Click a person or an object to see which ideas explain it.
Anthropology, sociology, social psychology: 21 ways of seeing
What you didn’t see at the party
An ordinary Friday night in one flat. Scroll, and the same rooms show you what anthropologists, sociologists and social psychologists have learned to see: the debt hidden in a gift, the backstage in every kitchen, the stranger people confide in. At the end, five famous ideas the data have put in doubt.
Marius Comper · 28 September 2026
Did something happen to you this week?
Pick a moment and jump straight to the idea that explains it.
On the stairs, the neighbour taking out the rubbish meets the courier. They look at each other for a moment, then both look away. Nobody is being rude. Goffman called this civil inattention: a brief look followed by withdrawal says ‘I’ve seen you, and I’ll leave you alone.’
What it rests on
In 1983 three American psychologists tested the idea in lifts where one of the passengers belonged to their team. About half of the 320 riders observed gave that passenger one short look at the start and then none; 35% glanced once or twice more. A stranger who kept to the rule was rated more favourably than one who stared, or one who never looked at all.
Later studies on video found glances shorter than a tenth of a second, too quick for Goffman’s notebook. The details of the rule vary from place to place.
The guest at the door is no longer someone on the street, and not yet someone at the party. Shoes come off, the coat goes on a hook, someone hands over a glass. Van Gennep noticed in 1909 that passages from one state to another tend to come in three steps: separation, an in-between, and joining. He also noticed how often the middle step happens literally on a threshold.
What it rests on
His examples run from weddings and funerals to the front door, which he describes as the boundary between the family’s world and the world outside. Among the rites of joining he lists offering salt and sharing a meal. Sixty years later Victor Turner built on the middle step: people in it are ‘betwixt and between’, and those who pass through it together can feel a bond that ignores rank. He called it communitas.
It is a pattern for describing transitions, not a law that every transition obeys.
Nobody minds the heap of shoes by the door. Put one pair on the coffee table and the whole room would flinch. Mary Douglas took an old saying, that dirt is matter out of place, and made it an argument: what we call dirty is whatever breaks our order. ‘Shoes are not dirty in themselves,’ she wrote, ‘but it is dirty to place them on the dining-table.’
What it rests on
‘Where there is dirt there is system’: a set of places where things belong. Her own list from 1966 reads like a flat after a party: cooking pots in the bedroom, food on clothes, clothes lying on chairs, outdoor things indoors.
Douglas called it ‘the old definition’, not her own discovery. She did not say germs don’t matter; she set hygiene aside to look at order. Anthropologists still argue with her habit of treating a culture as one coherent whole.
The courier on the landing smiles as the host opens the door. The smile is part of the job. Arlie Hochschild, who studied Delta’s flight-attendant training in the early 1980s, called this emotional labour: feelings managed for a wage. Trainees were told their smile was their biggest asset. The host’s partner, beaming with a tray, may be doing the unpaid version, which Hochschild called emotion work.
What it rests on
She separated surface acting (showing a feeling you don’t have) from deep acting (working yourself into actually feeling it). A 2011 meta-analysis of 95 studies found surface acting goes with more exhaustion and less job satisfaction; deep acting was linked to better performance and happier customers.
Most of these studies ask people once, at a single moment, so they show a link, not a cause.
Nobody asked for the bottle of wine or the little red parcel. Everyone brought something anyway, and everyone would notice who didn’t. In his 1925 essay on the gift, Marcel Mauss argued that gifts come with three obligations: the obligation to give, the obligation to receive and the obligation to give back. He didn’t find them only far away: in his own society too, he wrote, ‘the invitation must be returned, just like politeness’.
What it rests on
A field test of that pull: in 2007 the economist Armin Falk sent almost 10,000 fundraising letters in Zurich. With no gift inside, 12% of recipients donated; with one postcard, 14%; with four, 21%.
Mauss built his argument mostly from other people’s fieldwork, and later anthropologists disputed his reading of the Maori word hau, the ‘spirit’ he said makes a gift return. A 1976 study in which 20% of strangers answered a Christmas card from an unknown sender is often quoted; a repeat published in 2016, with 755 cards, got replies from 2%.
The joke didn’t land. Two of the listeners laugh anyway, a little too quickly. Goffman called ‘face’ the positive social value a person claims for themselves in an encounter, and noticed that keeping it is teamwork: each of us defends our own face and, quietly, everyone else’s.
