The Enneagram: where it came from and what research says

The figure and the types have different authors

The figure is older than the types. George Gurdjieff, a spiritual teacher who died in 1949, was the first to show it publicly in the West, and it’s the only known source for it. He probably showed it to students in Moscow and Petrograd in 1916, and later used it in the sacred dances he taught.

What he said about it was written down by his student P. D. Ouspensky in In Search of the Miraculous (1949). Gurdjieff called the enneagram a universal symbol and used it to explain two laws. The triangle joining points 9, 3 and 6 stands for the law of three. The jagged line 1-4-2-8-5-7 stands for the law of seven, and it comes from dividing 1 by 7, which gives 0.142857 repeating; the digits 3, 6 and 9 never appear in it.

There are no personality types in that text. Gurdjieff talks about levels of people, “man number 1”, “man number 2” and so on. He also says that knowledge is only what a person can put into the enneagram, and that the figure had long been kept secret. The words “Sufi” and “ancient” aren’t in the chapter.

The types came about fifty years later. Oscar Ichazo (1931–2020), who was born in Bolivia, built a system of nine “ego-fixations.” He first called the figure an “enneagon,” and the term “Enneagram of Personality” is his.

The Arica School dates from 1968, when he lectured on his theory in Santiago, and in 1970–71 he ran intensive training in the Chilean town of Arica. Arica Institute was registered in New York in 1971.

A lawsuit decided in 1992 shows how tangled the authorship is. Arica Institute had sued the writer Helen Palmer and her publisher over her book on the Enneagram. Palmer said she had studied with Claudio Naranjo from 1973, had never trained at Arica, and that Gurdjieff’s version of the figure was a bare star with no words on it.

The federal court found that Ichazo hadn’t created the nine-point figure. It called his way of assigning fixations to points minimally creative, and so protectable in principle, though the wording and the order of the fixations weren’t. Arica itself described the fixations as discovered, not invented, and the court treated them as facts, which copyright doesn’t cover. Palmer’s use was fair use and the suit was dismissed.

The ancient-Sufi story is easy to trace in the schools’ own words. The Enneagram Institute, founded by Don Riso and Russell Hudson, says the system of types goes back only to the 1960s, when Ichazo began teaching it, and calls crediting Gurdjieff or the Sufis a mistake of early enthusiasts.

Riso and Hudson’s own book (revised edition, 2000) says there is no body of Enneagram material that was handed down as a ready-made oral tradition. They describe the modern system as a synthesis of Ichazo, Naranjo and others, themselves included. Ichazo has been quoted as saying he never met a Sufi who knew the enneagram.

The Sufi line still has defenders. A 2024 article in Spirituality Studies argues that the figure reached the West through the Naqshbandi order and that Gurdjieff heard of it from a sheikh. The 2021 review discussed below mentions experts who trace the beginning to a Babylonian or Near Eastern oral tradition around 2500 BCE. Documents in support of these versions aren’t cited.

A different thread goes back to Evagrius of Pontus, a fourth-century monk who described eight “thoughts” that grow out of self-love. Their themes resemble the later list of passions, and Wiltse and Palmer, both Enneagram teachers, suggest that clues to the figure’s origin lie in Evagrius and in earlier texts. They write that answers aren’t there yet and call their paper a beginning of the search. A resemblance in subject is all that’s established.

The Church has also weighed in, on the religious side only. In 2003 two pontifical councils, writing about the “New Age,” named the Enneagram, when it’s used as a path of spiritual growth, as an example that introduces ambiguity into the teaching and life of the Christian faith.

How the Enneagram reached psychology and the public

Claudio Naranjo (1932–2019), a Chilean psychiatrist, studied with Ichazo around 1970 and then taught his own version, mostly in Berkeley and at Esalen. He tied the types to Western psychological ideas and wrote several books on them: Ennea-type Structures (1990), Character and Neurosis (1994) and Enneatypes in Psychotherapy (1995).

Ichazo objected to what Naranjo and his students did with the system. Arica made copyright claims against some of those students for not crediting him. Meanwhile Jesuits among Naranjo’s students, Robert Ochs among them, had been teaching the Enneagram at Loyola University Chicago since 1971, and Ichazo objected to that adaptation too.

