The Impact of 'S-N Blindness' on the Distribution of MBTI Types Across the Enneazones

It was our view that the MBTI types found in a zone could be thought of as clustering around 'issues' related to the inferior function associated with that zone. We demonstrated that the traditional descriptions of each enneatype contained significant references to the specific inferior function of the prototype associated with it. This is, in our view, what makes each pair of MBTI types prototypical of the enneazone to which it is connected.³

In this paper we will present the view that there is a second principle to be taken into consideration when attempting to understand the distribution of MBTI typeS across the enneazones, related to a phenomenon that we shall call 'S-N blindness'. We will show that these two principles, in combination, explain the EM survey data quite well.

The two principles can be summarized in the following way: 1) there is a clustering of MBTI types in each of the enneazones around issues related to the inferior function associated with a particular Jungian type; 2) there is a propensity, in the world of the enneagram, to treat the Jungian functions 'S' and 'N' as if they were one function - or, put in another way, there is a failure, in enneagram theory and testing, to adequately distinguish between S and N (ie, there is 'S-N blindness')

Section One: Anomalies

The EM survey data provides compelling evidence for our prototypical pairs in 12 or 13 out of the 16 MBTI cases (or in 6 out of 8 cases, if the data is analyzed with respect to Jungian type). In this section we will review the supporting data for the theory that we constructed on the basis of the first principle above, answer the criticisms lodged by Tom Flautt and identify possible anomalies for that theory. In subsequent sections principle 2 will be invoked to account for anomalous data.

Flautt identifies the five (out of sixteen) MBTI types which, in his view, offer special difficulties for our theory. These five, according to him, have special associations with enneazones different from the zones to which we assign them as prototypes:

... To be specific [Fudjack and Dinkelaker's] theory does not predict the following SRTT significant correlations: 3-ESFJ & ENTP, 4-INFP, 5-INTJ, 7-ESFP.

Flautt invokes Baron and Wagele's [B&W] conclusions in support of his assignments, but does not seem to recognize that in a number of cases B&W's conclusions contradict his assignments, and agree with ours. As a matter of fact, if one takes B&W's correlations as the 'standard' to which both our assignments and Flautt's may be compared (in order to do this one can use Flautt's chart in the Sept issue of EM), one finds that both Flautt and we have exactly the same number of 'hits' (agreements with B&W): eleven.

Lets look at some of the MBTI types that Flautt identifies as problematic for our theory. Although B&W do list the ESFP as one of the five 'most common' types in zone 7 (agreeing with Flautt), also included in their list of 'most common' are the two types that we take as prototypical of zone 7, ENFP and ENTP!

Furthermore, for zone 3, B&W identify ESFJ and ENTP (Flautt's choices) not as 'most common', but only as 'quite common' and 'less common' (respectively). This poses a problem for Flautt's assignments in zone 3.

Also, we are not convinced that the INTJ's concentrated presence in zone 5 produces problems for our theory. Here's why - using the EM data, we calculated the I-value for INTP, ISTP, and INTJ in zone 5, and this is what we discovered: that the I-value for both the INTP and ISTP (our choice of prototypes for zone 5) were greater than the I-value for the INTJ (Flautt's choice): INTP = 4.3, ISTP = 2.7, INTJ = 2.6. But Flautt chose to take the INTP and INTJ as representative of zone 5, ignoring the higher I-value of the ISTP!!!

In other words, 3 of the 5 assignments that Flautt points to as providing problems for our theory may actually be problematic for him: 3-ESFJ & ENTP, and 5-INTJ. And the studies that he invokes (from B&W) in support of his 4th assignment (ESFP-7) may also be used in support of our alternative assignments (ENFP-7 and ENTP-7). Finally, although Flautt's analysis, as the reader may recall, generates only one type representative of zone 9 (the INFP), we would like to point out that the ISFP (which we selected as prototypical of zone 9, along with the INFP) has a value of 2.6 in zone 9, the highest in that enneazone. Furthermore, an analysis of the data according to Jungian type reveals that the IF pair has its highest I-value in zone 9, not in zone 4!!! And, if this were not enough, B&W also conclude that the INFP and ISFP pair are the two 'most common' MBTI types in zone 9.

