What Are the 5 Stages of a Breakup?

Somewhere between the breakup and the first honest night of sleep, most people end up in the same place: phone in hand, a list on the screen, checking themselves against it. Denial. Anger. Bargaining. Depression. Acceptance. Five tidy boxes with arrows between them, and the quiet question underneath — which box am I in, and am I behind schedule?

It's a strange thing to do to a wound: grade it. But that's what the stages list invites, and that's why it needs to be examined before it's used. Because here's the thing nobody tells you when you find that list — it wasn't built for you. It wasn't built for breakups at all. And the research that followed it suggests the real shape of heartbreak looks very different from five boxes in a row.

1. Where the five stages come from

The list has a birthday. In 1969, the Swiss-American psychiatrist Elisabeth Kübler-Ross published On Death and Dying, a book about people facing terminal illness. Based on her work with dying patients, she described five emotional responses she kept observing: denial, anger, bargaining, depression, and acceptance. She originally called it "The Five Stages of Death" [1].

Notice what's not in that description: heartbreak. The model was an observation about people confronting the end of their own lives, and the people around them. It was never a study of relationships ending. It wasn't about someone reading old texts in bed, or the specific horror of watching a shared future dissolve.

That didn't stop it from becoming the most popular framework in the world for any loss. The stages got renamed along the way — "The Five Stages of Loss," "The Kübler-Ross Grief Cycle," "The Change Curve" — and applied to everything from job loss to divorce to the end of a relationship [1]. Somewhere in that chain, a model built for dying patients became the checklist people measure their heartbreak against.

Even Kübler-Ross's own model wasn't as tidy as the infographic. The book included other responses — shock, anticipatory grief, hope — and later editions and later writing emphasized that the stages were never meant to be a strict sequence everyone walks through in order [1].

2. The empirical problem with stages

Here's the uncomfortable part: when researchers actually tried to test whether grief moves through predictable stages, the evidence didn't cooperate.

The most influential critique came from Camille Wortman and Roxane Cohen Silver, whose 1989 paper "The Myths of Coping with Loss" argued that the stage expectation was not just unsupported but actively harmful [2]. Their central observation is quietly devastating: if you expect grief to unfold in a particular way, you start evaluating people — including yourself — against that expectation. Someone whose grief doesn't follow the sequence gets judged as coping "abnormally" or "inappropriately." The person checking themselves against the list at 2 a.m. is running the exact mechanism the paper warns about. Their research found enormous diversity in how people grieve: some show little distress early and recover gradually, some show distress that worsens before it improves, some move through phases in order and many don't [2].

Then came the first large empirical test. In 2007, a Yale team led by Paul Maciejewski published an analysis of bereaved people in JAMA — often cited since as evidence that the stages exist [3]. Some patterns were consistent with the stage idea, but others directly contradicted it — most strikingly, "acceptance" was the most frequently endorsed response at the first time point measured, and at every time point after that. Acceptance wasn't the hard-won final stage; it was where people started [3].

The same journal published letters from critics, including Silver and Wortman themselves, pointing out serious methodological problems: roughly 40% of the originally recruited participants were excluded, the data were never analyzed as individual trajectories over time, and the sample's cultural and geographic limits weren't controlled for [4]. One person's single randomly-selected data point can't show a sequence — that requires watching the same person move through time, which is exactly what the study didn't do [3][4].

That's the fate of the five stages: a model that was observation-based to begin with, tested, and found to describe some people some of the time — but nowhere near a universal law of loss.

3. What breakup research shows instead: two clocks

If fixed stages aren't the answer, what is? The breakup-specific research gives a more useful picture — not a sequence of boxes, but two clocks running at once.

Clock 1 — acute distress, weeks to months. The emotional flood after a breakup — the inability to eat, the replaying, the physical heaviness — has a measurable shape. In a 2020 study of recently separated adults, separation distress fell in a nearly straight line across five months: from 30.42 in month one to 24.15 in month three to 20.69 in month five, on a 0–100 scale [7]. Other work found people consistently overestimate how bad the first weeks will be — the forecasting error is concentrated right at the start — and that for most people, the worst fades within about ten weeks [8]. This clock runs on the scale of weeks and months.

Clock 2 — the attachment bond, years. Meanwhile, the quieter pull — the "I wonder how they're doing," the background sense of them — runs on a different clock. A 2025 study of people from long-term relationships measured the attachment bond's half-life at about four years [9]. Language patterns changed by a breakup normalize within about six months [10], but the bond itself fades on a scale of years, not weeks [9].

