Key Points:
- Denial, anger, bargaining, depression and acceptance came from Elisabeth Kübler-Ross's work with terminally ill patients facing their own deaths. They were never a map for grieving a loss, and certainly not for divorce.
- When researchers finally tested the model, it mostly didn't hold. In the
Yale Bereavement Study, disbelief was not the dominant first reaction, acceptance was the most endorsed response from the earliest measurement onward, and yearning, not anger or depression, was the dominant negative feeling throughout.
- Grief oscillates. You don't complete anger and graduate to bargaining. You move between loss-focused and life-focused states, sometimes several times in an afternoon.
- The stage model does one thing well: it names five real feelings. Treat it as a vocabulary list, not a schedule.
- The most useful timing finding in that study is a flag, not a stage. Negative grief indicators generally peaked within about six months, and the authors suggested that people still scoring high beyond that point might benefit from further evaluation.
Someone hands you the five stages within about a week of your separation. A friend, a blog, a well-meaning aunt. And it's a relief at first, because a thing with stages is a thing with an end.
Then you notice you're doing them wrong. You hit acceptance in March and denial again in April. You're bargaining and furious in the same hour. Somebody tells you that you must still be in denial, which is a claim that can't be disproven and helps nobody.
So let me clear this up, because the model does more harm than good when it's used as a timeline.
Where the Stages Came From
Elisabeth Kübler-Ross published On Death and Dying in 1969, based on interviews with terminally ill patients. Her subject was how people come to terms with their own approaching death. It was humane, influential work, and it changed how medicine treats dying people.
It was not a study of bereavement. It was not a study of divorce. It wasn't longitudinal, it didn't have a comparison group, and Kübler-Ross herself was clear that the reactions weren't a fixed sequence everyone marches through.
The stages got borrowed anyway, first for grief, then for job loss, then for breakups, then for essentially every unwanted event, and the caveats fell off along the way. Somewhere in the borrowing they turned from "here are five things people commonly feel" into "here is the order in which you will feel them."
What Happened When Someone Tested It
Maciejewski and colleagues ran the
first real empirical test in 2007, following 233 bereaved people for two years and measuring disbelief, yearning, anger, depression and acceptance.
Some of it held up. Rescaled, the five indicators did reach their peaks in roughly the predicted order.
But the substance mostly didn't. Disbelief was not the dominant opening reaction. Acceptance was the most frequently endorsed item, and it was climbing from the very first measurement, which is not what a model that puts acceptance last would predict. And the dominant negative feeling across the whole two years was yearning, which isn't one of the five stages at all.
That last finding matches what I hear in my office almost word for word. People rarely say "I'm angry" or "I'm in denial." They say "I miss the house." "I miss having someone to tell." "I keep reaching for my phone to text him something funny." Yearning is the main event, and the popular model doesn't even list it.
Later work has been more skeptical still. What most grief researchers now describe is oscillation: people move back and forth between confronting the loss and getting on with life, and the movement between those two states is the process itself, not a failure to progress.
The Five Feelings, Used Properly
Keep the five words. Drop the numbering. Here's what each one actually is and what to do when you're in it.
Disbelief is your brain refusing to update a model it spent years building. You reach for the second toothbrush. You plan around a person who isn't there. You catch yourself thinking "when this is over." That's not weakness or avoidance, it's a system rewriting itself, and it gets loudest right after the separation becomes concrete. What helps is contact with facts. Sign the thing. Change the address. Move the furniture. Small acts of physical reality do more than any amount of telling yourself it's real.
Anger shows up when an expectation has been violated, which in a divorce is roughly everything. It's not a defect and it doesn't need to be dissolved. It needs somewhere to go that isn't your ex's inbox. Physical exertion helps genuinely. So does identifying the boundary the anger is demanding and then setting it, calmly, once. What doesn't help: social media, group texts, and any message composed after 9 p.m. Write those, save them as drafts, send them to nobody.
Bargaining is your mind's attempt to find the version of events where this didn't happen. If I'd noticed sooner. If I'd gone to counseling. If I change now, maybe. It's rumination in a hopeful costume. Notice that every bargain requires rewriting the past, which is the one thing that's off the table, and that noticing the structure is more effective than arguing with the content.
Depression in the ordinary sense, meaning the heavy, flat stretch where getting up is a project, is expected and usually temporary. What matters is whether it's grief or a depressive episode. Grief comes in waves, lets pleasure through in gaps, and leaves your self-worth roughly intact. A depressive episode is flat, constant, and turns on the self. Two weeks of most-of-the-day low mood plus real impairment in functioning is the line for getting evaluated.
