It's an accidental structure because it's based on an observed correlation that may or may not be a causation. The structure of Socionics is intrinsic to its make-up i.e. derived throughout from its axioms with the meaning of each part being explained from its structure.
I actually prefer Enneagram in terms of that. I mean, if Socionics axioms were less unfounded, more closely connected to reality in a way, then I would be fine with a derived system from such axioms.
Anyway, what I originally meant by Socionics structure being no better than the Enneagram one, it was more about how the basic axioms in Socionics do not lead to such a comprehensive theory on their own. Meaning, correlations do possibly get confused with causation in Socionics. Let me know if you disagree about that but then do please show me how you can derive everything from the axioms without adding observations. Afaik even the Reinin dichotomies came from observation. Or Gulenko's cognitive styles, they also originate from observation. And let's not even talk about Reinin or Gulenko but stick with the definitions of functions and then derivation of Model-A from that. I see several logical jumps there too, adding in stuff that isn't directly derived from the definitions in a strict sense. Do note that associations in general are not considered logical by me.
It's true that energy and happiness can be explained by the chemical composition of our brains but how in any way does is this chemical composition intrinsic to our fears and desires. How can my serotonin level determine whether I'm afraid of being controlled or of being deprived? An explanation may or may not be provided but it's in no way intrinsic to the Enneagram theory.
Socionics has similar jumps in logic.
It doesn't mean serotonin has nothing to do with any of that but clearly it's more complex than just a direct causal link and that's a problem for me yep

I prefer seeing the whole chain of causations...
There is a model to the Enneagram, namely, the Enneagram itself, the 9 pointed star. This is a structure although one may wonder why it's derived the way it is. Why does a 7 grow to 5 and disintegrate to 1? I can't see any explanation from the site. I might be wrong, in which case I would like to hear these explanations.
I have never seen an explanation for that one, it just seems to build on common sense logic or something. Not that Socionics does not make that mistake in places. (It does.)
The sites I linked to were just examples to show some of the structures Enneagram theory has, though the Horney one is definitely not that mainstream

.
Because information is a more basic thing than a person or physical object. We can say it's either abstract or concrete, objective or personal etc. but how else can we really classify information in such a fundamental way? I'm not talking about arbitrary classifications like distinguishing blue information from ticklish information. Now a person gets up to all sorts of things, there are lots of things a person can do or be that we can use to classify them.
I see a problem here. When you attempt to classify information as e.g. objective or personal, it will no longer be just direct low level analysis of information in a basic sense but it will be involving higher level concepts that relate more to the complex workings of a person. We might as well start classifying information in the same complex way as we classify people etc.
That's actually an issue I have with Socionics
Looking at it from another side, there are dichotomies for the information aspects listed here
http://wikisocion.org/en/index.php?title=Information_element
They are nice and basic, satisfying my requirement above. Now, you can build more than just 8 information aspects from those, no? Why are there only 8?
Uh and sure enough, another big jump is how we go from information aspects to supposedly existing processing modules in the brain, so-called information elements
I like structures, but this is kinda moving against a lot of what I thought I know about Enneagram, I'd need to know how it can be reconciled with the rest of the information, whether it is legitimate given what is the case. It looks like a new set of triads that are different to the triads I already know so I would need more research before I know what I think about it.
Okay. Where do you see issues with reconciling it with the other stuff about Enneagram?
Socionics Fi doesn't seem bad at all to me. A keen sense of interpersonal/psychological distance? Internal statics of fields? Morality? Sounds good to me.
Better than the common interpretation of MBTI Fi as .. well, completely selfish
Yet, none of it makes sense to me.
Not the MBTI one either
I'll expound, because why not
One of the goals of any typology system is to be as comprehensive as possible (for its domain of application)--to explain as much as it can. The number of factors is determined from there. The belief of Socionics is that it can be captured in 2x2x2x2=16 types,* MBTI also in 16, Enneagram in 9-ish types,** and Big Five in .. .. well, 5.. traits.
They'd require more factors if they weren't comprehensive; and they'd (likely) require less if one factor is correlated with (that is, not independent to) one or more of the others.
I see what you mean but I don't think it's a good idea to pre-determine number of factors in this way.
Btw... Big Five theory actually isn't about just 5 traits, it's just the five ones at the top of all analysed personality traits structured in a way.
As for the issue of correlations between factors, MBTI has that issue. Maybe the others too.
** Enneagram also expresses its types as points on a circle--a continuum of mental disorders (as extreme versions of the types) that loops back around. If it were to lose its points of integration/disintegration, the only thing that really ties it down to 9 types, it could subdivide that circle in a number of ways that'd make sense--it could have more than 9 if more would be useful. If the circle itself isn't comprehensive, then other disorders would need to be able to be mashed into that circle in order for the Enneagram to describe what it's intended to describe.
You could still add more types without changing the currently existing integration/disintegration lines, just make a new group of the new types... like 3-6-9 is a separate group from the other 6 types.
What did you mean about subdividing the circle in ways that would make sense?