Earned secure attachment
Arriving at security later in life — what the term means, what the evidence says about how it happens, and what it does not require.
Relationship science
Earned secure attachment means arriving at a secure attachment pattern in adulthood without having had a secure childhood. The term comes from the developmental literature, where researchers noticed a group who described difficult early caregiving and nevertheless functioned like securely attached adults.
It is the most important finding in this whole area, and it is usually reported badly. The claim is not that attachment style is easy to change, or that it changes by deciding to. It is that attachment style is a learned expectation about what other people do with your needs, and expectations update when the evidence changes. Everything else follows from that.
What the research established
Roisman and colleagues gave the category its rigour, and their finding is more interesting than the popular version.
Earned-secure adults — identified by coherent, reflective accounts of difficult childhoods — functioned like continuously secure adults in their romantic relationships. Observers rated their partnerships equivalently. But they reported more depressive symptoms, and they were more affected by stress.
That is worth sitting with. Earned security is not a state where the history stopped mattering. It is a state where the history no longer determines how you behave in a relationship. The cost did not vanish; the pattern changed.
The other robust finding: attachment security increases over the course of psychotherapy, across modalities and measured as a secondary outcome. Levy's meta-analytic work finds this consistently. No therapy is licensed for attachment, because attachment style is not a disorder — but security moves when the underlying expectation gets repeatedly disconfirmed, and therapy is one of the more reliable places that happens.
The three routes
A relationship with a secure partner. The most common route and the slowest. Repeated experience of bringing a need forward and having it met, over years, does update the model. This is not a rescue and cannot be assigned to a partner as a job — but the direction of influence is real and shows up in the couples literature.
Therapy. The fastest reliable route. What matters more than the modality is the length of the course and the alliance — which, for anyone insecurely attached, is also the first place the pattern appears and therefore the first place it can be worked on. Emotionally Focused Therapy is built on attachment theory directly; psychodynamic, schema and mentalization-based work all show movement.
Deliberate practice. Underrated and genuinely effective in combination with the others. It is behavioural: saying the thing at a five out of ten instead of waiting for a nine, staying in the room ninety seconds past comfortable, asking directly rather than testing. The belief changes after the behaviour, not before it — which is why waiting until you feel secure enough to act securely does not work.
What it does not require
Four things people assume are prerequisites and are not.
Reconciliation with your family. Roisman's earned-secure group were characterised by a coherent account of what happened, not by having repaired the relationships. Making sense of it and making peace with them are separate.
Remembering everything. Coherence is the marker, not completeness.
Being single first. No evidence supports the idea that you must resolve your pattern alone before you are entitled to a relationship. Relationships are one of the main places it resolves.
Feeling secure. Security is a set of expectations expressed in behaviour, not a feeling of confidence. Plenty of earned-secure people still feel the old alarm and no longer act on it.
The timescale, honestly
Months to years, not weeks. A prediction assembled from thousands of childhood data points does not revise on three good conversations.
Two things about the shape of the change are worth expecting. Progress is uneven, and stress reliably brings the old pattern back — that is a return of an old habit under load, not evidence that nothing changed. And the first thing that shifts is usually the gap between the impulse and the action: you notice yourself about to withdraw, or about to escalate, slightly before you do it. That gap is small, it is where everything else gets built, and it is a much better early marker than feeling different.
One more: security is more partner-specific than most descriptions admit. Someone can be measurably secure with one person and not with another. That is not a failure of the work — it is information about the relationship.
sources
- · Roisman, G. I., Padrón, E., Sroufe, L. A., Egeland, B. (2002). Earned-secure attachment status in retrospect and prospect. Child Development.
- · Levy, K. N., Johnson, B. N., Kivity, Y., et al. (2018). Attachment style and psychotherapy outcome. Psychotherapy Research.
- · Mikulincer, M., Shaver, P. R. (2016). Attachment in Adulthood, 2nd ed. — Chapter 14.
- · Johnson, S. M. (2019). Attachment Theory in Practice.