first of all thank you for actually tackling the challenge this thread was supposed to pose instead of agreeing. this is the core debate this thread aims to stimulate, with a seemingly high degree of failure (what's with all the agree'ers?).
That's the first time anyone has thanked me for disagreeing with them..
You're most welcome, it's something that comes naturally to me.
so fast forward:
within this structure:
EP: Pe>Ji>Je>Pi
IP: Ji>Pe>Pi>Je
EJ: Je>Pi>Pe>Ji
IJ: Pi>Je>Ji>Pe
we see two core principles:
J>P>P>J (EJ & IP)
P>J>J>P (EP & IJ)
The second 2(6) functions vary in different models. I do not recognise fixed function orders. There is no support for this speculation. As I said before, it's unnecessary in order to derive type. We all use 8 functions with varying degrees of facility.
in other words, we are assuming that these:
J>P>J>P
J>J>P>P
P>J>P>J
P>P>J>J
as well as these:
I>I>E>E
I>E>E>I
E>E>I>I
E>I>I>E
are impossible.
why would they be impossible?
Who said they are?
The 16 types are
idealised patterns of
normal development. In healthy individuals balance between "introverted" functions and "extroverted" functions is important. Imbalance is pathological. This should be self-evident. Homeostasis is the definition of health.
However, unbalanced development is entirely possible and not uncommon. When people talk about Dom-tert loops, they are talking about a person whose auxiliary function isn't functioning normally (to provide balance) which means they are operating in, e.g. ultra-introvert mode. (Effectively, I,I [E,E]) But since we regard imbalance as pathological, then it would be wrong to type a person by pathological manifestation in a system which is aimed at delineating
normal personality types. (MBTI is not helpful where an individual has a personality disorder or other mental illness.) In short, there is an implied assumption that I,I,E,E is an aberration, but that doesn't make it "impossible".
and please remember we're talking psychology here, so "innate for the system" doesn't quite meet the benchmark, yes the system is internally consistent, but that doesn't mean it's principles are applicable to it's theoretical subjects (people's personalities).
Personality is a system. Brains are systems. They do not have logic or physics-defying properties. They cannot do two mutually exclusive things simultaneously. Evidence suggests that they specialise and some of those specialisations are important for personality.
Also, you have to remember that all of this is metaphorical. It doesn't have a concrete implementation (that we can prove). It's simply a useful model for describing personality differences. It's usefulness deteriorates as you introduce more and more subcategories (which is an argument against so doing). How useful it is, as you rightly imply, is a function of how closely it matches real people and generates valid predictions. Limited support for the Dom/aux relationship is provided by the individual's own responses (+ implied dichotomies*). E.g. if you score ENTP, you know that you are a thinking extrovert who prefers perception. Given your answers, (and assuming the existence of cognitive functions with attitudes) this gives you Ne+Ti. This says nothing about tertiary or other function use (although by implication, you will have *least* facility with Si and Fe, since these work in direct opposition to your preferences). Beyond that, there is no support for the arbitrary ordering schemes of type dynamics, and so it's possible (and probably sensible) to disregard them. (See
James H. Reynierse, “
The Case Against Type Dynamicsâ€, Journal of Psychological Type, Issue 1 2009)
Type dynamics has persistent logical problems and is fundamentally based on a series of category mistakes; it provides, at best, a limited and incomplete account of type related phenomena; relies on anecdotal evidence, fails most efficacy tests, and does not fit the empirical facts.
The prediction that provides the greatest support for type dynamics, that the dominant > auxiliary > tertiary > inferior never occurred for 270 direct tests of the Grant-Brownsword model and occurred only once for 270 direct tests of the Manual model.
So, only once in 540 cases was the dom>aux>tert>inf order validated.
*It's also possible to dispute the existence of these dichotomies. However, they *are* useful heuristics. The E/I distinction (while a continuum and not a dichotomy) is real (has a physiological basis) and is useful in terms of predicting behaviour. N v S is useful in predicting interests and aptitudes. T v F is useful in predicting interaction and decision-making styles. J v P is similarly useful in predicting general outlook and response to novelty/change. Combined, they are useful in predicting career interests and aptitudes, which is, after all, what the instrument was designed for, it's probably asking too much to expect more from it.