The research · Attachment

John Bowlby: the man who showed that attachment is a need, not a weakness

British psychiatrist and psychoanalyst, father of attachment theory · 1907-1990

Before John Bowlby, needing someone was seen as immaturity. He showed that it is a survival system, wired into every one of us, that never switches off: it only changes its target. Your relationship runs on that system, every single day.

The attachment systemThe secure baseInternal working modelsFrom the cradle to the grave

By Laurène · updated September 27, 2026 · 8 min read

Who was John Bowlby?

John Bowlby was born in London in 1907 into a well-off family where children were raised at a distance from their parents, as was customary in his circle. He studied at Cambridge, then medicine, psychiatry, and psychoanalysis, becoming a qualified analyst in 1936. In the 1930s he worked at a London clinic for troubled children. There he made his first striking observation: the children who were most cut off from their emotions had often lived through early, prolonged separations.

After the war, he took charge of the children’s department at the Tavistock Clinic in London, where he set up a research unit on mother-child separation. In 1951, the World Health Organization asked him for a report on homeless children in postwar Europe, “Maternal Care and Mental Health.” In 1952, his colleague James Robertson filmed a two-year-old girl spending eight days in the hospital without her parents, and the film upended hospital practice. Then Bowlby wrote his great trilogy, “Attachment and Loss” (1969, 1973, 1980), followed by a final book that pulled his thinking together, “A Secure Base” (1988). He died in 1990.

Why does a child psychiatrist matter for your relationship? Because the system he described in babies does not disappear in adulthood. Bowlby himself said that attachment stays with us “from the cradle to the grave.” It only changes its main figure: first your parents, then, for most of us, the person we share our life with.

The central idea

In Bowlby’s day, two explanations dominated for why a baby bonds with its mother. For psychoanalysis, the child attaches because she satisfies its drives. For behaviorists, it is because she feeds it. Bowlby rejected both. Drawing on ethology (the study of animal behavior, including Harry Harlow’s baby monkeys, who preferred a cloth mother to a wire mother that gave milk), he proposed that attachment is a system in its own right, shaped by evolution, whose only goal is to keep the child close to someone who protects it.

This system follows a simple logic. It sleeps when all is well. It wakes up in the face of danger, fatigue, illness, or separation, and pushes you to seek out your attachment figure: crying, calling, clinging, following. It settles when that figure responds. The consequence is huge: needing someone is not a weakness, it is how a healthy human being normally works. Today this idea sounds obvious, but it was revolutionary.

The key concepts

The attachment system

Bowlby describes attachment as a regulation system, a bit like a thermostat. It tracks a single variable: how available the person who matters is. As long as that availability seems good enough, the system stays calm and makes room for other systems, including exploration and play. As soon as it seems threatened, the system takes priority over everything else. In an adult relationship, it is the one that kicks in when a message goes unanswered, when the tone hardens, when the other person seems far away. And it reacts before thought.

The secure base

The phrase was coined by Mary Ainsworth from her observations in Uganda, and Bowlby made it the title of his last book. A secure base is the bond you set out from to explore and come back to when things get rocky. A baby who knows someone is waiting ventures farther and plays more freely. So does an adult: we don’t dare in spite of the bond, we dare because of it. A good relationship doesn’t hold you back. It makes the world feel within reach.

Internal working models

Through repeated interactions, a child builds mental representations: an image of themselves (am I worthy of love?) and an image of others (can I count on them?). Bowlby calls these internal working models. They run in the background without our noticing, and they shape what we expect, what we notice, and how we interpret things. They are the raw material of what we now call an attachment style: secure, anxious, avoidant, or disorganized. The word “working” matters: these models do their job, and they can be reworked.

Protest, despair, detachment

Working with James Robertson, Bowlby observed that a young child who is separated from a parent goes through three phases: protest (crying, calling out, anger), then despair (withdrawal, sadness), then detachment (apparent indifference when the parent returns). This sequence replays in miniature in adult couples. Pursuing and getting angry is protest. Closing off and acting as if nothing happened is detachment. Understanding that both behaviors come from the same system changes how you read an argument.

From the cradle to the grave

For Bowlby, seeking closeness to a trusted person when you are hurting is not a leftover from childhood that you should outgrow. It is a sign of healthy functioning at any age. He took direct aim at a culture that treats autonomy as the absence of need. Real autonomy, he said, grows out of trusting that you can count on someone. That is exactly what the MyLoveLab program calls becoming a secure base for your partner.

What the research says

Bowlby was first and foremost a theorist. He pulled together clinical observations, ethology, and systems theory. The empirical testing of his ideas came largely from other people, starting with Mary Ainsworth. Here are the milestones that matter.

  • 1944, the 44 juvenile thieves. Bowlby compares 44 children referred for stealing with 44 children referred for other difficulties. Early, prolonged separations are far more common among the thieves he describes as “affectionless.” A retrospective study, and a small one, but a founding one.
  • 1951, the WHO report. A review of the evidence on children in institutions, concluding that a lack of continuous maternal care harms development. Its conclusions changed how children were cared for and when parents could visit in hospitals.
  • 1952, Robertson’s film. “A Two-Year-Old Goes to Hospital” makes the sequence of protest, despair, and detachment visible in a girl separated from her parents for eight days. Pediatricians who watched it could no longer say that “children forget quickly.”
  • The 1970s, the Strange Situation. Mary Ainsworth observes that babies differ in how they handle a brief separation, and that these differences track the parent’s sensitivity. A 1988 meta-analysis of 32 samples from 8 countries finds a majority of secure children everywhere (50 to 75%).
  • 1987, adult couples. Cindy Hazan and Phillip Shaver show that adult love has the same features as an attachment bond, and that the three styles described in children show up in adults in similar proportions.

