โ† Blog
PersonalitySeptember 20, 2026By Johnson

Nurturer, Defender, Protector: Why Your Type Has a Different Name on Every Site

One set of four letters carries a different noun on every site that reports it, and the nouns come from three unrelated naming systems. Where each one came from, and which part of a result is worth keeping.

A result comes back as the Nurturer. Two years later, on a different site, the same person answers a similar set of questions and comes back as the Defender. Nothing in the profile looks familiar enough to be reassuring or different enough to explain itself.

Both readings are ISFJ. Nothing changed except the noun.

16

four-letter codes in every version of the system

3

naming sets in wide circulation, only two of which have a published author

0

nicknames in the documentation of the instrument the codes came from

5

personality aspects on the largest free site, one more than the tradition it borrowed the letters from

Where the names actually come from

Isabel Briggs Myers and Katharine Briggs built a questionnaire that reports four letters. The published type descriptions attached to that instrument use the letters and a paragraph. No Nurturer, no Defender, no Architect. Whatever else is said about the four-letter tradition, the nicknames are not part of it.

They arrived later, from two people who were building something else.

David Keirsey's Please Understand Me II organised the sixteen codes into four temperaments, called Guardian, Artisan, Idealist and Rational, and gave each code inside them a role name. ISFJ became the Protector. INTJ became the Mastermind. Keirsey was explicit that his temperaments came from an older tradition of observable behaviour rather than from Jung's functions, so his names describe a different theory that happens to be indexed by the same letters.

The second set is newer and much more widely read. The largest free personality site rebuilt its instrument on five-factor trait research rather than on cognitive functions, kept the four-letter shorthand because everyone recognises it, added a fifth letter of its own, and named the sixteen results itself. ISFJ became the Defender. INTJ became the Architect.

The third set has no author at all. Nurturer, Thinker, Giver, Idealist and a dozen others circulate across free test sites, get copied into listicles, and appear in search results with no publication behind them. That is why a search for a nurturer personality returns confident descriptions that agree with each other in tone and disagree about almost everything else.

The word that proves the problem

Keirsey's Architect is INTP.

The most-read free site's Architect is INTJ.

One word, two codes, both published, both in current use. Somebody who reads a Keirsey profile of the Architect and then takes the free test and comes back INTJ has just been handed the same name for a different result, and there is nothing in either page to warn them.

Four lettersKeirsey's role nameThe largest free siteThe original instrument
ISFJProtectorDefenderno nickname, letters only
INTJMastermindArchitectno nickname, letters only
INTPArchitectLogicianno nickname, letters only
ENFPChampionCampaignerno nickname, letters only
INFJCounselorAdvocateno nickname, letters only

Read down the Architect row and the point lands. Read across it and the point lands harder.

Why Thinker returns nothing consistent

Thinker is not in Keirsey's list and not in the free site's list. It is not in the instrument's documentation either. It is a word that fits several types loosely and belongs to none of them, which is exactly the condition under which search traffic scatters across a dozen pages that each assign it somewhere different.

The nearest thing to a home for it is INTP, whose published names are Architect in one set and Logician in the other, and who gets called the Thinker by sites working without either. Enneagram material uses the same word for its fifth type, which is a different framework measuring a different thing, and one whose own labels are less stable than they look: two free Enneagram tests will often hand the same person two different numbers, for a reason that has nothing to do with either test being badly made.

So a person searching for the thinker personality type is not searching for a type. They are searching for a word somebody once used in front of them, and the honest answer involves asking where they read it.

The name is the part that persuades and the part that was never measured

There is a reason the nouns spread faster than the letters. A code is four capital letters with nothing to hold onto. Defender is a story about a person, and the story arrives feeling true before any evidence is offered for it.

Forer demonstrated the mechanism in 1949 by handing a class of students an identical personality sketch assembled from a newsstand astrology column and asking each of them to rate how well it described them personally. The average rating was 4.26 out of 5. The sketch fitted everyone because it was written to fit everyone, which is a property that type nicknames share by construction. A name chosen to make a profile feel recognisable is optimised for exactly the wrong thing.

None of this makes the underlying letters worthless. McCrae and Costa found in 1989 that the four dimensions map onto four of the five factors in mainstream trait research, which is a real finding and better news for the system than most of its critics allow. Pittenger's review reached the awkward middle position that the dimensions hold up while the categorical claim does not. The measured part survives. The noun on top of it was never in the measurement.

A different name
A different result
Two sites, same four letters, different nouns
Two sittings, different letters
The profile disagrees in vocabulary and agrees in substance
The profile disagrees about the direction of a dimension
One of the sites uses a different model underneath
Both used the same instrument
The change appeared the moment the site changed
The change appeared after something in life changed
Nothing to explain
Worth reading, and usually worth a percentile rather than a letter

What to keep out of a result

Three things, in descending order of how long they stay true.

The dimensional score is the part with the most information in it, because it says how far from the middle a person sits rather than which side of a cut point they landed on. A score of 52 against 48 produces a capital letter with the same confidence as 89 against 11, and the two situations are not alike. The reason a result moves between sittings is usually that a near-middle position got rounded twice in different directions.

The four letters are the part that travels. They survive the trip from one site to another, they index the same cell of the same grid everywhere, and they let two people who took different tests compare something.

The nickname is the part that does not survive anything. It changes with the site, it carries a theory the reader has not been told about, and in at least one documented case it points at two different types depending on who wrote the page.

The question underneath a search for the nurturer personality is rarely about vocabulary. It is someone checking whether they are still the thing they were told they were. The letters answer that. The names were never in a position to.

Sources

Frequently asked questions

Can I convert my Enneagram number into a four-letter type?

Not reliably, because the two systems are not measuring the same thing. The Enneagram sorts by core motivation and characteristic fear, while the four-letter systems sort by stated preference for where attention goes. Correlations between the two exist in the published literature but they are modest and none of them supports a clean lookup table. Anyone offering an exact conversion is describing a pattern they noticed rather than a measured relationship.

What does the fifth letter mean in a result like INFJ-T?

It comes from one site's own model rather than from the four-letter tradition, and it marks what that site calls Identity, split into Assertive and Turbulent. Turbulent scores describe more self-consciousness and more sensitivity to stress, which in five-factor terms sits close to neuroticism. Because no other naming set includes it, an INFJ-T and an INFJ from another test are not two grades of the same result.

Are groups like Guardian, Diplomat and Sentinel measuring something extra?

No, they are ways of bundling the same sixteen codes into four families, and different authors bundle them differently. Keirsey's four temperaments group by sensing and intuition crossed with other letters, and the group names carry a theory of temperament that predates the questionnaire. A group label adds vocabulary and a sense of belonging. It does not add a measurement.

Which version should I give when someone asks for my type?

The four letters, because they are the only part that survives the trip between systems. A colleague who hears Defender will look it up and may land on a profile written by a site that uses a different model underneath, while four letters point at the same cell of the same grid everywhere. Adding the name of the test that produced the result makes it more useful still.

Can an employer use a four-letter type to decide who gets hired?

The instrument's own publisher advises against it, and that guidance has been in place for decades. The objection is not that the result is meaningless but that a preference measure was never validated as a predictor of job performance, so using one as a filter puts weight on a number that was not built to carry it. Development conversations and team vocabulary are the intended uses.

See this pattern in your own numbers

โœฆ

This article, about your situation

Luna has read the same research โ€” and, if you let her, your results.

โœฆ Talk it through

More notes on people