A relaxation group is a set of four types built from two dual dyads, chosen so that all four lead with one of the same two valued elements. Reinin called the structure a square, from the Russian квадрат, and the arrangement is sometimes described as a double duality. The name this page leads with is the one attached to the claim that made the grouping interesting: that four people assembled this way relax when put in a room together.
Square is not quadra, although quadra literally means square and the two are constantly conflated. A quadra is four types who share one complete set of valued elements and who all sit inside that single quadra. A square group straddles two quadras, and its members share only half their valued elements. The two groupings overlap in one respect only: both are built from two dual dyads.
It is also one of three. Join a dual dyad to a dual dyad of the same rationality from any other quadra and you get this grouping, the blocking group, or a third that has no page of its own. All three have everybody's dual present. They are set out further down.
The four relaxation groups
The Judicious Irrationals
Two dual dyads that value openness and comfort, moving at whatever pace the situation sets. Shared valued elements: Ne and Si. The relations inside this group are Dual, Kindred and Semi-dual.
| Type | Code | Title |
|---|---|---|
| ILE | ENTp | The Searcher |
| SEI | ISFp | The Mediator |
| SLI | ISTp | The Craftsman |
| IEE | ENFp | The Psychologist |
The Subjectivist Rationals
Two dual dyads that value an agreed framework and a shared mood, and expect a room to have both. Shared valued elements: Ti and Fe. The relations inside this group are Dual, Kindred and Semi-dual.
| Type | Code | Title |
|---|---|---|
| LII | INTj | The Analyst |
| ESE | ESFj | The Enthusiast |
| LSI | ISTj | The Inspector |
| EIE | ENFj | The Actor |
The Decisive Irrationals
Two dual dyads that value force and timing, reading the moment and moving on it. Shared valued elements: Se and Ni. The relations inside this group are Dual, Kindred and Semi-dual.
| Type | Code | Title |
|---|---|---|
| SLE | ESTp | The Marshal |
| IEI | INFp | The Romantic |
| ILI | INTp | The Critic |
| SEE | ESFp | The Ambassador |
The Objectivist Rationals
Two dual dyads that value what works and who can be trusted, and judge a room by both. Shared valued elements: Te and Fi. The relations inside this group are Dual, Kindred and Semi-dual.
| Type | Code | Title |
|---|---|---|
| LIE | ENTj | The Pioneer |
| ESI | ISFj | The Guardian |
| LSE | ESTj | The Director |
| EII | INFj | The Humanist |
How the groups are derived
Start from the leading function. An irrational type leads with a perceiving element and a rational type with a judging one, and each element is valued alongside one partner: Ne with Si, Se with Ni, Ti with Fe, Te with Fi. A square is the four types who lead with one of those pairs.
That gives the Judicious Irrationals (Ne or Si, from Alpha and Delta), the Decisive Irrationals (Se or Ni, from Beta and Gamma), the Subjectivist Rationals (Ti or Fe, from Alpha and Beta) and the Objectivist Rationals (Te or Fi, from Gamma and Delta). Every member leads with a value the other three share. That is a narrower thing to have in common than a quadra, and it is doing most of the work.
It is not the intersection of two dichotomies. The irrational squares split on Judicious/Decisive and the rational squares on Subjectivist/Objectivist, and no single dichotomy does both. That is why the grouping is not one of the 35 dichotomy systems.
The membership also falls out of the relations themselves. Take any type, find its dual, then its kindred, and the kindred's dual completes the four. Start from any of the four and you get the same group, which is what makes this a real structure rather than a list.
Why the structure produces ease
Four things are true of every relaxation group.
Every member has their dual present. That alone distinguishes the arrangement from almost any other group of four, and it means nobody in the room is short of the support they are least able to supply themselves.
Every relation is symmetric. There is no supervision and no benefit anywhere in a square group, which are the two relation families that install a status gradient. Nobody is structurally above or below anybody. Whatever else is happening, the group has no pecking order built into it. Note that the four members are four different types, so identity never arises between them: it is the relation each member has to themselves, and it is not one of the six pairs described below.
None of the difficult relations appear. No conflict, no super-ego, no extinguishment, no quasi-identity. The four hardest relations in the system are all absent by construction.
Every pair shares at least half its valued elements. The minimum overlap across all six pairs in all four groups is two elements out of four, because both dyads sit in quadras that share a perceiving axis or a judging one.
Where Reinin's rational list goes wrong
Reinin gives the relations inside a square as duality, semi-duality and kindred, and lists two squares: ILE, SEI, SLI and IEE, and LSE, EII, ESE and LII. The first matches his formula. The second doesn't. LSE, EII, ESE and LII are related by duality, business and mirage. They lead with Te, Fi, Fe and Ti, so no two of them share a leading function, and without that there is no kindred.
