Why the decision to stay or leave so rarely arrives as a decision

The conventional explanation is fear: people stay because they are frightened of the alternative. That is sometimes true and it is nowhere near sufficient. The deeper reason the decision resists being made is that it is not really one decision, and the parts do not resolve at the same time or point in the same direction.

There is the question of the relationship as it is now. There is the question of the relationship as it might be if something changed, which requires an estimate of another person’s capacity to change and of your own tolerance for waiting to find out. There is the practical arithmetic of housing, money and geography. There is what it would mean for children, if there are children. There is what it would mean about you — about the years already spent, about the promises made in front of people, about the version of your life you had assumed you were living.

Each of those is answerable, slowly. What is not answerable is all of them at once, which is what people attempt at three in the morning. The sense of impossibility is not a failure of nerve. It is the accurate feeling of trying to solve six problems simultaneously with incomplete information about at least four of them.

There is a further complication. Deciding requires being able to think, and the condition most people are in by this point — poor sleep, sustained low-level dread, a nervous system that has been braced for months — is precisely the condition in which thinking is hardest. This is described in more detail in the guide to relationship crisis, and it is the reason The Untangled Method™ works with steadiness before it goes anywhere near a decision.

Money is rarely only money in these conversations. It can stand for safety, freedom, dependence or the practical question of whether leaving is possible at all. Kajal explores the relationship and emotional patterns around money; she does not provide financial advice, and decisions about property, income or settlement belong with a solicitor or a qualified financial adviser.

There is a term for the particular kind of loss involved here, and it is one of the more useful pieces of vocabulary in the whole field. Ambiguous loss describes a loss without a clear boundary: the person is still present, the relationship is still nominally intact, and yet something has gone that cannot be publicly named or formally mourned.

Its defining feature is that it is unwitnessed. When someone dies there are rituals, an acknowledged occasion, people who know to ask. When a marriage becomes something you no longer recognise while continuing to function from the outside, there is no occasion at all. You attend the same family lunch, answer the same questions about the summer, and go home to a house where the central fact of your life is not spoken about.

This is why so many people in this position describe feeling slightly unreal, and why they often feel they have no right to be as affected as they are. Nothing has technically happened. Losses that cannot be named are harder to carry than losses that can, not easier, and it is worth knowing that the disorientation is a property of the situation rather than a deficiency in you.

Anticipatory grief is grief that starts while the thing being lost is still there. In this context it usually begins at the moment separation becomes genuinely thinkable rather than at the moment it happens, which means many people are well into mourning a relationship they are still inside.

It shows itself in odd, specific ways. A sudden difficulty with the small domestic rituals — the way tea is made, a particular walk — because you have started noticing them as things that will end. An unexpected tenderness towards a partner you are also considering leaving. Weeping at something trivial while remaining composed during the serious conversations.

People frequently read this tenderness as proof that they should stay, and read the preceding month’s certainty as proof that they should go. Neither is proof. Grief moves through affection and resolve in turn, and the movement itself carries very little information about the right outcome. What it does indicate is that something real is being lost, which is worth taking seriously whichever way the decision eventually falls.

The period before a decision has its own texture, and it is frequently harder than the outcome that follows. Life becomes provisional. Plans are made at arm’s length. Invitations for next summer feel like a question rather than an invitation. Both people may be doing this at once, without either saying so.

The strain here is not primarily sadness; it is the cost of holding every future open simultaneously. Nothing can be adjusted to, because nothing has settled. It is why people in this stage often report that they could cope if they simply knew, and why some make abrupt decisions largely to end the suspense — a route that tends to produce its own regrets, because the decision was made to relieve the uncertainty rather than on its merits.

Two things help, and neither is a resolution. The first is separating the practical unknowns from the emotional ones and answering only the practical: what the housing situation would actually involve, what the finances actually look like. Vague dread is heavier than specific difficulty. The second is choosing a horizon — a period during which you are deliberately not deciding, and something particular is being tested — so that the waiting has a purpose instead of being an indefinite condition.

Both sentences are said with equal conviction, often by people in almost identical situations. The first holds that children need the household intact. The second holds that children should not grow up learning what this marriage is teaching them about how adults live together.

The evidence is less obliging than either. What emerges most consistently is that the structure of the household matters considerably less to children’s outcomes than two other things: how much sustained, unresolved conflict they are exposed to, and how workable their relationship with each parent is afterwards. A home that stays together in hostility does not protect children. A separation conducted with care, where both parents remain present and the child is not asked to carry adult information, is not the catastrophe it is often imagined to be.

