The Wheat Bargain

Part 00The sentence itself, and what it actually says

Before you can judge a claim, you have to know exactly what was claimed. Most fights about this sentence are fights about words nobody defined.

0.1Where the words come from

Yuval Noah Harari is an Israeli historian. He wrote a book called Sapiens: A Brief History of Humankind. It came out in Hebrew in 2011 and in English in 2014. It has sold tens of millions of copies. Chapter 5 of that book is titled “History’s Biggest Fraud.” The whole chapter is about the Agricultural Revolution — the moment humans stopped wandering to find food and started growing it.

Inside that chapter sit the two lines that made the book famous:

That is it. Two short lines. They have probably done more to change how ordinary people think about human history than any other sentences written this century.

0.2The three words that carry all the weight

0.3A fraud with nobody to blame

Normally a fraud needs three things: a cheat, a victim, and a lie. Here there is no cheat. Wheat is a grass. It has no plan. It cannot lie. There was no contract, no salesman, no fine print.

So what does Harari mean? He means this: the outcome had every feature of a scam except the scammer. People made a series of small, sensible choices. Each choice looked like an improvement. Added together, over hundreds of years, they made life worse for the people making them — and by the time anyone could have noticed, going back had become impossible.

This is the single point most readers miss, and most critics attack. When a scholar says “calling it a fraud is silly, nobody defrauded anyone,” they are usually attacking the literal meaning of a word Harari was using as a picture. When a fan says “see, we were robbed,” they are usually doing the same thing in the other direction.

0.4The scale, so you can feel it

Picture the whole human story as a 24-hour day. Farming starts at about 11 minutes to midnight. Everything you think of as history — pyramids, empires, religions, wars, money, cities, the internet — happens in those eleven minutes. Harari’s claim is that those eleven minutes were bought at a price, and that we never read the price tag.

Now we take the sentence apart.

Part 01The eleven axioms hidden inside the claim

An axiom is a thing you have to assume before your argument can even start. It is the floor you stand on. Harari’s one sentence stands on eleven floors. If any one of them gives way, the sentence changes meaning — and some of them are shakier than others.

Axiom 1The thing worth measuring is the individual life, not the species

This is the foundation stone. Without it, the whole argument collapses immediately, because by any measure of species success, farming was a colossal win. Human numbers went from maybe 5–8 million foragers to about 8 billion people. If “success” means “more of us,” farming is the greatest success in the history of large animals.

Harari refuses that scoreboard. He says: who cares how many of us there are, if each of us is more miserable? A hundred people living well beats a million people living badly.

Why he needs it: without this axiom, “fraud” makes no sense. You cannot be defrauded if the deal made you numerous and powerful.

What would break it: if you think existing at all is good — that a life barely worth living is still better than no life — then farming created billions of net-positive lives and the deal was excellent. This is not a fringe view. It is a serious position in moral philosophy, and we come back to it in Part 05.

Axiom 2You can meaningfully compare wellbeing across twelve thousand years

This sounds obvious. It is not. Nobody in 13,000 BC left a diary. We cannot ask them how they felt. We have bones, teeth, tools, seeds, rubbish heaps and burial sites. From that we build a guess about how their days went.

Harari writes with total confidence: “The average person in Jericho of 8500 BC lived a harder life than the average person in Jericho of 9500 BC or 13,000 BC.” That is a very specific claim about people whose names, thoughts and feelings are gone forever.

What would break it: if wellbeing is mostly about meaning, belonging and expectation rather than calories and joint pain, then bones tell us almost nothing. A person with backache who feels their life has purpose may report a better life than a healthy person who feels lost. We have no way to measure ancient purpose.

Axiom 3Wellbeing leaves marks on the body, and those marks can be read

This is the evidence engine of the whole argument. The field is called bioarchaeology or paleopathology — the study of disease and stress in ancient bones. The signals it reads are things like: adult height, tooth decay, tooth enamel defects (which record childhood hunger), iron-deficiency marks in the skull, arthritis, spinal damage, and bone infections.

The landmark work is a 1984 book edited by Mark Nathan Cohen and George Armelagos called Paleopathology at the Origins of Agriculture. It pooled more than twenty separate studies of populations moving from foraging to farming. In 19 of 21 cases, health markers got worse.