What it rests on
His tools are small and familiar: tact, diplomacy, a ‘tactful blindness’ that doesn’t ‘see’ that someone has stumbled, a glance turned away for a moment to give them time to recover. Mutual acceptance in an encounter, he wrote, is typically a ‘working’ acceptance, not a ‘real’ one. The linguists Penelope Brown and Stephen Levinson later built a whole theory of politeness on his idea of face.
Goffman wrote from observation, not experiments. How universal the politeness theory built on his idea really is remains disputed.
Look at the gaps between people. The two talking by the speaker stand closer than they would in daylight: the music is loud, and noise, Hall noted, draws people closer. The old friends sit shoulder to shoulder on the couch; the newcomer keeps an arm’s length. Edward Hall described four zones, intimate, personal, social and public, each with a close and a far phase.
What it rests on
Hall’s figures (intimate up to 18 inches, personal to 4 feet, social to 12) came, as he warned, from middle-class adults on the north-east coast of the United States. In 2017 a survey of 8,943 people in 42 countries asked where a stranger, an acquaintance and a close person should stop. The average answers were 135, 92 and 32 centimetres, with real differences between countries. Romania was among them, with 56 participants.
The survey records what people say on paper, not how they actually stand.
Lazarsfeld and Merton, 1954 · Thomas Schelling, 1971
Nobody drew a seating plan, and yet by eleven the flat has sorted itself: dancers by the speaker, talkers on the couch, smokers outside. Sociologists call the tendency of our contacts to resemble us homophily. Thomas Schelling showed in 1971 how far a small preference can go: on his board, if everyone wants merely half their neighbours to be like them, a mixed pattern sorts itself into sharply separated clusters.
What it rests on
The standard review, from 2001, puts it in three words: similarity breeds connection. In the American data it gathers, race and ethnicity divide people’s networks most, followed, ‘in roughly that order’, by age, religion, education, occupation and gender. As early as 1954 Lazarsfeld and Merton had separated similarity of status (age, education, occupation) from similarity of values.
Schelling’s result is about a model. It shows that sharp separation doesn’t require strong prejudice; it doesn’t show that real neighbourhoods separated that way. And homophily mixes choice with opportunity: we meet the people who are around us.
Every guest reads the bookshelf and the record player, and the host knows it. Bourdieu argued that taste works as a social marker: it classifies things and, in doing so, classifies the person judging them.
What it rests on
Eighty years earlier Veblen had written about conspicuous consumption. A 2009 study of American data found that Black and Hispanic households spend nearly a third more than comparable white households on visible goods, clothes, jewellery, cars, and that most of the gap disappears once you account for how poor their reference group is: status is measured against the people around you.
Bourdieu’s survey, 1,217 people questioned in 1963 and 1967–68, was about the France of those years. Later studies found high-status people with ‘omnivorous’ taste across high and popular culture, but that trend doesn’t look stable either.
The man by the drinks table has held the same full glass for an hour. He isn’t enjoying the drinking much; he assumes everyone else is. Psychologists call this pluralistic ignorance: privately, each person doubts a norm and thinks they are the only one.
What it rests on
At Princeton in the early 1990s, students rated themselves less comfortable with campus drinking habits than they thought the average student was. Over one semester, the men shifted their own attitudes toward the norm they had imagined. The women did not.
Telling students the real norm makes them drink only very slightly less: a 2015 review of 70 randomised trials found drops of under one drink a week, from a starting point of nearly 14 drinks.
When the song everyone knows comes on, the room changes: people who were talking in threes are suddenly one crowd. Durkheim called this collective effervescence, ‘a sort of electricity’ that comes from the mere fact of gathering, and he believed the religious idea itself was born of it.
What it rests on
Durkheim worked from other people’s accounts of Aboriginal Australian ceremonies. In 2011 researchers recorded heart rates at a fire-walking ritual in a Spanish village: the heartbeats of the fire-walkers rose and fell in step with those of relatives watching them, but not with those of unrelated spectators.
That was 38 people at one ritual, and the study measured synchrony, not ‘bonding’.
How you sit, how you hold a glass, how you dance: none of it is natural. Mauss called these techniques of the body, ‘the ways in which from society to society men know how to use their bodies’. Ill in a New York hospital, he noticed that his nurses walked like the women he had seen at the cinema; back in Paris, he saw young women walking the same way. He suspected the films had spread a way of walking.
What it rests on
He also noted that a child squats naturally and that European adults no longer know how to, and remembered Australian soldiers in the war resting on their heels in the mud while he had to stay standing.
It was a lecture, not a study. He used the Latin word habitus, taken up later above all by Bourdieu.
