Four-box personality models keep getting invented. Jung did a version in 1921, Marston another in 1928, and the North-South-East-West wording most people recognise today was published in 1998.
None of them borrowed from the others in any straightforward way, which is the interesting part. The same four-way cut keeps reappearing because it sits exactly where a model is still small enough to remember and already big enough to say something.
The recurrence is not evidence that the four types are real. It is evidence that four is roughly the number of categories a human being can hold in working memory while arguing with a colleague.
Before Psychology: The Four-Element Habit
Long before anyone measured a trait, the four-way split was already the default way to describe people. The classical humours — sanguine, choleric, melancholic, phlegmatic — assigned temperament to bodily fluids, and the scheme survived in European medicine for well over a millennium on essentially no evidence.
The mechanism was nonsense. The observation underneath it was not entirely: people really do differ in baseline energy, in how quickly they anger, in how much they brood. What the humours got wrong was the cause and the hard boundaries between categories, and those are the same two things every later four-type model has had to defend.
1921: Jung Turns It Into a Structure
Carl Jung's Psychologische Typen is the hinge. Jung proposed that people differ along an attitude (introversion or extraversion) and in which mental function they lead with — thinking, feeling, sensation or intuition. Crucially, he crossed the dimensions rather than listing types, which is what turns a taxonomy into a map.
Jung himself was careful in a way his popularisers were not. He described the types as tendencies with enormous individual variation, not as boxes people belong to. Most of the damage done by four-type models since has come from dropping that caveat.
1928: Marston and the Behavioural Quadrant
William Moulton Marston's Emotions of Normal People introduced the four-way behavioural scheme that became DISC — dominance, inducement, submission, compliance in the original wording. Marston was not building an assessment; the instruments came later, from other people.
What DISC contributed, and what the compass inherits, is the workplace framing. Marston's four were about how a person acts on their environment when they perceive it as favourable or hostile — which is a much more testable question than "what kind of soul do you have", and much easier to talk about at work.
Jung gave the four-type family its structure. Marston gave it its job. Almost every business-facing quadrant model since is some recombination of the two.
The Counter-Movement: Traits, Not Types
While the type models were spreading through workplaces, academic personality research went the other way entirely. Factor-analytic work through the middle of the twentieth century converged on continuous dimensions rather than categories, and by the time Lewis Goldberg described the structure of phenotypic personality traits in 1993, the five-factor model was the mainstream position.
The two traditions collided directly in 1989, when McCrae and Costa reinterpreted the MBTI through the five-factor model. Their finding was double-edged, and it is still the fairest summary of the whole family:
- The dimensions are real. What the type instruments tap corresponds reasonably well to established trait factors.
- The cutting is the problem. Turning those dimensions into types throws information away and introduces instability — a person near a boundary flips categories on a small change in answers.
1998: The Compass Wording Arrives
The specific North-South-East-West framing was published by Diane Turner and Thelma Greco in The Personality Compass: A New Way to Understand People. That book is where the directional vocabulary in common use today comes from, and it is cited here for exactly that reason — as the origin of the wording.
The directional metaphor caught on because it solved a presentation problem the earlier models had. "High D, low S" means nothing to someone who has not been trained. "North" is instantly spatial: it implies an opposite, it implies neighbours, and it implies that you can be somewhere between two of them. The map does work that a letter code cannot.
Why Compass Models Keep Being Rediscovered
Every few years a four-quadrant framework is presented as new. It usually is new in its wording and genuinely not new in its structure. Three forces keep regenerating it:
- Cognitive load. Two crossed axes produce four cells, which people can learn in a minute and still use a year later. Sixteen types need a reference card; five continuous scores need a chart.
- Conversational utility. A type gives colleagues a shared word for a friction they already felt. That is a real service, independent of whether the categories carve nature at its joints.
- The middle is crowded. Because most people sit near the centre of any trait distribution, a four-box model flatters almost everyone with a plausible description — the same reason it is unstable on retest.
What the History Should Change About How You Read Your Result
Two things, both practical.
First, treat your direction as a summary of where you sit on two spectrums, not as a category you belong to. The evidence has pointed at continuous distributions for seventy years and has not wavered.
Second, take the blend seriously. The recurring weakness of every model in this lineage is the boundary case, and naming the second direction is the cheapest available fix — which is why this test reports a leading direction and the neighbouring one that colours it rather than a single winner.
You can see how that plays out in your own answers with the Arabic Personality Compass, or read how the scoring actually works and what the four directions mean.
The Bottom Line
Directional personality models are a century-old idea in a hundred costumes: Jung's crossed functions, Marston's behavioural quadrant, Turner and Greco's compass wording. They persist because four is a usable number of ideas, and they are limited for exactly the same reason. Read them as vocabulary, not as measurement, and they are genuinely useful.