Judea Pearl is a scholar on the math of cause and effect. His work focuses on taking what humans understand intuitively and determining if and how it can be expressed mathematically for computers or AI to use. What’s really interesting is how his way of describing problems can also describe the flaws in our own thinking.*
Imagine the weather app on your phone warns you it’s about to rain and then it actually happens. You, as a human, understand that the app didn’t cause the rain, it just was relaying some information to you. A computer, however, doesn’t carry that same intuition. It has to be programmed. What’s the difference between a predictive warning and a causal factor? This is part of what makes Pearl’s work so interesting.
Before we mock our future robot overlords too harshly, we should remember that humans also aren’t the best cause and effect collection machines either. We’re quite good at drawing faulty conclusions on our own. Have a lucky number? A special routine? A deep understanding of economic history? Cause and effect may seem simple, but their relationships are often quite complex.
There’s a lot going on in the world, and there’s a veritable choose your own adventure book of interpretations. Narrative history isn’t our friend when we’re looking for cause and effect relationships. To paraphrase Alex Rosenberg in How History Gets Things Wrong, “Many Things don’t have answers.” It’s hard to say we don’t know, especially when we’re supposed to be the professionals, but often times it’s the only correct option.
What we can do is focus on explaining the process for thinking about the questions. Yes, it leads to one-handed economist jokes (“stop saying on the other hand already”), but like the weather forecast, it acknowledges a distribution of possibilities. When we look backward we can describe what we know that’s happened, how we think things are related, and why we think so. Perhaps most importantly, given our own exposures and time frames – how much should we care or worry as we consider a wide range of scenarios moving forward?
The job of the professional isn’t to adjust the picture, it’s to adjust the frame. The frame can’t encompass the entire world, we have to move it around to look through it. We’ll have blind spots and we’ll have things that seem painfully obvious in hindsight. What counts is that we have context – clearly understanding the things we can control, the things that actually matter, and the narrowly overlapping area of what we should be focused on.
With clients, we should bring less cause and effect and more causes and effects in context.
*Listen to Judea Pearl speak with Sam Harris on the Making Sense podcast episode #164 – “Cause & Effect.”