The Stages of a Breakup

The stages of a breakup are accurate as a list of states and wrong as a sequence. Shock, craving, private bargaining, flatness, anger, and stretches of ordinary interest show up in most breakups, but they arrive out of order, overlap each other, and come back after you were sure they were finished. What people want from a stage list is a position. There isn’t one.

Key Takeaways:

• The recognizable states are real; the numbering and the arrows between them were added by other people later
• Grief research moved off sequences in 1999: Stroebe and Schut’s dual process model has no positions in it at all, only movement back and forth between the loss and the life the loss changed
• The famous five-step version was written about people facing their own deaths, not about people who have been left
• Flatness is routinely read as recovery, and anger often lands months after the sadness rather than before it
• Two states usually run at once, which is why no honest answer exists to “which stage am I in”

Why People Look Up the Stages of a Breakup

Nobody types “stages of a breakup” at one in the morning because they want a taxonomy. They want two separate things: a position on a map, and a verdict on whether what is happening to them is normal. A numbered list appears to hand over both in nine seconds. That is why the format has outlived the evidence for it by several decades and will outlive this article too.

One of those two questions has an honest answer. Is this normal: almost certainly, including the parts you have not said out loud to anybody. Where am I: unanswerable, because the thing has no fixed route for you to be partway along, and how long the whole thing takes has no fixed answer either.

The reassurance was doing all the work anyway. Numbering was decoration on top of it, and the decoration charges you later, when week nine feels like week two and the ladder says that shouldn’t happen.

Where the Numbered Version Came From

The staged model everybody half-remembers did not come out of breakup research. It came from observing people who were facing their own deaths, and it was offered as a description of things that turned up in those conversations, not as a schedule with a pass mark at the end. Denial, anger, bargaining, depression, acceptance got borrowed into bereavement, then into divorce, then into relationship advice, collecting arrows between the boxes along the way that nobody had put there originally.

Grief researchers left the sequence behind a long time ago. Margaret Stroebe and Henk Schut published the dual process model in 1999 partly because phase accounts did not match what bereaved people were observed to do. People moved back and forth instead of forward. They faced the loss hard for an evening, then deliberately did not think about it for three days, and both halves looked like the process working rather than one of them being a failure at the other.

Read a stage list as an inventory of what can happen and it earns its place. Read it as a route and it will tell you, incorrectly, that you are behind schedule.

The States You Will Recognize, in No Fixed Order

Six states cover most of what people describe after a breakup, and not one of them holds a slot in the calendar.

Unreality tends to come first, and it is physical rather than philosophical. You reach for the phone to report something completely ordinary, the neighbor’s ridiculous new car, and you are three lines into the message before it lands again. Nobody has measured how long that runs. What people describe is that it loses its grip before the others do, not that it clears on a date.

Craving is the state that gets misread as weakness. In 2010 Helen Fisher and colleagues scanned people who had been rejected recently and still called themselves in love, and watched what happened when a picture of the ex came up: activity in reward-related regions, the ones tied to wanting rather than to sadness (Fisher et al., 2010). That is a finding about a brain in a scanner and it stops there. It does not instruct you to delete the number or to keep the photograph, and anybody using it to recommend either is supplying the instruction themselves. What it does puncture is the idea that still wanting a person who left says something about your character. It spikes at a song, or at a photograph an app surfaced because it thought it was being kind, and it never gets permanently crossed off a list.

Bargaining in a breakup is rarely done with the other person. It gets done with a version of yourself from four months ago, the one who could still have answered differently in that kitchen in March. Hours go into rewriting a conversation with somebody who no longer exists, which is why the bargaining state feels so private and so stupid to describe to a friend. It is a close cousin of the anxiety that follows a breakup, the mind running one loop with no new input.

Flatness is the one people get wrong most often. Nothing hurts, nothing interests you, and the entire business feels like something that happened to a colleague. That is when people announce they are over it. Then it lifts, on no schedule anyone can give you in advance, and the sadness is sitting precisely where it was left, next to the flight they booked in the meantime.

Anger arrives late for a lot of people, which almost nobody is warned about. It can turn up months after the crying has thinned out, by which point everyone around you has filed the breakup under old news, so the fury lands as if it came from nowhere.

The sixth is duller and much better: an hour where you are genuinely absorbed in something trivial and the subject does not come up at all. Most people notice it only afterward, and a surprising number feel a flicker of guilt about it.

One group gets a completely different opening. If you had been quietly leaving the relationship for a year before the conversation happened, craving may never show up and flatness comes first, which reads as coldness to everybody watching and worries them more than crying would have.

