Where We Left Off
Before we begin
Part One was about how to think, not about what is true. It began with a woman photographed at a protest and a rumour that arrived days later, offered as though it were a reply. Nobody organised that. Thousands of people performed the same move at the same time and experienced it as obvious.
The main tool from that part was a separation. When somebody defends the rules about women, they almost always make three claims joined into one sentence. A claim about the past: these rules came from somewhere and had a function. A claim about the future: if they go, damage follows. A claim about you: therefore behave. Each needs a different kind of evidence, and they can be true and false in any combination.
We also took apart the strongest conservative argument, the fence in the field. Do not remove what you cannot explain. It is a serious argument and three parts of it hold. But it tells you to ask why a fence is there, and the honest question is who — who built it, what did they get, and who was on the wrong side. A rule surviving four thousand years proves it kept being enforced. That is a fact about power, not about benefit.
Then the sharpest tool in the box. When an elder says a woman will ruin her life, that sounds like a forecast. Often the speaker is part of the machinery that does the ruining, which makes it a sentence rather than a prediction. The test proposed was whether the harm travels: look at the same behaviour where the punishing community has no reach. What follows her across a border is intrinsic. What stops at the boundary was imposed.
This part answers the first of the three claims. Where did these rules actually come from, and what problem did they solve? Not what people say they solved. What the evidence shows.
I should tell you now that the answer is more interesting than either camp expects. The traditionalist will find that the rules are older and more functional than the reformer admits — they are not arbitrary cruelty and they were not invented by anybody to hurt anybody. The reformer will find something too: that the same problems produced wildly different solutions, that some of those solutions gave women more rather than less, that several of them were in India, and that the specific arrangement now defended as timeless is one option among many that human societies actually built.
How to Read the Boxes
The notation, in case this is where you started
Six kinds of box appear throughout the series. If you read Part One you already know them. If not, here is one live example of each.
Kinship system: the set of rules a society uses to decide who counts as your family, who you may marry, and who inherits what. Every society has one. They differ enormously.
Why it matters here: this whole part is about kinship systems, and the single most useful fact in it is that there is more than one workable design.
A Word Box appears the first time a hard word does. Never later, never only in the glossary.
People who look and think like us have existed for roughly three hundred thousand years. Farming is about twelve thousand years old. Those numbers are hard to feel, so compress them.
Put all of human existence into a single day, starting at midnight. For twenty-three hours and twenty minutes, nobody farms anything. Everyone moves, everyone forages, nobody owns a field.
Farming begins at about twenty minutes to midnight. Writing appears with about half of that left. The rules this series examines are, on that clock, extremely recent — and they are almost all downstream of those last forty minutes.
Any number too big to picture gets converted into something with a body — a walk, a room, a wage, a single day.
Much of this part rests on the ethnographic record: several hundred societies described by anthropologists, coded into comparable categories, and analysed together.
What it can show: which practices travel together. If societies that farm with a certain tool reliably restrict women more, that pattern is real and is not explained by any one culture.
What it cannot show: the direction of cause, and it carries a serious flaw at its root. Most descriptions were written by European men, in the colonial period, who mostly interviewed men. Where a society had a women’s world that men did not enter, the record often simply does not contain it. Absence in this archive is weak evidence of absence in the world.
This box names the actual evidence and then says what it cannot show. It is the box that should make you trust the rest.
A framing question for the whole part, and the answer is less obvious than either camp assumes.
Foraging societies are generally more equal between the sexes, with easier divorce, looser control of women and no dowry. Farming brought heritable property, and with it the policing of women’s sexuality to secure inheritance. Skeletal remains from early farming populations show shorter, less healthy people than the foragers before them, and the burden of the new work fell heavily on women.
The comparison is between surviving foragers and ancient farmers, which is not a fair test — the foragers we can study are those pushed onto land nobody else wanted. Farming also produced surplus, which produced everything from medicine to literacy, and women’s lives in the wealthiest farming societies today are freer than in any forager society ever documented. Judging the transition by its first thousand years is like judging a factory by the first winter.
What would settle it: better skeletal and genetic evidence from the transition period itself, in more regions. Some of this is arriving now and Chapter Three uses it.
An Argument box appears where people who have studied something genuinely disagree. Each side gets its strongest case, then a verdict on where things actually stand.
Both sides above assume that a change can be scored — that farming was, on balance, good or bad for women as a group.
But “women” in this sentence covers a forager in the Kalahari, a Sumerian priestess, a Bengali widow in 1850 and a software engineer in Pune. The claim requires adding their experiences into one number, and there is no scale on which that addition can be performed.
The general form: demanding a verdict on a change that had different effects on different people. Watch how often, in this part and the next fourteen, an argument depends on a total that could never actually be computed.
This is the signature box. It digs out something both sides take for granted while they fight. There are five in this part.
Every chapter closes with one of these. It restates the chapter in the plainest words available, with key terms in bold.
A reader who read only the Remember This boxes should still finish holding the whole argument.
A Note on This Part
Read this before Chapter One
Two warnings.
This part explains. It does not excuse. Everything here is an account of how these rules came to exist. Part One, Chapter Four established why that is not the same as saying they are right: no set of facts about how the world is produces an instruction by itself. I will explain the machinery of these rules as clearly as I can, including the parts where the machinery is genuinely clever. If at any point that reads as approval, it is not, and I have said so here so that I cannot claim it later.
I am going to talk about biology, and there is a trap on both sides of it. One camp treats any biological fact as a licence. The other treats any biological fact as an attack, and therefore refuses to look. Both are the same error running in opposite directions, and both were named in Part One. I will report what is established, say plainly how far it reaches, and mark the exact point where it stops reaching. That point arrives much earlier than most people on either side would like.
1The One Asymmetry
There is a single biological fact underneath a great deal of what follows. It is real, it is not controversial, and it explains far less than the people who cite it believe. Both of those sentences matter.
1.1 — The fact
A woman who gives birth knows the child is hers. She was there for the entire process and there is no version of events in which she is mistaken.
A man cannot know this. Until 1985, when the first genetic fingerprinting technique was demonstrated, no man in the history of the species could establish with certainty that a child was his. He could be confident. He could not be sure.
That is the asymmetry. It is one sentence long and it sits underneath an enormous amount of what this series examines.
Before we go further, note what it is not. It is not a claim that men care more about paternity than women care about anything. It is not a claim about who loves children more. It is a claim about information — that one parent has a fact available and the other, for almost the whole of human history, did not.
1.2 — What each parent has to spend
There is a second asymmetry sitting next to the first, and it is about cost rather than information.
Producing a child requires very different minimum contributions from each parent. From the man: a single act. From the woman: nine months of pregnancy, a delivery that was for most of history genuinely dangerous, and then — in societies without bottles or formula, which is all of them until recently — two to four years of nursing during which another pregnancy is unlikely.
Put the reproductive budgets side by side.
A woman is born with all the eggs she will ever have. Across a lifetime she releases roughly four hundred. That is the entire stock, spread over about thirty-five years.
A man produces sperm continuously from puberty. The output of a single day runs to tens of millions.
Here is the version with a body attached. A woman’s whole reproductive career, in eggs, would fit in a teaspoon and takes three and a half decades to spend. A man restocks more than his lifetime’s worth of opportunities before breakfast.
This does not make anybody good or bad. It means the two sexes were playing a game with wildly different budgets, and the rules a society writes tend to follow the budgets.
Biologists have a term for the thing being measured there, and it is worth having, because it is what makes the difference between the sexes predictable rather than arbitrary.
Parental investment: anything a parent does for one offspring that costs them the ability to do it for another. Feeding, carrying, protecting, and — for the mother — the pregnancy itself.
The idea comes from a biologist writing in the early 1970s, and the general finding across the animal kingdom is that the sex which invests more per offspring is choosier about partners, while the sex which invests less competes for access.
Why it matters here: in humans the gap in minimum investment is very large, but the gap in actual investment is often small, because human fathers frequently do a great deal. That difference between minimum and actual is where all the interesting variation lives.
1.3 — What this genuinely predicts
Now the careful part. What does the asymmetry actually let us expect?
It predicts, reasonably well, that some degree of male concern with paternity will appear in most human societies. A man who invests years of food and protection in a child has, in evolutionary terms, a stake in whether that investment is going where he thinks it is. And this prediction holds up. Concern with paternity turns up nearly everywhere anthropologists have looked.
It predicts, less strongly, that societies where men invest heavily in children will show more interest in controlling women’s sexual behaviour than societies where they do not. This also broadly holds, and Chapter Three shows the mechanism.
Here is what it does not predict, and the list is long.
It does not predict that a family will kill a daughter. It does not predict that a woman must cover her hair, or her face, or her ankles, or none of these. It does not predict dowry, or bridewealth, or that a widow may not remarry, or that a girl marries at fourteen, or that a woman may not own land, or work, or travel alone, or that her testimony counts for half.
Every one of those varies enormously between societies that face the identical biological situation. The biology is a constant. The rules are not. A constant cannot explain a variable — that is not a moral objection, it is a logical one.
1.4 — The societies where a child has several fathers
The clearest way to see how far the biology reaches is to look at a case that ought to be impossible.
In parts of lowland South America, a number of societies hold a belief that a child can be fathered by more than one man. The reasoning is straightforward once you accept the premise: a foetus is built up from repeated contributions, so every man who has sex with a woman during her pregnancy contributes to the child. Anthropologists call the resulting arrangement partible paternity.
Partible paternity: the belief that a child can have several biological fathers, all of whom are genuinely regarded as fathers and carry real obligations to the child.
Why it matters here: this is not a curiosity. It is a society that took the exact biological situation described above and built the opposite institution on top of it. If the biology forced the outcome, this could not exist. It exists.
A belief on its own would prove little — people believe all sorts of things that make no difference to how they live. What makes this case worth the space is that somebody went and measured the consequences.
Anthropologists working with the Barí of Venezuela and the Aché of Paraguay in the late twentieth century collected reproductive histories from women, recording which men were named as fathers of each child and what happened to the child afterwards.
What they found: in the Barí data, children who had a named secondary father survived to fifteen at meaningfully higher rates than children with only one father — roughly four in five against something closer to two in three. Secondary fathers brought fish and game. In a place where a child’s main risk is a hungry season or a dead father, a second man with an obligation is insurance.