Indeed, when one calculates the I-values for each MBTI type in each enneazone, one finds that in 12 cases, the highest I-value for the MBTI type is found in the enneazone to which we have assigned it. This is important, as it means that in these 12 cases the MBTI type 1) has a greater propensity to gravitate toward the zone that we find it prototypical of, and 2) has its highest concentration in that zone. The four exceptions to this rule are: ESFP, ISTJ, ENTP, INTJ. It is with these four 'anomalous' types that we wish, in this paper, to deal in more detail.

Section Two: 'Data Lenses' and Their Importance

As the prototypical MBTI pairs that we assigned to zones 2 through 9 are equivalent to the eight Jungian types, it was only natural to wonder how the EM data would appear if seen in this light. It is useful to view data through various 'lenses' which bring into relief different 'levels' of specificity of information, thereby revealing patterns that might not otherwise be noticed. In fact, in the article originally presenting the EM raw data (Isaacs and Fudjack, EM, March 96), the information was intentionally grouped in such a way as to make it convenient to view the data from three 'levels' of interest: 1) the 16 MBTI types; 2) eight Jungian types; and 3) the four functions.

In our July 96 article we suggested taking the same tool (SRTT) that Flautt was applying exclusively to the first (MBTI-specific) level and applying it to the second (Jungian-specific) level, thereby creating an alternate 'lense' through which the same data could be viewed. So we combined the relevant cells (eg, INFJ and INTJ, to get 'IN') in each enneazone, and calculated the associated 'I-values'. This analysis brought into relief the important fact that in 7 out of 9 cases, the highest I-value for the Jungian type in each enneazone is the one that we previously discerned as the prototype for that zone. The 'IS', for instance, has its highest I-value in zone 6 (meaning that the IS is most likely to gravitate toward 6, and has its highest concentration there). [The two exceptions were ES and IN. With these two exceptions, and the four MBTI types listed above, we will deal below].

But before going further, we would like to digress by pointing out that there is benefit to be derived from the creation of additional 'data-lenses' with which to analyse the EM survey data (two of which we shall propose herein). One could, for instance, further collapse the data by combining certain Jungian types together [eg, so that EF plus IF = F]. By assuming the perspective offered by this 'widest' angle lense (which sees the data as defining four types - the feeling type, thinking type, sensing type, and intuitive type), one discovers that in the EM study one has data which can be used to evaluate the claims of various enneagram theorists regarding the so-called 'triads'.

Here's how that can be done. In addition to adding the appropriate cells together in the way mentioned [eg, EF and IF = F] also combine the S and N types, so that the data distinguishes only between Thinking types, Feeling types, and 'Other'. Calculating the the I-values for each new cell, one arrives at the following table of I-values, which one can use to evaluate various triad theories:

                  N or S (=C)           F               T   

 ZONE 2              0.93             1.59*           0.18 

 ZONE 3              1.28*            1.03            0.89 

 ZONE 4              0.96             1.46*           0.08 

 ZONE 5              0.99             0.47            2.00*

 ZONE 6              0.88             1.25            0.81 

 ZONE 7              1.45*            0.38            0.73 

 ZONE 8              1.05             0.23            2.54*

 ZONE 9              0.67             1.65*           0.70 

 ZONE 1              1.10*            0.75            1.07* 

We have put an asterisk beside the three highest I-values in each column. The EM data tells us, then, that the 'thinking' types (as assessed by MBTI standards) gravitate most strongly toward zones 8, 5, and 1 (in that order); the 'feeling' types toward 9, 2, and 4; and so forth. Compare this data to Palmer's view that the 2,3,4 triad is the 'feeling' triad (comprised of 'feeling' enneatypes), the 5, 6,7 triad contains 'intellectual' types, and the 8,9,1 comprise a third category of type (which, from the MBTI perspective, looks like a conglomeration of S and N - Palmer calls it the 'gut' triad). Only four of these nine assignments agree with the EM data (which show enneatypes 2 and 4 as 'feeling' types, 5 as a thinking type, and 1 as 'other'). Incidentally, although Wright identifies enneatype 9 as a feeling type, he is also apparently caught up in the need to maintain the (traditional) 891, 234, 567 triadic groupings, and so makes the rather dubious (judging from the perspective of the EM data) decision to conceive of the 8 and 1 as 'feeling' types also. Hurley and Dobson 'correctly' (again, judging from the perspective of the EM data) identify the same four types as Palmer, but they (like Palmer) miss the fact that type 9 is best characterised as a feeling type (it indeed 'collects' the greatest concentration of MBTI 'feeling types', according to this data), that 8 is a thinking type (with the highest 'I-value' amongst all of the thinking types), and 7 and 3 are predominantly 'other' (which H&D call 'creative').

