thomascolthurst

I guess I view it as both a philosophical issue and a practical one. The way I phrased it above ("What can data tell us about causes?") may make it sound exclusively philosophical, but you could also phrase it a purely practical way: What can data tell you about what to do, given that what you care about is almost always different from what you can directly effect? (Here I'm construing "directly effect" very narrowly -- basically just muscle control -- and what you care about very broadly: health, wealth, happiness, etc. for yourself and others.)

But here's a question for you: if data can't tell us about causes, what can? If humans infer causes incorrectly pretty often as opposed to all of the time, then sometimes they must get causes right -- how?