Kindred means the same leading function. LII's kindred is LSI and its semi-dual is EIE, both in Beta. So the rational square that matches Reinin's formula is LII, ESE, LSI and EIE, which is also how Wikisocion defines the group: four types who lead with the same pair of quadra values. The grouping and its description are Reinin's. The list on this page is corrected, and an earlier version of it followed his.
Thanks to a reader who pointed out the rational/irrational distinction.
Three ways to join two dual pairs
Take a dual pair and join it to a dual pair of the same rationality from one of the other three quadras. That gives three possible groups, and for two of them the relations inside depend on rationality. From Alpha:
| Alpha pair joins | Irrationals (ILE, SEI) | Rationals (LII, ESE) |
|---|---|---|
| Delta | + IEE, SLI: kindred, semi-dual | + LSE, EII: business, mirage |
| Beta | + SLE, IEI: business, mirage | + LSI, EIE: kindred, semi-dual |
| Gamma | + SEE, ILI: extinguishment, super-ego | + LIE, ESI: extinguishment, super-ego |
Across all four quadras the pattern is the same:
| Relations across the two pairs | Irrational pairings | Rational pairings | Grouping |
|---|---|---|---|
| Kindred, Semi-dual | Alpha–Delta, Beta–Gamma | Alpha–Beta, Gamma–Delta | Relaxation (this page) |
| Business, Mirage | Alpha–Beta, Gamma–Delta | Alpha–Delta, Beta–Gamma | No page |
| Extinguishment, Super-ego | Alpha–Gamma, Beta–Delta | Alpha–Gamma, Beta–Delta | Blocking |
What decides the pairing is shared values. Beta and Delta both sit next to Alpha, but Alpha and Delta share Ne and Si, and Alpha and Beta share Ti and Fe. An irrational Alpha type leads with Ne or Si, so its kindred and semi-dual are in Delta. A rational Alpha type leads with Ti or Fe, so they are in Beta.
Blocking is the one member of the family that doesn't change with rationality: Alpha joins Gamma and Beta joins Delta on both sides of the line, and the relations are always super-ego and extinguishment. The business and mirage grouping is the square's rationality mirror. It takes the square's rational pairings for its irrationals and the square's irrational pairings for its rationals. I haven't found a name for it in the literature, so I am not giving it one.
That difference matters, because the two profiles are not interchangeable. A semi-dual gives partial dual support and a kindred shares your leading function aimed at different values. A mirage partner is comfortable in a way that never quite resolves, and a business partner offers competence without much warmth. Both are easy. They are not easy in the same way, and only one of them is Reinin's square.
What Reinin claimed, and where we disagree
Reinin reports in Lecture 8 of his socionics course, on small groups, that these groups relax spontaneously when assembled, that they suit stress relief and meditation, and, to his credit, states openly that the mechanism is unknown: he calls the effect a medical fact and declines to speculate about why that formula produces it.
The following is our analysis rather than classical Socionics, and it is offered as a prediction that could turn out to be wrong.
We think the claim is strongest for the Judicious Irrationals and weakest for the Decisive Irrationals. With the membership corrected, all four squares hold the same relations, so the relation profile can't be what separates them. What separates them is the pair of values they share.
The irrational squares share a perceiving pair. Judicious groups share Ne and Si, which makes rest itself a valued state: an unstructured hour is what those four types would have chosen anyway. Decisive groups share Se and Ni, which pushes toward mobilisation. Put four Decisive types in a room with nothing to do and the shared instinct is not to unwind but to find the thing that needs doing. Whatever happens there, calling it relaxation is a stretch.
The rational squares share a judging pair instead, and each holds one Judicious dyad and one Decisive dyad, so on stance they sit between the two irrational squares. The Subjectivist Rationals share Ti and Fe: an agreed framework and a mood everyone is in together. The Objectivist Rationals share Te and Fi: what works and who can be trusted. Neither is a value about rest, so we expect both to settle more slowly than the Judicious Irrationals and more readily than the Decisive ones.
This is falsifiable. Assemble each of the four groups, give them an unstructured hour, and measure afterwards. If Reinin is right, all four report roughly the same thing. If we are right, there is an ordering, with the Judicious Irrationals at the top and the Decisive Irrationals at the bottom.
Practical use
The practical question a square group answers is not who will enjoy each other most. It is who can be in a room together for a long time without anybody having to manage it.
Use one when the stakes need to be low: a long journey, a recovery period, a group that has to share a space without a task to organise them. Because there is no status gradient, nobody spends energy on position, and because every member has their dual present, nobody is running on empty in the area they are weakest.
Do not reach for one when you want output. That is what a quadra is for. A quadra gives shared values and shared energy, and it will generate more in an afternoon than a square group will in a week. A square group gives symmetry and low stakes, which is a different good and usually the wrong one if something has to get built.
If you are unsure which of these your own type belongs to, the small group finder will place you from three questions.