Why does this matter for the stages question? Because most of the "stages" people report after a breakup — the bargaining, the despair, the sudden anger — are symptoms of Clock 1. And the ones that linger — the slow detachment, the identity confusion — belong to Clock 2. The stage list forces both clocks into one linear sequence, which is why it always feels slightly wrong: you can be past the acute despair (clock 1, mostly done) while the attachment pull quietly persists (clock 2, still running). That's not being stuck in a stage. It's two clocks, and they aren't synchronized.

You can see where you are on both clocks with the Breakup Timeline Calculator, which plots your weeks since the breakup against the research curves.

4. The phases people actually report

So if not five stages, what do people go through? The research describes phases — overlapping, looping, negotiable — rather than stages. Here's a rough map of the evidence-backed shape:

The critical difference from the stages list: these phases don't require a fixed order, and they don't require you to pass through all of them. Some people skip phases. Some repeat them. Some hit "absence becomes routine" and then crash back into the measured dip for a week. That variability isn't a defect in you — it's the finding [2].

5. What the stages list gets right

Before we throw the list out entirely, credit where it's due. The five stages got as popular as they did for a real reason: they name the emotions that actually show up.

Grief after a breakup is real grief. A romantic breakup has been cited among the worst events people experience [12]. So the anger, the bargaining ("if I just explain myself clearly, they'll understand"), the despair, the strange desire to believe it isn't happening — these aren't invented. They're the normal vocabulary of loss, and the list gives people permission to feel them without shame.

That's the genuine gift of the model. It tells you: these feelings are recognized, they're part of the territory, you're not broken for having them. As long as you read the list as a description of possible weather, it can be genuinely comforting.

The trouble starts when you read it as a forecast — a schedule you're supposed to follow. The same emotions become evidence of failure when they show up "out of order" or "too late." The list flips from permission to grading the moment it becomes a timetable [2].

6. What the stages list misses: the growth part

The most important thing the five-stage list omits is the part that comes after — and it's the part with the best evidence behind it.

In a 2003 study, researchers asked 92 young adults about their experience after a romantic breakup. Growth was not a side effect; it was the headline. Participants reported, on average, about five kinds of personal growth they believed would improve their future relationships — "I learned relationship skills," "I'm a lot more confident." Women reported more growth than men. People who attributed the breakup to situational factors — timing, circumstances, incompatibility — reported more growth than people who spiraled into self-blame [5].

A second study found growth was especially likely when the relationship itself had been low-quality — the kind that offered little room for your own self to expand. After those relationships ended, people reported positive emotions and growth tied to rediscovering parts of themselves [6]. It's addition through subtraction: sometimes the relationship was the subtraction, and its ending is what let the self you'd set aside come back [6].

None of this is a cheerful "everything happens for a reason" spin. Growth and distress are independent outcomes — you can be genuinely devastated and still, months later, notice that you learned things you wouldn't have traded away [12]. The honest version isn't "breakups make you better." It's "the distress and the growth both happen, on separate tracks, and neither one cancels the other."

7. How to use the five stages you found

So what do you do with the list that's probably still open in a browser tab?

Use it as vocabulary, not a schedule. "I'm in the anger phase" can be a useful way to name what's happening — it externalizes the feeling and makes it manageable. The problem is the second half of that sentence: "…and therefore I should move to bargaining soon." Drop the second half [2].

If the list is making you feel behind, that's the list working against you. Feeling "stuck in depression" for a month is often just clock 2 doing its slow work — or clock 1's decline not feeling like progress yet [7][10]. Neither is a failed stage.

Use the two clocks instead of the five boxes. The useful question isn't "which stage am I in?" — it's "which clock is doing what?" Is the acute heaviness still declining (clock 1)? Is the attachment pull still strong (clock 2)? Those two answers tell you more about where you are than any stage name [7][9].

Watch what you're doing, not what you're feeling. The stages are emotions. The research on recovery is largely about behaviors: contact with your ex slows the decline [7], monitoring them on social media predicts more distress [10]. Feeling angry is weather. What you do with the anger is the actual work.

8. When "stuck" is real

None of this means "feeling bad forever" is normal. There's a real line between slow recovery and being stuck — and the research gives you markers.

