I was once a strong advocate for MBTI. But that was before I expanded my knowledge-banks to neuroscience, not to mention my own collective research on the functions. Now, I put forward the proposition that typology is an inaccurate, and rather flawed, theory that attempts to simplify a very complex and diverse system that has only recently been explored in-depth within the mid to late 90's (you could even pinpoint rapid advances to the early 2000's) and we still don't know much!
A theory is a theory if it follows two distinctive traits:
(1) It must accurately describe a large class of observations on the basis of a model, and
(2) It must make definite predictions about the results of future observations.
These are generally held to standard in physics, but I don't see why it couldn't hold weight to theories in other scientific fields.
MBTI often has a hard time meeting those two criteria.
Typology theories, like MBTI to use as the prime example here (most familiar with) attempts to configure personalities that overlap, and often contradict, the core of each established identity. I could point out that the brain is a complex system of chemical and electrical reactions/impulses, ones that can be quantized, and would therefore fit in to the uncertainty principle. If you wanted to expand the uncertainty principle to a larger concept, you could propose that the more you attempt to define and pinpoint a person's personality, the less certain you can predict how it is moving. On the flip side, the more you attempt to predict where a person's personality is going, the less certain you can predict where it is at.
That may be entirely untrue though, as quantum mechanics is quantized, whereas in physics the level of personalities would fall under the threshold of General Relativity (if the only connection to General Relativity is the abstract view.) But I wouldn't the uncertainty principle maintains an important connection to personalities.
Furthermore, in the realms of neuroscience, the brain's structure can change over time, which then presents the dilemna of fixated personalities. Some neuroscientists and psychologists have argued against typology as the changing structure can throw typology into realms of questionable limits.
To add to the frustration is a myriad of subjective definitions of each function, with a unified definitization of each function non-existent, if not entirely vague. Few people like to refer to Jung, as his results are questionable. Even Myers's results could be questioned.
Edit - You can't discount that people like MBTI because it does have aspects of simplicity to it, and can be used with some ease. It is like comparing Newton's Laws to Einstein's General Relativity: Newton's laws, like gravity for example, have proven inaccurate, but it is also easy to use and not fully innacurate! General Relativity is far more precise, but Newtons's is not too shabby.
It is interesting to hear about, and discuss, the different angles of the theory of MBTI.
First of all, bravo on an excellent, thoughtful, objective post.
Have you done any research on Jon Niednagel? He's best known for his work in sports, but he's also working behind the scenes to identify the 16 types through genetics. He believes we are born as one of the 16 types, and that it predicts not only much of our behavior and cognition, but even our motor skills (he usually types people based on the latter observations). He prefers the term "Brain Typing" to personality typing because people's personalities can change based on a number of things - upbringing, personal morality, brain health (neurotransmitters, etc.).
I agree that people can easily get caught up in the vagueness of function theory - how does one apply something that's pretty nebulous? I showed a lot of what may be termed "Ne" as a child, but some people here swear that my video appears Ni. I'm speaking abstractly as can be, so couldn't one come to the conclusion that I am, in essence, extroverting Ne and not introverting Ni?
Niednagel's Brain Typing has about the clearest methodology I have found so far, but it's still not completely provable. As a school teacher, I would say that the majority of my students are NATURALLY extroverted, spacy and daydreamy, not too worried about how they come across to others, and tend to be far more open-ended than organized, in keeping with Niednagel's theory that ENTP is way way WAY more common than MBTI fans think - and that their motor skills tend to be rather loose, etc. in keeping with his motor connections with type - but even so, it's not like we have a blood test right now that can confirm an inborn type. People can always counteract observations with ones that contradict them, after all - the brain is too complex to be strictly extroverted or introverted all the time, motor skills can be developed with practice, etc.
Whatever the truth may end up being about typology - we're just at the tip of the iceberg right now in neuroscience - I look forward to learning more about how the brain works. We have SO MUCH to learn.