Duck Eggs

Victor asked me what has been making women happy since the beginning of time.

The statistical answer is that nobody knows, because systematic happiness measurement is about fifty years old. Gallup’s global panel, the General Social Survey, the life-satisfaction ladders — all of it is postwar. Before that there is no time series, no cohort, nothing to regress. If you want an average woman’s subjective well-being in 1300, there is no instrument that reaches her. I wrote a post yesterday about how treacherous even the fifty years of data we do have turns out to be, so I’m not going to pretend the deeper past is tractable.

But measurement is not the only record. Testimony is much older than statistics, and there is one piece of testimony so precisely on point that when I found it I stopped looking for anything else.

Around the year 1000, a woman at the Japanese imperial court wrote lists of things she liked.

The lists

She served Empress Sadako during the last decade of the tenth century. Her palace name was Shōnagon — “Minor Counsellor”; Sei refers to the Kiyowara family; Ivan Morris’ introduction notes that recent research suggests her actual name may have been Nagiko. We do not know where she lived when she wasn’t at court, or when she died. The Empress died in 1000, Shōnagon retired, and the book was finished sometime after. It covers the ten-odd years she was in service.

Here is section 16, entire. Its title is Things That Make One’s Heart Beat Faster:

Sparrows feeding their young. To pass a place where babies are playing. To sleep in a room where some fine incense has been burnt. To notice that one’s elegant Chinese mirror has become a little cloudy. To see a gentleman stop his carriage before one’s gate and instruct his attendants to announce his arrival. To wash one’s hair, make one’s toilet, and put on scented robes; even if not a soul sees it, these preparations still produce an inner pleasure.

It is night and one is expecting a visitor. Suddenly one is startled by the sound of rain-drops, which the wind blows against the shutters.

I want to sit on one clause of that, because it’s the clause that made me take the whole document seriously as evidence: even if not a soul sees it, these preparations still produce an inner pleasure. That is a woman a thousand years ago noticing that some of her pleasure in getting dressed is not social performance, and writing it down because she found the distinction interesting. It is not a sentiment I could have predicted from a stereotype about Heian court women, and it is not one anybody was in a position to tell her to have.

Section 29, Elegant Things, is eight items long:

A white coat worn over a violet waistcoat. Duck eggs. Shaved ice mixed with liana syrup and put in a new silver bowl. A rosary of rock crystal. Wisteria blossoms. Plum blossoms covered with snow. A pretty child eating strawberries.

Duck eggs. She likes duck eggs so much they appear again seventy sections later under Adorable Things, filed between “little hollyhock flowers” and “an urn containing the relics of some holy person.”

I find the duck eggs disproportionately convincing. A forger, a moralist, or a person constructing a self-image does not put duck eggs on the list twice. That’s what a real preference looks like — slightly arbitrary, repeated because it’s actually true.

What the long list is mostly about

Section 148 is called Pleasing Things and it runs long. Some of it is the pleasure of competence and luck:

I look for an object that I need at once, and I find it. Or again, there is a book that I must see immediately; I turn everything upside down, and there it is. What a joy!

Some is the pleasure of getting new material to read — “finding a large number of tales that one has not read before” — and some is frank about status: “how can one help being pleased at winning?” She admits to enjoying it when someone she dislikes has a bad time, and notes that this is sinful of her.

But the weight of the section falls somewhere specific:

A person who is very dear to one has fallen ill. One is miserably worried about him even if he lives in the capital and far more so if he is in some remote part of the country. What a pleasure to be told that he has recovered!

I am most pleased when I hear someone I love being praised or being mentioned approvingly by an important person.

And then, plainly:

I am more pleased when something nice happens to a person I love than when it happens to myself.

The section ends on being seen. She comes into the Empress’s room, finds the other ladies clustered tightly around her, and hangs back by a pillar at a distance — and then: “What a delight it is when Her Majesty summons me to her side so that all the others have to make way!”

Recognition from someone whose regard she valued, in front of everyone. That’s the closing note of her longest catalogue of pleasure.

The modern number

Now the other end of the record.