That is genuinely not an argument for separating. It is an argument for moving the question. “Should we stay together for the children” has no general answer. “What are the children currently living inside, and what would we each need to do to make either outcome bearable for them” can be answered, by these particular parents, about these particular children. The wider pressures around this — extended family, community and cultural expectation — are taken up in the chapter on family, parenting and cultural pressures.

One further point is worth stating plainly, because it is so often left unsaid. Children usually know. Not the details, but the atmosphere — the tone of the house, the careful politeness, the rooms people leave. The choice is rarely between protecting them from knowledge and exposing them to it. It is more often between an unexplained atmosphere and an honestly managed situation.

Where the question is not shared equally, two very different experiences run in parallel under the same roof, and each is routinely misread by the other.

The asymmetry of time is the crucial part. One person has had two years to arrive somewhere; the other is at day one on the day it is finally said. Expecting a shared pace at that point is one of the most reliable ways to turn a painful separation into an embittered one.

It is also why protective silence so often backfires. Not saying anything until you are certain is meant kindly, and it removes from the other person any opportunity to understand, to respond, or to have been part of what happened. Where the relationship might survive, that opportunity is the thing most likely to matter; where it does not, it is what allows the ending to be borne without contempt.

Guilt and shame get used interchangeably here and they are not the same. Guilt is about an act: I am doing something that will hurt someone who has not deserved it. Shame is about a self: I am the kind of person whose marriage ended. The first is painful and largely proportionate. The second distorts thinking badly, because it turns a decision about a relationship into a verdict on a life.

Shame is also where other people arrive. Almost everyone in this position is conducting a silent argument with an audience — parents, a faith community, colleagues, a version of themselves at twenty-five who believed they would not be here. Some of that audience is real and will have views. A good deal of it is imagined, and the imagined portion is usually harsher and more unanimous than any real group of people would be.

It is worth noticing which is which, because they call for different things. Real consequences within a family or community deserve to be thought about seriously and planned for. An internalised jury does not deserve to make the decision, and it frequently does.

“What if I regret it” is the question that keeps most people still. It deserves a straight answer: yes, regret is possible, in both directions, and no process can rule it out. People regret leaving. People regret the eleven years they spent not leaving. The fantasy of a risk-free option is the single most immobilising idea in this whole subject.

What can be influenced is not the outcome but the ground. When people look back with least regret, it is rarely because they chose correctly in some cosmic sense. It is because they can say three things: that they understood their own contribution rather than only their partner’s, that the relationship was given a fair test rather than a resentful one, and that they behaved in a way they can still recognise as themselves.

Those are all available before the decision, and all three are more achievable than certainty. It is a reframing worth taking seriously: the aim is not to find the choice that cannot be regretted, but to reach a point where either choice could be lived with.

This is the central claim of the chapter, and it is worth stating directly. People wait for certainty in order to act, and certainty does not turn up first. What turns up first is clarity: an accurate, unembellished account of what the relationship is, what has been tried, what each person is genuinely able to offer, and what you would be choosing in either direction.

Clarity is unglamorous and it is achievable. It comes from getting the picture properly described rather than argued about — from noticing that the relationship you are weighing is sometimes the one from eight years ago and sometimes the one from last Tuesday, and that they are not the same relationship. Once the account stops shifting, most people find that the decision has quietly resolved itself without ever being made.

This is also why so much of the useful work happens before any question of staying or leaving is addressed. Steadying comes first, because a person who cannot sleep cannot think. Understanding comes second. The decision, when it comes, tends to arrive as a recognition rather than a choice — which is a considerably better foundation than a resolution taken at midnight to end the suspense.

Everything above assumes a relationship that is unhappy, exhausted or in question, and that both people are able to think and speak freely. Where there is fear, intimidation, control or violence, the emphasis on unhurried reflection does not apply in the same way, and this page is not the right source of help.

Safety comes before relationship work of any kind, and specialist services are far better placed to help than a general counselling conversation. In the UK, the National Domestic Abuse Helpline, run by Refuge and Women’s Aid, is 0808 2000 247 and operates twenty-four hours a day. The Men’s Advice Line, run by Respect, is 0808 8010 327. In an emergency, the number is 999. Nothing written elsewhere on this site should be read as a reason to delay contacting them.

Where the answer turns out to be that the relationship should end, there is a second question immediately behind it, and it receives far less attention than it deserves: what kind of ending is this going to be.

The difference between separations is enormous, and much of it is decided in the first few months. Endings conducted with some care leave two people who can be in a room at a child’s wedding, who can make a practical arrangement without a solicitor translating it, and who do not have to revise their entire shared history into a story about a villain. Endings conducted badly cost years, and where there are children they are paid for twice.

