[wpietri]

Your theory is that all social commentary is handwavey general remarks? If so, yes, let's invalidate it.

I think effective social commentary is mostly grounded in the personal, the specific, and the (thoughtfully) statistical. Otherwise it's very easy just to project one's biases onto a complicated picture. E.g., consider whatever it is that people think is "ruining the kids" these days. In the 1700s, you know what was ruining the kids? Novels: https://www.economist.com/1843/2020/01/20/an-18th-century-mo...

It's similar to how arguments about "what God wants" go wrong. Researchers showed that when people talk about what their god wants, they're unconsciously consulting their own preferences: https://www.discovermagazine.com/planet-earth/creating-god-i...

With broad social argument, it's easy to introduce all sorts of fallacy. Humans aren't really equipped to reason intuitively about things at this scale. Especially if the point is to persuade and/or entertain, simplifying into broad "logical" arguments can paint false pictures. Andrew Gelman wrote about this today: https://statmodeling.stat.columbia.edu/2022/04/24/the-perils...

[simulate-me]

The error in your line of thinking is that you assume humans are better interpreting statistical evidence than they are broad logical arguments. I don't see any evidence of that.

[wpietri]

It's not a error until a) you show that's the case, and b) it's material to my point.

Honestly, I don't think people are "better" at that (if such a fuzzy construction can have a useful meaning). But I do think they are more careful with it because it's harder to get right and more easy to visibly get it wrong. It's fine with me if expecting more concreteness just drastically reduces the total amount of social commentary. Even if it dropped by 99%, we'd still have more than we needed.