In the living room, the host’s partner is on stage, beaming, with a tray. In the kitchen, the host leans on the counter and exhales. Goffman described social life as a performance with a front region and a back region, a place where ‘the impression fostered by the performance is knowingly contradicted as a matter of course’.
What it rests on
His model came partly from fieldwork in a hotel in the Shetland Islands. In the hotel kitchen, half-used pats of butter were ‘rerolled to look fresh’, and the door between the kitchen and the dining room was fought over every day. The waitresses, another observer noted, swept into the kitchen like actors coming off stage into the wings.
It is a way of seeing, built from observation and anecdote. Goffman did not claim that everyone fakes all the time; the back region is where a team relaxes and repairs.
The guest who opens the fridge without asking has broken no written rule, and yet the kitchen goes quiet. Garfinkel argued that everyday life rests on expectations so ordinary that we only see them when someone breaks them.
What it rests on
He had his students spend up to an hour at home behaving like polite lodgers: formal, speaking only when spoken to. Of 49 reported attempts, 9 failed or were refused. In most of the rest, families reacted with astonishment, bewilderment, embarrassment or anger: ‘What’s gotten into you?’ In every case, the situation could be restored once the student explained.
These were class assignments written up by students, not controlled experiments. Garfinkel’s point was how hard families worked to make the odd behaviour make sense.
Who pays for the pizza? Close friends don’t keep count. Acquaintances split it exactly. The host decides when the music stops. Offer the host money for dinner and feel the room go cold. Alan Fiske proposed that people run their relationships on four models: communal sharing, authority ranking, equality matching and market pricing.
What it rests on
Fiske even gives a party example: when everyone drinks from the same punch bowl, ‘no one keeps track’ of who drinks how much. Dinner invitations owed, ‘you may owe me an invitation’, belong to equality matching. Money is the typical medium of market pricing. His 1992 paper reviews 19 experimental studies, with 7 different methods and participants from 5 cultures.
Many of those studies were by Fiske and his collaborators. The models are a theory of how we think about relationships, not a list of rules everyone follows.
The sink fills with glasses nobody will own up to. Garrett Hardin’s 1968 essay imagined a pasture open to all, where each herder gains by adding one more animal, and concluded: ‘Freedom in a commons brings ruin to all.’ Elinor Ostrom spent her career showing that people often manage shared things themselves, if they make and enforce their own rules.
What it rests on
Her principles include clear boundaries on who may use the resource, rules that fit the place, monitoring by the users themselves, sanctions that ‘start very low’ and grow only with repeated offences, and quick, cheap, local ways to settle disputes. A 2010 review of 91 studies found her principles well supported. In 2009 Ostrom shared the Nobel memorial prize in economics with Oliver Williamson.
Ostrom studied fisheries, forests and irrigation systems, not flatshares; the sink is an analogy. Hardin’s critics say his pasture describes open access with no rules, not a commons that has owners and rules.
W. I. and Dorothy Thomas, 1928 · Robert Merton, 1948
Someone told the group by the window that the newcomer is a bit odd. So nobody talks to him, and by midnight he is standing alone, looking odd. ‘If men define situations as real, they are real in their consequences,’ wrote W. I. Thomas and Dorothy Swaine Thomas in 1928. Robert Merton turned it into the self-fulfilling prophecy: a false belief that produces the behaviour that makes it true.
What it rests on
Merton’s example was a bank that is sound until a rumour says it isn’t, then ruined because depositors believe it and withdraw their money. His bank is a parable, not a real event.
How strong such effects are in real life is disputed. The most famous case, teachers whose expectations supposedly raised pupils’ IQ, turned out much smaller than its legend (see below).
The friend-of-a-friend nobody knows stands by the bookshelf. On the balcony, two people who met ten minutes ago are telling each other things they haven’t told their families. Simmel wrote in 1908 that the stranger is the one who comes today and stays tomorrow: near and far at once, and for that very reason often given confessions.
What it rests on
Milgram’s students photographed commuters on a railway platform and a week later asked them whom they recognised. As two researchers later reported it, 89% recognised at least one person they had never spoken to, four on average. Milgram called them familiar strangers, like the neighbour on the stairs. A 2004 repeat in Berkeley found 78% and three.
Simmel’s point about confessions is an observation, not a measured result.
Mark Granovetter, 1973 · LinkedIn experiment, 2022
Your closest friends are on the couch. They are wonderful, and they know most of what you know. The acquaintance on the balcony knows other people, and so hears other news. In 1973 Mark Granovetter called this the strength of weak ties.