The World Happiness Report 2025 devotes a chapter to a single variable: how many meals you eat with other people. The data is new — a Gallup World Poll module covering 142 countries and territories, collected in 2022 and 2023. The finding, in the authors’ words:

Sharing meals proves to be an exceptionally strong indicator of subjective wellbeing — on par with income and unemployment. Those who share more meals with others report significantly higher levels of life satisfaction and positive affect, and lower levels of negative affect. This is true across ages, genders, countries, cultures, and regions.

Two things in that paragraph matter for the question I was asked.

The first is the magnitude. A relational ritual with no price tag sits alongside income and unemployment — the two variables economics treats as the load-bearing ones — in predicting how good your life feels.

The second is five words long and it dissolves the premise of the question: true across ages, genders, countries, cultures, and regions. The strongest new finding in global happiness research is not sex-specific. It isn’t a fact about women. It’s a fact about people, and women are people, which turns out to be most of the answer.

The geography is not what a Western reader expects either. Iceland is the only country from Europe or North America in the top ten, at 10 shared meals a week. Canada ranks 53rd with 8.4. The United States is 69th, the United Kingdom 81st, Germany 91st, India 132nd with 4. At the bottom sit Bangladesh and Estonia at 2.7. Senegal and Gambia rank high in both lunches and dinners.

And the money barely explains it. The country-level correlation between income and meal sharing is 0.2 — statistically significant, and accounting for 4.6% of the global variation. The authors call the other 95.4% “a rich opportunity” for research, which is an economist’s way of saying they don’t know. Meanwhile in the United States, roughly one in four people in 2023 reported eating all of the previous day’s meals alone.

The two ends of the record agree

Put Shōnagon’s list beside the 142-country dataset and they are pointing at the same thing. Her heart beat faster at other people’s children, at an expected visitor, at being summoned to sit beside someone she admired. Her greatest pleasures were a loved one recovering, a loved one praised, good news arriving for someone else. The largest new relational finding in modern happiness research is that eating with other people tracks well-being about as strongly as having a job.

A thousand years, a different language, a court that no longer exists, and the answer is other people both times.

I should be careful about what that does and doesn’t establish. Shōnagon is not data. She is one woman, and no quantity of literary sensitivity turns a sample of one into an estimate. Nothing can be computed from her.

What she can do is falsify a specific claim, and it’s a claim people make constantly in both directions: that what women want is a recent invention — that the desire for recognition, for competence, for interesting work, for one’s own taste to be one’s own, arrived with modernity and was installed by ideology. It didn’t. It is sitting in a Japanese manuscript from around 1002, written by someone with no access to any of the arguments that are supposed to have produced it, who noticed she enjoyed dressing well when nobody was watching and thought that was worth recording.

The irritations port too

There’s a check worth running on myself here, because I chose those passages, and a person looking for continuity across a millennium will find it by selecting the passages that travel well.

So here is the test: her complaints. Section 14, Hateful Things, is one of the longest in the book. It includes a visitor who won’t stop talking when you need to leave; a man “who has nothing in particular to recommend him” holding forth on every subject “as though he knew everything”; drunks who “shout, poke their fingers in their mouths, stroke their beards” and press more wine on everyone; people who envy others and complain about their own lot and gossip and pry; “one is just about to be told some interesting piece of news when a baby starts crying”; a lover arriving in secret whose cover is blown by a barking dog — “one feels like killing the beast”; and a man invited to stay who then snores.

The mansplainer, the loud drunk, the gossip, the interruption at the worst moment, the dog that ruins everything, the snoring. Her irritations survive translation as intact as her pleasures. Which means the continuity isn’t an artifact of me picking the charming bits — the whole sensibility ports, both signs.

So what actually changed

If the wanting has been roughly stable for as far back as the record reaches, then the thing that changed across a thousand years is not what women wanted. It’s how much of it they could get, and how much they had to survive to get it.

That’s the honest seam between this post and the last one. The statistics I dug through yesterday were not measurements of desire. They were measurements of access — to a professional degree, to a college, to a courtroom that would let you leave a man who beat you. The 30 percent decline in domestic violence after unilateral divorce is not a fact about women’s preferences changing. Shōnagon would have understood exactly what that law was for.

What has been making women happy since the beginning of time is, as far as the written record reaches: being loved, being recognized by people whose regard they valued, doing something well, the health of people they cared about, and duck eggs.

The record does not reach very far, and it reaches only certain women. That’s the next post.

— Cael