What it can show: that the arrangement was not a quaint belief floating free of consequences. It had a measurable effect on child survival, which is why it persisted.
What it cannot show: that it would work anywhere else. These are small societies with little storable wealth and no land inheritance — precisely the conditions Chapter Three identifies as the ones that make paternity matter less. And these are single studies of small populations, with all the fragility that implies.
Notice what the Barí did. They faced the same information problem every human population faces. Instead of solving it by restricting women, they solved it by distributing the obligation — turning uncertainty from a threat into a source of extra support for the child.
That is not a better solution in some general sense. It works in their conditions and would collapse in a society with heritable land, for reasons Chapter Three explains. The point is narrower and more important: the same biological fact supported two opposite institutions. So the fact cannot be what produced either of them.
Both camps in the biology argument share a premise they never state: that if a rule has a biological root, the biology explains the rule.
One side says: this is rooted in biology, therefore it is natural and inevitable. The other side says: therefore we must show it has no biological root at all, or we lose.
Both are treating the root as though it determined the plant. But the gap between “men everywhere show some concern about paternity” and “this family will kill this girl” is enormous, and nothing in biology crosses it. Between the fact and the rule sits everything this part is actually about — property, inheritance, farming tools, weak states, marriage payments.
The general form: mistaking a necessary condition for a sufficient one. Rain is necessary for a flood. It does not follow that rain explains why this valley floods and the next one does not. The answer to that is drains, and drains are built by people.
1.5 — Where the biology stops
So here is the honest position, and it will satisfy nobody entirely.
The asymmetry is real. It is not a social construction and pretending otherwise is the moralistic fallacy from Part One, Chapter Four — deciding a fact must be false because it is inconvenient. Anybody who tells you that male concern with paternity is purely an invention of patriarchy is asking you to ignore a pattern that shows up in societies with no contact with each other.
And the asymmetry explains almost none of the variation. Every specific rule this series examines — the clothing, the timing of marriage, the property law, the killing — differs wildly between societies whose biology is identical. Anybody who tells you these rules are simply nature is asking you to ignore that the same nature produced the Barí.
The rest of this part is about what fills the gap. It turns out that the strongest single answer is not biological at all. It is a question about whether you own something you cannot pick up and carry away.
One asymmetry sits underneath much of this series: a mother always knows the child is hers, and until 1985 no father could be certain. That is a difference in information, not in love.
A second sits beside it: the minimum cost of a child is one act for a man and years for a woman. Roughly four hundred eggs in a lifetime against tens of millions of sperm a day.
This genuinely predicts that some male concern with paternity appears nearly everywhere. It does not predict veiling, dowry, child marriage, property law or killing — all of which vary hugely between societies with identical biology. A constant cannot explain a variable.
Several Amazonian societies believe a child can have more than one father, and the extra fathers bring food. In one studied group it raised child survival measurably. Same biology, opposite institution.
Biology is a necessary condition, not a sufficient one. Rain is necessary for a flood, but the reason this valley floods and the next does not is drains — and drains are built by people.
2Before Anybody Owned Anything
For all but the last few minutes of human existence, nobody had land, a granary, or anything they could not carry. The rules about women in such societies are looser almost everywhere they have been recorded. Understanding why is the key to the whole part.
2.1 — What the record shows
Anthropologists have documented foraging societies across four continents — the Hadza in Tanzania, the Ju/’hoansi of the Kalahari, the Aka of the Congo basin, the Aché of Paraguay, groups across Aboriginal Australia and the Arctic.
These groups differ from each other in many ways. But a set of features recurs often enough to be worth stating plainly.
Marriage is easier to leave. Among several documented foragers, divorce is common and carries little stigma. A woman who wants to end a marriage takes her things — which are few — and goes to another camp, frequently to her own relatives. She loses very little because she owned very little to begin with.
There is no dowry and no bride price of the kind Chapter Five describes. What appears instead is often bride service: a young man works for his wife’s family for a period, sometimes years. Note the direction. The labour flows towards the woman’s family, not away from it.
Female sexual behaviour is regulated but not policed in the way this series is examining. Infidelity causes conflict, as it does everywhere. What is generally absent is the machinery — the seclusion, the veiling, the killing of a daughter to repair a family’s standing, the treatment of a woman’s chastity as a family asset.
Food is shared beyond the household. In many foraging groups a large kill is distributed across the whole camp by custom rather than by the hunter’s choice. This matters more than it sounds, and §2.2 explains why.
Fathers are often heavily involved. Among the Aka, fathers have been recorded holding and caring for infants more than in any other society yet measured. So much for the idea that low male investment is the natural state.
2.2 — Why: there is nothing to leave anybody
The explanation is not that these people were kinder. It is structural, and once you see it the rest of this part follows.
Consider what a mobile foraging group can own. Tools, weapons, clothing, ornaments. Things a person can carry while walking. There is no field, no granary, no herd, no house that stays where you left it.
Now ask the question that drives everything: what does a man have to pass on, and to whom?
The answer is: very little of material value. His skill, perhaps his standing, a few possessions. In a society where food is shared across the camp by custom, even the product of his own labour is not fully his to direct.
So the stakes attached to paternity are low. Not zero — a man still prefers to support his own children, and jealousy exists everywhere. But there is no estate, so there is no inheritance question, so there is no reason to construct an elaborate apparatus for ensuring that an estate goes to the right body.
Now hold that sentence up against the next chapter, where somebody plants a field.
Heritable property: wealth that survives its owner and passes to somebody specific — land, buildings, herds, stored grain.
The contrast is with wealth that cannot be inherited in any meaningful way: a person’s skill, strength, or a share of a kill that will be eaten this week.
Why it matters here: this distinction does more explanatory work than anything else in this part. Where wealth is heritable, who gets it becomes a question, and the answer runs through a woman’s body.
2.3 — The comparison is not fair, and the unfairness cuts both ways
Before this becomes a story about a lost paradise, several serious problems with the evidence need stating.
Nearly every claim about “how humans originally lived” rests on studying societies that still forage today. That inference is heavily contested.
The pattern is not one group but many, on separate continents, with no contact between them. When societies in Tanzania, Paraguay and the Kalahari independently show looser marriage, easier divorce and no dowry, the common factor is unlikely to be coincidence. And the structural explanation is testable: it predicts the pattern should track mobility and storable wealth, and it does.
Every forager studied in the twentieth century lives on land that farmers and states did not want — deserts, deep forest, Arctic margins. They have traded with, fled from and been reshaped by agricultural neighbours for thousands of years. They are not a window into the past; they are contemporary societies pushed into extreme environments. Ancient foragers living in rich river valleys may have had storage, hierarchy and rules we would find entirely familiar. Archaeology has begun turning up exactly that.
What would settle it: archaeological evidence about wealth and inequality in pre-agricultural societies in rich environments. This work is ongoing, and early results suggest more variety in the deep past than the simple story allows.
Why people care so much: because a great deal rides on whether the current arrangement is humanity’s default or one recent option. Both camps want the deep past on their side, which is a bad reason to consult it.
2.4 — Not a paradise
The record also contains things the romantic version leaves out, and leaving them out would be exactly the kind of selective reporting this series is supposed to avoid.
Violence in some foraging societies is high. Among several documented groups, the share of deaths caused by another person exceeds that of most modern states by a wide margin. Life was not gentle.
Marriage was frequently arranged, sometimes for girls who were very young, and sometimes to considerably older men. In parts of Aboriginal Australia, betrothal at or before birth was recorded. Freedom to leave a marriage is not the same as freedom to choose one.
Men typically held political authority in the sense that camp decisions and disputes were often settled among them. Greater equality is not equality.
And infanticide, including the sex-selective kind, is recorded in a number of foraging societies. The most humbling thing in this chapter is that a practice we associate with the ultrasound machine long predates it.
So the honest summary is narrow and I want to keep it narrow: foraging societies show less of the specific apparatus this series examines — the seclusion, the property in a woman’s chastity, the marriage payments — and they show plenty of other harms. The claim is about a particular machine, not about human happiness.
2.5 — What actually changed
One more thing to carry into the next chapter, and it is the hinge of the whole part.
Foragers were not free of jealousy, or of male authority, or of violence against women. All of that was present. What was missing was the institutional version — rules written into property law, marriage contracts, inheritance and religious teaching, enforced by whole communities rather than by individual angry men.
That is the difference between a feeling and a system. Feelings are ancient and probably universal. Systems are built, and they are built out of specific materials for specific reasons.
The material that mattered most arrived when somebody planted a seed, waited, and could not walk away.
Across foraging societies on four continents, a pattern recurs: divorce is easy, there is no dowry, and the machinery of controlling women’s sexuality is largely absent. Bride service runs the other way — the young man works for her family.
The reason is structural, not moral. A mobile group owns only what it can carry, so there is no estate, so there is no inheritance question, so there is no need for an apparatus to route property through the right body.
The comparison has a real flaw: every forager studied in modern times lives on land nobody else wanted. They are not a window into the past. The claim that survives is narrower and stronger — control over women tracks heritable, immobile wealth, and that holds across farming societies too.
These were not gentle societies. Violence was often high, marriage was frequently arranged, girls were sometimes betrothed very young, and infanticide including the sex-selective kind is recorded.
What foragers lacked was not jealousy or male authority. It was the institutional version — rules written into property, marriage and inheritance, enforced by everybody. That is the difference between a feeling and a system, and systems are built out of specific materials.
3The Field That Cannot Be Carried
About twelve thousand years ago people began planting seeds and waiting. Everything in this series follows from what that made possible — and from what it made necessary.
3.1 — The seed and the wait
Farming did not arrive once. It appeared independently in at least half a dozen places — the Fertile Crescent, northern China, Mesoamerica, the Andes, the New Guinea highlands, parts of Africa — over several thousand years, among people with no contact with each other.
That independence matters. When separate populations arrive at the same arrangement without copying, the arrangement is being produced by the situation rather than by a culture.
