Actually, correlation kinda does imply causation
There’s a famous rule: “correlation does not imply causation.” Unfortunately, it’s wrong, *as normally stated.
It’s missing an important word. It really should be: “correlation does not necessarily imply causation.” Because correlation actually does “imply” causation, and many (if not most) events that occur in sequence that appear to be causally related are in fact causally related. Human brains are dazzlingly good at correctly inferring causal relationships from observed correlations: clapping makes noise, braking stops cars, hot coals burn fingers. This mental super power served us well as we grew up as a species.
The problem is that we’re so good at it, and it’s such an essential mental skill, that we tend to overdo it and perceive causation in all kinds of situations where causality detection is much harder… like evaluating the results of medical treatments.
Complex causal relationships are as tricky to infer from simple observations as simple ones are easy. And we are just pathetically bad at figuring out exactly how events are causally connected — “mechanism of action.” Because of all the unknown variables. What’s really going on in a casual relationship almost always turns out to be different and waaaaay more complicated than we thought.
Nature: defying “common sense” since the dawn of intelligence.
But humans are causality bloodhounds: we smell it everwhere, even when we don’t understand what’s really going on (which we usually don’t). For instance, if someone who’s been limping and grimacing for days walks out of a massage appointment with a grin and a light step, then, yeah, massage probably did cause that result, one way or another.
[Update: see several important clarifications in the next post.]