Naming the state you are in beats placing it. Calling a given Thursday flatness rather than recovery is the difference between waiting it out and making a decision on the strength of it.

Oscillation Instead of a Sequence

What replaced stages in the research has no positions in it at all. Stroebe and Schut described two modes running alongside each other: loss-oriented coping, where you deal with the loss itself, and restoration-oriented coping, where you deal with the changed life around it, the accounts and the empty Sundays and the question of who you are as one person. Healthy coping oscillates between them. It does not advance through them.

The part worth taking from that is where avoidance sits. Deliberately not thinking about it, for a Saturday or for two weeks straight, lives inside the model rather than outside it as a symptom. Stage lists put denial on the bottom rung and quietly mark everything short of acceptance as work you still owe.

Which changes the question you can usefully put to yourself. Not which stage, but which of the two modes has gone missing lately. Three weeks without a single evening where the subject stayed shut is one kind of stuck; three weeks without once letting it in is the other kind, and both get corrected by booking the missing half rather than waiting for it to arrive on its own, which is most of how you get over a breakup once the ladder is gone.

Try it on the last seven days. Name one evening that went to the loss and one that went to building the week around it, and if you cannot supply one of the two, you have found the half to schedule.

When an Earlier State Comes Back

Recurrence is the ordinary shape of this, not evidence of relapse. Week nine can feel like week two because a wedding invitation arrived, or the weather turned, or somebody used a phrase your ex used. The state was never crossed off. Only quiet.

George Bonanno’s bereavement research found resilience to be the most common trajectory after major loss, and also that trajectories differ substantially from person to person. That second finding is the one that matters here. If people demonstrably do not all follow the same path through a loss, a single ladder was never going to fit any particular one of them.

What does not work is measuring your week against a curve. The specific damage the stage frame does is to convert one bad Tuesday into proof of being back at the beginning, and people act on that proof, sending the message, driving past the building, in a way they would not if they read the same Tuesday as a spike on a line that is otherwise thinning out.

Worth saying plainly at this point, because a ladder invites people to wait a bad week out. This article describes experience and mechanism, not care. If things have turned toward not wanting to be here, that belongs with a clinician or a crisis line; in the US, the 988 Suicide and Crisis Lifeline.

Something concrete instead of the ladder, closer to the practical register of what to do after a breakup: write down the date every time one of these states hits hard, nothing else, just the date. Six weeks of dates tells you what a stage list only pretends to, and it tells you about yourself rather than about an average.

FAQ

What are the 5 stages of a breakup? There is no validated five-stage sequence for breakups. The list that circulates, denial through acceptance, was adapted from writing about people facing their own deaths, where it was a description rather than a timetable. The states it names are recognizable in breakups; the order and the count came from somewhere else.

How do I know which stage of a breakup I am in? You cannot, and the problem is in the question rather than in you. A version that does work: what did the past week consist of, hours spent facing the loss or hours spent rebuilding around it? Whichever of the two is close to zero is the one to move, and the check takes about two minutes on a Sunday.

Is it normal to feel nothing after a breakup? Yes. Flatness is one of the common states and it lands at unpredictable points, sometimes within the first 48 hours, sometimes in month three after weeks of crying. It gets mistaken for being finished more than any other state, which is why so many large decisions get made during it.

Why am I angrier now than I was when it happened? Anger commonly shows up after the sadness rather than before it, and months later is not unusual. Everyone around you will have filed the breakup as settled by then, so the anger tends to be met with mild surprise, which reads as one more sign that you are doing this wrong. You are not.

Stage models keep getting rewritten because they answer a question nobody stops asking, which is how far along am I. Distance is the wrong unit. What changes over a year is not your position on a line but how much of an ordinary week the loss can still claim, and that share moves unevenly, upward in some months, invisibly in others. Anybody who tells you where you are is describing a diagram rather than you. The states are real. The staircase was drawn on afterward. If what aches most is the reflex of narrating something ordinary to a person who was keeping up, you can now hand those small reports to software instead. Lona is an AI built for that kind of daily check-in, and it remembers where yesterday left off, so you are not restarting the whole story every time. It is not the person who left. It does not stand in for the friends still keeping up with you either. It just gives the reflex somewhere to go.

Sources

• Stroebe, M. & Schut, H., “The Dual Process Model of Coping with Bereavement,” 1999
• Fisher, H. E. et al., neuroimaging research on romantic rejection, 2010
• Bonanno, G. A., research on resilience and trajectories in bereavement