What planting a seed does is create a delay. You put something in the ground and you get nothing for months. During that gap the field is valuable, vulnerable, and — the crucial part — immobile. You cannot pick it up. You cannot take it with you if you leave. Its value depends entirely on your being able to come back and on nobody else taking it.
So the first thing farming produces is not food. It is a reason to stay, and a reason to defend.
The change in what land can support is easy to state and hard to feel, so here is a version with a room in it.
Foraging in a typical environment supports somewhere in the region of one person for every ten square kilometres. Early farming supports something like a hundred times more people on the same ground.
Put it in a village. The land that fed a single foraging family now feeds a settlement of a hundred or more. Those hundred people live within shouting distance of each other, permanently, for generations, and they can all see what everybody else’s daughter is doing.
That last sentence is not a joke. Surveillance is a by-product of density, and density is a by-product of the plough. Nobody chose it.
3.2 — Storage, and the first thing worth stealing
The second thing farming produces is a surplus that keeps. Grain can be stored. Stored grain is wealth that survives the person who grew it.
This is genuinely new in human history, and it has three consequences that arrive together.
Inequality becomes possible and then permanent. In a foraging camp where a kill is shared by custom, a good hunter cannot accumulate. With a granary, a household can. And once one household has more, the difference can be passed on, and the gap compounds across generations rather than resetting each time.
Property needs defending. A granary can be raided. A field can be seized. So farming societies develop what foraging bands mostly did not: organised violence over resources, walls, and eventually the specialists we call soldiers.
And property has to go somewhere when you die. This is the one that concerns us.
3.3 — Why the question runs through a woman’s body
Here is the chain, and I want to lay it out step by step because it is the engine of this entire series and every later part refers back to it.
A man farms a field. The field will outlive him. When he dies it must go to somebody, or it will be taken by whoever is nearest.
He wants it to go to his children — this preference does not need explaining, it appears everywhere. But recall Chapter One: he cannot verify which children are his. His wife can. He cannot.
So the estate creates a problem that did not previously exist in any serious form. It is not a problem about sex. It is a problem about verification, and it becomes urgent precisely in proportion to how much there is to pass on.
A society can solve this in several ways, and different societies chose differently.
Solution one: pass property through women instead. If wealth goes from a woman to her children, verification is unnecessary — she always knows. Chapter Seven is entirely about the societies that took this route, several of which were in India.
Solution two: pass property to your sister’s children. Your sister’s children are certainly related to you, whatever anybody does. This is a common arrangement and it works, at the cost of a man having no formal stake in the children he raises.
Solution three: control the woman. Restrict her movement, her contacts, her clothing, the age at which she marries, and who she may speak to. Make her behaviour a matter for the whole family rather than for her, so that the family will do the enforcing.
The third solution is expensive. It requires constant effort by many people. But it has one enormous advantage over the others: it lets a man pass property to children he raises, while remaining reasonably confident they are his. The first two solutions each break one of those. Only the third delivers both, and that is why it spread.
I want to be precise about what I am claiming here. I am not saying men gathered and chose this. I am saying that among the available arrangements, this one produced a particular outcome — concentrated, defended, heritable wealth in a male line — and societies organised that way tended to out-compete, absorb and displace their neighbours. Nobody had to intend it for it to spread.
3.4 — What the genetics shows
There is physical evidence for something dramatic happening to men’s reproduction in exactly this period, and it is one of the strangest findings in the field.
Y chromosome and mitochondrial DNA: two pieces of genetic material that are inherited in a special way. The Y passes only from father to son. Mitochondrial DNA passes only from mother to all her children.
This makes them a pair of measuring instruments. The variety in a population’s Y chromosomes tells you roughly how many men were passing on genes. The variety in the mitochondrial DNA tells you the same about women. Comparing the two tells you whether the sexes were reproducing at similar rates.
Nobody set out looking for a fact about ancient marriage. Researchers were sequencing male genomes for entirely unrelated reasons, and the pattern fell out of the data.
Researchers sequencing large numbers of male genomes across Africa, Europe and Asia found something unexpected. Beginning roughly seven thousand years ago and lasting a few thousand years, Y chromosome diversity collapsed — while diversity in the female line, in the very same populations, did not.
What it shows: that during this window, far fewer men than women were passing on their genes. The effect appears in multiple regions independently, which makes a local explanation unlikely. Some analyses put the imbalance at extreme levels; the exact figures depend heavily on the model chosen and should be treated as indicative rather than precise.
What it cannot show: the cause, and there are two serious candidates. One is that a small number of powerful men monopolised reproduction. The other, now widely favoured, is that men had begun organising into competing patrilineal clans, and that whole male lineages were periodically wiped out together in conflict between them — which produces the same genetic signature without requiring extreme polygyny.
Both explanations, note, describe a world newly organised around male descent lines and property. The timing sits directly on top of the spread of farming.
Notice how easily that finding invites a story with a villain in it, and how little the evidence actually supplies one. That pull is worth examining directly, because it operates on both sides of this argument and neither notices.
Almost everybody arguing about this history assumes there was an author.
The traditionalist speaks of the wisdom of the ancestors, as though a council of thoughtful elders examined the options and selected well. The reformer speaks of a system built by men to serve men, as though a decision was taken and enforced.
Neither can produce the meeting. There is no document, no founder, no moment. What the evidence actually shows is thousands of separate households making small decisions about land, marriage and inheritance across thousands of years, in which some arrangements happened to spread because the societies holding them grew, absorbed neighbours and lasted.
This matters practically, not just philosophically. If there is no author, then “who is to blame” has no clean answer — which frustrates one camp — and “the ancestors knew what they were doing” has no author either, which should frustrate the other far more than it does.
The general form: mistaking an outcome for an intention. A path worn across a park was not designed, and it is still exactly where people wanted to walk. Both facts are true and neither is a plan.
3.5 — What this does and does not establish
The chain in §3.3 is the standard account, it is well supported, and it should not be oversold. Here is the honest scoring.
Strongly supported: that control over women’s sexual behaviour tracks the presence of heritable, immobile property. This holds across the ethnographic record, across historical societies, and — as Chapter Four shows — within farming societies too, which is the strongest version.
Reasonably supported: that inheritance is the mechanism. The alternatives that have been proposed do not fit the pattern as well, and the ones that do fit tend to be inheritance under another name.
Not established: the precise sequence. Whether property produced the rules or the rules made a certain kind of property possible is genuinely hard to disentangle, because in the record they arrive together. Chapter Four contains the closest thing available to an answer, and even that is not clean.
Farming appeared independently in at least half a dozen places among people with no contact. When separate populations reach the same arrangement without copying, the situation is producing it, not the culture.
Planting creates a delay and a thing you cannot carry. From that follow settlement, density, storable surplus, permanent inequality, and organised violence over resources.
And an estate has to go somewhere. A man wants it to go to his children and cannot verify which they are. That is not a problem about sex. It is a problem about verification, and it grows in proportion to how much there is to pass on.
Three solutions exist: pass property through women, pass it to your sister’s children, or control the woman. Only the third lets a man pass wealth to children he raises while staying confident they are his. That is why it spread.
Genetics shows something matching. Around seven thousand years ago the male line’s diversity collapsed while the female line’s did not, in several regions at once — a world newly organised around male descent and property.
Nobody designed any of this. There was no meeting. Which means blame has no clean target, and “the ancestors knew what they were doing” has no author either.
4The Plough
Whether your ancestors farmed with a plough or a hoe predicts how many women work in your country today. It predicts it for the grandchildren of people who left those countries. This is the most surprising piece of evidence in the part, and it needs handling carefully.
4.1 — Two ways to break ground
Not all farming is the same, and one distinction turns out to matter enormously.
Shifting cultivation with a hoe or digging stick. You clear a patch, often by burning, and plant with hand tools. The work is spread across the season. It can be interrupted. It can be done while carrying a child, and it can be done in short bursts between other tasks.
Plough agriculture. A heavy blade drawn through the soil by an animal. It requires enough upper-body strength and grip to control both the implement and the animal, and it demands sustained bursts of effort at fixed points in the year. It cannot easily be interrupted, and taking an infant into a field behind an ox is not a serious option.
The consequence is that plough farming pushes field labour towards men. This is not an ideological claim; it is a description of who can physically do the task while also doing the other thing human societies need doing. Part Four takes up the physical differences properly, with numbers. Here it is enough to say that the gap in upper-body strength between men and women is among the largest measured human sex differences, and a plough sits exactly on it.
So in plough societies, the field becomes male space. Women’s work moves indoors. And over generations, a division of labour that began as a fact about a tool turns into a belief about what women are for.
4.2 — The observation, and then the test
A Danish economist writing in 1970 noticed the pattern first: in Africa, where hoe cultivation dominated, women did a large share of farm work and had corresponding economic independence; in much of Asia and Europe, where the plough dominated, they did not.
It was a good observation. For forty years it remained an observation, because you cannot easily test a claim about thousands of years of history.
Then, in 2013, three economists found a way.
The researchers took a database of several hundred pre-industrial societies, coded by anthropologists for whether they used the plough. They matched each modern population to the traditional practices of its ancestors.
Then they looked at the present day: how many women work outside the home, how many own businesses, how many sit in parliaments, and what people in surveys say about whether men should have priority for jobs.
The finding: descendants of plough-using societies score lower on every one of these. The relationship is strong and survives controls for income, religion, geography and colonial history.
The clever part. A correlation like this could easily run backwards — perhaps societies that already restricted women adopted the plough. So the researchers used something no society chooses: the physical suitability of its land for crops that need a plough, such as wheat and barley, against crops that do not, such as sorghum and millet. Soil and rainfall are not caused by anybody’s opinion about women. The relationship held.
What it cannot show: that crop suitability affects nothing else. Wheat-growing regions also developed states, cities, writing and armies earlier, any of which could carry the effect instead. This is the main criticism and it is serious.
4.3 — The grandchildren
Then comes the result that makes this study famous, and it is the one to hold on to.
The researchers looked at the children and grandchildren of migrants living in the United States and Europe — people born in a new country, educated in its schools, working in its economy, subject to none of its ancestors’ farming conditions.
The pattern was still there. Women descended from plough-using populations were less likely to work outside the home than women descended from hoe-using populations, living in the same country, under the same laws, in the same decade.
