And yet, if you're interested in studying these "moral effects" rigorously, you'll attempt to model them mathematically. Not necessarily with from-first-princi

And yet, if you're interested in studying these "moral effects" rigorously, you'll attempt to model them mathematically. Not necessarily with from-first-principles equations -- sometimes the big data / machine learning approach is best. Plenty of (most? all?) economists understand the need to model human irrationality in order to properly predict economic behavior, but if you ask me the solution is not to somehow adopt a less rigorous humanities approach to economics, like you seem to be implying.