So it can be very helpful to look at the data through lenses that bring into relief different levels of specificity. Of course one should not choose more general levels to the exclusion of information available at more specific levels, and it was not our intention, in invoking a Jungian-specific data lense, to do this. Indeed, we feel that in order to actually test our theory it would be necessary to go to an even finer level of detail than is present in the EM data at the MBTI-specific level which Flautt prefers. The EM data, as it stands, does not distinguish, for instance, between INFPs for whom the fourth function, as the 'inferior' one, provides the 'issue' which attracts the individual to his/her enneazone, and those INFPs for whom an underdeveloped 3rd function will present the 'issue' which attracts her/him to the enneazone in which he/she resides. The EM survey, of course, did not attempt to obtain information about individuals at this level of specificity. But it is only with this sort of (currently missing) information that our theory could actually be adequately evaluated. In our view any future 'definitive' study of the relationship between MBTI and Enneagram must consider lenses at this more detailed level - ones which would generate information of the sort that could confirm/disconfirm hypotheses about why some individuals in one MBTI type gravitate towards one enneazone and other individuals of the same type gravitate toward another zone.

The INFJ, for instance, has only two cells with I-values above 1: in zone 4 [I = 2.3] and zone 1 [I = 1.4]. We suspect that those INFJs who wind up inhabiting zone 1 will have comparatively higher 'J' preferences than those who wind up inhabiting zone 4 - and that the 'issue' that attracts them there will have more to do with the concerns typical of the 'J' types prototypical of zone 1 and with a struggle to develop and utilize their 'third' function ('thinking'), than with the issues revolving around the fourth-function 'sensing' concerns that attract other INFJs to zone 4. But this is just conjecture, in need of data, and the EM survey did not have the specificity to provide us with information regarding this matter. [Similarly, we would conjecture that for INFPs those that gravitate toward zone 9 are being pulled there by their inferior fourth function (thinking) which they share with ISFPs, who are also pulled there, while INFPs who gravitate toward 4 will be pulled there by their underdeveloped third function, and have issues that are more similar to the INFJ's inferior Sensing: money problems, problems dealing with the body, and so forth.] These are conjectures that can be 'tested', but not by applying the EM data per se (nor by simply invoking the SRTT)- as the raw EM data contains no pertinent information regarding these matters!

Section Three: Measuring 'Spread'

In order to deal more satisfactorily with the four anomalous MBTI types mentioned above, we turned our attention to a different type of analysis of the data. When one looks at the (March 1996) EM raw data chart, one gets the impression that certain MBTI types distribute through the enneazones in a manner that parallels certain other MBTI types. Some of the MBTI types seem to mimic each other's distribution behavior, concentrating heavily in similar zones, and together avoiding other zones. Could similarities in distribution between any two MBTI types be quantitatively measured so that they might be compared to the similarity of distribution in other pairs?

The first thing to be noticed in this regard is that the little triangles in the EM data chart might lead to misleading conclusions regarding patterns of distribution (for the reasons pointed out by Flautt, which - after all - is the principle motivation for adopting the SRTT methodology). So it is the 'I-values', not the number of triangles, that must be compared. We did so in the following way. The INFJ's distribution through the enneazones can be represented by a curve, in which the I-value (or, more accurately, a percentage derived from the I-value) is plotted with respect to the nine zones. If we wish to compare two MBTI types and their distribution, the two curves could be superimposed on each other, as in the following two examples:

Now, in the above example, one can see that the distribution of the first two MBTI types (ISFP and INFP) appear to resemble each other, and the distribution of the latter two (ISFP and ENTJ) are more dissimilar.