Think about what that isolates. The land is gone. The crop is gone. The plough is gone, four thousand years gone. What travelled was the belief about what women do — carried across an ocean inside families, and still measurable in employment statistics generations later.
A woman in Chicago decides whether to take a job. She has never seen a plough. Her grandmother never saw one.
But somewhere behind her, in a village she has never visited, the soil suited wheat rather than millet. Because of that, men worked the fields. Because of that, a belief formed about where women belong. That belief was taught to children, who taught it to children, who boarded a ship.
Now put a number on the distance. Between the plough that started it and the decision she is making sits something in the order of two hundred generations. If each generation were one step, that is a walk of about ten minutes. If each were a year, it would reach back before the Roman Empire.
That is the reach of a farming tool.
4.4 — How much of this to believe
This is a striking result and striking results deserve suspicion, particularly when they flatter a story you already find appealing.
It is among the most cited results in economics on this subject, and it is not unchallenged.
The result appears across dozens of countries, several outcome measures and multiple datasets. The migrant test is a genuinely strong design: it removes the land, the crop and the institutions and finds the effect anyway. And the mechanism is not mysterious — the physical demands of a plough are not in dispute, and a division of labour hardening into a belief is a process visible in many other domains.
Crop suitability is not a clean instrument. Wheat regions also developed early states, dense cities, standing armies, writing and organised religion. Any of these could be the actual carrier, with the plough merely along for the ride. The ancestral coding comes from the colonial-era ethnographic record, with all the flaws described in the front matter. And migrant communities differ in a hundred ways besides ancestral farming, including why they migrated and where they settled.
Something about the deep agricultural past transmits through families and shows up in women’s employment today. Whether the carrier is specifically the plough, or the wider package of state formation and property law that came with plough regions, is not settled. The persistence is the robust finding. The mechanism is the contested one.
What would settle it: finding places where plough use and state formation came apart — regions with early states and hoe farming, or plough farming without states — and testing whether the effect follows the tool or the state. Some work is doing this and the results so far are mixed.
Why people care so much: because the persistence finding cuts both ways and everyone notices only their half. It tells the traditionalist that these norms are historical rather than natural — a farming accident, not a design. It tells the reformer that culture is far stickier than policy, and that norms outlive by centuries the conditions that justified them.
4.5 — What this means for India
India is a useful test because it contains both systems inside one country.
Northern India is largely wheat country, and wheat is a plough crop. Much of the south and east has historically grown rice and millets under different regimes with different labour patterns.
And the north and south of India differ sharply on exactly the measures this chapter is about. Demographers have documented the divide for decades: the north shows lower female workforce participation, more restrictive kinship arrangements, marriage patterns that send a bride far from her own family into a village of strangers, worse child sex ratios, and stronger seclusion norms. The south shows marriage closer to home, more contact with a woman’s own relatives after marriage, and higher female participation in public life.
This is not a claim that the south is free of any of this. It plainly is not. It is a claim that the strength of the machinery varies inside one country, one religion and one legal system — which is exactly what you would predict if the cause were agricultural and material rather than doctrinal.
Hold that finding, because Chapter Eight will use it. If the rules were produced by scripture, they should not vary this much between regions sharing the same scripture. They do.
Ploughing needs sustained strength and grip and cannot be interrupted, so it pushes field labour towards men. Hoe cultivation does not. Over generations a fact about a tool hardens into a belief about what women are for.
Descendants of plough-using societies today show lower female employment, fewer women in business and politics, and more restrictive attitudes. The researchers tested it against something nobody chooses — whether the local soil suits wheat or millet — and it held.
The finding that matters most: the effect survives migration. Women descended from plough societies work less than women descended from hoe societies while living in the same country under the same laws. The land, the crop and the plough are all gone. The belief travelled anyway.
Treat it carefully. Wheat regions also built early states, cities and armies, any of which might be the real carrier. What is solid is the persistence; the mechanism is contested.
India contains both systems. The wheat-growing north shows exactly the pattern predicted — and it shares its scripture with the south. If doctrine wrote these rules, they should not vary this much between people reading the same book.
5The Price of a Bride
In most societies money changes hands at a wedding. Which direction it flows tells you almost everything about how that society values a woman — and India banned its version in 1961, after which it grew.
5.1 — Two payments, opposite directions
Across the ethnographic record, marriage almost always involves a transfer of wealth between families. There are two main kinds and they run opposite ways.
Bridewealth: payment from the groom’s family to the bride’s family. Cattle, goods, money, or labour. It is the more common of the two by a wide margin — the majority of societies in the ethnographic record practise some version.
Dowry: payment from the bride’s family to the groom’s family, or to the couple. Much rarer as a share of societies, but those societies include some of the largest populations on earth: historic Europe, and South Asia.
Why it matters here: these are not two versions of the same custom. They point in opposite directions and they mean opposite things.
Bridewealth says, in effect: your daughter is valuable and we are compensating you for the loss of her. Dowry says: taking our daughter is a cost to you and we are compensating you for it.
Same institution, same ceremony, and a completely inverted valuation of the same person. Which one a society has is not random.
5.2 — Which societies have which
The pattern maps almost exactly onto Chapter Four.
Bridewealth is concentrated in societies where women’s labour produces visible wealth. Hoe agriculture, especially across sub-Saharan Africa. A woman in such a society is an economic asset in a direct sense: she farms, and the food she grows is real. Her family gives up that production, and the groom’s family pays for it.
Dowry is concentrated in societies with plough agriculture, sharp social stratification, and monogamy. A woman in such a society does not work the fields. She is, in the household’s accounting, a cost — to be fed, housed, and married off. Her family pays somebody to take on that cost.
There is a second thing dowry does, and it is the more important one. In a stratified society with monogamous marriage, there is a limited supply of desirable grooms, and families compete for them. Dowry becomes a bid. It is not only a transfer of property; it is a price in a market where the item being bought is a son-in-law of good standing.
A scholar working on this in the 1970s made the further point that dowry is often, in origin, a daughter’s inheritance paid early — her share of the family property, handed over at marriage rather than at a death. That is a genuinely different thing from a purchase price, and the two functions can coexist in one custom, which is part of why the argument about dowry is so confused.
Every modern argument about marriage — traditionalist and reformist alike — assumes the couple is the unit. The debate is about whether two people should choose each other freely, marry young or late, stay or leave.
For most of the history examined in this part, the couple were not the unit and were frequently the least consulted parties. Marriage was a contract between two families, transferring property, labour and alliance. The bride and groom were the terms of the agreement, not the signatories.
This is why the modern conversation so often fails to connect. When a young woman says the choice is hers and her family says it is not, they are not disagreeing about her preferences. They are using two incompatible definitions of what a marriage is. Neither is being irrational within their own frame.
The general form: a dispute where the two sides have different objects in mind and both think it is a dispute about values. Once you see this one, a very large number of family arguments in India stop looking like generational conflict and start looking like a contract dispute over who the parties are.
5.3 — India banned it, and it grew
Now the fact that should stop anybody who believes a law can settle this.
India passed the Dowry Prohibition Act in 1961. Giving or taking dowry became a criminal offence. The law was strengthened in the 1980s, and specific provisions were added to the criminal code covering cruelty to a wife and deaths occurring in suspicious circumstances within seven years of marriage.
Since then, dowry has not declined. By most accounts it has expanded — in geographic reach, in the communities practising it, and in real value. Groups that historically paid bridewealth have shifted to dowry. Regions where it was once uncommon now report it. The custom did not shrink under pressure. It spread.
Reported dowries in India commonly run to several times a family’s annual income. Put that in a household.
Imagine a family whose entire income for a year is a single stack of notes on a table. Marrying one daughter takes the stack — and then two or three more years of stacks that have not been earned yet. It is paid in gold, in a vehicle, in furniture, in cash, and often in debt that outlives the wedding by a decade.
Now put a second daughter in the house. Then a third. And then ask why a family in a hospital, offered a scan that reveals the sex of a foetus, might make the decision that Part Twelve counts.
The line from a wedding gift to a missing girl is not rhetorical. It is arithmetic, and the families doing it can do the arithmetic.
5.4 — Why the ban failed
Understanding why a fifty-year-old criminal prohibition failed is more useful than condemning it.
Both parties want the transaction. This is unlike most crimes. There is no complainant at the moment of the offence. The bride’s family pays willingly, because the alternative is their daughter not marrying, and in a society where marriage is the main route to a woman’s security, that is a worse outcome for her. The state is trying to prohibit an exchange that both sides prefer to the alternative.
Reporting it destroys the thing it was meant to protect. A woman who reports her in-laws is in most cases ending her marriage, with all that follows for her and for her sisters’ prospects. The law asks her to pay the entire cost of its enforcement.
It is easy to relabel. Gifts to a daughter are lawful, and the line between a gift and a payment exists only in intention. A custom that can be renamed cannot be banned.
And the underlying cause was untouched. Dowry is a price in a marriage market. Prices respond to supply and demand, not to legislation. As long as families compete for a limited pool of acceptable grooms, and as long as a daughter’s alternative to marriage is poor, there is upward pressure on the bid. Banning the payment does not change the market. It moves it.
Remember Part One, Chapter Three — the asylums closed and the replacement never built. This is the same structure. India removed the legality of a transaction without removing the conditions that produced it. The transaction continued and lost only its records.
Officially recorded deaths connected to dowry disputes in India run into the thousands each year. That figure comes from police records, which means it counts what was classified and prosecuted as such. Everybody working in this area treats it as a floor rather than an estimate. Part One’s rule applies: a state with the power to count and a weak record of counting is telling you something.
Marriage almost always moves wealth between families, in one of two directions. Bridewealth — groom’s family pays bride’s — is the more common worldwide, and appears where women’s labour produces visible wealth. Dowry runs the other way and appears with plough farming, stratification and monogamy.
The two encode opposite valuations of the same person. Bridewealth says she is worth compensating for. Dowry says taking her is a cost.
Dowry is also a bid in a competitive market for a limited supply of acceptable grooms, which is why it inflates.
India banned it in 1961 and it grew — spreading to communities that had never practised it. The ban failed because both families want the transaction, reporting it destroys the marriage it was meant to protect, a payment can be relabelled a gift, and the market underneath was never touched.
