Y11W27RC Repairing after fights

This week’s reading focuses on relationship recovery.


Stage 1 of 4

Prior knowledge activation

  • After a conflict with someone you care about, what helps the relationship feel restored?
  • Do you think avoiding fights or recovering well from them is more important?

Stage 2 of 4

Purpose-setting statement

This article focuses on relationship recovery. It argues that the key to relationship health isn’t avoiding conflict—it’s recovering well from it. You’ll learn about repair attempts, non-verbal communication, and the rupture-repair cycle that successful couples follow.


Stage 3 of 4

Prediction or discussion prompt

What matters most for relationship longevity: (A) Avoiding conflict (B) Fighting fairly (C) How well you repair after conflict (D) Not getting angry?

As you read, track: What repair attempts are, examples of successful repairs, and why recovery matters.


Stage 4 of 4

A question to carry into the reading

Notice how this article shifts the focus from ‘what’s wrong in conflict’ to ‘what’s right in recovery.’ It’s building on Gottman’s earlier work but emphasizing the active skill of repair rather than just avoiding the four horsemen.


Now read

Repairing after fights

~14 min read · ~2,100 words

Two people who care about each other will fight. This is not a failure of the relationship. It’s a feature of the fact that any relationship close enough to matter will eventually collide with the edges of two separate people’s needs, expectations, histories, and bad days. The research on long-term relationships — couples, families, close friendships, long working partnerships — is not really about whether fights happen. They happen in almost every relationship worth having. What the research is about is what happens after.

Here is the finding that matters most, from a generation of studies on how relationships survive or don’t. It’s not how often you fight. It’s not even how well you fight. It’s how well you repair.

Couples who regularly have difficult conversations and recover well from them do better, in the long run, than couples who argue rarely but never quite get past the arguments they do have. The specific skill of repair — of getting back to a good place after something has gone wrong — is, across the research, one of the strongest single predictors of whether a close relationship lasts and flourishes.

The repair research

The most detailed research on this specific question comes out of the same laboratory we looked at in the four-horsemen article — John Gottman and Julie Schwartz Gottman’s work at the University of Washington. Gottman observed, early in his research, that couples who stayed together weren’t couples who avoided conflict. They were couples who, in the middle of conflict, made specific moves that interrupted the escalation.

He called these repair attempts, and he catalogued them carefully. Some were verbal: explicitly acknowledging the other person’s feelings, admitting fault, suggesting a break, offering a compromise, asking for clarification, shifting the frame of the argument. Some were non-verbal: a softer tone, a reduction in intensity, a touch, a small smile, a physical reset. A few were idiosyncratic: an inside joke, a silly phrase that both parties understood meant I’m still with you even though we’re fighting, a signal that both knew to take as a peace offering.

What distinguished the couples who stayed together was not the presence of these repair attempts — many couples made them — but whether they were received. In distressed relationships, repair attempts were frequently ignored, dismissed, or used as fresh ammunition. In stable relationships, even imperfect repair attempts got acknowledged, and the acknowledgement itself began to defuse the conflict. The repair wasn’t magic. It was just: one person tried, and the other person let it work.

This is a small thing, and it’s also most of the research. A relationship in which repair attempts are made and received can absorb enormous amounts of conflict. A relationship in which repair attempts are made but rejected — or aren’t made at all — tends to deteriorate even under modest stress.

What makes a repair attempt actually work

The research on what makes repair effective, drawn from the Gottmans and from related clinical traditions, converges on a few specific elements.

Timing matters. Repair attempts made too early in an argument, before the other person is ready to receive them, often fail. Repair attempts made far too late, after the argument has fully escalated and both people are physiologically overwhelmed, also fail. There’s a window — often short, often identifiable by emotional intensity — where the repair can land. Skilled couples develop a sense of when that window is open.

Specificity matters. Vague repair — I’m sorry for whatever I did — tends to land poorly. Specific repair — I shouldn’t have raised my voice when you were telling me about the meeting; I was tired and I took it out on you — tends to land much better. The difference is not just performative. Specificity communicates that you’ve actually understood what went wrong, not just that you want the conflict to end.

Turning toward rather than turning away matters. A repair attempt that maintains connection — a small physical gesture, the use of the other person’s name, eye contact — works better than a repair attempt that accompanies emotional withdrawal. The repair isn’t just the words; it’s the signal that the underlying connection is still there. A perfect verbal apology delivered while looking away or leaving the room often fails. A clumsy verbal apology delivered while visibly present and attentive often works.

Taking responsibility without over-explaining matters. There’s a specific pattern where an apology gets undone by the explanation that follows it. I’m sorry I was short with you, but it’s because I’ve had such a hard week. The but cancels the apology. A cleaner version would be: I’m sorry I was short with you. I’ve had a hard week, and it came out sideways at you, but that doesn’t make it okay. The structure is subtle but the effect is large. The first version sounds like an apology; the second version is one.

The anatomy of an effective apology

A separate tradition of research — drawn mostly from a psychiatrist named Aaron Lazare at the University of Massachusetts — has examined what makes apologies actually repair rather than inflame.

Lazare’s research, summarised in his 2004 book On Apology, identified four components of a genuinely effective apology, any of which can be missing and cause the apology to fail:

Acknowledgement — explicitly naming what you did and recognising its impact. Not defending it, not explaining it, not softening it. Just naming it clearly.

Explanation — offering, when possible, a context that helps the other person understand the offence without excusing it. This is optional and often best kept brief, because it tips easily into defensiveness.

Remorse — communicating, in a way the other person can receive, that you genuinely feel bad about what happened. The absence of remorse is often what people mean when they say an apology didn’t feel sincere.