How can the similarity of distribution be quantified, so that all 16 types can be numerically compared with each other? [There are 128 possible comparisons to be made!] A simple way of doing this is to calculate the distance (the 'spread') between the two lines at each of the nine points [Z1 through Z9] and average them. This will result in a number that we call the 'average spread' or, more simply, 'Sp'. The number can range from zero upwards. In the case in which two MBTI types have identical I-values in each enneazone, and would appear, if graphed like above, as one line (since the two identical curves are, in effect, perfectly superimposed) - the Sp-value would be zero.

As one would expect, the MBTI types that constitute Jungian pairs will usually have relatively low Sp-values. One would expect that the ENTJ and ESTJ, for instance, since they have a similar preference pattern [T-N-S-F and T-S-N-F] would frequent the same zones. And in this sample, the Sp-value for ENTJ/ESTJ is 5.25, which is very low. In comparison, one would expect 'shadow types' (eg, ENTJ/ESTJ) to have higher Sp-values, in comparison, and in this data sample they usually do. For instance:

MBTI PAIR	Sp-value
ENTJ/ESTJ	5.25
ENTJ/ISFP	18.39
ESTJ/INFP	13.80

We calculated the Sp-values for all 128 MBTI pairs, and analyzed them. We discovering that for each of 12 MBTI types the corresponding 'Jungian' pair was amongst the three lowest Sp-values for that type. For the ENTJ, the lowest Sp-value is 5.25, for the ENTJ/ESTJ pair (the 'Jungian' pair). In other words, the MBTI type with which the ENTJ has the most similar pattern of distribution across the enneagram is the ESTJ, its 'Jungian' partner. It turns out, by the way, that the 12 MBTI types that have a low Sp-value for their Jungian pairs are the same 12 MBTI types that have their highest I-values in the zones to which we assigned them as prototypical!

But some unexpected patterns also appeared. For six out of the sixteen MBTI types, the MBTI pair with the lowest Sp-value was one in which the dominant and inferior functions were switched, everything else remaining the same: out of all of the MBTI types, ESFP prefered to be with ENP, and vice versa. Similarly with the ESTP, ENTP, INTJ, and ISTJ.

Why would this occur? Why would the way that the ISTJ is distributing through the enneazones parallel the way the INTJ is distributing? This is an especially curious phenomenon in that the INTJ has 'intuition' as its dominant function and the ISTJ has 'sensing' (the opposite of intuition) as its dominant function. One would not expect the ISTJ and INTJ to 'pal up' in this way and frequent the same zones.

Indeed, if an ISTJ individual were experiencing an inner struggle in which N and S, as polar opposites, were actively vying for the first place in that individual's preference order, one could only imagine what great tension this might create in that individual - as replacing the 'S' in the ISTJ with 'N' is tantamount to making that individual's inferior function their dominant function and vice versa.

Enneagram theorist Larry Gabbard calls this phenomenon a 'dominant-function struggle' , and, offering an MBTI-enneagram theory alternative to ours ⁴, hypothesizes that this (and not the inferior function) is the issue around which MBTI types cluster in certain enneazones (7, for instance). Although, in the end, we do not agree with all of his specific conclusions, we credit Larry with bringing this critical concept of a 'dominant-function struggle' clearly into relief for us, and with being the first to point out that the phenomenon of the 'dominant-function struggle' seems to characterize certain enneazones (more about this below). Incidentally, we believe that the character of Colonel Slader in the movie 'Scent of a Woman' vividly illustrates a 'dominant-function struggle' in an individual. Slader, an ISTJ, is suddenly forced to 'become' an INTJ, when his sensory functioning (his sight) is suddenly taken from him. We would take the ISTJ /INTJ struggle as characteristic of enneazone 6, however, not zone 4.

Although Flautt correctly registers the presence, in high concentrations, of both the ESFP and ENFP in zone 7, he did not seem to appreciate that they constituted a 'dominant-function struggle' pair, or notice that their is a tendency toward such pairs in the distribution of MBTI types across the enneagram.