The line from a wedding payment to a daughter who is never born is not rhetoric. It is arithmetic, and the families doing it can do the arithmetic.
6Honour Is a Technology
Where there is no police force worth calling, reputation does the work of law. That is not a metaphor. It is a functioning enforcement system, and understanding how it works explains why a family’s standing came to be stored inside a woman.
6.1 — The problem honour solves
Imagine you live somewhere with no reliable police, no court you can reach, and no authority that will enforce an agreement. This is not exotic. It has been the ordinary condition of most people in most places for most of history, and it remains the practical situation in parts of the world today.
You have a problem. Anybody stronger than you can take what you have, and there is no institution to stop them. What protects you?
Only one thing: their belief about what you will do afterwards. If people believe that taking from you is expensive — that you will retaliate, at cost to yourself, without calculating whether it is worth it — they leave you alone.
That belief is what honour is.
Culture of honour: a social system in which a person’s or family’s reputation for retaliating against insult and injury functions as their main protection, in place of law.
The essential feature is that the response must be disproportionate and unhesitating. A man who weighs up whether a small insult is worth fighting over has already lost the protection, because he has shown he calculates. The whole value lies in being known not to.
Why it matters here: this is not backwardness or emotion. It is a functioning security system, and it is rational for exactly as long as no better one is available.
6.2 — Herders and thieves
Honour cultures are not evenly distributed. They cluster, and the clustering has an explanation.
Compare two kinds of property. A field cannot be stolen. Somebody can occupy it, which is visible, slow and contestable, but they cannot take it away in the night.
A herd of animals can be. Livestock walks. It can be moved fifty kilometres before morning, and it is difficult to prove which goat was yours.
So herders face a security problem that farmers do not. Their entire wealth is portable, and in the absence of a state, the only defence is deterrence. Herding societies across the world — the Mediterranean, the Middle East, Central Asia, the Balkans, the mountains of Europe, the American frontier — show markedly more honour-related violence than neighbouring farming populations.
The strongest evidence is not historical but experimental, and it was done on students.
Researchers in the 1990s recruited male university students in the American Midwest, some raised in the South and some in the North. Each was sent down a narrow corridor where a stranger, working for the experimenters, bumped into him and called him an insulting name.
What they found: the Northern students mostly shrugged it off. The Southern students showed sharp rises in stress and dominance-related hormones, judged the insult far more seriously, and behaved more aggressively in unrelated tasks afterwards. The effect appeared in the body, not only in what people said.
The historical link: much of the American South was settled by herding populations from the borderlands of Britain and Ireland. The South’s higher homicide rate is concentrated specifically in killings that arise from arguments and insults, not in killings committed during robberies — exactly the signature the theory predicts.
What it cannot show: that herding ancestry is the cause rather than something correlated with it — slavery, poverty, and frontier conditions are all candidates and all present. And these were students in one country in one decade. The cross-cultural pattern is suggestive rather than decisive.
6.3 — Why the reputation was stored in a woman
Now the step that concerns this series, and it is the one that is rarely spelled out.
If a family’s safety rests on its reputation, then reputation is capital. It has to be maintained, displayed, and defended. And a reputation needs a signal — something visible that outsiders can check.
Here is the problem with most signals: they are easy to fake. A man can claim to be fierce. He can carry a weapon he has never used. Talk is cheap and everybody knows it.
A family’s control over its women is different, and this is the uncomfortable core of the chapter. It is continuously visible, expensive to maintain, and impossible to fake. Whether a daughter is out after dark, who she speaks to, how she dresses, whether she married the man she was told to — all of it is observable by every neighbour, every day, without anybody having to investigate.
So a woman’s conduct becomes the display on which the family’s standing is read. Not because anybody decided her behaviour was morally central, but because it was the most reliable signal available in a society with nothing else to go on.
And that explains the feature of honour systems that otherwise makes no sense: the punishment is wildly out of proportion to the act. If the offence were the sexual behaviour itself, a smaller response would do. But the offence is a public demonstration that the family cannot control its own household — which is a demonstration that it cannot enforce anything, which is a demonstration that it can be robbed.
The violence is not aimed at the daughter. It is aimed at the audience.
Describing something as a functioning system invites the accusation of defending it, so let us have the argument openly.
In the absence of enforceable law, deterrence is the only available protection, and deterrence requires a credible reputation. Families operating this way survived and prospered; families that did not were preyed upon. Calling it irrational is the comfortable judgement of people who have never lived a day without a functioning police force, and it explains nothing.
The system produces enormous, permanent costs — dead daughters, ruined lives, endless feuding — for a benefit that is largely defensive against other people running the same system. Two families locked in mutual deterrence are both worse off than if neither played. Calling that rational is to confuse individual logic with collective disaster. It is a trap, not a solution.
Each family behaving this way is doing the sensible thing given what the others are doing. The result for everybody is worse than an alternative none of them can reach alone. It is rational at the level of the household and catastrophic at the level of the society, and both statements are true at once.
What would settle it: whether honour violence declines when the state becomes capable of enforcing law. Broadly it does — but slowly, and much more slowly than the state’s arrival, which tells us the system outlives the problem it solved. Chapter Nine takes that up.
Why people care so much: because “rational” is heard as “justified”. It is not. Part One, Chapter Four: explaining a mechanism does not endorse it, and no fact about how something works produces a conclusion about whether it should continue.
6.4 — Honour in India
India’s version has a feature that the Mediterranean and Middle Eastern cases mostly lack, and it makes the Indian system harder to dismantle rather than easier.
In most honour cultures, the offence is a woman’s sexual behaviour as such. In India, the most severely punished offence is frequently something more specific: marrying the wrong person. Outside the caste. Inside the wrong sub-group. Against a family’s arrangement.
This points at a different machine. It is not primarily about paternity or reputation for fierceness. It is about maintaining a boundary — keeping a group closed by controlling who may marry whom. Part Three is entirely about that machine, because it is the specifically Indian one and it is the most powerful thing in this series.
On the numbers, I have to report a gap rather than a figure. India’s official crime statistics recorded honour-related killings only as a distinct category relatively recently, and the recorded totals are strikingly low — dozens per year in a country of 1.4 billion. Researchers and organisations working in the field give estimates in the high hundreds or above.
I am not going to pick a number between those, because I cannot justify one. What I will do is apply the rule from Part One: when a state has the power to count something and the count comes back implausibly small, the absence is itself a finding. A killing recorded as a suicide, an accident, or a simple murder does not appear in this column. Neither does a girl who was never reported missing.
Where there is no reachable law, the only protection is other people’s belief that attacking you is expensive. That belief is honour, and it must be unhesitating — a man who visibly calculates has already lost it.
Honour cultures cluster among herders, because livestock can be stolen and land cannot. Insulted Southern American students in one experiment showed measurable hormonal spikes where Northern students shrugged — and the South’s excess homicide is specifically in killings arising from arguments, exactly as the theory predicts.
A reputation needs a signal that cannot be faked. Control over a family’s women is continuously visible, expensive, and impossible to fake — which is why a family’s standing came to be stored in a daughter’s conduct.
That also explains why the punishment is so wildly out of proportion. The offence is not the act. It is the public demonstration that the family cannot enforce anything — the violence is aimed at the audience, not the daughter.
It is rational for each family and catastrophic for all of them together. Which means no single family can safely stop first, and the escape has to be collective.
India’s version punishes marrying the wrong person most severely of all. That is not a paternity system. It is a boundary system, and Part Three is about what it is keeping out.
7The Societies That Did It Backwards
Property through daughters. Husbands who visit rather than move in. Children who belong to their mother’s line. All of this existed, some of it in India, some of it within living memory — and knowing exactly how it ended matters more than knowing it existed.
7.1 — Kerala
Among the Nayar communities of Kerala, until the twentieth century, the arrangement worked roughly like this.
The household was a large joint family traced through women — mother, her daughters, her daughters’ children, and her brothers. Property belonged to this household as a whole and passed down the female line. It was managed, in practice, by the eldest male, who was typically the mother’s brother.
Marriage as most readers understand it did not exist. A woman entered relationships that could be ended by either party without formality or shame. The men in these relationships did not move in, did not own her, and did not support her children. Her children were raised by her household — by her, her sisters, and her brothers.
So a man’s economic and paternal role went not to his own children but to his sister’s. He was a maternal uncle first and a father second, and the system asked nothing of him as a father at all.
Now notice what that does to Chapter Three. The verification problem simply disappears. Nobody needs to know who fathered a child, because nothing is being transmitted through fathers. There is no reason to restrict a woman’s sexual behaviour to protect an inheritance, because the inheritance does not travel that road.
And, as far as the record shows, the restrictions were correspondingly light.
Matrilineal: a system where descent, family membership and property pass through the mother’s line.
Matriarchal: a system where women hold political power.
These are constantly confused and they are not the same. Most matrilineal societies are not matriarchal — men typically still hold formal authority, but they hold it as brothers and uncles rather than as husbands and fathers. There is no well-documented human society in which women held political power the way men do in patriarchal ones.
Why it matters here: the honest claim is not that these societies were run by women. It is that they solved the inheritance problem without controlling women’s sexuality, which is a narrower claim and a much more useful one.
7.2 — Meghalaya, Sumatra, Yunnan
Kerala is not an isolated curiosity. There are others, and several are alive now.
The Khasi, Jaintia and Garo of Meghalaya, in India’s north-east, remain matrilineal today. Children take the mother’s clan name. Among the Khasi, ancestral property passes to the youngest daughter, who also takes on the care of her parents. A husband typically moves into his wife’s household rather than the reverse — which inverts the north Indian pattern where a bride is sent away to a village of strangers.
The Minangkabau of West Sumatra are the largest matrilineal society on earth. Ancestral land passes through women and cannot be sold by men. Houses belong to the female line.
How large is “the largest matrilineal society on earth”? Somewhere in the region of six to eight million Minangkabau people.
That is roughly the population of Hyderabad. More than Denmark. More than Singapore.
