Prediction vs. Control

A large part of science has grown around the ideas of theorizing on causality and experimental verification of the predictions. But with complexity becoming increasingly interesting, perhaps it’s time to admit that prediction is no longer the key defining characteristic of science. Rather it has to be control. Control means that you don’t fully know what a process does, but you do know enough about its global properties, which you monitor, so that while its inner workings remain opaque, you get the macro results you desire. All the most complicated fields — all social sciences, medicine, meteorology — they all suffer from the lack of predictability. But they’re all controllable to various degrees, which is the key characteristic of these systems, as opposed to a pulley-lever system, or gravitational attraction, which is predictable in human timescales.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.