"I'm way too anxious in relationships." "He's avoidant, obviously." The vocabulary of attachment has taken over social media and conversations between friends. Behind these labels lies a scientific theory born more than seventy years ago from watching babies, which took a long time to reach couples. Back to the source.
A psychiatrist and separated children
It all starts with the British psychiatrist and psychoanalyst John Bowlby. In the 1940s, he studied young offenders and noticed that many of them had experienced long or repeated separations from their mother as children. In 1951, he wrote a report for the World Health Organization on children deprived of maternal care, orphaned or displaced by the war.
At the time, the dominant idea was simple: babies become attached to their mother because she feeds them. Bowlby did not believe it. He drew on ethology, the study of animal behavior, and in particular on Konrad Lorenz's work on young birds that follow the first moving creature they see. For him, the need to stay close to a protective figure is a need in its own right, shaped by evolution: a little one who stays near an adult has a better chance of surviving.
In the late 1950s, the experiments of the American psychologist Harry Harlow pointed the same way. His baby monkeys, raised with two artificial "mothers", one made of wire that gave milk and one made of soft cloth that gave none, spent most of their time clinging to the second. Comfort mattered more than food.
Bowlby brought his ideas together in a trilogy, Attachment and Loss, published between 1969 and 1980.
The "Strange Situation"
It was a Canadian-American researcher, Mary Ainsworth, who gave the theory its measuring tools. After observing mothers and their babies in Uganda and then in Baltimore, she designed a laboratory procedure in the 1960s lasting about twenty minutes: the Strange Situation.
A child of around one year old explores a room full of toys with their mother. A stranger comes in. The mother leaves, then returns; she leaves again, this time leaving the child alone, then comes back once more. What interests researchers is not so much the crying when the mother leaves as the way the child greets her when she returns. Ainsworth identified three patterns:
- the secure child protests when she leaves, is quickly comforted when she returns, then goes back to playing: the mother is a "secure base" from which to explore the world;
- the avoidant child seems indifferent when she leaves and ignores her when she returns, even though their heart rate shows they are indeed stressed;
- the ambivalent (or resistant) child is very upset, clings to the mother on her return while pushing her away, and struggles to calm down.
In her sample, about two thirds of the children were secure. In 1986, researchers Mary Main and Judith Solomon added a fourth category, disorganized, for children with no coherent strategy: they approach and then freeze, as if the person meant to reassure them also frightened them.
The big leap to couples
In 1987, two psychologists at the University of Denver, Cindy Hazan and Phillip Shaver, asked a bold question: what if adult romantic love was an attachment bond too? They published a "love quiz" in a local newspaper, the Rocky Mountain News. Readers chose, from three short paragraphs, the one that best described how they experienced their relationships.
The result: about 56% recognized themselves in the secure description, 25% in the avoidant one and 19% in the anxious one, proportions close to those observed in babies. The paper became one of the most cited in relationship psychology and opened up a huge field of research.
Four styles, two dimensions
In 1991, psychologists Kim Bartholomew and Leonard Horowitz proposed a four-style model built on two questions: do I have a positive view of myself? Do I have a positive view of others?
| Others are reliable | Others let me down | |
|---|---|---|
| I deserve to be loved | Secure | Avoidant (dismissing) |
| I doubt myself | Anxious (preoccupied) | Fearful |
The fearful style, which longs for closeness while dreading it, is in a way the adult extension of childhood disorganized attachment. This four-style model is the one used in our attachment style quiz.
Researchers then simplified things further. Since the late 1990s, most of them measure adult attachment on two continuous scales, anxiety (fear of abandonment, the need for reassurance) and avoidance (discomfort with intimacy and dependence), notably with the Experiences in Close Relationships questionnaire by Kelly Brennan, Catherine Clark and Phillip Shaver (1998). Statistical analyses suggest that people do not fall into neat boxes: everyone sits somewhere along these two axes. The styles are handy landmarks, not watertight categories.
What attachment changes in a relationship
People who are mostly secure have, on average, more satisfying and more stable relationships. They ask for support when they need it, give it readily, and get through conflicts without seeing them as a threat to the relationship.
Anxious attachment often shows up as high alert: an unanswered message becomes a sign, a silence a worry. Avoidant attachment tends to lead people to play down their needs, keep part of themselves out of reach and pull back when the other person comes closer. Popular books such as Attached by Amir Levine and Rachel Heller (2010) have popularized the "anxious-avoidant trap", where one partner's demand for closeness feeds the other's withdrawal, and vice versa.
A style is not a life sentence
This is probably the most important message of the research: attachment style is moderately stable, not fixed. The relationships we have, a reassuring partner, life events or therapy can change it. Researchers speak of earned security for adults who, despite a difficult childhood, have developed secure attachment.
A few caveats are worth keeping in mind. Each person's temperament, partly innate, plays a role the theory long underestimated. The proportion of each style varies from one country to another, even though the secure style remains the most common in most studies. And no online questionnaire, whatever it is, replaces a conversation with a professional.
In short
- Attachment theory was born with John Bowlby and Mary Ainsworth: the need for a protective figure is a basic need, not a side effect of being fed.
- In children, the "Strange Situation" distinguishes secure, avoidant, ambivalent and disorganized attachment.
- Since 1987, the theory has been applied to couples; today it is measured on two axes, anxiety and avoidance.
- An attachment style can change. It is a tendency, not a destiny.
To go further, our article on what science says about love compatibility shows what, beyond attachment, keeps a couple together.