A woman in her late sixties is making coffee on a Saturday morning. Her husband of forty-one years sits at the kitchen table reading the paper. She knows how he takes the coffee. She knows the cup he prefers, the order in which he reads the sections, the small noise he makes when something irritates him in the news. She could anticipate his next ten movements without looking up. He could do the same for her.
This is not the same thing as knowing what he is actually thinking about, this morning, while he reads. It has been some years, and she could not say how many, since she felt sure she knew that. She does not, in any given moment, particularly mind. The not-knowing has become its own kind of furniture. It is when she stops to notice it that something in her chest tightens slightly, and she does not always stop to notice it, because nothing follows from the noticing that she would know how to do.
What follows is an essay about a pattern many long marriages settle into, and what the research does and does not say about it. It is not a diagnosis of any particular marriage, and it is not advice. People who recognise themselves in what follows will, mostly, have already recognised it before reading this.
What the research actually shows
The popular story about long marriages is the U-shape: high satisfaction early, a long dip through the child-rearing years, recovery in late middle age. The dip is real. The recovery, on closer reading, is more complicated than the popular version suggests.
The cleanest single piece of research on the recovery is Sara Gorchoff, Oliver John, and Ravenna Helson’s 2008 paper “Contextualizing Change in Marital Satisfaction During Middle Age,” in Psychological Science. The authors followed 123 women across 18 years, from their early forties to their early sixties, and tracked how their marital satisfaction changed. Satisfaction did rise. The empty-nest transition was a meaningful contributor. The detail that matters for the present piece is what the rise was made of.
The improvement was not driven by couples spending more time together. The amount of time together hardly mattered. What rose was the women’s enjoyment of the time they did spend with their partners. The active ingredient was the quality of the contact, not the quantity. Late-middle-age marriages got better because the moments together became more pleasant, not because there were more of them.
This finding leaves a particular question unanswered. Pleasant company is not the same as known company. A couple can enjoy each other’s presence on a Saturday morning, after decades of practice, without either of them currently knowing very much about the other’s inner life. The Gorchoff study does not distinguish between the two, and the broader literature on long-term marriage has often conflated them. The popular story of late-life marital satisfaction usually slides over this distinction. The thing the title is pointing at lives in the gap.
The two competences that drift apart
Long marriages produce two competences that look similar from inside the marriage and are actually different.
The first is the competence of practical familiarity. After enough years, a spouse can predict a great deal of the other’s behaviour. The taste in restaurants. The mood at the end of a particular kind of week. The small physical signs that something is wrong. The reliable bad jokes. The order of operations in a kitchen, in a car, in a hospital waiting room. This competence is real and is one of the genuine pleasures of a long marriage. It is the kind of knowing that makes a long-married couple look, from outside, like a single organism.
The second is the competence of current intimacy. This is the much smaller and more demanding skill of knowing what the other person is actually thinking about, now, in the parts of their life that the household routines do not surface. Whether they are afraid of something they have not mentioned. What they make, on reflection, of how the years have gone. What they are quietly grieving. What they would do differently if they could. This competence is not automatic. It requires the same kind of curiosity that produced the original courtship, applied to a person the courting partner already thinks they know.
The two competences can drift apart over a long marriage without either spouse noticing. The first competence keeps growing. It compounds with every year. The second can plateau early and then quietly decline, because most of the conditions that would force a couple to keep practising it are gone by the time the marriage is twenty years old. They are not flirting any more. They are not negotiating who they are to each other. They are managing a household, raising children, dealing with parents, dealing with money. The questions that once required real listening have been settled.
The household runs. The two of them are warm with each other. Neither of them is currently asking the other what they think, because neither of them is currently expecting to be surprised.
What the title overstates, and what it gets right
The title’s strong form, that what the couple called intimacy was really just two people who had stopped being curious at the same time, overstates in one direction. Comfortable familiarity is not nothing. The forty-year practical knowledge is its own kind of love, and many long-married couples who would not currently pass a curiosity test about each other would still walk through fire for each other in any concrete situation. The Gorchoff finding, that the late-life rise in satisfaction is real, is consistent with this. Couples who have stopped being curious have not necessarily stopped loving.
What the title gets right is narrower and harder to argue with. The two competences are different. A marriage can be warm, comfortable, mutually supportive, and reliable across decades, and the partners can know each other’s habits down to the millisecond, and still have very little current information about what is happening inside the other person. The discovery that this is the case, when it arrives, is usually not the discovery of a fraudulent marriage. It is the discovery of an asymmetry between two real things that the couple had been calling by the same word.
What the discovery actually looks like
The moment of recognition rarely arrives as a dramatic crisis. In the marriages where it arrives at all, it usually comes quietly, in some unremarkable interval. The husband mentions, in passing, a thought about his father that the wife has never heard before, in forty years. The wife is asked by a friend what her husband is like, and finds that the answer she gives is mostly a list of habits. One of them is unexpectedly funny in a way the other has not seen in a long time. A grandchild asks a question that neither of them knows how the other would answer.
What happens next varies. Some couples treat the recognition as information and do something with it. They begin, slowly, asking each other questions they have not asked in decades. The asking is awkward. The first answers are short. Both of them are out of practice. After a few months of trying, in the couples who keep trying, something shifts. The household still runs. The habits are still the habits. There is also, now, a thin layer of current information moving between them, of the kind that used to be there in the early years and that they had not realised had thinned.
Other couples notice the recognition and let it pass. The marriage continues in its current form, which is companionable and warm and, by most external measures, a good marriage. The interior loneliness inside it is real but is not, for either of them, the kind of loneliness that justifies action. They have other things to be doing. The recognition becomes part of the furniture of the marriage, sitting alongside the cup he prefers and the order in which he reads the sections.
Neither response is wrong. The Gorchoff data suggests that even the second kind of marriage is, on average, getting more pleasant rather than less in late life. The first kind, where the discovery becomes a project, is rarer and more demanding and is not always the right call for every couple. People who try to retrofit intimate curiosity into a forty-year marriage sometimes find that the marriage prefers what it had, and that the project disturbs more than it heals.
If the recognition is heavier than the piece describes
For most readers who recognise the pattern, the recognition is the lighter version: a thing noticed, considered, set down. For some, it is heavier. If the recognition is paired with a sustained sense that the marriage is no longer one the person wants to be in, with persistent low mood, or with a feeling that the gap between the two people has become unbearable rather than merely visible, the right resource is a couples therapist or counsellor, not an essay. The patterns described above are at the population level. What is happening inside any particular long marriage is not, and the work of figuring it out belongs in a setting designed for it.
What the woman at the kitchen table is doing, when she stops to notice, is not making a discovery that her marriage was a mistake. She is noticing that two competences have drifted apart inside it, and that the one she practised for forty years is not the one she is now wishing she had more of. Whether she does anything about that, this Saturday or in the next ten years, is up to her and to the man at the table. Most of the time, on most Saturdays, she will pour the coffee, hand it to him in the cup he prefers, and the noticing will not happen at all. That is also part of what a long marriage is, and the part the popular version of the story tends not to mention, because it is harder to sentimentalise than the parts about practical love.