The researchers behind the six-month language study put it plainly: if you're still in deep emotional distress six months to a year after the breakup, it's a good time to seek professional help. Their slow-recovery markers: prolonged preoccupation with the ex, heavy reference to them in your speech, and difficulty untangling your identity from the relationship [10].

A useful distinction: lingering attachment alone is not the same as being stuck. The attachment bond has a four-year half-life for a reason [9]. But deep distress that isn't declining — that's the signal. If the acute phase is still running full-force at six months, that's worth professional attention, not a stage on a list.

And if you're in crisis — struggling with thoughts of self-harm or feeling unsafe — please reach out right now. In the US, call or text 988 (Suicide & Crisis Lifeline), or contact your local emergency services. This article is structured support for everyday recovery, not crisis care.

FAQ

Do all breakups follow the five stages?

No. The five stages were never validated for breakups, and grief researchers have repeatedly found that people don't reliably move through a fixed sequence [2][3]. Some people skip stages, some repeat them, some experience them all at once. What research shows instead is two overlapping clocks — acute distress declining over weeks and months, and an attachment bond fading over years [7][9].

Is it normal to skip stages or go through them out of order?

Yes — that's the finding, not a failure. Wortman and Silver's influential critique found enormous diversity in how people grieve, and even the study most often cited in support of stages found acceptance was highest at the first time point measured [2][3]. There's no required order.

Which stage of a breakup is the hardest?

For most people, not the first week. People overestimate the initial blow, and actual distress fades within roughly ten weeks for many [8]. The stretch that often feels hardest is later — when the shock is gone and the absence has become daily life, usually around months two and three [10]. Feeling worse then than in week one is the data's expectation, not a personal failing.

How long does each breakup stage last?

There's no honest fixed answer, because stages don't run on a fixed clock. What research does measure: acute separation distress declines across five months (30.4 → 20.7) [7], breakup language normalizes within about six months [10], and the attachment bond has a half-life of about four years [9]. If you want a number for where you are, the Breakup Timeline Calculator maps your situation onto those curves.

Why do I feel fine one day and terrible the next?

Because recovery isn't linear — phases overlap and loop. The same variability that makes the stages list unreliable is what makes your week feel like a roller coaster.

Is it normal to feel nothing at all after a breakup?

Yes. Some people show little distress shortly after a loss and recover gradually — that was one of the central findings of the myths-of-coping research [2]. Numbness in the first days is also a common response to shock. Neither means you didn't care.

Can a breakup make you a better person?

Growth and distress are separate outcomes, and both are real. People report an average of about five kinds of personal growth after a breakup — more confidence, better relationship skills, clearer boundaries — and growth is especially common after low-quality relationships [5][6]. It doesn't cancel the pain, and it isn't a reason to be grateful for the breakup. It just happens, on its own track.

Conclusion: the honest version

The five stages will keep showing up. They're too neat, too memorable, and too reassuring to disappear. And now you know their real history: a model built from dying patients in 1969, applied to everything since, tested, and found to be a description of possible weather rather than a forecast.

So here's the reframe, and it matters: you're not behind schedule. You never were. The heaviness you're feeling is one clock doing its measured work, and the slow pull toward them is another clock entirely — and neither one is a grade.

What the research offers instead of a checklist is something both less tidy and more honest: two clocks, a set of phases you'll move through in your own order, and a real possibility on the other side — not the tidy "acceptance" at the end of an arrow, but a rebuilt life that feels like yours again. That's the honest 90-day plan, and the full timeline breakdown is in our guide to how long it actually takes to get over a breakup. The 72-hour rule and the 3-week rule are the same folklore wearing round numbers.

The list was never the point. The point was always that the emotions have names, that you're not alone in them, and that — on no fixed schedule, in no fixed order — they pass.