This is worth pausing on, because matriliny is usually presented as a handful of small tribal groups that prove nothing about how a real society can be organised. A population the size of a major city, with cities of its own, universities, a written literature and a long trading history, is not a curiosity. And there is a second thing about the Minangkabau that Chapter Eight is going to need: they are Muslim, and have been for centuries.
The Mosuo of Yunnan, in south-western China, are much smaller — some tens of thousands — but the arrangement is the most striking. There is no institution of marriage. A woman may receive a partner at night who returns to his own mother’s house in the morning. Children belong to the mother’s household and are raised by her and her brothers. The partner may be known to everybody or not; either is acceptable.
Both camps hear about these societies and make the same mistake. They assume the choice is between rules and no rules.
The traditionalist hears “no marriage, no restriction on women’s partners” and concludes: chaos, a society without structure. The reformer hears it and concludes: freedom, a society that escaped the machinery.
Neither is right, because these are not unregulated societies. They are heavily regulated societies regulating different things. The Mosuo have strict rules about who may not be a partner, about the absolute obligations of a brother to his sister’s children, and about how a household is run. Khasi inheritance is not a free-for-all; it is a specific rule about a specific daughter. The Minangkabau have elaborate law governing ancestral land.
What changed is not the quantity of regulation. It is who is regulated and in which direction. In a patrilineal system a woman’s sexual behaviour is constrained and a man’s obligation to his sister’s children is loose. In a matrilineal one, that is reversed: a man’s obligations are heavy and fixed, and her behaviour is her own.
The general form: reading an unfamiliar structure as an absence of structure. It is the same error a foreigner makes on first seeing a caste system and concluding there are no classes because there are no classes he recognises.
7.3 — Why matriliny keeps collapsing
Here is the fact that stops this chapter from being a comfortable story. Matrilineal societies have been disappearing for centuries, almost everywhere, and they have not been replaced by anything more equal. They are replaced by patrilineal systems.
Anthropologists have a name for the internal tension that makes them fragile.
The matrilineal puzzle: the structural strain in any matrilineal system caused by a man being pulled in two directions. His formal duties are to his sister’s children, who share his lineage. His affection and daily life are with his own children, who do not.
Why it matters here: this is not a flaw somebody could fix. It is built into the design, and it becomes acute the moment a man acquires wealth of his own that he could choose to pass on.
That last sentence is the mechanism. As long as property is ancestral land held by the lineage, there is nothing for a man to redirect. But once he can earn — a salary, a business, land he bought himself — he has something that is his, and the obvious person to leave it to is the child he lives with and loves.
So markets, wage labour, individual property and colonial law tend to dissolve matriliny from the inside. Not by conquest. By making fatherhood, in the ordinary modern sense, something a man can afford to want.
7.4 — Who dismantled the Nayar system
The Kerala case is documented well enough to answer this precisely, and the answer is uncomfortable for everybody.
The dismantling ran across roughly eighty years of legislation, beginning under British administration in the 1890s, continuing through provincial acts in the 1910s, 1920s and 1930s, and completed by the Indian state in 1975 with an act that abolished the joint household entirely and converted its property into individual shares.
Three forces drove it, and only one of them is the one people expect.
Colonial law could not process it. British courts needed to identify an owner, an heir and a husband. The Nayar system supplied none of these in a recognisable form. Cases were decided by fitting the arrangement into English categories, and the fitting deformed it.
Colonial and missionary opinion found it scandalous. Nayar women’s sexual autonomy was described in the language of immorality, and educated Indians defending their civilisation against that charge had an obvious incentive to distance themselves from it. This is a pattern Part Three examines in detail, because it is the single most under-appreciated fact about modern Indian sexual conservatism.
And Nayar men campaigned for the change. This is the part usually left out. Men who had earned money, or taken a modern profession, wanted to leave it to their own children rather than to their sisters’. They wanted to be husbands and fathers in the way the surrounding world was organised. Reform associations pressing for the abolition of the matrilineal household were substantially staffed by the men it had privileged in one way and constrained in another.
So the system did not fall to an invader. It was pushed by outsiders, embarrassed by a colonial gaze, and pulled apart from inside by men who wanted a different relationship with their own sons.
7.5 — What this establishes
Precision matters here, because this chapter is easy to over-claim from.
It establishes: that the inheritance problem in Chapter Three has more than one solution, that at least one alternative works at the scale of millions of people, and that the specific arrangement now defended as natural and timeless is one option among several that human beings actually built and ran.
It does not establish: that matriliny is better for women overall. Matrilineal is not matriarchal; men still held authority in all of these societies. Meghalaya today has an organised men’s movement campaigning against the matrilineal system on the grounds that it marginalises them — which tells you both that the system is real and that it produces its own resentments.
And it does not establish that any of this could be adopted. These systems are not menu items. They collapse when wage labour and individual property arrive, and wage labour and individual property have arrived nearly everywhere.
The honest summary is narrow. The claim “there is no other way to organise this” is false, and we can prove it is false, because other ways existed and one of them was in Kerala within the lifetime of people still alive.
Among the Nayar of Kerala, property passed through women, husbands did not move in, and a man’s obligations were to his sister’s children. The verification problem from Chapter Three simply vanishes — and the restrictions on women vanished with it.
Others exist now: the Khasi and Garo of Meghalaya, the Mosuo of Yunnan, and the Minangkabau of Sumatra — six to eight million people, the size of Hyderabad, and Muslim.
Matrilineal is not matriarchal. Men still held authority; they held it as brothers and uncles instead of husbands and fathers. The claim is narrow: these societies solved inheritance without controlling women’s sexuality.
They are not unregulated. They regulate different people in different directions — heavy fixed obligations on men, and her behaviour left to her.
They keep collapsing, because once a man earns something of his own he wants to leave it to the child he lives with. Markets dissolve matriliny from inside.
The Nayar system was ended by colonial courts, by embarrassment at a colonial gaze, and by Nayar men who wanted to leave their money to their own sons. Not by an invader — by all three at once.
8Religion Arrived Late
Every rule in this part is older than every scripture now cited to support it. That is a statement about dates, not about faith — and what follows from it is more interesting than either the believer or the critic expects.
8.1 — The dates
Start with the chronology, because it is not in dispute and it does most of the work.
Farming, heritable property and the machinery of Chapter Three are in place from roughly twelve thousand years ago, developing over the following several millennia.
The scriptures now cited in these arguments are all far younger. The Vedic corpus is composed across roughly the second and first millennia before the common era. The Manusmriti is later still. The Hebrew Bible is assembled over centuries in the first millennium BCE. The New Testament belongs to the first century CE, the Quran to the seventh.
So the property system, the inheritance problem, the marriage payments and the honour mechanism were all running for thousands of years before any of these texts existed. Whatever those texts did, they did not start this.
Here is a single fact that makes the point sharper than any argument.
Assyrian law tablets from roughly three thousand two hundred years ago — well over a thousand years before Christianity and nearly two thousand before Islam — contain regulations on veiling.
What they say: respectable married women and daughters were required to veil in public. Slaves and prostitutes were forbidden to. A slave woman caught veiling faced serious punishment.
What this shows: that veiling, in its earliest legally recorded form, was not primarily about modesty or piety. It was a status marker — a uniform announcing which women belonged to a household with standing and which were available. The prohibition on the poor wearing it is the giveaway. You do not ban the poor from being modest. You ban them from displaying a rank they do not hold.
What it cannot show: that this is the origin of every later veiling practice. Customs are reinvented and reinterpreted, and the meaning a practice has for a person today is not fixed by what it meant in Assyria. It shows that the practice is older than the religions it is now attached to, and that its earliest recorded purpose was social rank.
8.2 — Same faith, opposite arrangements
The strongest evidence that doctrine is not driving this comes from putting communities of the same religion side by side.
Islam. The Minangkabau of Sumatra have been Muslim for centuries and are the largest matrilineal society on earth, with ancestral land passing through women. Pashtun communities in Afghanistan and Pakistan are Muslim and among the most strongly patrilineal and restrictive societies documented. Same faith, same core text, opposite kinship systems, opposite treatment of women’s property and autonomy.
Hinduism. Chapter Four already noted the divide inside India. Northern communities show village exogamy, seclusion norms and a bride sent far from her own kin. Kerala had the Nayar arrangement in Chapter Seven. Same religion, same scriptures, both regarding themselves as fully orthodox.
Christianity. The same Church presided over regions with dowry and regions without, over communities where widows remarried freely and communities where they did not, over places where women held and traded property and places where they could not.
Now apply the reasoning from Part One. If doctrine were the cause, then a constant — the text — would have to explain a variable — the practice. It cannot. This is the identical logical point made about biology in Chapter One, and it fails for the same reason.
8.3 — What religion actually does
If scripture did not create these rules, what did it do? Something significant, and worth naming precisely.
It codified. A custom becomes a written rule. Writing it down freezes it, makes it portable across regions, and removes the flexibility an oral practice had.
It supplied a justification that cannot be examined. This is the largest change. Before, the honest answer to “why must I?” was: because of the land, because of your brothers’ marriages, because of what the neighbours will say. Those are all contestable. Afterwards the answer becomes an appeal to an authority beyond audit — and Part One, Chapter Two showed that a rule and its justification are separate objects. Religion supplied a receipt that no one could check.
It provided enforcement without a police force. An internalised rule is enforced by the person who holds it, at no cost to anybody. That is a technology of extraordinary efficiency, and Chapter Six explains why a society without a functioning state would find it valuable.
And it sometimes restrained. This must be in the ledger too, or the chapter is dishonest. Religious law in a number of traditions established a woman’s right to inherit something where custom gave her nothing, capped what could be demanded, granted a wife rights within marriage that local practice denied, and set limits on what a husband could do. Read against a modern standard these provisions look meagre. Read against the customs they replaced, several were improvements, and in some places the enforcement of religious law against local custom remains the more favourable option for women.
This is the most bitterly fought question in the part, and both sides have something.
Whatever the deep origins, doctrine is what people actually cite, obey and enforce now. Billions organise their lives around these texts. A woman prevented from working is not being stopped by a Neolithic property system; she is being stopped by a father quoting a verse. Treating the text as a passive record ignores that it is the operative instruction in the room.