What would break it: a problem archaeologists call the osteological paradox, described by James Wood and colleagues in 1992. It is counter-intuitive and important, so read it twice: a skeleton can only show a disease if the person lived long enough with the disease for the bone to change. Someone who dies fast from an infection leaves clean bones. Someone who survives it for ten years leaves scarred bones. So a graveyard full of damaged skeletons can mean a healthier population that survived illness — and a graveyard of clean skeletons can mean people who died before their bones could record anything. We come back to this in Part 05, because it is the most under-reported problem in the entire debate.

Axiom 4Foragers had it good — the “before” picture

Harari calls foragers’ knowledge and bodies superior, and treats the pre-farming Middle East as a place where “the natural resources of the area were enough to support its human population.”

This idea has a name and a birthday. In 1966, at a conference called Man the Hunter, the anthropologist Marshall Sahlins presented a paper he later titled “The Original Affluent Society.” His claim: foragers were the first affluent people, not because they had much, but because they wanted little and got it easily. He cited figures suggesting three to five hours a day of food work — and for the Hadza of Tanzania, under two hours.

What would break it: the numbers were wrong. Richard Lee’s original !Kung study counted only time spent hunting and gathering. It left out food processing, tool repair, firewood, water, childcare and camp moves. When Lee himself later added processing time, the working week came out at roughly 40 hours for women and 44 for men — an ordinary modern job. David Kaplan’s paper “The Darker Side of the Original Affluent Society” lays out how the pleasant number became famous while the corrected number did not.

Axiom 5Early farmers had it bad — the “after” picture

Here Harari is on his firmest ground. The specific charges:

• Worse diet. Foragers ate dozens of species. Farmers ate mostly one or two grains. Wheat has no vitamin C, little iron the body can use, and is poor in several essential nutrients. Diseases of narrow diets — anaemia, rickets, beriberi, pellagra — arrive with agriculture.
• More work. Clearing, ploughing, sowing, weeding, watering, guarding, harvesting, threshing, grinding, storing. Grinding grain alone was hours of daily labour, usually by women, and it wrecked knees, toes, backs and wrists.
• More disease. People now lived crowded, beside their own waste and their animals. Most of the great killers — measles, smallpox, flu, tuberculosis — came to us from domesticated animals.
• New fragility. A forager who loses one food source switches to another. A farmer who loses the harvest loses everything. Famine is an agricultural invention.
• New violence. Stored grain is worth stealing. Jericho built a wall. Walls need guards. Guards need feeding. Feeding guards needs more grain.

Axiom 6The switch was made blind — nobody could see where it ended

This is the gradualism axiom, and it is the cleverest part of Harari’s argument. The change took thousands of years. Each step was tiny. Plant the seed a bit deeper. Pull a few weeds. Stay three weeks longer at harvest camp. Each step visibly paid off. None of them looked like a civilisational decision.

This axiom is what makes “fraud” work without a fraudster. The deception is not in any one step. It is in the gap between the step and the sum.

Axiom 7The switch was one-way — the boats were burned

Harari puts it plainly: if ploughing raised a village from 100 people to 110, which ten volunteer to starve so the rest can return to the old way? Nobody. The door locks behind you.

He calls this the luxury trap. A luxury becomes a habit, the habit becomes a need, the need becomes a chain. It is not that people were too weak to go back. It is that going back had become mathematically impossible.

There is a second lock. Farming does not need everyone to agree. It only needs one band to start. That band grows faster, so it outnumbers its neighbours, so it takes their land — and the neighbours must either farm or leave. The old life did not lose an argument. It lost a headcount.

Axiom 8Evolutionary success and personal happiness are two different scoreboards

In biology, “success” means one thing: more surviving copies of your genes. It has nothing to do with pleasure, dignity, or a good life. Wheat won by that measure — it went from a wild grass in a small patch of the Middle East to roughly 2.25 million square kilometres of the planet’s surface. Cattle, pigs, sheep and chickens won by that measure too, and each individual animal pays for it in a cage.

Harari’s move is to point out that we are on the same list. Humans multiplied. Individual humans got shorter, sicker and more overworked. Both facts are true at once, and only one of them is normally taught in school.

Axiom 9Wheat can be described as the one doing the acting

This is a rhetorical device — a way of speaking, not a scientific finding. Wheat has no intentions. But the inversion does real work: it forces you to look at the relationship from the outside, without assuming humans are the heroes.

