Origins
Where the theory came from
In 1951 the British psychiatrist John Bowlby wrote a report for the World Health Organization on children separated from their families after the Second World War. He noticed that these children grieved in a recognisable sequence: protest, then despair, then a kind of detachment. Bowlby argued that humans are born with an attachment system, a set of behaviours that keeps a vulnerable child close to a protective caregiver. He laid this out fully in Attachment (1969), the first volume of his trilogy.
Around the same time, Harry Harlow's controversial studies with infant rhesus monkeys showed that comfort, not only food, drives early bonding: the infants clung to a soft cloth "mother" even when a wire one provided milk. Today those experiments are seen as ethically troubling, but they helped shift psychology away from the idea that love is simply learned through feeding.
Mary Ainsworth, who worked with Bowlby in London, took the theory into the field, first in Uganda and then in Baltimore. Her "Strange Situation" procedure, published with colleagues in 1978, watched how one-year-olds responded when a caregiver briefly left a room and returned. Most infants protested and then settled quickly on reunion. Others seemed unbothered on the surface, or stayed distressed and hard to soothe. These became the secure, avoidant and ambivalent patterns; Mary Main and Judith Solomon later added a disorganised category.
Adulthood
From infants to romantic partners
In 1987 Cindy Hazan and Phillip Shaver published a now-famous study in which readers of the Rocky Mountain News answered a "love quiz." People described their romantic relationships in ways that echoed Ainsworth's infant categories. Romantic love, they proposed, is itself an attachment process: partners become each other's safe haven in distress and secure base for exploring the world.
Kim Bartholomew and Leonard Horowitz (1991) refined the picture into four styles built on how people see themselves and others: secure, preoccupied, dismissing-avoidant and fearful-avoidant. Later work, notably the Experiences in Close Relationships scale by Brennan, Clark and Shaver (1998), showed that attachment is best measured on two continuous dimensions rather than in boxes:
- Attachment anxiety: worry about rejection and a strong need for reassurance.
- Attachment avoidance: discomfort with dependence and a preference for self-reliance.
Where you fall on each dimension can shift with context. Many people are fairly secure with friends and more anxious with a new partner. That is normal, and it is one reason researchers are cautious about labels.
Everyday life
What the patterns can look like
| Pattern | Common inner story | Under stress you might… |
|---|---|---|
| Secure | "I'm worth loving, and people are mostly reliable." | Ask for support directly and accept it. |
| Anxious | "I need closeness, but I fear it will slip away." | Seek reassurance, check in often, read silence as rejection. |
| Avoidant | "It's safer to rely on myself." | Withdraw, change the subject, focus on tasks. |
| Fearful | "I want closeness, and closeness feels dangerous." | Move toward and away in quick succession. |
A common pairing is one anxious and one avoidant partner, sometimes called the pursue-withdraw cycle. The more one reaches, the more the other retreats, and both are trying, in their own way, to feel safe.
Change
Earned security: a template, not a destiny
The most hopeful finding in this field is that attachment can change. Researchers using Mary Main's Adult Attachment Interview describe "earned security": adults who had difficult childhoods but can now tell their story coherently, with perspective and without being swept away by it. A reliable partner, a good friendship or therapy can all contribute.
Mario Mikulincer and Phillip Shaver have shown in laboratory studies that even brief reminders of a supportive person ("security priming") make people more compassionate and less defensive for a time. Sue Johnson's Emotionally Focused Therapy, described for general readers in Hold Me Tight (2008), helps couples name the attachment fears beneath their arguments, and outcome studies report meaningful improvement for many couples.
Secure love is not the absence of need. It is the confidence that need will be met often enough.
Practice
Gentle ways to work with your pattern
- If you lean anxious: name the need rather than testing for it. "I'd love a quick text when you land" is clearer than waiting to see if one arrives.
- If you lean avoidant: signal your return. "I need an hour to decompress, then I'm all yours" keeps space from looking like rejection.
- Together: describe the cycle as the shared problem, not each other. Johnson calls this finding the "demon dialogue" you both get caught in.