Disorganised attachment: when the same person is both comfort and fear

“When she was closest to me, I needed her most. And that was exactly when I was most afraid.”

The woman who said this apologised the moment it was out of her mouth, because it struck her as illogical. It wasn’t. It described precisely what the literature calls disorganised attachment — a state in which two systems that are sequential for most people fire at once and point at the same person.

This is the most complicated and also the most misused idea in all of attachment theory. It is applied far too loosely, sold online as a personality type, and some people label themselves with it after fifteen questions in a browser. So I will be careful here: I will say what the term means, where it comes from, how it shows up in adults — and very plainly what it does not mean and what cannot be measured by a quiz. I have written separately about how attachment patterns repeat at all.

What it means to have no pattern

The first three attachment patterns are three different but working solutions. The child who shows distress directly, the child who shows it louder, and the child who mutes it — each has a plan. Each knows what to do when frightened.

Disorganisation is something else. Mary Main and Judith Solomon described it in the 1980s after reviewing recordings of children coders could not place anywhere: a movement towards the mother that stopped halfway; approaching with the back turned; a brief stillness; a sudden change of posture for no reason. These children did not have a poor plan. They had two that contradicted each other.

The phrase that has stuck for this state is fright without solution. The attachment system carries a frightened person towards someone close. The defence system carries them away from danger. If the source of fear is also the source of comfort, both routes lead to the same address and neither can be carried through. The behaviour that results looks strange because it is: it is the outward expression of an inner position that has no exit.

Where it comes from

The commonest assumption I meet is that it always means abuse. That is not accurate, and for many people it is unfair. It is true that disorganisation is more frequent where there has been severe neglect or violence. It is also true that it appears in families where nothing of the sort took place.

Researchers describe another route: a parent who is themselves frightened. If an adult has lost someone close and never worked it through, if they carry their own trauma that occasionally returns, they may become unreachable for a moment, or fear may cross their face that has nothing to do with the child. The child cannot interpret this. All they register is that something happened, and that the person who is meant to settle them is in that instant in distress themselves. If it repeats, the system does not organise.

Circumstances that are nobody’s fault also count: long hospital separations in the first year, severe postnatal depression left without help, repeated changes of carer, chronic illness in the family. All of it can create an unpredictability the child’s system reads as danger.

How it shows up in an adult

In adulthood it is rarely visible as an odd movement. It is visible as two truths that hold at once and alternate faster than a person can catch themselves.

People describe genuinely wanting a relationship and being seized by panic when closeness actually arrives. Attaching quickly and deeply, then at some point turning the whole thing over, sometimes by leaving, sometimes with an argument they do not understand themselves. Feeling like two people in a relationship: one who wants to stay and one who is already packing.

Often there is something the first three patterns do not carry: a sense that in bad moments a person steps slightly outside themselves — as if watching from a distance, or as if the world had moved back for a second. This is not madness and it is not rare. It is an old way of protecting oneself when there was nowhere to run. I wrote about what the body keeps in the post on trauma in the body.

And there is almost always shame — not about what happened, but about one’s own reactions. “No normal person behaves like this.” I wrote about that layer in the post on shame.

Why online tests do not measure this

Here I will be completely plain, because it is the most practical part of the piece.

Disorganisation in children is established by observation coded by a trained rater who has completed a course and demonstrated reliability. In adults, the related state is established through an hour-long semi-structured interview that is transcribed verbatim and then rated — and what interests the rater is not so much what the person lived through, but whether the account comes apart while they speak about hard things: whether tense shifts mid-sentence, whether the thread is lost, whether the speech becomes oddly precise in the wrong place.

None of that can be done with a self-report questionnaire, because it concerns something a person by definition does not know about themselves. An online quiz measures what you think about yourself. That procedure measures what happens to your account while you are telling it. So the label you get after fifteen statements is not the same thing — even when the name matches.

What the term is not

It is not a diagnosis. It appears in no diagnostic manual and no professional may use it as a label for a person. A group of eminent researchers in this field issued a joint statement some years ago precisely because the term had started to be used in decisions about children, where it has no business: its predictive value for an individual is far too small to build anything on. I wrote about the difference between a name and a map in the post on what a diagnosis tells you and what it does not.

Nor is it evidence that something happened. You cannot reason backwards from an adult’s pattern to events in childhood. A pattern is not a memory and it is not proof.

And it is not a permanent position. In adults the most promising marker is precisely whether the account of hard things eventually holds together — and with slow, hard work that often happens.

What actually helps in therapy

I will not promise speed, because I could not deliver it. What helps is as a rule duller than people imagine.

First, pace. With this pattern the fastest route is almost always the wrong one. If too much opens at once, the person is not relieved but flooded, and next week they do not come. So the first part of the work is not about the story but about how a person returns to the present when they are carried off.

Then predictability. The same slot, the same chair, the same length of session, a clear statement of what happens next time. This is not ritual but medicine: a system that broke down around unpredictability repairs around predictability.

Then the relationship forming in the room. That is where the pattern shows — and where it can shift, because for the first time it is possible to find out what happens when you tell someone that at this moment you are afraid of them.

And finally, patience with the fact that it will go both ways. Progress with this pattern is almost never a straight line. Withdrawal after a good session is not a relapse; it is the most expected part of the road.

What this means if you have children

The question I hear most often, and which always carries the same tension in the voice, is: will I pass this on.

The answer is more encouraging than people expect. Transmission across generations exists, but it is partial — a large share of the differences between children cannot be explained by it. Research also shows something very useful: what matters most in passing on a pattern is whether the parent can think about the child’s inner world as something of the child’s own and separate, not whether the parent had a nice childhood themselves.

In daily life that asks less than it sounds. It means a parent does not have to be calm twenty-four hours a day. They have to notice when they were carried off, stop, and afterwards tell the child something simple: that they were on edge, that it was not about them, that they are back now. Repair is what organises a child’s system — not the absence of mistakes. A child who sees that contact returns after a break learns exactly what disorganisation prevents: that there is a way out of distress, and that someone comes down it.

What does harm a child is an adult denying distress the child can plainly see. Silence around the obvious is more confusing to a child’s system than a difficulty named out loud.

Where I have to hold back

Now the part that argues against my own account.

First, the concept is reasonably well researched in children, while its adult counterpart is less clear than it sounds. The terms differ between research traditions and are not equivalent. When you read online that “disorganised attachment” is an adult type, you are reading a simplification that merges two different measurement traditions.

Second, other things can produce a similar picture. An untreated thyroid condition, long-term sleep loss, substance use, severe depression, life in precarious conditions — all of it can look like swinging between closeness and withdrawal. So it matters to me that the physical is ruled out first, and that circumstances are examined, before I explain anything by childhood.

Third, when this framework is used to explain violence in an adult relationship, it becomes harmful. “We are both disorganised” is not an explanation but a way of staying. Where there is no safety, safety is the first question.

Fourth, a lot of people recognise themselves in the description because the description is broad enough. Recognition is not a finding. If it feels as though you are reading about yourself, that is a good reason to talk to someone who works with this — I wrote about finding such a person in the post on choosing a therapist.

What to remember

Disorganised attachment is not a fourth personality type but a description of a position in which a child could not choose between going to someone and moving away from them. It does not always stem from abuse; it also arises with a frightened parent and in circumstances that are nobody’s fault. In an adult it shows as two truths at once, often with a stepping outside oneself and with shame about one’s own reactions. It is not measured by an online test and it is not a diagnosis. It shifts slowly, through pace, predictability and relationship — and precisely because it shifts slowly, it is worth starting with someone who knows what they are doing.