Ask a couple how they met, and within about ten minutes you will hear their story of us — the shared account of who they were, what they went through, and what it all meant. It sounds like small talk. It isn’t. In one of the most quietly unnerving findings in relationship science, researchers listened to couples narrate their own history and predicted, three years in advance, which marriages would end. They were right more than nine times out of ten. Not from watching a fight. Just from listening to a story.
The story of us is a diagnostic, not a keepsake
In 1992, Kim Buehlman, John Gottman and Lynn Katz sat 52 married couples down for what they called the Oral History Interview. How did you meet? What was courtship like? What were the good times, and the hard times? No conflict task, no arguing about money in a lab. Just a couple telling their own story, together, out loud.
Coders then rated the narratives on a handful of dimensions — how much fondness came through, how much the couple spoke as a “we” rather than two separate “I”s, how much disappointment leaked in, whether their lives sounded chaotic or purposeful, and whether they seemed to take pride in what they’d survived together. From those ratings alone, the team correctly classified over 94% of the couples as still married or divorced three years later.
That number gets repeated a lot, and it deserves an honest caveat: 52 couples is a small sample, and the model was tested on a population it partly learned from, which inflates accuracy. Later replications with larger samples landed lower. But the direction of the finding has held up across decades of work, and it points at something worth taking seriously: a couple’s narrative is not a neutral record of what happened. It is a live readout of how they currently feel about each other, projected backwards onto the past.
The past doesn’t change. The way we narrate it does — constantly. Photo: Laura Fuhrman / Unsplash.
Memory is not a hard drive. It’s a press office.
Here is the mechanism that makes this more than a party trick. Human memory doesn’t store the past and play it back. It reconstructs the past each time, and it reconstructs it in a way that supports how you feel right now. Psychologists call this mood-congruent recall, and it’s one of the sturdiest effects in cognitive psychology.
So when a marriage is in good repair, the retelling of the past gets quietly edited in the couple’s favour. The disastrous first apartment becomes the funny one. The year he was unemployed becomes “the year we figured out we could handle anything.” When a marriage is corroding, the same raw footage gets re-cut. The proposal that was once romantic becomes “he took forever.” The move for her job becomes “I gave up everything.” Nothing that happened has changed. The soundtrack has.
This is why the story of us is such a good early-warning system. Couples are very good at hiding dissatisfaction when you ask them directly — “how’s the marriage?” gets a socially acceptable answer. But nobody censors the way they tell a story about a holiday from nine years ago. The disappointment slips out sideways, in an adjective.
What a healthy narrative actually sounds like
The signal isn’t happiness. It isn’t even the absence of hard chapters — some of the most stable couples in the research had brutal histories. The signals were structural:
We-ness. The story is told in the plural. “We didn’t have any money.” “We decided to take the risk.” A deteriorating narrative fragments into two parallel accounts, each with its own protagonist: “I was working nights.” “I basically raised the kids.”
Glorifying the struggle. The hardship is framed as proof of the bond rather than evidence against it. The commute, the miscarriage, the immigration paperwork, the awful in-laws — a healthy story treats these as the crucible the relationship was forged in. A failing story treats them as the receipts.
Expansiveness. Detail is offered freely. Someone lights up and adds colour to the other’s version rather than clipping it short. Terse, minimal answers — “it was fine, we met at work” — often signal that the well of fondness has gone dry, not that the person is shy.
Fondness and admiration. Small, specific compliments survive in the telling. She was so funny. He was unbelievably patient with my mother. Gottman has long argued that fondness and admiration are the antidote to contempt, the single most corrosive thing that can enter a marriage.
How the story of us goes bad without anyone noticing
Narratives rarely collapse dramatically. They get revised by a thousand tiny amendments, usually under the pressure of unresolved daily friction — the argument you keep half-having, the resentment that never quite gets named, the accumulation of invisible labour nobody has thanked anyone for.
What happens is that the couple’s ratio of positive to negative feeling flips. Gottman’s lab described the tipping point as negative sentiment override: a state in which neutral behaviour starts getting read as hostile by default. Once you’re there, the historical record gets audited with the same suspicious eye. Old evidence of love is downgraded to “he was just being charming.” Old evidence of neglect gets promoted to “the pattern was always there.”