Reparation — offering or taking some action that addresses the harm. This doesn’t need to be dramatic. For small offences, the reparation can be implicit in the acknowledgement and the resolve not to repeat. For larger ones, something more concrete is usually required — a change in behaviour, a making-good of what was damaged, a sustained effort to rebuild trust.

Many apologies that feel inadequate are missing one or more of these components. I’m sorry you feel that way is missing acknowledgement (of what you actually did) and remorse (about your own behaviour). I’m sorry, but I had good reasons is overweight on explanation and underweight on remorse. I said sorry, what more do you want is missing reparation — the recognition that the apology itself isn’t the complete repair. Understanding the four components lets you diagnose, in yourself and in others, why a given apology is working or not.

The attachment angle

A related and complementary tradition comes from the Canadian psychologist Sue Johnson, whose development of Emotionally Focused Therapy has drawn on attachment theory to understand what happens in conflicts that feel worse than the content of the conflict should warrant.

Johnson’s observation is that many fights between close partners are, underneath the surface argument, about attachment security — the deep and often unconscious sense that the other person is emotionally available and will respond when needed. When one partner feels attachment-insecure in a moment of conflict, even a minor disagreement can trigger what Johnson calls an attachment injury — a wound that’s out of proportion to the specific incident but that connects to deeper fears about whether the relationship can be counted on.

The repair for attachment injuries, in Johnson’s framework, is not really about the content of the specific argument. It’s about restoring the sense of attachment security. Her clinical protocol includes what she calls Hold Me Tight conversations — structured exchanges where each partner explicitly communicates their attachment needs and the other acknowledges having heard them. These conversations can feel uncomfortably vulnerable for couples who aren’t used to them. They also reliably produce repair at a depth that surface-level conflict resolution doesn’t reach.

Johnson’s work has been influential in couples therapy and is especially useful for relationships where fights seem to keep returning to the same core tension regardless of the specific trigger. If every argument seems to end up at the same underlying place — you’re not really there for me, or you don’t really accept me — the actual repair needed probably isn’t about the current argument. It’s about the attachment tension the argument is surfacing.

The counter-thread that matters

A genuinely important caveat about all this research: not every conflict should be repaired, and some relationships use repair cycles as a mechanism that reinforces harm rather than healing it.

Clinical psychologists working with abusive relationships — particularly those following the work of researchers including Leonore Walker, who developed the concept of the cycle of abuse — have documented a specific and disturbing pattern. An abusive incident occurs. The perpetrator then engages in what looks, superficially, like repair: apology, expressions of remorse, promises of change, sometimes romantic gestures or gifts. The victim, responding to what looks like genuine repair, re-engages with the relationship. The abuser feels relieved of pressure, tension accumulates, and another incident occurs. The repair cycle, in this pattern, is not repair. It’s part of the machinery of continued abuse.

This is important to name, because popular writing on repair can slide into the assumption that any relationship’s troubles can be fixed if both parties are willing to engage in the right conversations. Some relationships genuinely can be. Some can’t. Some shouldn’t. The skill is knowing the difference, which is hard in the middle of it and often requires outside perspective — from friends, from family, sometimes from professionals.

A rough diagnostic: healthy repair is followed by genuine change. The apology is accompanied by behaviour that, over the following weeks and months, is different. If the apologies keep coming and the behaviour never changes, the repair isn’t real, regardless of how sincere the moment of repair felt. The pattern, not the moment, is the information.

What to actually do

For most relationships — the ordinary, worth-repairing kind — a few concrete moves are worth practising.

Notice when you need to repair rather than waiting for the other person to name it. The party that gives way first, even slightly, is usually the one who enables repair. Waiting for the other to apologise first is a common relationship-ender; one person always eventually gives up. Repair that comes slightly sooner than feels comfortable tends to work better than repair that comes slightly later.

Use specific language. Name what you did. Name why it landed as it did. Name what you’ll try to do differently. Specificity is the difference between an apology that lands and one that doesn’t.

Receive repair well. This is as important as offering it. When the other person apologises, try to accept rather than deflecting, minimising, or adding new grievances. An apology received grudgingly and then followed by three more complaints trains the other person not to apologise again. An apology received warmly and then followed by a small repair from your side trains the relationship to resolve conflict fast.

Don’t make every fight a referendum on the whole relationship. Specific arguments should stay specific. Generalising small conflicts into this is what our relationship is like usually intensifies the fight and delays the repair. The practice of saying this particular thing went wrong and we need to address it is more useful than we always do this.

Pay attention to patterns over time. One failed repair is nothing. A string of them is a signal. If the relationship keeps running into the same unresolved tensions, the repair needed is probably at a different level than the one you’ve been attempting.

The question that remains

The deepest thing the repair research teaches is probably that long relationships are not built by people who avoid conflict or even by people who resolve conflict quickly. They’re built by people who have developed, between them, a set of moves that let the relationship survive the inevitable breakages — small and large — that being close to another human produces. Those moves can be learned. They can be practised. They can be improved across years.

And the opposite is also true. A relationship without functional repair, no matter how much love is initially present, tends not to last. Love, alone, is not enough. What makes love sustainable, across decades, is the unglamorous skill of fixing what keeps going wrong.

The question to carry, about a relationship that matters to you:

The last time something went wrong between you — how was it repaired, and would it repair the same way today?

Key research referenced: John and Julie Gottman’s research on repair attempts at the University of Washington; Aaron Lazare, On Apology (2004); Sue Johnson, Hold Me Tight (2008) and Emotionally Focused Therapy research; Leonore Walker’s cycle-of-abuse research on pathological repair dynamics.