Of course I hold myself to the same standards--they're just conditional standards, and you probably haven't figured out the conditions yet for when I think what level of support is called for along with the original claim, what I think is appropriate to ask for, and what I think is inappropriately burdensome, unnecessary, irrelevant, or counterproductive. Feel free to ask questions and I'll explain. I don't think any of it is unusual; it's a fairly ordinary though high standard for back-up-important-claims-with-something (whether evidence or argumentation depends mostly on the nature of the claim).

Also, although I try my best not to be biased, it is probably the case that I have made mistakes about "this is too obvious or tangential to be worth supporting at all" in my own writing and "surely you need to support that central outrageous claim" in others’, but it's not that I'm not trying to apply the same standards. It's just that it's hard to understand others' perspectives as well as I understand my own, so I don't always perceive the parallels in conditions perfectly. This is why communication is valuable.

Anyway, it's all easily resolved with a few questions back-and-forth.

But your claim that I have no data whatsoever in the article is wrong. Firstly, I cite a study on IQ--not very important to the overall argument, but it absolutely is data so technically your charge is false. Furthermore, the examples I give are, in fact, data; case studies aren't very good data, but they are data.

Regarding "I believe in science", I admit that it's possible that I'm simply out of date. My internal sample size is large and varied enough for it to be fairly reliable (though admittedly not shareable, which is always a downside), but almost all of it comes from before Covid. If this is now something commonly said, it is entirely possible given my current circles that I have missed it said with the meaning Giles ascribes. So that aspect of my critique might be incorrect. I am not sure how to find out, however. Neither asking LLMs nor searching Google was particularly helpful in this regard with the queries I tried.