The most sobering part is that this rewrite feels like insight. It arrives as a sudden clarity — “I see it now, I was fooling myself for years.” It is very hard, from the inside, to tell honest reappraisal apart from a narrative that has simply been recoloured by present unhappiness. That’s not an argument for staying in a bad marriage. It’s an argument for noticing that the story is moving, and asking why.
Fondness survives in how a story is told, not in whether it was a happy one. Photo: Omar Lopez / Unsplash.
You can audit your own narrative in twenty minutes
Do this with your partner, out loud, and resist the urge to turn it into a state-of-the-union conversation. Take turns. Don’t correct each other’s facts.
- How did we meet? Tell it as if to a stranger who has never heard it.
- What do you remember about the early days that you liked about me?
- What was the hardest year we’ve had — and how did we get through it?
- Why do you think some marriages don’t make it, and what’s kept ours going?
Then listen to the shape of it, not the content. Did either of you say “we”? Did anyone volunteer a detail nobody asked for? Did the hard year come out as something you survived together, or as something one of you did to the other? Did either of you use the word lucky?
If the answers come back thin — clipped, factual, told in the first person singular — that’s information, not a verdict. It usually means there’s unmetabolised resentment sitting between you, quietly editing the archive. The productive move is not to argue about whose version is accurate. It’s to go find what the resentment is attached to. Very often it’s a specific injury that never got properly repaired, and repairing it late still works: our guide to repairing after a fight covers how those conversations actually go.
Can you rewrite it on purpose?
Partly, and this is where the research gets practical rather than fatalistic. A well-known study led by Eli Finkel had couples spend just 21 minutes a year writing about a recent conflict from the perspective of a neutral third party who wants the best for everyone — seven minutes, three times a year. Couples who did it stopped the decline in marital quality that the control group experienced over the same period.
Twenty-one minutes did not fix anyone’s marriage. What it did was interrupt the automatic, self-serving edit — the one where you are the reasonable narrator and your partner is the antagonist. The story got a second reader.
The practical version, applied to your history rather than your last argument: deliberately go looking for the generous reading of an old chapter you’ve been telling bitterly. Not a lie — a different true version. He didn’t move you across the country to sabotage your career; he was frightened about money and handled it badly. She wasn’t cold after the baby; she was running on four hours of sleep and nobody had asked her how she was. Both accounts fit the same facts. You get to choose which one you keep telling — and the one you keep telling will shape what you notice tomorrow.
What this means if you’re already at the exit
There is a version of this article that says “just tell a nicer story and your marriage will heal,” and it would be dishonest. Sometimes the darkening narrative is accurate. Sometimes what looks like rewriting the past is really the slow, painful arrival of a truth that was being suppressed — that the relationship was contemptuous, or lonely, or unsafe, all along. A story of us that has curdled into two separate stories is often the honest end of something, and pretending otherwise helps nobody. If you are in that territory, our separation and divorce section is written for people making that decision with their eyes open.
But most couples aren’t there. Most couples are somewhere in the long middle, where the story is drifting a few degrees a year and nobody has looked up from the map. That drift is reversible, and it is much easier to reverse at three degrees than at ninety. The story of us is the cheapest instrument you have for checking your heading. It costs nothing, it takes twenty minutes, and it is very, very hard to fake.
Ask your partner how you met. Then listen to yourselves.
If any part of your relationship history involves fear, coercion or abuse, the framing in this article does not apply — reframing is not a tool for making an unsafe situation feel acceptable. Please reach out to a qualified professional or a domestic abuse support service in your country.
Further reading: Marriage · Building trust in a relationship
Written by
Elena Rostova
Elena Rostova is the Lead Editor and a Relationship Advocate at Relationship-99, where she combines empathetic insight with practical advice to help individuals and couples navigate the complexities of dating, marriage, and family dynamics. She holds a B.A. in Communications and writes professionally on relationships and wellness.