What the Theory Actually Says
John Bowlby was studying something narrower than romance: what happens to a small child when the person they depend on leaves the room. Mary Ainsworth built the observation into categories. Decades later, researchers noticed that the strategies people develop in infancy for handling closeness bear a striking resemblance to how adults behave when a partner does not reply.
The core claim is modest and well evidenced. Early on, each of us learns an answer to one question: when I need someone, will they come? The answer shapes a strategy. If comfort was reliable, you learn that reaching out works and that closeness is not dangerous. If it was inconsistent, you learn to escalate — because sometimes escalation worked. If it was mostly unavailable, you learn to need less, or to appear to.
None of these are flaws. Each was an intelligent adaptation to the environment a child actually had. The difficulty is that strategies outlive their conditions, and a tactic that once secured care from an unpredictable adult tends to work poorly on a partner who was going to be reliable anyway.
The Three Familiar Patterns
Anxious. Closeness feels good and precarious at once. Small ambiguities — a shorter message, a delayed reply — get read as evidence about the state of the whole relationship. The strategy is to close distance: to reach out more, seek reassurance, and sometimes to test. It is exhausting and it is not neediness. It is a nervous system that learned that vigilance was once genuinely necessary.
Avoidant. Intimacy is wanted and experienced as encroachment. As things get closer, something tightens; the person becomes busy, critical of small flaws, or subtly less available. This is rarely a lack of feeling. It is more often a learned association between depending on someone and being let down, which makes the approach of real closeness read as exposure.
Secure. Not a person without difficulty, but one for whom closeness is not itself a threat. They can say what they need without a strategy, hear a complaint without treating it as an attack, and tolerate the ordinary silences of another person having a life. Roughly half of adults look broadly secure, which is worth remembering when the internet implies everyone is damaged.
A fourth pattern, disorganised or fearful-avoidant, mixes the two: a strong desire for closeness with a strong alarm at receiving it, often producing a cycle of pursuit and retreat within a single relationship.
An attachment style is not a personality. It is a strategy — learned early, for managing the risk of needing someone.
The Trap That Forms Between Two of Them
The anxious-avoidant pairing is common enough to have become folklore, and the mechanism is unfortunately elegant. The anxious partner senses distance and moves closer. Increased closeness triggers the avoidant partner's need for room, so they withdraw. Withdrawal confirms the anxious partner's deepest expectation, so they escalate. Escalation confirms the avoidant partner's deepest expectation, so they withdraw further.
Each person is behaving rationally given what they believe, and each is producing exactly the evidence the other most fears. Nobody is the villain. The loop simply runs, and it produces enormous intensity — the highs of reunion, the lows of distance — which is very easily mistaken for depth of feeling.
This is why the pairing is so sticky. The physiological experience of an activated attachment system is close to the experience of falling in love: preoccupation, urgency, relief when contact resumes. Calm partners can feel unexciting by comparison, not because there is less between you, but because nothing is triggering alarm.
What This Framework Is Not
It has become a horoscope. People announce a style as an identity, use it to explain every behaviour, and occasionally to license it — I am avoidant, this is simply what I am like. That is a misreading in a way worth being precise about.
First, these are patterns, not categories. Most people sit on a spectrum and move along it depending on who they are with and what else is happening in their life. Second, they are context-dependent: the same person can be broadly secure with one partner and markedly anxious with another. Third, and most importantly, a style is a description of a strategy, not an excuse for its consequences. Explaining behaviour and being accountable for it are separate acts.
It is also a poor instrument for diagnosing other people. Applied to yourself, attachment theory produces useful self-knowledge. Applied to a partner two dates in, it mostly produces a confident story that stops you noticing who they actually are.
How Patterns Actually Change
The most reliable mechanism is unromantic: sustained contact with someone secure. Over time, a nervous system that expected inconsistency accumulates evidence to the contrary. Nothing dramatic happens; the alarm simply stops firing as often, because it keeps being wrong.
The second mechanism is catching the strategy mid-flight. Not preventing the feeling — that rarely works — but noticing it as it happens. I am about to send a fourth message, and the feeling driving it is old. I have just found three faults with someone I liked yesterday, and yesterday is when we got closer. Naming the pattern while it is running is what gradually loosens its authority.
The third is plain speech, which sounds trivial and is not. Anxious strategies rely on indirect signals; avoidant ones rely on ambiguity. Both are undone by saying the true thing simply — I noticed I got worried when I did not hear from you, or I want to see you and I also need a quiet evening. Directness removes the need for the strategy, because the strategy existed to get a need met without having to state it.
Why the Format of Dating Matters Here
Attachment strategies are activated by uncertainty, and modern dating manufactures uncertainty at scale. Unlimited options, ambiguous intentions, unexplained disappearances and the constant awareness that someone might be talking to twenty other people — these conditions would unsettle a secure person. For anyone anxious, they are a permanent alarm.
The effect is that the format makes people look less secure than they are. Vigilance you have never displayed in your life becomes reasonable when the environment genuinely rewards it. Withdrawal becomes reasonable when there is genuinely no defined relationship to withdraw from. It is worth being careful about diagnosing yourself from behaviour produced by conditions engineered to destabilise.
Which is an argument for choosing conditions deliberately. Fewer people, clearer intentions, a pace at which someone can actually become known — these do not cure anything, but they remove the constant low-level threat that keeps old strategies switched on. Security is partly disposition. It is also, more than we tend to admit, a property of the situation you agree to date inside.
Common questions
- What are the four attachment styles?
- Secure, anxious (preoccupied), avoidant (dismissive) and disorganised (fearful-avoidant). They describe habitual strategies for handling closeness and the fear of losing it, not fixed personality types.
- Why do anxious and avoidant people attract each other?
- Because each confirms what the other already expects. Distance activates the anxious person's pursuit, and pursuit activates the avoidant person's withdrawal. The intensity this produces is easily mistaken for chemistry, but it is a feedback loop rather than a bond.
- Can your attachment style change?
- Yes. Attachment patterns are learned and can be revised, most reliably through a sustained relationship with someone secure, and through noticing the strategy as it activates rather than after it has run its course. It is slow, and it is genuinely possible.