I have watched a quiet but powerful shift take place across services, education, therapy, and public discourse. A concept that was once nuanced, debated, and carefully researched has been flattened, simplified, and turned into something rigid, predictive, and deeply pathologising.
Attachment.
People ask me about this all the time, so I figured I would write an article for you all to use and share to have these conversations.
We now live in a culture where people introduce themselves as their attachment style. Where professionals speak about children as if their futures can be mapped from a single early pattern. Where women are told they are ‘anxiously attached’ in relationships, and men are labelled ‘avoidant’ as if these are fixed traits rather than context-dependent responses. Where children who have experienced trauma are quickly described as having ‘attachment disorders’, as though this explains them.
And I want to be very clear about this. This is not what the evidence says.
The evidence and theory has been twisted and contorted - like many of the theories used these days in psychology, social work and mental health, really.
The popular, oversimplified version of attachment theory that dominates practice today bears very little resemblance to what longitudinal and lifespan research actually demonstrates. The idea that we each have one attachment style, formed in early childhood, which then acts as a blueprint for all future relationships, is not supported in the way it is being used.
In fact, research has consistently shown the opposite.
Attachment is fluid. It is relational. It is context-dependent. And it changes over time.
Elizabeth Meins has been one of the key voices raising concerns about the way attachment is being misunderstood and misused. In her work, she argues that attachment has become overrated and over-applied, particularly in professional contexts where it is used to predict outcomes or to pathologise children. She highlights that a large proportion of the population could be categorised as having ‘insecure attachment’ under current frameworks, which should immediately raise questions about whether we are describing a disorder at all, or simply describing normal human variability.
More importantly, she emphasises that attachment is not a fixed, singular characteristic of a person. Children, and adults, form different attachment patterns with different people. A child may feel safe and secure with one caregiver, and fearful or avoidant with another. These patterns are not contradictions. They are adaptations.
When I lecture on this topic, I always say the same:
You can be securely attached to your partner whilst being simultaneously insecurely attached to your parents. You could even have different attachment styles with each of your parents. You could have different attachment styles with each of your exes.
This matters. Because what we are currently doing in practice is taking a flexible, relational concept and turning it into a static label.
And once that label is applied, it begins to shape everything.
Let me bring this into something real.
In my own work, I have seen countless examples of children and women who have been labelled with ‘attachment disorders’ or described as having problematic attachment styles. And when you actually look at their lives, at what has happened to them, their behaviour makes perfect sense.
Take a girl who has been abused by people who told her they loved her. Who learned, over years, that affection and care were precursors to harm. Who learned that trust was dangerous. When she later rejects adults who are kind to her, when she refuses closeness, when she pushes people away before they can hurt her, this is not evidence of a disordered attachment system.
It is evidence that she has learned. It is evidence that she is protecting herself. It is rational, in my view.
But instead of understanding this as a trauma response, we label it. We call it ‘attachment disorder’. We call it ‘borderline personality traits’. We locate the problem inside her.
And in doing so, we completely ignore the context that created those behaviours in the first place. This is the core problem with the current obsession with attachment. It allows us to pathologise adaptation.
From an anti-pathology perspective, this is not a small issue. It is a fundamental misinterpretation of human behaviour. When we take coping strategies that have developed in response to harm and reframe them as symptoms of disorder, we shift the focus away from what has happened to the person and towards what is ‘wrong’ with them.
We also begin to make dangerous assumptions about their future.
If a child is labelled as having an attachment disorder, what are we really saying? We are implying that their ability to form relationships is impaired. That their future connections will be unstable. That their capacity for love, trust, or parenting may be compromised. These are not neutral statements. They are predictive, limiting, and often self-fulfilling.
And yet, they are not supported by the evidence.
Longitudinal research shows that attachment patterns can and do change across the lifespan. People form new relationships. They experience safety where there was once danger. They develop trust in contexts where it was previously impossible. They adapt, again and again, to new environments and new people.
We do not have one attachment style. We have many.
We can feel secure with a partner and avoidant with a parent. Anxious with one friend and completely at ease with another. Guarded in professional settings and open in intimate ones. These are not inconsistencies. They are reflections of our experiences with those individuals and environments.
Human beings are not static blueprints. We are responsive, complex, intelligent systems.
What concerns me most about the current discourse is how quickly attachment language is being used outside of its original context. It is now everywhere. Social media, therapy rooms, schools, court reports. It is being used to explain everything from relationship conflict to criminal behaviour. And with that expansion has come distortion. Attachment has become shorthand for blame.
Women are told they are too anxious, too needy, too insecure. Children are described as disordered rather than traumatised. Survivors of abuse are framed as having dysfunctional relationship patterns, rather than being recognised as people who have learned, through experience, that closeness can be dangerous.
This concerns me, because it changes the question we ask.
Instead of asking, “What has happened to you that has led you to respond this way in your relationships?”, we ask, “What is wrong with you? Do you have attachment issues or something?”
And once we ask that question, we begin to design interventions to fix the person, rather than address the harm. From a trauma-informed perspective, this is the exact opposite of what we should be doing.
We should be recognising that behaviour is meaningful. That it is rooted in experience. That it serves a purpose. We should be creating environments that are genuinely safe, consistent, and respectful, where people can, over time, form new patterns of relating if they choose to.
Not diagnosing them. Not predicting their futures. Not reducing them to a category.
Attachment theory, at its core, was never meant to be a tool for labelling and limiting people. It was meant to help us understand the importance of relationships, safety, and responsiveness. Even then - the original research was debatable. But what we have done is take that theory and turn it into another mechanism of control.
Another way to pathologise. Another way to locate the problem inside the individual.
We need to step back. We need to stop telling people that who they are in relationships is fixed, disordered, or predetermined.
Because it isn’t. Attachment is fluid.