Don Riso, then a Jesuit seminarian, met the system in Toronto in 1974 in the form of nine one-page sketches of the types. He left the order in 1975, worked out levels of development for each type in 1977, and published Personality Types in 1987.

Hudson joined him in 1991 to build a questionnaire. It became the RHETI, which gives a person 144 pairs of statements and picks the type with the highest score. The Enneagram Institute opened in New York in 1995.

Other writers had reached readers earlier: Palmer’s book came out in 1988, and in 1994 Stanford hosted the first international Enneagram conference, with about 1,400 people attending, support from its psychiatry department, and Palmer and David Daniels as co-chairs. Daniels later co-wrote a different kind of test: the reader goes through nine paragraphs, one per type, and chooses the one that describes them best.

The schools don’t even agree on what to call the types. Riso and Hudson name type 5 the Investigator and type 8 the Challenger; other sources say Observer for 5 and Protector for 8, Enthusiast or Epicure for 7, Loyalist or Loyal Skeptic for 6. A reader who looks up a type in two books may meet two names. The number stays the same.

How it’s built: types, centers, wings and lines

The system has three layers, and they differ in how well they’re documented. The first is Gurdjieff’s: nine points on a circle, a triangle through 3, 6 and 9, and the jagged line 1-4-2-8-5-7-1. The second is the numbering with its types, which comes from Ichazo and the schools after him. The third, wings and lines, is the one with the thinnest evidence.

Each type is described through a basic desire, a basic fear and a characteristic passion. For Type 2 the desire is to be needed and the fear is to be unloved. The numbers don’t rank anything; Type 9 isn’t better than Type 1. The types are also grouped into three centers. In Riso and Hudson’s version, 8-9-1 is the instinctive center (tied to anger), 2-3-4 the feeling center (shame) and 5-6-7 the thinking center (fear). Helen Palmer calls them body, heart and head, and the 2021 review notes that teachers don’t group the triads the same way.

Wings are the two types next to a person’s main type on the circle. A person with main type 6 who also draws on 7 is written 6w7. Riso and Hudson say most people lean on one neighbor, and some schools don’t use the idea at all.

Lines connect each type to two others. In Riso and Hudson’s version Type 1 moves toward 7 in growth and toward 4 under stress. Teachers disagree about what that means: some say a person takes on the unhealthy traits of the other type under stress and the healthy ones in growth, others say either can show up. Naranjo and some other teachers rejected the lines or changed them.

The description of each type is on the Enneagram system page.

What the studies found

The most complete summary is a systematic review published in 2021 by Hook and colleagues in the Journal of Clinical Psychology. It found 104 independent samples. Only 49 were published; the rest were mostly doctoral dissertations. Of the 40 peer-reviewed publications, only nine appeared in mainstream psychology journals. Most studies were questionnaires given once, and only a handful were experiments. The authors’ verdict: the data on reliability and validity are mixed, and the field is at an early stage.

Some scales show the expected relationships with the Big Five traits, and several papers found that the Enneagram helped people with personal or spiritual growth.

Other results are less favorable. Factor analysis usually produces fewer than nine factors, and no study has found nine types through cluster analysis. The skeptic’s reference work, the Skeptic’s Dictionary (Carroll, updated 2015), goes further and puts the system with astrology, saying there’s no way to test it. In a 2006 poll of American psychologists by Norcross, Koocher and Garofalo, the Enneagram was among the methods that a sizable share of the experts called discredited.

The RHETI shows why the tests are hard to evaluate. Every item is a forced choice. Take a pair like this: “I tend to take charge when things get disorganized” against “I tend to make sure everyone feels included.” A person might do both, but one has to be picked, and the other goes unscored.

The total score is the same for everyone, so a high score for one type means a lower one for another. Scores of this kind can be compared inside one person, but not between people, and the review’s authors link the RHETI’s low internal consistency partly to this format. Its scales’ reliability coefficients ran from 0.35 to 0.84 depending on the study, while two studies of a version that lets people rate each statement separately found every scale above 0.70.