Acceptance is the most misunderstood word in this list. It doesn't mean you're at peace, you've forgiven anyone, or it stopped hurting. It means you've stopped arguing with the fact. You can hold "this happened, and I hate it, and it's my life now" in one sentence without flinching. And it isn't a finish line you cross once. In the Yale data, acceptance was present early and grew steadily throughout. It's less an arrival than something that accumulates.
The Timing Marker Worth Knowing
If you take one number from the research, take this one, with its caveats attached.
In the Yale Bereavement Study, the negative grief indicators generally peaked within about six months, and the authors noted that people still scoring high beyond six months might warrant further evaluation.
Two caveats matter. That study was about bereavement following a natural death, not divorce, and divorce differs in important ways: the person is still alive, there's often ongoing contact and litigation, and the losses keep arriving in installments rather than all at once. Second, "warrants a look" is not "you're doing it wrong." Grief has no deadline.
Still, I use six months as a soft prompt, and I'd rather you have a prompt than an open-ended permission to wait. If you're a year out and it's exactly as heavy as month one, that's worth a conversation with a professional.
Why Any of This Matters
Because the stage model, used as a timeline, produces two specific failures I see repeatedly.
The first is people concluding they're grieving incorrectly. They swing from acceptance back into rage and decide they've regressed, which adds shame to grief and slows everything down. Nobody regresses. That's just the shape of it.
The second is people waiting. If you believe the stages will carry you through on their own, you sit still and expect delivery. Recovery is not passive. The people who come out of this in reasonable shape are doing things: keeping a structure, seeing humans on a schedule, moving their bodies, interrupting the replay loop, getting treatment when it's warranted. Time is a condition, not a mechanism.
Feel whatever shows up, in whatever order it shows up. Then get up and do the next thing on the calendar.
This article is for general informational purposes and is not psychological or medical advice, and it does not create a treatment relationship. If you're struggling with thoughts of suicide or self-harm, the
988 Suicide & Crisis Lifeline is available by call or text at 988, 24 hours a day.
Frequently Asked Questions
What are the five emotional stages of divorce? The five commonly cited are denial, anger, bargaining, depression and acceptance, borrowed from Elisabeth Kübler-Ross's 1969 work on terminally ill patients facing their own deaths. They describe five feelings people genuinely have during a divorce. They do not describe a sequence, and treating them as one causes more trouble than it solves.
Do you have to go through the stages in order? No. When the model was tested empirically, acceptance was present and rising from the earliest measurement, disbelief was not the dominant first reaction, and the strongest negative feeling throughout was yearning, which isn't one of the five. Grief oscillates between confronting the loss and getting on with life, often several times a day.
How long do the emotional stages of divorce last? There's no fixed timeline. In bereavement research, negative grief indicators generally peaked within about six months, and the study's authors suggested people still scoring high beyond that might benefit from further evaluation. Divorce differs from bereavement in that the person is still present and the losses arrive in installments, which can stretch things out.
Can you go back to an earlier stage? There's nothing to go back to, because they aren't stages. Feeling calm in March and enraged in April is the normal pattern, not a relapse. Anniversaries, the final decree, a new partner appearing, and children's milestones all reliably reopen feelings you thought were finished.
What does acceptance after divorce actually mean? That you've stopped arguing with the fact of it. Not peace, not forgiveness, not the absence of pain. You can say "this happened, I hate it, and it's my life now" without flinching. It builds gradually rather than arriving, and research suggests it starts building far earlier than the stage model implies.
How do I know if my sadness is normal grief or depression? Grief comes in waves, tends to have triggers, allows moments of pleasure in between, and leaves your sense of self-worth largely intact. Depression is flatter and more constant, blocks pleasure across the board, and turns on the self with thoughts like "I'm worthless." If low mood or loss of interest has lasted most of the day, most days, for two weeks or more with real impairment in functioning, get a professional evaluation.
Related reading:
Divorce and Depression: Telling Grief From Something That Needs Treatment |
Emotional Neglect in Marriage: The Loneliest Way to Not Be Alone |
Why Am I So Angry? The Emotions of Divorce |
10 Signs You're Ruminating on Your Ex and How to Stop |
How to Tell Your Kids About Divorce: An Age-by-Age Guide |
Moving Past Heartbreak