Sources

1. Kübler-Ross, E. (1969). On Death and Dying. Macmillan. — The five stages (denial, anger, bargaining, depression, acceptance) as originally "The Five Stages of Death," based on work with terminally ill patients; model described as widely used but empirically unsupported. Origin and history corroborated via the Wikipedia article "Five stages of grief" (accessed 2026-08-24; book itself not read directly — see Methodology).
2. Wortman, C. B., & Silver, R. C. (1989). The myths of coping with loss. Journal of Consulting and Clinical Psychology, 57(3), 349–357. — Stage expectations unsupported; diverse grief courses normal; the expectation itself can make people feel they're coping wrongly. Full-text PDF read directly 2026-08-24.
3. Maciejewski, P. K., Zhang, B., Block, S. D., & Prigerson, H. G. (2007). An empirical examination of the stage theory of grief. JAMA, 297(7), 716–723. — Acceptance most-endorsed at first and every time point; ~40% of recruited participants excluded; data analyzed cross-sectionally. Abstract + criticism letters read directly 2026-08-24.
4. Silver, R. C., & Wortman, C. B. (2007). The stage theory of grief [Letter]. JAMA, 297(24), 2692. — Methodological critique of Maciejewski 2007. Read via JAMA record 2026-08-24.
5. Tashiro, T., & Frazier, P. (2003). "I'll never be in a relationship like that again": Personal growth following romantic relationship breakups. Personal Relationships, 10(1), 113–128. — 92 undergraduates; average of ~5 reported growth types; women more growth; correlates with situational attributions and Agreeableness. Abstract read directly 2026-08-24.
6. Lewandowski, G. W., Jr., & Bizzoco, N. M. (2007). Addition through subtraction: Growth following the dissolution of a low quality relationship. The Journal of Positive Psychology, 2(1), 40–54. — 155 undergraduates; growth and positive emotion after dissolution, especially from low-quality relationships; linked to self-expansion and rediscovery of self. Abstract read directly 2026-08-24.
7. O'Hara, K. L., et al. (2020). Clinical Psychological Science. — Separation distress declined linearly 30.42 → 24.15 → 20.69 across five months; in-person contact predicted higher distress two months later (no shared children). Full text read directly 2026-08-24 (verified for our timeline article).
8. Eastwick, P. W., et al. (2008). Journal of Experimental Social Psychology. — People overestimate the initial blow; the worst usually fades within ~10 weeks; N=69, 26 breakups. Author PDF read directly 2026-08-24 (verified for our timeline article).
9. Chong, S. C., & Fraley, R. C. (2025). Social Psychological and Personality Science. — Attachment bond to an ex has a half-life of ~4 years (midpoint 4.18 years); contact can keep the bond strong for a decade or more. Abstract + authors' summary read directly 2026-08-24 (verified for our timeline article).
10. Seraj, S., Blackburn, K. G., & Pennebaker, J. W. (2021). PNAS. — Breakup language patterns normalize ~6 months; slow-recovery markers = prolonged preoccupation, heavy ex-reference, identity entanglement; deep distress at 6 months–1 year → professional help. Authors' summary read directly 2026-08-24 (verified for our timeline article).
11. Slotter, E. B., Gardner, W. L., & Finkel, E. J. (2010). Personality and Social Psychology Bulletin. — Self-concept clarity drops at a breakup and stays lower, predicting distress. Full text read directly 2026-08-24 (verified for our timeline article).
12. Frazier, P., & Hurliman, C. (2001) — romantic breakup among the worst life events; plus "Making Sense and Moving On: The Potential for Individual and Relational Growth Following Romantic Relationship Dissolution" review (PMC6051550) for growth/distress as independent outcomes in emerging adults. Review read directly 2026-08-24.

Methodology

How this was compiled. This article was researched and written on 2026-08-24. The anchor claims were verified against primary sources: the Wortman & Silver 1989 paper (full text read directly), the Maciejewski 2007 JAMA study (abstract + published criticism letters), the Tashiro & Frazier 2003 and Lewandowski & Bizzoco 2007 abstracts, and the PMC review on growth following dissolution (read directly). The Kübler-Ross origin claim is corroborated through the Wikipedia "Five stages of grief" article and multiple secondary accounts; the 1969 book itself was not read in full, so the origin is stated as the model's documented history rather than a direct quotation. Timeline evidence (O'Hara 2020, Eastwick 2008, Chong & Fraley 2025, Seraj 2021, Slotter 2010) was verified in full when we produced our timeline article on 2026-08-24 and is reused here.

What is not claimed. We are not claiming the five stages have no value — they name real emotions. We are claiming they are not a validated sequence for breakup recovery, which the research supports. We are not claiming every person experiences every phase described in section 4. No first-hand clinical experience is claimed; the author is an editorial lead, not a therapist. If you are in crisis, call or text 988 (US) or your local emergency services — this article is not crisis care.

Update schedule. This article will be reviewed monthly against the cited research for changes or corrections, consistent with the site's freshness policy for recovery content.