What all this really shows: early bonds matter, the quality of the responses a child receives predicts security, and the system stays active for life. What it does not show: that the beginning fixes what follows. Longitudinal studies find real but moderate stability in attachment from childhood to adulthood. The past weighs on you, but it doesn’t lock you in.

Limits and criticisms

The first major criticism came from Michael Rutter in 1972, in “Maternal Deprivation Reassessed.” Bowlby, he said, lumps together three different things: separation from an attachment figure, losing that figure, and having no bond at all. He also confuses correlation with cause: the separated children often lived in families full of conflict, and it is family discord, more than the separation itself, that predicts later difficulties. The study of the 44 thieves, meanwhile, was retrospective, tiny, and Bowlby made the diagnoses himself.

The second criticism concerns the role of the mother. Bowlby’s early writings stress a single, maternal figure, and were used to make working women feel guilty. Research has since shown that children attach to several figures, that fathers matter, and that good-quality childcare does not harm attachment. Bowlby himself softened his position over the years.

Finally, the theory may underestimate the child’s temperament and varies across cultures. And its popularization raises a problem that Bowlby didn’t create but that deserves a name: turning a probabilistic theory into destiny. “I have an anxious attachment, that’s just how I am” is not Bowlby. It is the opposite of what he wrote about models that can be reworked.

How MyLoveLab uses it

The Attachment and security dimension rests entirely on this framework. The test measures your style on the two axes that researchers drew from Bowlby’s theory, abandonment anxiety and avoidance of intimacy, with questions inspired by the ECR-R scale, without being that scale. The program takes three of Bowlby’s ideas across ten levels: your alarm system and its triggers (level 2), the security that gets built when a signal receives a response (levels 5 to 7), and the final role of secure base for your partner (level 10). To follow the rest of the story, read the pages on Mary Ainsworth and on Cindy Hazan and Phillip Shaver, or the article on the four attachment styles.

The MyLoveLab test measures your Attachment and security dimension and your attachment style in about fifteen minutes, no sign-up needed. Your partner can take it on their own, and you compare your two systems.

Measure my Attachment dimension

Further reading

  • Forty-four juvenile thieves: Their characters and home-life · Article, 1944The founding study, small and retrospective, where the intuition linking early separations to emotional difficulties was born.
  • Maternal Care and Mental Health · Book, 1951The WHO report that changed how children separated from their parents are cared for.
  • Attachment and Loss, vol. 1 : Attachment (in French: Attachement et perte, tome 1 : L’attachement (PUF, 1978)) · Book, 1969The heart of the theory: the attachment system, its evolutionary origin, how it works.
  • Attachment and Loss, vol. 2 : Separation (in French: Attachement et perte, tome 2 : La séparation, angoisse et colère (PUF, 1978)) · Book, 1973Anxiety and anger in the face of separation: protest, explained.
  • Attachment and Loss, vol. 3 : Loss (in French: Attachement et perte, tome 3 : La perte, tristesse et dépression (PUF, 1984)) · Book, 1980Grief and depression seen through the lens of attachment.
  • A Secure Base: Parent-Child Attachment and Healthy Human Development · Book, 1988A collection of lectures, and Bowlby’s most accessible book for understanding the secure base.

Frequently asked questions

What is Bowlby’s attachment theory, in a nutshell?

It is the idea that every human being is born with a survival system that tracks how available the important people in their life are, and pushes them to seek closeness when there is danger or distress. This system is tuned by the responses received in childhood and stays active for life, including in a relationship.

What are the four attachment styles according to Bowlby?

Bowlby didn’t define the styles: he described the system. The first three (secure, anxious-ambivalent, avoidant) come from Mary Ainsworth in 1978, and the fourth (disorganized) from Mary Main and Judith Solomon in 1990. In adults, we talk about secure, anxious, avoidant, and disorganized, measured on two axes rather than in boxes.

According to Bowlby, is everything decided before age three?

No. Bowlby talks about internal working models: representations built in childhood that guide relationships but get updated with experience. Research finds moderate stability in attachment across a lifetime and documents earned security in adulthood.

Why was Bowlby criticized?

For lumping together separation, loss, and lack of any bond, for early studies that were too small and retrospective, and for centering the theory on the mother to the point of making women feel guilty. Research has since widened the framework to include several attachment figures and the child’s temperament, without overturning the central idea.

References

  1. Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home-life. International Journal of Psycho-Analysis, 25, 19-53.
  2. Bowlby, J. (1951). Maternal Care and Mental Health. World Health Organization.
  3. Bowlby, J. (1958). The nature of the child’s tie to his mother. International Journal of Psycho-Analysis, 39, 350-373.
  4. Bowlby, J. (1969). Attachment and Loss, vol. 1 : Attachment. Basic Books.
  5. Bowlby, J. (1988). A Secure Base: Parent-Child Attachment and Healthy Human Development. Basic Books.
  6. Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759-775.
  7. Rutter, M. (1972). Maternal Deprivation Reassessed. Penguin.
  8. van IJzendoorn, M. H. & Kroonenberg, P. M. (1988). Cross-cultural patterns of attachment: A meta-analysis of the Strange Situation. Child Development, 59(1), 147-156.

See the full bibliography

Related authors

Read more on the blog