In contrast, Larry has hypothesized that certain zones will actually attract certain individuals who indeed are personally undergoing a 'dominant-function' struggle (for instance, persons who might best be described as EXFPs). Although the zones characterized by such pairs may attract individuals with dominant-function struggles, we would argue that those zones also attract 'pure' types - ESFPs and ENFPs, and that their attraction of EXFPs is an artifact of the failure to discriminate between S and N in the enneagram testing process. Although the narrative profiles for zone 7, for instance, seem to contradict themselves (from an MBTI point of view) by implying that 7s are highly refined sensory types but also subject to distinctly inferior sensate functioning (gluttony), this does not mean that only EXFPs will be attracted there - both ENFPs and ESFPs will also be attracted (as the EM data shows).

Looking more closely at the Sp-value data for the EM survey, it becomes apparent that it is not all dominant-function-struggle pairs that have low Sp-values - only those with a dominant-function S-N struggle! The INFP/INTP pair, for instance does not have a low Sp-value. Indeed, whereas six of the eight MBTI types that have S or N as their dominant function register their absolutely lowest 'Sp-value' with their 'dominant-function opposite' (eg, ESFP/ENFP), of the eight MBTI types that have either T or F as their dominant function, none have their lowest 'sp-value' with their dominant-function opposite. In fact, none come even close ! ⁵

Also, it is not the case that other 'opposite' arrangements (such as the three types of possible 'opposite' that we previously described as pertaining between MBTI types) characteristically draw down low Sp-values.

So why is it that the Sp-values for the S-N dominant-function-struggle pairs are the only group with unexpectedly low Sp-values? One possible explanation is simply that S and N are not being discriminated in the ennea-testing processes. This would not be surprising - in our first series of articles we pointed out that ennea-theory appears to distinguish only three functions - clearly discriminating thinking and feeling, and conflating S and N together into a (third) 'other' function (or 'center'). [Click here to see that argument].

Section Four: The Consequences of 'S-N' Blindness

If ennea-theory assumptions regarding the functions are 'blind' to the S-N distinction, this would be tantamount to what (from the MBTI perspective) would look like the collapsing of the 16 MBTI types into the following 8 combination types, by failing to distinguish between the members of each pair. While four of these pairs are identical to Jungian types, four are not. The eight types are:

INFJ+ISFJ=IXFJ
INTJ+ISTJ=IXTJ
ENFJ+ESFJ=EXFJ	or EF (Jungian Extraverted Feeling Type)
ENTJ+ESTJ=EXTJ	or ET (Jungian Extraverted Thinking Type)
INFP+ISFP=IXFP	or IF (Jungian Introverted Feeling Type)
INTP+ISTP=IXTP	or IT (Jungian Introverted Thinking Type)
ENFP+ESFP=EXFP
ENTP+ESTP=EXTP

We can create a new 'data-lense' by combining MBTI cells according to the above pairs, and calculating I-values for each pair in each enneazone, the following chart results. We have placed an asterisk next to the highest I-value in each column.

         IXFJ    IXTJ    IXFP    IXTP    EXFP    EXTP    EXFJ    EXTJ

                         (=IF)   (=IT)                   (=EF)   (=ET)

ZONE 2   0.95    0.00    0.84    0.42    1.99    0.65    2.96*   0.00
ZONE 3   0.84    0.57    0.15    0.00    1.31    3.59*   2.36    1.37
ZONE 4   1.83*   0.50    2.01    0.20    0.95    0.00    0.52    0.00
ZONE 5   0.79    2.24*   0.72    3.86*   0.00    0.34    0.00    0.64
ZONE 6   0.87    1.02    1.47    0.82    0.91    0.43    0.88    0.82
ZONE 7   0.39    0.36    0.14    1.29    3.90*   1.35    0.83    0.32
ZONE 8   0.15    0.99    0.11    0.00    1.61    2.12    0.44    4.30*
ZONE 9   0.86    0.69    2.06*   0.94    0.34    1.11    0.92    0.53 
ZONE 1   1.41    1.52    0.43    0.54    0.35    1.14    1.41    1.50