And it survives translation into hard science. In evolutionary terms, wheat and humans entered a partnership in which wheat’s reproduction rose enormously and its human partners changed their entire way of life to serve it — clearing its competitors, watering it, guarding it, carrying it, and rearranging their bodies and settlements around its calendar. Whether you call that “domestication of humans” is a choice of words. What the words describe is real.

What would break it: nothing, exactly — because it is a metaphor, it cannot be false. That is also its weakness. A claim that cannot be wrong is not doing scientific work. It is doing poetic work. Both matter; they should not be confused.

Axiom 10Almost everything we dislike about civilisation grew out of the grain surplus

The logic runs: farming creates a surplus → surplus can be stored → stored things can be taken → whoever takes them becomes powerful → power needs record-keeping, soldiers and a story that justifies it → and now you have a state.

The strongest scholarly version of this is James C. Scott’s Against the Grain (2017). Scott asks a sharp question: why is every early state built on grain, never on potatoes, cassava or lentils? His answer is beautiful. Grain ripens all at once, above ground, on a predictable date. It can be seen, counted, divided, stored, carried and rationed. Root crops sit underground and can be dug up whenever you like — invisible to a taxman.

Axiom 11This is a repeating pattern, not a one-off accident

Harari calls it “one of history’s few iron laws”: luxuries become necessities and spawn new obligations. He gives his own example. Email was supposed to save time. Instead of a few careful letters a month, he now gets dozens of messages a day, each expecting an instant reply. The tool did not give him free time. It raised the speed of the treadmill.

This axiom is why the chapter matters to a reader in 2026 who has never touched a plough. If the trap is a law and not an accident, then you are inside a version of it right now, and Part 08 is about finding out which one.

Part 02Why Harari said it — motive, lineage, and the long game

Nobody writes a sentence like that by accident. It is aimed at something. Understanding what it is aimed at explains most of its shape — including its exaggerations.

2.1The thing he is shooting at

There is a story every school teaches without saying it out loud. It goes: humans were once poor, dirty and ignorant. Then they got clever. They invented farming, then writing, then machines, then medicine. Life got better and better. We are the lucky ones at the end of the line.

Historians call this the progress narrative — the idea that history is a staircase going up. Harari’s sentence is a stick of dynamite placed at the bottom of that staircase. If the very first step down there was a mistake, the whole staircase becomes questionable.

2.2Where the idea came from — the family tree

Harari did not invent this. He inherited it, and he did not always say so clearly. The chain runs like this:

This matters for a practical reason. When people say “Harari is wrong,” they are often criticising a 1966 anthropology paper and a 1987 magazine article without knowing it. And when scholars have updated those older claims, the update rarely reaches the reader of Sapiens.

2.3Why Chapter 5 has to be shocking

Look at where the sentence sits in the book. Sapiens has a spine, and Chapter 5 is a vertebra near the base. The book’s real thesis, stated over and over, is: humans are ruled by stories they made up. Money is a story. Nations are stories. Human rights, companies, gods, castes — all shared fictions that work because enough people believe them.

Progress is one of those stories. So Harari needs an early, visceral demonstration that a story everyone believes can be flatly untrue. The Agricultural Revolution is the perfect demonstration, because everybody already agrees it was the good bit.

2.4The part almost every reader skips: the animals

Chapter 5 is not only about people. A large part of it is about sheep, goats, cattle, pigs and chickens. Harari argues that domestication was, from the animal’s point of view, the beginning of a catastrophe that has never ended — and that the number of animals involved makes it, on a simple headcount, the largest source of suffering humans have ever created.

Harari is a vegan and has said publicly that industrial farming is one of the worst crimes in history. That is not a hidden agenda; it is stated. But readers tend to remember the line about wheat and forget the line about cows. When Part 04 tries to find a version of the argument that no critic can answer, this is where it ends up.

2.5The long game — Chapter 5 is a rehearsal

Harari’s later books (Homo Deus, 21 Lessons, Nexus) are about the next set of bargains: artificial intelligence, biotechnology, surveillance, the automation of judgement. His whole warning about those bargains has the same shape as Chapter 5:

• Each step will look like an improvement.
• No single step will be the moment to refuse.
• The people who refuse will be outcompeted by the people who don’t.
• By the time the cost is visible, the old way will be gone.

