tl;dr of my point: we know dangerously more about technology than about people, their needs as individuals and their needs as a society. Somewhere along the line we should stop throwing technology at people just because we can, and start to focus on the right solutions - technological or not - to real problems.
Currently "social studies" in school is a bit of a trash bucket into which gets thrown every personal ideology and pet project that some teacher got her feminism or social work or grievance-studies BA in. It's the worst subject in most curricula, even worse than English. The problem is there's plenty of those nonsense degree holders and far too few college graduates who have studied the humanities.
Those are, despite the "social studies" stigma they might carry, real substantive disciplines which should not only contribute but predominantly shape the technological products and services we build.
Race/ethnic/gender/subculture (what you seem to call "grievance") studies are humanities (academic disciplines that study human culture) not sciences (soft or otherwise.)
History, Philosophy and Civics are still in their pristine condition.
However, psychology, sociology, economics, anthropology and especially communication have been systematically taken over, and have demanded self-censorship of those fields. When a field self-censors it can no longer be considered to be doing intellectually honest work. So they are dismissed as not Science.
You don't consider sociologists to be the least bit concerned about human cultures and experiences?
Regarding the need for a good grasp of statistics in social scientists, I fully agree. But I think it's something that already exists, up to a point at least. I work in software development with many software engineers around, but I also did a PhD with a strong social sciences background. You would be amazed with how much the guys with psychology backgrounds know about stats and methodology (and how much the common STEM guy doesn't).
The most interesting and potentially enlightening experiments are also probably illegal.
That's why it's hard.
And you ALSO need a branch of mathematics, called statistics, and probably many more. You can't simply pour lots of money into it without any hard science and expect useful results.
tl;dr: it's not about quantity, it's about quality and the political landmine of social sciences.
I fully agree with your points. However, regarding this one in particular, I would like to add that this shouldn't stop the harnessing of already existing knowledge in the social sciences to better shape the products we build.
I mean, you're totally right; there's a lot of knowledge "locked" in the difficulties related to studying humans. However, we know some stuff about people and their behaviour. A relevant amount of stuff I would say. And that stuff isn't being applied when we create a technical solution to a human problem; the solution is almost solely defined by technical aspects. We don't know much, but we know enough to do better than what we do with technology today, human-wise.
(At least, better than defining "UX work" as pixel-pushing in Photoshop or Illustrator)
It did not used to be that way. It started to change after WW2 and even stronger after 1967. Social "sciences" were made political because a group of radical influential thinkers were interested in furthering studies that would disprove the role of genetics, biology and darwinism in social studies. Before this shift, social studies were not politicized.