Our firm ran a restructure in March. For about three weeks I did not know whether I still had a job.

I told my wife the evening it was announced, and a few more times over the following fortnight. She was good about it. It got talked through properly, at the kitchen table, more than once. Then in April, when it had resolved and I still had the job, I worked out something that had taken me a month to notice: she was the only person who knew. Not the person I had leaned on hardest. The only person who knew it had happened at all.

Loneliness inside a good marriage is what it looks like when one relationship is quietly doing the work that six or seven used to do, and doing it well enough that nothing appears to be wrong. There is no fault in it and no failure in it. There is just a very small distribution list.

I want to be careful with this subject, because almost every version of it I read while trying to understand my own case came out sounding like a complaint about wives. Mine is not that, and the research does not support that. It supports something narrower and more useful.

What I had assumed, and had backwards

My working theory was that I was overloading her.

It is the standard story, and I had absorbed it without checking: men have nowhere else to put anything, so they put all of it on their partner, and she carries a load her husband's friends should be sharing. I went into the numbers half expecting to find that I was a statistic in that story and should feel slightly bad about it.

The numbers say something different, and the difference matters, because the two versions lead to completely different things to do on a Tuesday.

The one figure that changed how I think about it

Pew Research Center's Men, Women and Social Connections, published in January 2025 from a survey of 6,204 US adults fielded between 3 and 15 September 2024, asked people who they turn to for emotional support.

Men and women are equally likely to name a spouse or partner as someone they turn to for emotional support: 74 per cent and 74 per cent.

That is the whole thing, in one line. It does not tell us how much is being asked of her, because naming somebody is not a measure of how hard you lean on them. What it does tell us is that men are not unusual in naming her at all. The difference is in what else gets named.

Every other column in that table is where men come up short:

  • A close friend: 38 per cent of men, 54 per cent of women.
  • Their mother: 42 per cent against 54 per cent.
  • Another family member: 26 per cent against 44 per cent.
  • A mental health professional: 16 per cent against 22 per cent.

So the deficit is real, and it sits in how little stands behind the marriage rather than in how much is asked of it. Nobody's wife is being asked to carry an unusual amount. She is simply, in a lot of houses, the only one being asked at all.

Being the first call

There is a second finding that puts a finer point on it.

The Survey Center on American Life's The State of American Friendship, from its May 2021 American Perspectives Survey of 2,019 adults, asked who people go to first with a personal problem. Among married people, 85 per cent of men named their spouse first, against 72 per cent of married women.

Across all adults, 53 per cent named a spouse or partner first and 16 per cent named a friend. The report puts the friend figure at 26 per cent in 1990, though that comparison is drawn from a different survey three decades earlier, so I would treat the direction as solid and the exact size of the fall as approximate.

Thirteen points is not a scandal. It is a small, consistent tilt: married men are more likely than married women to have their spouse be the first stop, and the survey above says they are considerably less likely to have a second one.

Isolated is not the same as lonely

The Australian data makes this easier to see, because it separates the two things.

The AIHW's reporting on social isolation and loneliness, drawing on the 2024 HILDA survey of more than 17,000 Australians, found men more socially isolated than women, 18 per cent against 12 per cent, while loneliness ran at 15 per cent for men and 16 per cent for women.

Men are more isolated and are not lonelier. Those two facts sit together perfectly well once you notice that the marriage is answering the loneliness question.

And it genuinely is answering it. In the Pew data, 8 per cent of married adults said they felt lonely all or most of the time, against 21 per cent of people living with a partner they were not married to and 24 per cent of unpartnered adults. Among adults aged 50 to 64 the figure was 11 per cent. Pew does not publish the partnered split by sex, so that is a statement about married adults rather than married men. Either way, that is a marriage doing precisely what it is supposed to do.

The married man is just the sharpest version of the pattern. He has the one relationship that answers the loneliness question, and, if he is like I was, very little standing behind it. He will tell a survey he is fine, and he will be telling the truth.

One load-bearing wall

This is the part that made me actually do something.

Kapelle and Monden published a study in the Journal of Health and Social Behavior on 27 January 2024, using 21 waves of HILDA to follow Australians whose relationships ended at or after 50. Widowhood raised loneliness by 1.53 points on their scale for men, against 0.91 for women. Separation raised it 0.87 for men against 0.67 for women.

What struck me was the shape afterwards rather than the size of the jump. Women's elevated loneliness declined over the following years. Men's stayed elevated, in the data, for up to nineteen years.

Read that alongside the Pew table and it stops being mysterious. If the spouse is one of several sources and she goes, the others take the weight. If she is the only source, there is nothing to take it.

I am not writing that as a warning, and it says nothing about anybody's marriage lasting. It is a structural observation. A building held up by one wall is completely fine, right up until the day it is not, and nobody thinks it is an insult to the wall to add a second one.

What I did, and what I did not say

I said nothing to my wife about it for months, which in hindsight was the wrong call. There was nothing noble in the silence. I was dodging a conversation whose subtext sounded like a grievance, when the actual content was closer to noticing the guttering needed doing. When I finally raised it she had already worked it out and had been waiting for me to get there.

The practical part was smaller than the research made it feel. I picked three men I had genuinely liked and had simply stopped contacting, and I started sending them things. Small things. A photo of the pitch where the old five-a-side team used to play. A link to something one of them would find funny. No agenda and no request to meet, at first, because a request to meet is a much bigger ask than most men will answer.

One of the three took about a year to turn into something I would call a friendship. One became an every-few-months coffee. One never replied to anything, and I stopped reading that as information about me.

If the friendships in your case dissolved without anything happening, the mechanism is usually structural rather than personal, and I have set out the five containers that tend to close in your forties. If you want the experience rather than the mechanism, this is what it actually feels like from inside.

One boundary, plainly. I am not a therapist or a counsellor and have no training in any of this. If what you are carrying is flat all the time, present even in company, or has taken the enjoyment out of things you used to like, that belongs with a doctor and not with an article. The social problem is the one I went and worked on, and it is the only one I can speak to.