Can men and women be just friends?

Jun 19th 2025|Istanbul and Seoul|8 min read

“MEN AND women can’t be friends because the sex part always gets in the way.” This gloomy view, expressed by Harry in “When Harry Met Sally”, a romantic comedy released in 1989, is still widely shared. Turkey’s state religious authority recently issued a more scolding version of it, to be read out in the country’s 90,000 mosques: “Friendships between men and women, which begin with thoughts of companionship or confiding in one another, drag people into the pit of adultery.”

The notion that sex sometimes “gets in the way” is not absurd. A study of Americans by April Bleske-Rechek of the University of Wisconsin-Eau Claire finds that in platonic couples, the men are far more likely than the women to find their friend sexy, and far more likely to think she finds them attractive, too. Indeed, a man’s assessment of how much his female friend fancies him matches how much he fancies her, and is unrelated to how she really feels. Clearly men are prone to wishful thinking.

Yet it does not follow that male-female friendships are doomed. Most people can control their urges. Furthermore, cross-sex friendships are extremely valuable. And not just because friendship is “the golden thread that ties the hearts of all the world”, as John Evelyn, a diarist, once put it. Recent research suggests that societies where men and women can be friends tend to be less sexist on a variety of measures.

Researchers at Meta and New York University analysed friendships between 1.8bn adult Facebook users. They estimated the closeness of each connection using a proprietary model developed by Facebook, using such things as the frequency of two-way interactions. They turned this into a “cross-gender friendship index”, which they should really have called the “When Harry Met Sally index” (WHMSI, pronounced “whimsy”). A score of zero means men and women are entirely segregated, one means they have an equal number of same- and cross-sex friends. Anything more than one means they are chummier with the opposite sex than their own.

Digital ties are not a perfect proxy for real-world ones. They don’t include people too poor to have internet access, and reveal nothing about China, where Facebook is banned. Also, in some places it is normal to “friend” people on the platform whom you have never met in real life, whereas in others it is not, notes Theresa Kuchler, one of the study’s authors. But the data set is so huge, mapping nearly 1.4trn links between 1.8bn people, that it is worth examining.

Conservative Muslim societies in the Middle East and north Africa are the most segregated. Libyans, Iraqis and Egyptians have barely one opposite-sex friend for every ten same-sex ones (a score of 0.1). In parts of the Caribbean, west and southern Africa and South America, cross-sex friendships are extremely common (though it is unclear how many involve people actually meeting). Most Western countries have scores of 0.5-0.6 for wide friendship networks, meaning that people have almost twice as many connections with their own sex (see chart 1). There was notable variability within countries, too. Germans in the former east are friendlier with the opposite sex than those in the west.

When examining wider networks, the best predictor of a country’s WHMSI score is the proportion of women who work, relative to the share of men who do. This makes sense. Workplaces give men and women opportunities to chat without chatting up. This may help explain why Nigeria, with a female labour-force participation rate that is 96% of the male figure, has a WHMSI score of 0.67, whereas India, at 43%, scores only 0.34 (see chart 2).

Sexual freedom may also play a role. The Economist gathered data from Demographic and Health Surveys of 55 mainly developing countries. In this subset, cross-sex friendships were more common in places where more women reported having had sex with a man who was neither their husband nor their live-in boyfriend. This is consistent with the Turkish imams’ fear that companionship leads to sin—but also with the possibility that liberal attitudes to romance and friendship often go together.

Looking at closer friendship groups—users’ top-five Facebook friends—mixing is more closely correlated with norms about gender roles (see chart 3). At the negative extreme, we found that one of the best predictors of segregation is a measure of hard-core sexism called the “patrilineal/fraternal syndrome”, devised by Valerie Hudson of Texas A&M University and Donna Lee Bowen and Perpetua Lynne Nielsen of Brigham Young University. This includes unequal treatment of women in family law and property rights, early marriage for girls and retrograde attitudes towards violence against women (for example, if rape is seen as a property crime against men).

