from the weather machines. Minute variations gave rise to wildly different chains of events.