Support at this stage is not a rescue attempt. It is a place for two people to say the things that would otherwise be said through lawyers or through children, to understand what happened well enough that neither has to carry a distorted version of it, and to agree how they will speak about each other from now on. Some couples do this together; others do it separately, which is often more realistic. Ending well is genuine work, and it is worth as much attention as the deciding was.

A long relationship organises more than a household. It organises the week, the friendships, the assumptions about the next decade, and a surprising amount of a person’s sense of who they are. When it ends, what people frequently report is not only sadness but disorientation: not knowing what they like, or how they want their evenings to go, or who they are when nobody is expecting a particular version of them.

That stage is uncomfortable and it is not permanent. It is also, for a good number of people, where something genuinely new begins — not the triumphant reinvention of the magazines, but something smaller and more durable: a clearer sense of what they will and will not live with, and a better instinct for what a workable relationship actually looks like.

It is worth saying that this stage is real work rather than simply time passing, and it is usually done alone rather than as a couple. Support for it is described in Come Alone, and the wider question of what changes so that the next relationship is different rather than merely new is taken up under personal growth and healthier relationships.

How do I know whether to stay or leave?

Usually not by asking the question directly. People who ask it repeatedly are often trying to reach certainty before they have reached clarity, and certainty is the later of the two. The more answerable questions are what this relationship is actually like now rather than at its worst or its best, what has already been tried and under what conditions, and what would have to change for staying to be tolerable rather than merely survivable. A decision tends to become obvious once those are answered, and rarely before.

Why can I not seem to make up my mind?

Because most people in this position are not indecisive; they are holding two accurate accounts at once. The relationship genuinely does contain the history, the affection and the ordinary life you would lose, and it genuinely does contain whatever brought you to the question. Oscillating between them is what an honest mind does with incomplete information under sustained stress. It is a stage rather than a character flaw, and it usually resolves as the picture fills in.

Should we stay together for the children?

The research is more nuanced than either slogan allows. What appears to matter most to children is not the household structure but the level of sustained, unresolved conflict they are exposed to and the quality of their relationship with each parent afterwards. A low-conflict home that stays together and a well-handled separation both do far better than a home held together in hostility. That is not an instruction in either direction; it moves the question from whether to separate towards how the adults conduct themselves either way.

Is it normal to grieve a relationship that has not ended?

Yes, and it is one of the least recognised experiences in this whole subject. Anticipatory grief begins as soon as the possibility becomes real, which means many people are mourning a marriage while still living inside it, making the packed lunches and attending the birthdays. It is disorienting because there is no acknowledged occasion for it and nothing anyone can offer condolences about.

How long should I give it before deciding?

There is no defensible number, and anyone offering one is guessing. What is worth doing instead is being specific about what the time is for. Time spent waiting for a feeling to arrive tends to pass without changing anything. Time spent testing something particular — a change that has actually been attempted rather than agreed to in principle, a period of steadiness, an honest conversation that has not yet happened — produces information, and information is what eventually makes the decision.

What if I regret it?

Regret is possible in both directions and cannot be designed out. What can be influenced is the ground the decision was made on. People tend to look back with least regret when they can say they understood their own part, gave the relationship a fair account of itself, and behaved in a way they still recognise. Those are within reach even when the outcome is not.

Can counselling help if we are probably separating?

Often more than couples expect. Support at this point is not an attempt to talk anyone out of anything. It is a place to think without the conversation collapsing, to separate what is being decided from what is simply being feared, and — where a relationship is ending — to end it in a way that is bearable for both people and for any children. Ending well is real work and is not the same as ending quietly.

What if the relationship is not safe?

Where there is fear, intimidation, control or violence, this page's emphasis on unhurried reflection does not apply in the same way, and the priority is safety and specialist support rather than relationship work. In the UK, Refuge and Women's Aid operate the National Domestic Abuse Helpline on 0808 2000 247, Respect runs the Men's Advice Line on 0808 8010 327, and in an emergency the number is 999. Please treat that as more important than anything else written here.

If this chapter has made your own position more legible — the oscillation, the grief for something that has not ended, the sense of being unable to decide something you have been thinking about for years — then it has done most of what a page can do. A great deal of the additional distress here comes from believing that other people manage this more cleanly. They do not.

Nothing on this page has an interest in which way you go. If it would help to think it through with someone who has no stake in the outcome, the free Discovery Call is a short, unhurried conversation about where things stand and what kind of support might suit. You do not need to have decided anything, you do not need your partner’s agreement, and you are very welcome to come on your own.