So Chapter 5 is not really history. It is a template. He is teaching you the shape of a trap using an example that is safely 12,000 years old, so that you can recognise the shape when it appears in something you are living through. Whether that is honest teaching or clever rhetoric is a fair question. It is certainly deliberate.

Part 03Steelman, level one: the strongest honest case

Here the argument is built at its best, using the best evidence available, with no cheap shots. A steelman means: if you are going to disagree, disagree with this version — not with a weaker one you invented.

3.1The bones

This is the hardest evidence there is, because bones do not have opinions. Three findings matter:

Cohen and Armelagos, 1984. Twenty-one populations across the world that made the shift from foraging to farming. In nineteen of them, health markers declined: more tooth decay, more childhood growth interruptions, more infections, more signs of iron deficiency.

Mummert and colleagues, 2011. A systematic review of the bioarchaeological record on height and bone strength across the agricultural transition. The pattern held across different continents and different crops — a broad decline in stature and robusticity.

Marciniak and colleagues, PNAS, 2022. This one is the sharpest, and it is recent enough that most readers of Sapiens have never heard of it. The team combined skeletons with ancient DNA. Genes carry information about how tall a person was “supposed” to be. Early European farmers came out shorter than their own genes predicted. That gap is the fingerprint of environmental stress — hunger and disease during growth — rather than of ancestry.

3.2The teeth

Teeth are the best witnesses in archaeology, because enamel does not remodel. Whatever happened to you as a child is still written there when you die.

Two signals appear with farming almost everywhere. First, cavities explode. Starchy grain sticks to teeth and feeds the bacteria that rot them; forager teeth are famously clean by comparison. Second, enamel hypoplasia — thin horizontal lines across the tooth, each one a record of a period in childhood when growth stopped because of hunger or illness. Farming populations carry more of them.

3.3The germs

Almost every famous human epidemic disease came to us from a domesticated animal, or from the conditions of settled life:

There is a crucial technical point behind this. Some diseases cannot exist in small mobile bands at all, because they burn through everyone and then have nobody left to infect. Measles, for example, needs a continuously supplied population in the hundreds of thousands to keep circulating. Farming did not just spread these diseases. It manufactured the conditions in which they could exist.

3.4The plate

A forager band in a decent environment might use dozens of plant species and many animals across a year. A grain farmer eats grain. The consequences are predictable and documented:

• Anaemia — plant iron is poorly absorbed, and grain contains compounds that block absorption further. It shows in skulls as a spongy, porous texture.
• Rickets — bent leg bones from vitamin D shortage.
• Beriberi and pellagra — diseases of polished rice and maize-only diets, which appear wherever one grain crowds out everything else.
• Seasonal hunger — the “hungry gap” between the last of the stores and the new harvest. Foragers do not have a hungry season in the same way; the landscape keeps producing something.

3.5The clock

Farming is not one job. It is a stack of jobs that never ends: clear, break the soil, sow, weed, water, fence, scare birds, harvest, thresh, winnow, grind, store, guard, repair. Then repeat, forever, on a schedule set by the plant.

Grinding deserves its own line. Turning grain into food you can digest took hours a day, usually done by women kneeling at a stone. The skeletal signature is unmistakable: worn knees, arthritic toes, damaged lower backs, deformed wrists. It appears in Neolithic female skeletons across the world.

3.6The boss

This is James C. Scott’s contribution, and it is the part of the case that has aged best. Follow the chain:

Grain is visible (it grows above ground), predictable (it ripens on a known date), divisible (you can take a tenth), storable, transportable and countable. That combination is not a description of good food. It is a description of a perfect tax base.

Compare cassava or potatoes. They sit in the ground. They can be dug up any week of the year. A tax collector cannot see them, cannot date them, and cannot easily carry them away. So no early state was ever built on tubers. Every single one — Sumer, Egypt, the Indus, China, Mesoamerica, the Andes — was built on a grain.

Once you can tax, you can pay soldiers. Once you have soldiers, you can compel labour. Once you compel labour, you need records, which is why the earliest writing anywhere is not poetry — it is receipts for grain and lists of workers. Then you need a story to explain why this is right, and you get priesthoods and divine kings.

3.7The soil

Southern Mesopotamia, the world’s first great grain civilisation, irrigated its fields for centuries. Irrigation water carries dissolved salt. When it evaporates, the salt stays. Over generations the fields went white.