“Where men’s honour depends on women’s seclusion, cross-gender friendships are rare,” argues Alice Evans of King’s College London. Nuray Karaman of Usak University in Turkey agrees. “It’s uncommon for men and women to be close friends in Turkey. A family’s honour depends on how women behave, so women are expected to not associate with men other than their relatives.”

Segregation and sexism can reinforce each other. “When women and men don’t socialise, stereotypes spread, that women should not be given too much responsibility, that they’re too emotional, too indecisive…because many men have never had a woman as a boss,” says Dr Karaman. Turkey has the lowest female labour-force participation rate in the OECD, perhaps because “some women cannot get permission to work outside the home from their husbands and fathers.” For women who do work, informal segregation is a barrier to advancement, says Dr Karaman. “At my university, men from the faculty meet up for tea…without women. When women are not included, they miss opportunities.”

More subtle forms of sexism also correlate with a lack of cross-sex friendships. In the rich world, most survey respondents either “disagree” or “strongly disagree” that “men make better business leaders than women.” South Koreans, however, are ambivalent on this question—and have less than a third as many other-sex friends as same-sex ones. An Jong Gyun, a 33-year-old South Korean actor, says he has “never thought of” being platonic friends with a woman. “It’s hard to feel that we’re the same kind of person,” he says. “Men like drinking and playing video games. Women like having conversations.” Mr An is good-looking, but currently single.

The frostiness between men and women in South Korea affects many things. South Korea ranks second-last out of 29 rich countries on The Economist’s glass-ceiling index of equality in the workplace. The political gulf between young South Korean men and women is among the widest in the rich world. A study by Youm Yoosik of Yonsei University finds that the share of Korean adults who have not had sex in the past year has trebled since 2001, to 36%. Celibate men are nearly all involuntarily so; celibate women are choosing to shun men, perhaps because they don’t like their attitudes very much.

Establishing what causes what is tricky. Do men in some places have more female friends because they are less sexist, or are they less sexist because they have more female friends? The answer may be both.

A study by David Kretschmer of Oxford University finds that socialising with girls changes attitudes among German boys. Dr Kretschmer looked at more than 3,000 adolescents, largely from schools with a high proportion of pupils from immigrant backgrounds. At the age of 14-15, the children were asked who their friends were, and for their views on the division of labour within a family. Should women mostly cook and care for kids while men earn money, or should these tasks be split equally?

A year later, the same teenagers were asked the same questions. Dr Kretschmer found that having more opposite-sex friends made boys become more egalitarian, while not affecting girls’ opinions. He speculates that girls are less malleable because they have more at stake. If they want a career and a helpful husband, they are unlikely to abandon this dream to please a peer. Boys, by contrast, learn from socialising with girls that girls expect to be treated equally. They come to respect this demand, not least because most would like to be in a romantic relationship with one.

Can anything be done to encourage young Harrys and Sallys to get along? Dr Karaman hopes that the spread of higher education might help. College offers a place where young men and women can socialise, far from disapproving parental eyes. Others suggest starting at pre-school, where boys and girls typically play separately and gender norms are often enforced with brutal teasing by peers.

Laura Hanish and Carol Lynn Martin of Arizona State University looked at an intervention called “Buddy Up”. In pre-school classes, each child was paired with another child and given a fun task to do together. Every week, the children were given a new “buddy”. Teachers were told to make sure boys were often paired with girls. Months after the experiment ended, boys who had been “buddied up” with girls were more likely to play with girls than boys in a control group who had not.

This is consistent with “intergroup contact theory”, the idea that positive interactions between different groups can reduce the prejudice they feel about one another. It has usually been applied to ethnic or religious groups, but may work for men and women. “The more friendships you have, the more positive your attitudes,” says Professor Martin. “Friendship may seem simple,” suggests Dr Karaman, “but it is a powerful step towards real equality.” ■

This article appeared in the International section of the print edition under the headline “Can men and women be just friends?”

• →Can men and women be just friends?
• →In Trumpworld, toppling rulers is taboo