Two further results come from the same format. In one study only 48% of people were clearly sorted by the RHETI, meaning their top score beat the second by at least three points. Everyone else got a type, but a neighboring one was nearly as close. In another, people who took both the RHETI and a second Enneagram test, the Wagner scales, ended up with the same type 42% of the time.

The claim about wings has been checked once. In 1991 Edwards gave participants descriptions of the nine types and asked them to sort the descriptions by similarity. Riso had predicted that neighboring numbers would end up together more often, and they didn’t land together more often than chance would give. The review’s authors call that design debatable and suggest a better one: find a person’s main type, then see whether the scores for one neighbor run higher.

Movement between types has been tested twice, in 1994 and 1995, by comparing anxiety scores with scores for the “stress” type. Neither found the expected link, and both measured people at one moment only, which is a weak way to test change over time. The review’s verdict: no support yet, but with one study on wings and two on movement, there’s very little to go on either way.

Picking your own type

Many Enneagram tests end by asking people to read descriptions and decide which one fits. In an unpublished study from 1995, reported in the 2021 review, participants were given detailed descriptions, and in 87% of cases the type they chose matched the type from the RHETI. The review doesn’t give the sample size. A match also doesn’t confirm that either answer is correct, because both are people describing themselves.

The test with nine paragraphs gives a clearer picture of how stable such choices are. Retested after four weeks, people picked the same paragraph with a moderate level of agreement (average kappa 0.59). Agreement between the paragraph they chose and the type later found through an interview or a course was lower, 0.53.

Interview-based typing doesn’t help much either: in a 1986 study, agreement between judges was low (kappa 0.20, and 0.25 for the experienced ones), and the review rates the evidence for interview typing as weak.

Descriptions pull people in. In the late 1940s the psychologist Forer gave his students the same text, assembled from an astrology book, as their personal portrait, and most rated it as accurate. An Enneagram description can work similarly, since it lists a desire, a fear and a habit that most readers can find some trace of in themselves. The same question comes up for any questionnaire where people answer about themselves, and the article on whether the 16 types are real looks at it for another system.

When the type comes from descriptions a person reads, the result compares a text with that person’s own picture of themselves.

There’s another way to look for a type. CoreTypes starts from the idea that an answer to a question about yourself shows how you see yourself, while the type itself comes through in the meaning of what you say: what you notice first and what you keep coming back to without being asked. The free CoreTypes test is built on the same idea. It has 36 questions, each with four answers: sets of words, ready-made phrases or a short scene. It takes about ten minutes and doesn’t ask you to sign up, and the result is one of 16 types with a free type page. Take the CoreTypes test.

How to use it and how to read a result

The Enneagram’s own strength is its language of motives. A type is described through what a person wants and fears rather than through a list of behaviors, and the review quotes teachers as saying the point is understanding inner motives, often unconscious ones, more than changing behavior. In the review’s example, Type 2 wants to be loved, adjusts to whatever others need and ends up forgetting its own needs. Most teachers say the point is to build what the review calls compassionate self-awareness, a step from repeating rigid patterns toward flexibility. The types are descriptive, and none is better or worse.

That suggests a habit: ask “what am I after?” instead of “what did I do?” Take someone who agrees to cover a colleague’s report on a Friday evening, though the weekend was already planned. The behavior is easy to describe, and a Type 2 description would point at a motive: staying needed, or avoiding the thought of being unwelcome. Another person might say yes for an entirely different reason, and that would point to a different type. The same act can have different reasons, and the reason is what the system tries to name.

A test result works best as a hypothesis. Given how often neighboring types come close in one test, and how often two tests of the same system disagree, it makes sense to read the description for the type that came out and the ones on each side of it. Wings are a hypothesis too, because the one study on them found nothing.

It also helps to check a result against something measured differently. In the review, Type 2 scores on average high in agreeableness and conscientiousness in the Big Five, and one study linked Types 2 and 4 to anxious attachment. A description of a type can be held up against measures like these.

The description of each type and the tests that use the Enneagram are collected on the Enneagram system page. A simple practice for a week: pick the two descriptions that fit best, note one moment each day when a reason stood behind an action, and see which description holds up.