You can read the collapse in the crop records: the region shifted steadily from wheat, which is salt-sensitive, to barley, which tolerates salt — and then yields fell anyway. The first civilisation built on grain was also the first to poison its own land with the method that grew it. That is Harari’s trap in a single landscape.

3.8The animals

Set the human ledger aside for a moment and count everyone else. Today the world raises and kills on the order of eighty billion land animals a year, most of them in conditions no honest person would defend. The mass of farmed livestock now vastly exceeds the mass of all wild mammals on Earth.

Domestication is the ancestor of that system. Wheat’s bargain with humans was mirrored by a bargain with sheep, cows and chickens, and their side of it has got worse in every century since.

3.9The consent problem

Finally, the point that has nothing to do with health at all. Nobody agreed to this. No assembly voted. No generation was told what the deal would cost. No one alive at any point in the process had enough information to judge it, and no one at any point had the option to decline on behalf of their descendants.

Part 04Steelman, level two: the argument that survives every attack

Now the hard part. We hand every good objection to Harari’s side and let them rebuild. What comes out is a stronger argument than the one in the book — and a more disturbing one, because it no longer depends on whether ancient farmers were happy.

4.1First move: give away the ground you do not need

A weak arguer defends everything. A strong arguer works out which parts of the claim are load-bearing and abandons the rest. So the level-two Harari concedes, freely and immediately:

• Foragers did not work two hours a day. Corrected figures put it near a modern working week.
• Forager life was often violent. Among some studied groups, violence caused a third to a half of all deaths.
• Forager child mortality was terrible. Roughly a third of children died before fifteen.
• Skeletal evidence is noisy, and the osteological paradox means it can be read backwards.
• The transition was not a single revolution and probably ran through thousands of years of experiment.
• Causation may run backwards — population pressure may have forced farming rather than farming enabling population.

That looks like total surrender. It is not. Every one of those concessions is about welfare — about whether life got better or worse. The level-two argument is not about welfare at all.

4.2The claim that survives: the Ratchet

Read that again, because it is the whole document. It is a selection argument, not a happiness argument. It says nothing about whether farming was pleasant. It says that farming’s spread carried no information about whether farming was good — the same way a virus spreading tells you nothing about whether the virus is nice.

4.3Why “fraud” is exactly the right word after all

Here the level-two argument recovers the word Harari chose, and does it better than he does. In law, a deal is fraudulent when two things are true: the terms could not be understood at the time of agreement, and there was no real ability to walk away. Both held perfectly. Nobody could see the terms — the costs appeared centuries later. Nobody could exit — population growth had removed the option.

Notice what this does. Whether the deal turned out well is irrelevant to whether it was a fraud. If a broker forges your signature and invests your money, and you happen to become rich, you were still defrauded. Harari’s claim survives even if farming made everyone happier.

4.4Absorbing Boserup: the objection makes the trap tighter

Ester Boserup argued the reverse of the standard story: population pressure came first and forced people to farm harder, rather than farming allowing population to grow. This is usually offered as a refutation of Harari.

It is the opposite. If Boserup is right, then farming and population are not cause and effect but a closed feedback loop: more people force more intensive farming, which supports more people, which forces more intensive farming. That is not a mistake anyone made. That is a machine that runs itself, with no steering wheel and no brake. Boserup does not free us from the trap; she removes the possibility that anyone was ever driving.

4.5Absorbing Graeber and Wengrow: they date the door closing

In The Dawn of Everything (2021), David Graeber and David Wengrow attacked Harari directly. Their evidence: there was no single Agricultural Revolution. People farmed a bit, then stopped. They moved in and out of cultivation for thousands of years. Some societies farmed seasonally and foraged the rest of the year. Early farming societies were often not more hierarchical, and sometimes less.

Take that evidence entirely at face value. What does it show? It shows that for roughly three thousand years, humans across many regions tried very hard not to commit — they kept their options open, they refused to specialise, they experimented with reversals.

And then, everywhere, independently, it ended. The flexibility disappeared. The door shut.

4.6Absorbing Göbekli Tepe: the trap is older than the surplus

At Göbekli Tepe in south-eastern Turkey, people built enormous carved stone pillars around 9600 BC — before they had domesticated a single crop. Klaus Schmidt, who excavated it from 1994, summarised it as: first came the temple, then the city. This is normally used to break Axiom 10, since it shows monumental organisation without an agricultural surplus.

