Notes on Nomenclature
Note: this article uses very heavily the standard abbreviation system for functions – T,F,S,N for thinking, feeling, sensation, intuition, then ‘i’ for introverted or ‘e’ for extraverted, i.e.
Te=Extraverted Thinking; Fe = Extraverted Feeling, Se = extraverted sensation, Ne = extraverted intuition; Ti = introverted thinking, Fe= introverted feeling, Ni = introverted intuition; Si = introverted sensation.
Also note again that I do not always agree fully with standard definitions of how each function works or is defined, particularly Ni and Fi.
The following is an attempt to bridge further between my existing ideas as to how cognitive function preferences seem to operate in the types theoretically, and how this tends to come out in practice. The theory I am proposing here suggests that types may have roughly a ‘set’ primary stack, which the other functions are somewhat individually variable or developable. I would propose that in many cases, a strong development of a function alongside the primary function provided by the type can obscure the ‘real’ type.
In writing this I am considering
1) the theoretical model we have so far, for the most part based on my own analysis of various types’ functions as shown by works they have written;
2) occasional unexplained results in that same data
3) The amount of variation we see in MBTI results, though some of this is probably explained by measurement errors of various kinds e.g. social vs. object-related introversion.
For prior posts relating to this, see here for discussion, as well as for the original hypothetical list of Jungian function preferences by type (although that’s also reproduced below, with elaborations). https://chrisblog662.wordpress.com/2017/02/16/mbti-exploring-the-sn-axis/
Inter- Function Relationships As Suggested by the Enneagram
Adjacency between 1/2 [Fe/Te] and 4/5 [Fi/Ti] suggests similarities there, for example – we can imagine relationships between functions like a Te-dominant person being able to store ‘facts’ about people’s normal preferences and social customs, letting them duplicate Fe to an extent (and eventually develop actual Fe).
Likewise, Ne and Se [7/8] being adjacent suggests that [extraverted] intuition can be used to find fun things, even if it is a ‘fuzzy’ way of seeing the environment that tends to overlook small details and so conflicts with ‘Si’, the 7 inferior function. Ni/Ne [6/7] suggest a relatedness between Ni and Ne as well, except that Ni is intrinsically geared toward possible worries or anxieties i.e. negatively-focussed instead of positively-focussed.
Another view is apparent looking at (IMHO) the ‘stacks’ that form when wings have orthogonal preference. For example, it seems 2w1s develop Ne tertiary instead of Te, hence secondary Si and likely an ESFJ pattern, or 8w9 where secondary Fe develops instead of Se – seems to suggest also development of Ne can convert into Te or vice versa, and Se to Fe.
Comparison of other theories
The MBTI generally uses effectively a 4-function model, by having Ti/Te, Ni/Ne, Si/Se and Ni/Ne be grouped as items. This is at odds with the probable relationships generally predicted between functions in the enneagram, where wing adjacencies suggest degrees of ‘relatedness’ suggesting how functions can develop (we wouldn’t expect two adjacent types to have entirely opposing mental structures as that would make a psyche with that type/wing unworkable).
Socionics generates an 8-function model, but again has function maps that conflict with the enneagram, e.g. where introverted thinking and ‘introverted ethics’ [its Fi] are deemed to be extreme opposites; it likewise disagrees with the structural relationships that seem to be indicated within the enneagram (here with 4w5).
A semi-freeform 8-function model
Here I am proposing instead a partly ‘freeform’ function model; the dominant and inferior functions are set based on both type and wing – giving basically two dominant and two inferior functions, although for some types the realities of needing both an extravert and introverted function will promote a function that should be weaker to a relatively prominent level. The other functions are presumably variable; rather than Ti and Te being similar they are actually ‘sold separately’.
This accounts for:
*a four-function model doesn’t fully explain the full individual differences in preferences – potential tritypes, instinctual variants, or the full spectrum of MBTI results.
*that noticeable ‘inferior functions’ were apparent e.g. to Jung. Fairly often, its assumed that a ‘stack’ of a,b,c,d indicates those are the main functions, like the remaining four are actually somehow worse; to be discernably inferior, the inferior function has to be THE actual worst function. If a stack predicts only how good 4 of the 8 are, then a freeform model can state that these are actually ‘not defined’ as either good or bad, instead suggesting that they will vary.
Consequently, it may be that while type predicts ‘primary stack’ functions, it doesn’t necessarily predict the ‘secondary stack’, or put another way, development of some functions isn’t particularly helped or hampered by type, leaving these to be developed or not as “free choice” by the individual. Type and wing can be expected to drive two functions, and conversely inhibit the opposing two; this leaves some in the middle that are theoretically 2nd and 3rd but can vary in development, and another two or three that could vary anywhere from wholly-undeveloped to extremely-high: early environment could help or hinder, or education and professional development could drive development of these to very strong levels [conversely, the inferior function is something people will shy away from developing, like the 7 avoiding Si-based tasks like checking government forms for misprints).
