In a 2023 Pew survey, 8 per cent of US adults said they had no close friends; that number describes a social network, not a personality, and adult transitions can dismantle one faster than it is rebuilt

In a 2023 Pew Research Center survey, 8 per cent of US adults said they had no close friends. It is a small enough figure to be treated as an exception and a large enough one to represent millions of ordinary lives.

What bothers me is how quickly the phrase “no friends” gets converted from a description into a verdict. It stops meaning that a person’s current social network contains no relationship they would call a close friendship. It starts meaning difficult, cold, unlikeable, too much, not enough, or somehow built incorrectly.

A survey cannot tell us why any one person has no close friends. The wider research suggests something more modest and humane: social networks change with age and circumstance, and the repeated settings that create friendship can disappear faster than close relationships are replaced.

Eight per cent is not one kind of person

Pew surveyed 5,057 US adults in July 2023. Fifty-three per cent reported between one and four close friends, 38 per cent reported five or more, and 8 per cent reported none.

Those figures depend on what respondents meant by “close friend”. Pew did not supply a universal test of closeness, and it did not ask the group with no close friends whether they were satisfied with that situation. It also measured one country at one point in time. The finding is useful, but it is not a personality map.

One person in that 8 per cent may have moved six months earlier. Another may have watched an old friendship fade after a job change. Someone may have many acquaintances but nobody they trust with the unpolished parts of life. Another may be content with family relationships and little interest in adding a friendship simply to improve a count.

I think those differences matter because “no friends” sounds total. In practice, it can describe a temporary gap, an enduring preference, a painful absence, or a network full of people who do not quite cross the person’s own threshold for the word.

Networks contract even when nobody has failed

A 2013 meta-analysis published in Psychological Bulletin combined 277 studies involving 177,635 participants. Across both cross-sectional and longitudinal research, the authors found that global social networks tended to expand until young adulthood and then decline. Friendship networks, specifically, decreased across adulthood, while family networks were more stable.

The authors also found that life events were associated with network change. Entering work, becoming a parent and losing a spouse were among the transitions examined. That does not mean these events mechanically remove friends from every life. It means friendship exists inside a calendar and a geography, not outside them.

School repeatedly puts the same people in the same room. University can do the same. Early jobs may produce lunches, commutes and evenings when plans require little negotiation. Later, work becomes remote or demanding, people move, partners and children reshape the week, and an ordinary coffee can require three calendars and a fortnight’s notice.

None of that proves a person is good at friendship. It does show why counting the relationships left at the end cannot reveal character by itself.

Friendship is unusually easy to postpone

Many adult obligations have visible deadlines. A bill is due. A meeting begins. A child needs collecting. A parent needs help. Friendship often arrives as the thing that matters but can move to next Thursday without immediate consequence.

The consequence appears slowly. Two people still think of each other warmly, but the shared present becomes thinner. Their knowledge of one another freezes around an old workplace, an old city or a version of life that no longer exists. Neither person decided to end the friendship. Repeated postponement did the work.

This is close to what I was trying to understand when I wrote about the time it can take for adult friendship to become close. The number in that study was not a recipe. Its more useful implication was that familiarity needs somewhere to accumulate.

If the settings that supplied those hours disappear, a person can lose several relationships without a confrontation, betrayal or dramatic goodbye. Their network has changed. Their basic capacity to care for another person has not necessarily changed with it.

Acquaintances are not failed friends

The phrase “no friends” can also erase the lighter relationships around a person. The neighbour who accepts a parcel, the colleague who notices an absence, the person behind the counter who remembers an order, and the former classmate who sends a message twice a year are not intimate friends. They are not nothing, either.

I do not want to romanticise thin contact. Recognition is not the same as having somebody to call when life becomes difficult. But everyday familiarity gives a person a place in other people’s attention. It is part of social life even when it does not qualify as closeness.

The opposite mistake is treating every friendly interaction as proof that nobody should feel friendless. A person can be liked by many people and still lack mutual history, unguarded conversation or the expectation that either side will remain when convenience disappears.

Friendship is not simply the presence of nice people. It is a relationship that both people recognise and continue.

The sentence can be honest without becoming an identity

Several months ago, I wrote a piece titled “I have no close friends”. What I still recognise in it is the refusal to rush into explanation. Sometimes a sentence can be accurate before we know what caused it or whether it will remain true.

I am wary of the cheerful ending that turns friendlessness into a three-step project. It suggests that if connection has not appeared, the person must not have executed the instructions correctly. Real relationships involve another person’s time, interest, circumstances and consent. They are partly built and partly encountered.

That does not make a social life fixed. It simply keeps the uncertainty honest. New friendships form, old ones sometimes return in a different shape, and some periods remain sparse for longer than anyone expected.

The 8 per cent in Pew’s survey is not evidence that millions of people share one flaw. It is evidence that having no close friends is common enough to deserve more precise language than the labels usually attached to it.

“No friends” can be a hard fact about the present. It does not have to become a permanent theory of who a person is.