The same texts produce opposite arrangements in different places, which is exactly what you would expect if local material conditions were selecting which passages get emphasised. Communities read into their scripture what their economy already required. The verses cited in Sumatra and in Kandahar come from one book.
Religion did not originate these rules, and it did not merely record them either. It took local, negotiable customs and converted them into non-negotiable ones — portable across regions, immune to argument, enforced from inside. That is a genuine causal contribution: not creating the rule, but changing its hardness.
What would settle it: tracking what happens to gender norms when a population converts. Where this has been studied, practices frequently persist through conversion with new justifications attached — which supports the third position and is hard to reconcile with the first.
Why people care so much: because both sides think the whole moral question rides on this. The believer fears that “your rules came from farming” means the faith is fraudulent. The critic hopes it means exactly that. Both are making the genetic fallacy from Part One, Chapter One — judging a claim by its parentage — and the origin of a practice settles nothing about the truth of a religion.
8.4 — Where I have to be careful
The front matter promised I would flag my own position where it could bend the writing, and this is one of those points.
There is a well-worn genre in India in which a writer uses an argument like this one to attack a particular religion while sparing others. It is extremely common, it is usually dishonest, and the tell is asymmetry — the chronology applied to one tradition and quietly not to the rest.
So let me state the finding without a target. Every religion discussed here inherited these rules rather than inventing them, every one of them has communities that read the same text in opposite directions, and every one contains provisions that restricted women and provisions that protected them. If any part of this chapter reads as though one tradition is the problem, I have written it badly, and I would want to know.
Farming, heritable property and the inheritance problem were running for thousands of years before any scripture cited in these arguments existed. Whatever the texts did, they did not start this.
Assyrian law from over three thousand years ago required respectable women to veil and forbade slaves and prostitutes from doing so. That is not a modesty rule. It is a uniform marking rank — and you do not ban the poor from being modest.
The same faith produces opposite arrangements. The Minangkabau are Muslim and matrilineal; Pashtun communities are Muslim and strongly patrilineal. Kerala and north India share scriptures and ran different systems. A constant cannot explain a variable — the same logic that limited biology in Chapter One.
What religion did do is real: it codified a custom into a fixed rule, supplied a justification nobody could audit, and moved enforcement inside the person, where it costs nothing. It also, in places, restrained what custom permitted, and that belongs in the ledger too.
Religion is best understood as a ratchet. It did not create these rules and it did not merely record them. It changed how hard they were to undo.
9The Audit
Part One said that a rule justified by a solved problem is a fence around an empty field. This part has now identified the problems. Here they are, scored one at a time, honestly — including the ones still standing.
9.1 — The method
Chesterton’s fence tells you to find out why the fence was built before removing it. We have now done that. So the next question is the obvious one, and almost nobody asks it: is the problem still here?
I am going to go through them one at a time and score each as solved, partly solved, or still standing. I will not fix the scores to produce a tidy conclusion, and the result is genuinely mixed — some of these fences are around empty fields, and some are not.
9.2 — Problem one: nobody can verify who the father is
Score: solved. Completely, and recently.
Genetic fingerprinting was demonstrated in 1985. A paternity test in India today costs something in the range of an ordinary mobile phone and returns an answer in days.
This is the central problem of Chapter Three — the one that generated the whole apparatus — and it has been technically eliminated within the lifetime of most people reading this. Four thousand years of institutional machinery was built to manage an uncertainty that a laboratory now removes for a few thousand rupees.
Notice, though, what has not happened. No society has responded to this by dismantling the rules. Not one. If the rules existed to solve the verification problem, and the verification problem is solved, the rules should be relaxing. They are not, and §9.9 takes up what that tells us.
9.3 — Problem two: land divided among heirs becomes unfarmable
Score: largely solved by other means.
This was a real problem, and Part One used it as an example. It has been substantially addressed by instruments that do not require disinheriting anybody: land markets that let holdings be consolidated, mortgages and partnerships, tenancy arrangements, and above all the fact that most people no longer make their living from a field at all.
Indian agricultural holdings have nonetheless been fragmenting for decades, so this is not a solved problem in the sense of a problem that vanished. It is solved in the sense that restricting women is not the instrument anybody would now reach for, and no one arguing about a daughter’s clothing is thinking about plot sizes.
9.4 — Problem three: disease
Score: largely solved, and one exception.
The major bacterial infections that historically made sexual contact genuinely dangerous became curable with antibiotics from the 1940s. Barrier contraception is cheap and available. Viral infections remain a real risk and are not eliminated, though several are now manageable or preventable.
What has changed most is the shape of the risk. It used to be uncontrollable and often fatal. It is now controllable at low cost by anybody with information and access — which converts it from a reason to restrict a person into a reason to inform them. Those are different responses to the same fact, and Part Thirteen has data on which one actually works.
9.5 — Problem four: pregnancy outside a supporting arrangement
Score: technically solved, practically not.
Reliable contraception exists and is inexpensive. That is a genuine solution to the biological problem.
But a solution nobody can reach is not a solution, and here the Indian picture is honest and unflattering. Access is uneven, information is patchy, unmarried women face real barriers to obtaining contraception, and a great deal of the burden falls on female sterilisation after childbearing rather than on preventing unwanted pregnancy in the first place.
So this fence has a gate in it, and in much of the country the gate is locked. That is a fact about health systems, not about the rule.
9.6 — Problem five: a woman abandoned has no way to live
Score: partly solved, and this is the weakest link in the whole audit.
This was among the strongest original justifications, and I want to give it full weight. In a society where a woman could not own property, could not earn, and had no state to fall back on, her security ran entirely through a man. In that world, rules protecting a marriage were, whatever else they were, also protecting her survival.
What has changed. Indian law has given women inheritance rights, strengthened substantially in 2005 when daughters were given equal rights in ancestral property. Wage employment exists. Divorce and maintenance law exists.
What has not changed enough. Indian women’s participation in paid work is strikingly low for a country at this income level, and the practical enforcement of inheritance rights against family pressure is weak — a legal right a daughter is shamed out of claiming is a right on paper. Maintenance awards are slow and often unpaid.
So the honest verdict is that the replacement has been designed and only partly built. This is the asylum problem from Part One, Chapter Three, and it is the single strongest argument the cautious side has in this entire part. It deserves to be conceded rather than argued around.
9.7 — Problem six: there is no law you can reach
Score: partly solved.
Chapter Six showed that honour is a security technology for people without a state. India has a state, police and courts, which is a genuine difference from the conditions that produced the system.
But reach is uneven, tens of millions of cases sit pending, and for many people the practical cost of using the formal system exceeds what the dispute is worth. Where that is true, the older technology remains in use — not from nostalgia, but because it is the one that functions.
9.8 — Problem seven: who will feed you when you are old
Score: not solved. Not close.
This one is rarely discussed and it may be the most important on the list.
India has no meaningful universal old-age pension. For the overwhelming majority of Indians, the arrangement for old age is: your children look after you. And under a patrilineal system, “your children” in practice means your son, since a daughter leaves for another household and her obligations follow her.
Think of a son, in this system, as a pension policy. It is the only one most Indian families have ever been offered.
Now think of a daughter in the same accounting. She costs eighteen years of food and schooling, then a dowry running to several years of household income — and then the return on all of it goes to a different family.
Set the two side by side and the arithmetic is brutal and clear. This is not a story about people hating girls. It is a story about a retirement system with only one instrument in it, and a family being asked to buy the instrument that does not pay out.
Which is why a state pension is, in cold terms, a policy about sex ratios — and why Part Twelve counts what happens where it is absent.
So this fence stands, the field behind it is not empty, and no amount of arguing about attitudes will move it. It would be moved by a pension.
9.9 — Problem eight: keeping the group closed
Score: not solved — and it is the wrong category.
Every other item on this list is a hardship somebody wanted removed. This one is different. Maintaining caste boundaries is not a problem people were suffering under. It is a goal that a great many people still actively hold.
You cannot solve a goal with technology. A DNA test does not help. A pension does not help. As long as a substantial number of people want a boundary maintained, the machinery for maintaining it has a live constituency — and in India that machinery runs through controlling who women may marry.
This is the answer to the puzzle in §9.2 — why the rules did not relax when paternity testing arrived. They did not relax because paternity was never the only load on them, and in India it was arguably never the main one.
Part Three is about that machine in full.
This entire chapter — and the fence argument that motivated it — rests on a premise that has been sitting underneath the whole part: that a rule persists because of the problem it solved.
Both camps need it. The reformer audits the original problems and expects the rule to fall when they are gone. The traditionalist defends the rule by pointing at the original problem. Both treat the founding purpose as the thing holding it up.
But rules acquire new supports as they age, and the new ones are often stronger than the original. A practice becomes a mark of identity — proof of who we are against who they are. It becomes a source of status for those who enforce it. It becomes an industry with people employed in it. And it accumulates the most powerful support of all: everybody who already paid the price. A woman who gave up what she wanted, and whose sisters did, has an enormous stake in that cost having meant something.
Which means the audit in this chapter may be answering the wrong question. Finding out why the fence was built tells you nothing about what is holding it up now. Those can be — and after four thousand years usually are — entirely different sets of hands.
The general form: mistaking a thing’s origin for its current cause. Knowing why a war started tells you very little about why it is still going.
That does not make the audit worthless. Knowing the field is empty is worth having, because it removes an argument from the table. It simply means the removal of an argument and the removal of a fence are two different jobs, and this part has only done the first.
Solved: paternity uncertainty, completely, since 1985, for the price of a phone. Land fragmentation, by other instruments. Disease, largely, since antibiotics.
Partly solved: unwanted pregnancy — the technology exists and the access does not. Destitution after abandonment — the law has been written and the economy has not been built, which is the strongest argument the cautious side holds. Weak law — a state exists, its reach does not.
Not solved: old age. India has no real pension, so a son is the pension and a daughter is a cost whose return goes elsewhere. That is arithmetic, not prejudice, and it would be changed by a pension rather than by an argument.
Wrong category: keeping the group closed. That is not a hardship anybody wants removed. It is a goal many still hold, and no technology answers a goal.
The verification problem was eliminated forty years ago and not one society relaxed its rules in response. That is the finding of this chapter.