What it rests on
Of people in a Boston suburb who had found a professional, technical or managerial job through a contact, 17% saw that contact often, 56% occasionally and 28% rarely. The sample was small: 54 people. In 2022 a team working with LinkedIn analysed experiments on more than 20 million users over five years: moderately weak ties helped people change jobs more than either the closest ties or the weakest.
A 2017 Facebook study adds a twist: most help comes through weak ties because we have so many of them; one for one, a strong tie is worth more.
In 1956 an anthropologist described the strange body rituals of the Nacirema: a shrine in every house, a charm-box, ‘holy-mouth-men’. Nacirema is American spelled backwards, and the shrine is the bathroom. Try it here: guests queue for a small tiled shrine, perform a purification before the mirror, leave their footwear heaped at the entrance and bring the head of the household offerings of fermented grapes.
What it rests on
Miner’s text mimics the way ethnographers of his day wrote about distant peoples. It has been set in introductory anthropology courses ever since, as an exercise in seeing the familiar as strange.
It is satire, not a study, and it is aimed just as much at the way anthropologists wrote about others.
Joseph Henrich, Steven Heine, Ara Norenzayan, 2010
Many of the ideas on this page were born in a few countries and tested on the people there. A survey of top psychology journals for 2003–2007 found that 96% of research participants came from Western industrialised countries, home to about 12% of humanity.
What it rests on
On some tasks those very samples sit at the extreme. In a classic study of the Müller-Lyer illusion, two equal lines that look unequal, American students were fooled the most; San foragers in the Kalahari were hardly fooled at all.
The lesson is about samples: ‘people’ in a study are always particular people.
Some of the most quoted ‘lessons’ about people didn’t survive when someone went back to the data. Here they are as they are disputed: who says what. Three more, Milgram’s obedience experiment, the Stanford prison study and the Kitty Genovese case, are told in full in What’s left of psychology.
The legend: the human brain can’t hold more than 150 relationships.
Robin Dunbar, 1993
Started from the link between neocortex size and group size in primates and extrapolated to humans: 147.8, with a 95% confidence interval from 100.2 to 231.1. He warned himself that extrapolating beyond the original data is frowned on.
Hill and Dunbar, 2003
A study of Christmas cards sent found networks averaging 153.5 people.
Lindenfors, Wartel and Lind, 2021
Redid the calculation on newer primate data with several methods. They got averages between 16 and 109, with confidence intervals from 2 to 520: ‘specifying any one number is futile’.
Dunbar, 2021 and 2023
Replied that the result depends on the regression method: on the same data, the method used by the Stockholm team predicts 71, and the one he used predicts 158. He defended the hypothesis in an academic essay too.
What remains: There is some limit to how many people you can keep close, but the figure of 150 doesn’t follow from the primate data.
The legend: at the Hawthorne plant in the 1920s, workers produced more after any change in lighting, simply because they were being watched.
The textbooks
Repeated the story for decades, although the original data seemed lost.
Stephen Jones, 1992
Reanalysed the relay-assembly room data, covering more than five years, and found only ‘slender evidence’ of such an effect.
Steven Levitt and John List, 2011
Found the lighting data in a microfiche room of a library in Milwaukee and in an archive in Boston. The famous patterns proved ‘entirely fictional’: output jumped on Mondays, but the lights were always changed on Mondays, and Monday output jumped with no change at all. They found only faint traces of a subtler effect.
What remains: The plant’s data don’t support the story that output rose simply because workers were watched; Levitt and List find at most traces of subtler effects.
The legend: if a teacher believes a child is gifted, the child’s IQ rises.
Rosenthal and Jacobson, 1968
Told teachers at an American primary school that randomly chosen pupils would ‘bloom’. After a year, their average edge across the school was 3.8 IQ points. In first grade it was 15.4 points, but the first-grade ‘bloomers’ numbered seven children.
Richard Snow, 1969
Found impossible scores in the data, such as a child with a reasoning IQ of 17 before and 148 after, and wrote that the study would not have passed a journal’s review.
Raudenbush, 1984
In 18 later experiments, the effect appeared only when teachers didn’t yet know their pupils; after two weeks of contact it vanished.
Jussim and Harber, 2005
Self-fulfilling prophecies in classrooms do occur, but they are typically small and don’t accumulate much.
What remains: Expectations matter a little, sometimes; the legend grew far beyond seven children.
The legend, in two versions: Samoan youth had an adolescence free of turmoil; or Mead was hoaxed by two joking girls.
Margaret Mead, 1928
Described a relaxed adolescence, including about sex, and used it to argue that adolescent turmoil is cultural, not biological.