Bringing the enneagram into it, makes it possible to narrow down specifically which are best/worst more specifically (see table).
Information from common MBTI typing
Note that if this is correct, it explains another subset of difficulty in getting workable results from the MBTI. The MBTI’s determination of ‘stack’ correctly requires all of the letters to be correct (e.g. E/I, difficult as it focusses on ‘social’ rather than subject/object relatedness – and P/J), which means that an unusual result will throw off all of the results. A test which tests all of the 8 functions separately is more likely to show that e.g. 8s are consistently good at Se, even if some are also strong in Ne (7 wing) or Te, or that 5s consistently have good Ti but could shift toward F if they also have introverted feeling (4 wing). Particularly strange results can be largely accounted for by combinations of mismeasurement and strong development of an unusual function – an Eight might score as an INFJ if they have stronger social introversion, combined with particularly strong Ne development; they wouldn’t be likely to have actual ‘Ni’ since that would mean a very high level of integration that moves directly against the “passion” of the type; an 8 shouldn’t have ‘inferior’ extraverted sensation like a genuine INFJ.
Functions labelled ‘inferior’ can be developed but this would be much harder, as this is energy spent moving directly against a types’ “fixation”. For some types [2/3, 5/6] the stacks design forces them to develop the wing’s inferior function as their second function.
Overall, it may be more difficult to develop ‘introverted’ functions than ‘extraverted’ ones – the former being probably more complex.
Functions marked ‘likely’ are ad hoc assumptions on my part, in part based on the usual MBTI patterns seen (or for 4, the literature notes above).
Overall by primary type/wing, we can predict a theoretical stack and probably what’s ‘free choice’, which I think is roughly as follows:
[note: pasted as image due to table formatting problems]
Note: The stack functions above assume that the genuine ‘inferior function’ is the worst for a given type. The other main model of how types might work, may be one that another of my friends has, that a given type actually has a ‘superinferior’ function, which is the opposite orientation of the main function (for instance, Ti-dominant 5 might have a Te ‘superinferior’, or 2 with its Fe preference might actually have an Fi ‘superinferior’). Currently I’m not in favour of that model, although it would explain the types that show ‘forced development’ of what appears to be an inferior function i.e. 5w6/6w5 and 2w3/3w2, since otherwise we might expect that function preference might ‘flip’ i.e. a 6w5 might be expected to gain a Te preference for secondary function, rather than using Fe (the five inferior) as secondary. There may however be other explanations, for instance a 6w5 may want ‘Ti’ specifically and even as a tertiary this is more ‘useful’ in development than Te.
The main data point apparently indicating against the ‘superinferior’ model would be the 6/7 [Ni/Ne] and 8/9 [Si/Se] functions being adjacent on the enneagram; note it would be possible that the judging [T/F] functions to work differently to the perceiving functions, however. Empirically, 4/5 and 1/2 change little regardless of which model, but switching to the ‘superinferior’ model suggests 5w6/6w5 will have bad Te rather than Fe, 2w3/3w2 bad Fi rather than bad Ti.
Supporting Information for the above from literature analysis:
While generally the literature analysis approach explains a fair bit of individual difference in terms of people and is overall fairly good, we do see people at times performing much better than expected at a function. Literature wise, for example:
*sexuality is expected to be a part of Se, but a couple of the 2w1s and at least one 8w9 [Piers Anthony, Geoff Ryman; Robert Silverberg] who we expect to have Si – secondary for the Twos and Si-primary for the Nine, seem very strong here].
*Harry Harrison [8w9] would be expected to be Fi-secondary, but as an SF author, seems to have quite good Ti.
*A few other miscellaneous people have competent Fe where it isn’t expected particularly [Dave Duncan for example, who would be Te-secondary [6w7] but looks to have reasonably good empathy/people skills].
*among some of the 4s, we see very good introverted sensation, at times unusually, For Donald Moffitt, I’m now considering 4 as the type, but most indications are that he has extremely strong introverted-sensation. As a hard SF author, he’s likely a 4w5 with 5 ‘Ti’ being quite strong, as well. The sensation as a 4w5 would be an unusually strong development of the tertiary function (‘tertiary temptation’). Of other 4s, Clark Ashton Smith [4w3] as noted has strong Se, but the richness/depth of description there suggest some Si development as well. Michael Ende [3w4 ESFP] likely also has quite good Si.
At the other end, there could also be problems generating ‘underperformance’ in a function that a type is normally good at. Presumably if this happens early in development, it would prevent someone becoming a type that utilizes that heavily, or result in type being locked in to the one that corresponds to that inferior function, but in other cases distortion of functions could occur. A couple of problems that might be of this kind would include high-functioning autism spectrum problems of the kind that hamper social interactions [affects Fe, but at least one 7 I know of has], or aphantasia [probably Si; despite Jungian predictions if anything is probably correlates more to 6 than 7, since 7s are apparently visuals?] – this last seems to have a genetic basis rather than be type-based and predispose toward N – 15/16 respondents to a survey of this on Reddit with the condition were Ns).