Overly-detailed steelman of MBTI

Aella (@Aella_Girl)

Everyone knows that MBTI is pseudoscience, barely better than astrology - with too-vague descriptions and "well most of these sorta fit me" personality descriptions.
Well, you're only halfway right! Buckle up kids, I'm gonna take us on an overly-detailed steelman of MBTI.
1. MBTI is basically astrology.
Only kind of right. You can expect to see a basically even distribution of astrological signs across everything - GPA, big 5 personality test, jobs, cities - but information from all of these categories is predicted by MBTI type.
How useful is this? Not incredibly useful - I don't know how much this is 'carving reality at the joints' - but this does mean that if someone has tested as X type, you can actually make some predictions about them that will turn out more right than if you just knew their sign.
Also: Did you know MBTI types (e.g., ESFP) is just shorthand for a *function stack*? The function stack of ESFP is Se, Fi, Te, Ni. There's a whole complicated world underneath here, which comes from a symmetrical system with pretty basic, stacking rules.
Here's a description of the individual functions: cognitiveprocesses.com
And a chart of how they stack for each type: 66.media.tumblr.com/3480bd5a39e0c9…
The theory is something like, everyone has to engage with different parts - the world outside and the world inside. There are different spectrums in how people do this, and people can be more focused outside or inside individually on different spectrums.
this is like infinitely better than Enneagram, by the way, which as far as I can tell is basically some guy stumbled on some mystic traditions and was like, "I'M GOING TO SHOVE PERSONALITIES INTO THESE ARBITRARILY CHOSEN BINS".
I sorta hate Enneagram.
So actually understanding Jung's ideas and how they relate to function stacks takes a while to do, and involves a theory of mind that focuses a lot on modes on internal and external sliding scales. This is why advanced MBTI stuff says you can't figure out your type based
on internet tests - ideally you'd interview with a trained professional. So - you took an internet test and it 'kinda fits'? But other descriptions fit too? MBTI steelman says you took a shit test and the descriptions are overly simplified, and the reality is a granular system.
Now, all of this being said, and even though I've read way too fucking much about MBTI, I still think it's got some weaknesses.
For example, there's a rule that one type of function must be followed by a certain type of other function (e.g., no "Ni-Fi-Se-Ne"). I don't understand.
I've never come across any reasoning for why this rule is the case. Maybe Jung wrote about it somewhere? But it seems kind of arbitrary to me, and possible explanations aren't compelling. And it makes up a big basis of the theory behind MBTI. I'm pretty suspicious.
And while different MBTI types do have pretty strong correlations to different stuff like GPA or income, it doesn't feel as strongly predictive as something like the Big 5 (even though it's also strongly correlated to Big 5!)
Anyway, in conclusion, MBTI doesn't really deserve all the shit it's getting. It's actually a pretty cool, predictive, and beautiful system, when you look deeply. Its actual flaws are that the beautiful system is closer to a work of art than our best tool for mapping personality.