Attached

Attachment research began with infants and caregivers and was extended to adult romantic relationships from the late 1980s. Levine and Heller summarise that literature into three working styles: secure, anxious and avoidant.

Secure people are comfortable with closeness and reasonably direct about needs. Anxious people want closeness and monitor for signs of its withdrawal. Avoidant people value independence and experience intimacy as encroachment. The book's practical claim is that these patterns predict conflict better than compatibility of interests does.

Most of what looks like incompatibility is two people running different rules about closeness.

Devmark’s reading — not a quotation from the book

The book's best section is on the anxious-avoidant pairing, where one person's pursuit triggers the other's withdrawal, which intensifies the pursuit. Naming that loop is genuinely helpful, because inside it both people experience themselves as reasonable.

It is also practical about what it calls effective communication: stating a need plainly and early, rather than testing whether the other person will guess it.

The three-category scheme is a simplification of a research literature that uses continuous dimensions rather than types, and attachment style is more changeable across relationships and time than the book's framing suggests.

It is also written in a self-help register that can encourage labelling partners rather than understanding them. Treat the categories as a vocabulary for noticing a pattern, not as a diagnosis.

Useful as a lens. Unreliable as a verdict about a person.

Devmark’s reading — not a quotation from the book

The book sorts adults into secure, anxious and avoidant, based on what happens to them when closeness increases or a partner becomes unavailable. Secure is presented as the largest group and the least discussed, because the people in it generate fewer problems and therefore fewer books.

The pattern the book exists to name: one partner pursues reassurance, the other withdraws to preserve independence, and each behaviour intensifies the other. It is stable precisely because it is miserable — the pursuit confirms the avoidant's fear of engulfment, and the withdrawal confirms the anxious partner's fear of abandonment.

It is a self-sustaining loop. Each person's coping strategy is the other's trigger.

Devmark’s reading — not a quotation from the book

The most practical claim is that anxious people disproportionately end up with avoidant partners because secure people pair off early and stay paired, while the other two recirculate. The recommendation is to look for security as the primary criterion, which is duller and more useful than most dating advice.

The book's actual technique is direct statement of needs early, without protest behaviour — no testing, no withdrawing to provoke a response, no hinting. Its argument is that this filters quickly: a secure partner responds well, and an avoidant one reveals themselves before years have been invested.

Adult attachment research is a real and substantial field with decades of work behind it, but the popular three-category version simplifies it considerably. Styles are dimensional rather than discrete boxes, they vary from relationship to relationship rather than being a property of a person, and they change over time and with circumstance. The book presents them as more fixed, more uniform and more predictive than the underlying literature supports.

The vocabulary has escaped the book and is now routinely used to diagnose partners, excuse behaviour and end relationships pre-emptively. Levine and Heller do not recommend any of that, and the most useful way to read it is as a description of a pattern you are in rather than a category someone belongs to.

Its value is almost entirely in recognition rather than prescription. Readers stuck for years in the pursue-and-withdraw cycle very frequently report that simply having a name for the pattern was the thing that made it discussable, and therefore negotiable, with the other person. Take the description of the cycle and the advice about stating needs directly, and hold the three-part typology considerably more loosely than the book does.