Derek Freeman, 1983 and 1999
Argued that Mead’s picture was wrong, and later that she had been ‘hoaxed’ one evening in 1926 by two young women, relying on 1987 interviews in which one of them, Fa’apua’a Fa’amū, appeared to confirm it.
Paul Shankman, 2013
Read the full transcripts of those interviews: asked directly, Fa’apua’a said Mead never asked her about sex, and those passages were missing from Freeman’s account. Mead’s field notes don’t list her as an informant; Mead had systematic data on 25 girls.
What remains: The evidence for a ‘hoax’ doesn’t hold; how accurate Mead’s picture was is still argued over.
The legend: send Christmas cards to strangers and one in five will write back, so strong is the norm of reciprocity.
Kunz and Woolcott, 1976
Sent cards to strangers and got replies from about 20%. The result went into textbooks as proof of the norm.
Brian Meier, 2016
Repeated the experiment with 755 cards: 2% replied. In a second study, people said they would be less likely to reply when they felt suspicious.
What remains: The norm of reciprocity is real (see the gift), but the stranger’s card is no longer evidence for it: 2% replied in Meier’s experiment, and people who reported suspicion said they would be less likely to reply.
An idea stayed only if we read the text that launched it or the study that tested it directly, and if someone other than its author has tested or used it. Where we couldn’t open the book, the sources say what we read instead.
Each idea carries a label. Way of seeing: an observational tool, not a proven law; Goffman, Mauss and Douglas ran no experiments. Tested finding: a regularity measured and checked by others. Contested: a famous idea whose evidence has cracked.
The drawing is an illustration. The party is invented; each label says what the idea would notice, not what was measured at any party.
Some famous ideas are left out because we couldn’t read them at the source, such as Norbert Elias’s civilizing process.
Hülsheger & Schewe, “On the costs and benefits of emotional labor: a meta-analysis”, Journal of Occupational Health Psychology 16(3), 2011. doi.org/10.1037/a0022876
Penelope Brown & Stephen C. Levinson, Politeness: Some Universals in Language Usage, Cambridge University Press, 1987 (bibliographic record). doi.org/10.1017/CBO9780511813085
Mark Granovetter, “The Strength of Weak Ties”, American Journal of Sociology 78(6), 1973, p. 1371. doi.org/10.1086/225469
Rajkumar, Saint-Jacques, Bojinov, Brynjolfsson & Aral, “A causal test of the strength of weak ties”, Science 377, 2022. doi.org/10.1126/science.abl4476
Gee, Jones & Burke, “Social Networks and Labor Markets”, Journal of Labor Economics 35(2), 2017. doi.org/10.1086/686225
McPherson, Smith-Lovin & Cook, “Birds of a Feather: Homophily in Social Networks”, Annual Review of Sociology 27, 2001. doi.org/10.1146/annurev.soc.27.1.415
W. A. V. Clark, “Residential preferences and neighborhood racial segregation: a test of the Schelling segregation model”, Demography 28, 1991. doi.org/10.2307/2061333
Prentice & Miller, “Pluralistic ignorance and alcohol use on campus”, Journal of Personality and Social Psychology 64(2), 1993. doi.org/10.1037/0022-3514.64.2.243
Foxcroft et al., “Social norms information for alcohol misuse in university and college students”, Cochrane Review, 2015. doi.org/10.1002/14651858.CD006748.pub4
Émile Durkheim, The Elementary Forms of the Religious Life (1912; trans. 1915), book II, ch. 7. gutenberg.org/ebooks/41360
Konvalinka et al., “Synchronized arousal between performers and related spectators in a fire-walking ritual”, PNAS 108(20), 2011. doi.org/10.1073/pnas.1016955108
Páez et al., “Psychosocial effects of perceived emotional synchrony in collective gatherings”, JPSP 108(5), 2015. doi.org/10.1037/pspi0000014
Cox, Arnold & Villamayor Tomás, “A review of design principles for community-based natural resource management”, Ecology and Society 15(4), 2010. ecologyandsociety.org/vol15/iss4/art38/
Henrich, Heine & Norenzayan, “The weirdest people in the world?”, Behavioral and Brain Sciences 33, 2010. doi.org/10.1017/S0140525X0999152X
R. I. M. Dunbar, “Coevolution of neocortical size, group size and language in humans”, Behavioral and Brain Sciences 16(4), 1993. doi.org/10.1017/S0140525X00032325
Brian P. Meier, “Bah humbug: Unexpected Christmas cards and the reciprocity norm”, Journal of Social Psychology 156 (2016). doi.org/10.1080/00224545.2015.1129306