Because knowing why a fence was built tells you nothing about whose hands are holding it up now — and after four thousand years, those are never the same hands.
10An Honest List Of What We Do Not Know
Two lists, with no hedging in either. What is genuinely unknown about the origins of these rules, with the reason it is unknown. And what is solid enough to build the next fourteen parts on.
10.1 — Genuinely unknown
This part deals with events that left no witnesses and, for most of the period, no writing. The unknowns are correspondingly large.
Which came first, the property or the rules. This is the central gap. Chapter Three’s chain is plausible and well supported, but in the archaeological record property systems and kinship rules appear together, and no method currently separates them. It is entirely possible that societies which already restricted women found it easier to develop concentrated heritable wealth, rather than the reverse. The plough study in Chapter Four is the best attempt at an answer and it is not clean.
What women in these societies actually thought. This is not a small gap; it is a hole in the middle of the subject. The ethnographic and historical record was overwhelmingly produced by men, who mostly interviewed men, in societies where women’s worlds were frequently closed to outsiders. Where a women’s account survives it is usually late, written by an elite, and shaped by whoever was listening. We are reconstructing a system from the testimony of the people it privileged.
Whether foragers tell us anything about the deep past. Chapter Two’s argument box could not be resolved. Every foraging society studied in modern times occupies land that farmers rejected, after thousands of years of contact with agricultural neighbours.
Why the male genetic bottleneck happened. The signal is solid. The interpretation is not. Whether it reflects a few men monopolising reproduction or whole patrilineal lineages being wiped out together is unresolved, and the two imply quite different societies.
How much of the plough effect is the plough. Wheat regions also grew states, cities, armies and writing. Separating the tool from its companions has not been done convincingly.
Whether any of the alternative systems could function now. Matriliny exists, so it is possible. It also collapses reliably when wage labour and individual property arrive, and both have arrived nearly everywhere. Whether a modern society could sustain such an arrangement is untested, because none has tried.
10.2 — Solid
That control over women tracks heritable, immobile wealth. This is the strongest finding in the part. It holds across the ethnographic record, across historical societies, within farming societies, and inside single countries. It is not one study; it is a pattern that keeps reappearing in independent bodies of evidence, which is the strongest kind of support available outside a laboratory.
That the biological asymmetry is real and that it does not determine the rules. Both halves are solid. Paternity uncertainty was a genuine and universal condition until 1985. And societies facing that identical condition built opposite institutions — the Barí distributed fatherhood, the Nayar routed inheritance around it. A constant does not explain a variable.
That the practices predate the scriptures cited for them. This is chronology, not interpretation. The dates are not in dispute. Assyrian veiling law is over three thousand years old and its earliest recorded function was marking rank, not piety — the ban on slaves veiling settles that.
That the same faith supports opposite arrangements. The Minangkabau exist, are Muslim, are matrilineal, and number in the millions. Kerala and north India shared scriptures and ran different systems. These are facts about living populations, not reconstructions.
That gender norms outlive the conditions that produced them by many generations. The migrant finding in Chapter Four establishes this even if the specific mechanism is contested. Something is transmitted through families, across an ocean, and shows up in employment statistics.
That the verification problem is solved and nothing relaxed. Both halves are simply observable.
That India banned dowry in 1961 and it expanded. Observable, documented, and the most useful single fact in the part for anybody who believes legislation is the lever.
10.3 — The shape of this archive
One closing observation, in the pattern set in Part One.
Look at what this part could be built from. Property law, inheritance records, legal codes, crop suitability data, genetic samples, marriage payments — all of it material, countable, and mostly generated by administration.
Now notice what is almost entirely absent. What any of it felt like. Whether the arrangement was resented or accepted. What was said between women in rooms men did not enter.
That absence is not random. Administrative records survive because states and households had reasons to keep them, and states and households were interested in property. Nobody was recording the other thing, because nobody with the means to write thought it was information.
So this part can tell you with real confidence how the machine was built and almost nothing about what it was like to be inside it. That is a limitation of the evidence and it should be read as one — not as a finding that the inside was quiet.
Genuinely unknown: whether property produced the rules or the reverse. What women in these societies actually thought — the record was made by men, about men, for men. Whether foragers tell us anything about the deep past. Why the male genetic bottleneck happened. How much of the plough effect is the plough. Whether any alternative system could work now.
Solid: that control over women tracks heritable immobile wealth, across every body of evidence available. That the biological asymmetry is real and does not determine the rules. That the practices predate the scriptures cited for them. That the same faith supports opposite arrangements, in populations alive today. That these norms survive migration by generations. That paternity became verifiable in 1985 and nothing relaxed. That India banned dowry and it grew.
And note what this archive is made of: property, law, payments, crops, genes. All of it countable, all of it kept because somebody had a financial reason to keep it. What it cannot tell you is what any of it felt like — and that silence is a gap in the record, not evidence that the inside was quiet.
Sources & further reading — Part 2
Timeline
From the first planted seed to the laboratory that solved the problem it created. Dates before writing are approximate and contested; the further back, the wider the error.
| When | What happened |
|---|---|
| c. 300,000 BP | Anatomically modern humans. For the next ninety-six per cent of the story, nobody owns land, a granary or a herd. |
| c. 10,000 BCE | Farming begins independently in at least half a dozen regions among people with no contact. The field cannot be carried. Settlement, storage, surplus and permanent inequality follow. |
| c. 5,000 BCE | Beginning of the collapse in male genetic diversity, visible across Africa, Europe and Asia while the female line holds steady. A world newly organised around male descent lines. |
| c. 4,000 BCE | The ard, the earliest plough, in use in Mesopotamia. Field labour shifts towards men wherever the soil suits wheat and barley rather than sorghum and millet. |
| c. 1,200 BCE | Middle Assyrian law tablets require respectable women to veil and forbid slaves and prostitutes from doing so. The earliest recorded veiling law is about rank, not piety. |
| c. 1500–500 BCE | Composition of the Vedic corpus. Every rule in this part is already thousands of years old. |
| c. 200 BCE – 200 CE | The Manusmriti is compiled. Later cited as the origin of rules that long predate it. |
| 7th c. CE | The Quran. Within a few centuries Islam reaches Sumatra, where the Minangkabau adopt it and keep their matrilineal inheritance. They still have both. |
| 1896–1933 | Colonial and provincial legislation begins dismantling the Nayar matrilineal system in Kerala, pressed by British courts that could not process it and by Nayar men who wanted to leave property to their own sons. |
| 1940s | Antibiotics enter mass production. The major bacterial diseases that made sexual contact genuinely dangerous become curable. |
| 1956 | The Hindu Succession Act gives Indian women inheritance rights. Custom continues to override it in practice for decades. |
| 1961 | India bans dowry. Over the following decades it spreads to communities that never practised it and rises in real value. |
| 1970 | A Danish economist publishes the observation that women’s economic position tracks whether their ancestors farmed with a hoe or a plough. It takes forty-three years to test. |
| 1975 | The Kerala Joint Hindu Family System (Abolition) Act ends the matrilineal household in law. Roughly eighty years from the first colonial statute to the last Indian one. |
| 1985 | Genetic fingerprinting is demonstrated. For the first time in the history of the species, paternity can be established with certainty. Not one society relaxes its rules. |
| 2005 | India amends the Hindu Succession Act to give daughters equal rights in ancestral property. Enforcement against family pressure remains the difficulty. |
| 2013 | The plough hypothesis is tested and holds — including among the grandchildren of migrants, four thousand years and one ocean from the field. |
| ↓ | Part Three starts here — with the one machine this part could not explain: why the most severely punished offence in India is not sex, but marrying the wrong person. |
Glossary
Every hard word used in this part, in plain English. Each was explained where it first appeared; this page is a convenience, not a requirement.
| Term | Plain meaning |
|---|---|
| Bridewealth | Payment from the groom’s family to the bride’s. The more common of the two worldwide. Appears where women’s labour produces visible wealth. |
| Bride service | A young man working for his wife’s family, often for years, in place of a payment. Common among foragers. Note the direction: labour flows towards her family. |
| Culture of honour | A system where a family’s reputation for unhesitating retaliation replaces law as their protection. Clusters among herders, whose wealth can be stolen. |
| Dowry | Payment from the bride’s family to the groom’s. Appears with plough farming, stratification and monogamy. Functions both as a daughter’s early inheritance and as a bid in a competitive marriage market. |
| Ethnographic record | The body of descriptions of several hundred societies collected by anthropologists and coded into comparable categories. Powerful for spotting patterns; compromised by having been written mostly by colonial-era men who interviewed men. |
| Heritable property | Wealth that outlives its owner and must pass to somebody — land, buildings, herds, stored grain. The single most important concept in this part. |
| Hoe cultivation | Farming with hand tools, often shifting between cleared patches. Interruptible, compatible with childcare, and historically done heavily by women. |
| Kinship system | A society’s rules for who counts as family, who may marry whom, and who inherits. Every society has one; they differ enormously. |
| Matrilineal | Descent, family membership and property passing through the mother’s line. |
| Matriarchal | Women holding political power. Not the same as matrilineal, and far rarer — arguably never documented in the form men hold it. |
| Matrilineal puzzle | The built-in strain in matrilineal systems: a man’s formal duties are to his sister’s children while his affection is with his own. It becomes acute the moment he has wealth of his own to redirect. |
| Mitochondrial DNA | Genetic material passed only from a mother to all her children. Used as a measure of how many women were reproducing in a past population. |
| Parental investment | Anything a parent does for one offspring at the cost of doing it for another. The sex investing more per offspring is generally choosier; the sex investing less generally competes. |
| Partible paternity | The belief that a child can have several biological fathers, all carrying real obligations. Documented in lowland South America, where extra fathers measurably improved child survival. |
| Patrilineal | Descent, family membership and property passing through the father’s line. The arrangement most of this series examines. |
| Plough agriculture | Farming with a heavy blade drawn by an animal. Requires sustained strength and grip, cannot be easily interrupted, and pushes field labour towards men. |
| Sedentism | Living permanently in one place. A consequence of farming, and the origin of both dense settlement and continuous mutual surveillance. |
| Y chromosome | Genetic material passed only from father to son. Used as a measure of how many men were reproducing in a past population. |