The level-two response: fine, and it makes things worse. If people were already locking themselves into enormous, unpaid, multi-generational collective obligations before agriculture, then the ratchet does not need grain at all. Shared belief alone can produce a commitment nobody can exit. Schmidt himself suggested that feeding the crowds who gathered there may have driven the first cereal cultivation. On that reading, farming was not the trap. Farming was the bill.

4.7Absorbing “what was the alternative?”

The most common objection in every comment section: so what should they have done? Stayed hunter-gatherers and stayed at eight million people? You would not exist. I would not exist.

The level-two answer is short: this is a diagnosis, not a prescription. A doctor who says “you have diabetes” is not saying “you should not have been born.” Identifying the mechanism of a trap does not commit you to any proposal about escaping it, and Harari never proposes going back. Demanding an alternative is a way of avoiding the diagnosis.

4.8The part with no answer at all

Every human-welfare objection can be argued both ways. One part of the ledger cannot. Whatever the balance for humans, the domesticated animal side of the bargain is not close. Tens of billions of individual creatures a year, bred to exist in confinement, is a consequence of the same event, and no amount of human wellbeing data touches it. On any moral view that counts animal suffering at all — even at a heavy discount — this alone is enough to call the Agricultural Revolution a bad deal for the planet’s sentient beings taken together.

Part 05Where it cracks — evidence problems, logic problems, and the hidden side

Now the attack. Some of these are technical problems with the evidence. Some are holes in the reasoning. And some are things that are almost never mentioned in popular discussion of this chapter, because they are inconvenient for both sides.

A — Problems with the evidence

5.1The osteological paradox: bones can lie in the opposite direction

This is the most important technical objection in the whole debate, and almost nobody outside the field has heard of it. It comes from a 1992 paper by James Wood, George Milner, Henry Harpending and Kenneth Weiss.

The problem is this. A skeleton can only show evidence of a disease if the person lived long enough with it for the bone to react. Bone changes slowly — months or years.

This does not prove Harari wrong. It proves the bone evidence is much weaker than it is usually presented. When you read “nineteen of twenty-one populations got sicker,” the honest translation is “nineteen of twenty-one populations show more skeletal lesions” — which is not the same sentence.

5.2Modern foragers are refugees, not fossils

Almost everything we “know” about hunter-gatherer life comes from studying groups who survived into the twentieth century: the !Kung of the Kalahari desert, the Hadza of the Tanzanian scrub, the Inuit of the Arctic, the Ache of the Paraguayan forest.

Look at that list again. Desert. Scrub. Arctic. Deep forest. These are the worst environments on Earth. Those groups are not there because foragers like harsh land. They are there because farmers took everything good, thousands of years ago.

So the “before” picture in Axiom 4 is built from survivors in marginal refuges. It is like judging what village life was like by studying only the people who fled to the mountains. Both directions of error are possible: real foragers in the fertile Levant may have lived far better than the Kalahari !Kung — which helps Harari — or the comparison may be meaningless — which hurts him. Either way, the confidence is not warranted.

5.3Forager life was not one thing, and some of it was extremely violent

The single most damaging fact for the romantic picture is the enormous variation between forager groups.

The rate of violent death per year lived was more than ten times higher among the Ache and Hiwi than among the Hadza and !Kung. There is no such thing as “the forager condition.” Harari’s “before” picture quietly averages over a range so wide that the average is meaningless.

Add child mortality. Across studied hunter-gatherer groups, roughly 38% of children died before the age of fifteen. Whatever else foraging was, it was not gentle.

5.4The famous work-hours number was wrong, and the correction never travelled

Sahlins’ “three to five hours a day” is the number everybody quotes. It counted only time spent hunting and gathering. It did not count food processing, tool-making, firewood, water-carrying, or camp moves. Richard Lee’s own later work, adding processing, produced roughly 40 hours a week for !Kung women and 44 for men.

That is a normal job. The entire “original affluent society” contrast — the emotional heart of Axiom 4 — largely dissolves at that number, and it has been known to dissolve for decades.

B — Problems with the reasoning

5.5The non-identity problem: who exactly was cheated?

This comes from the philosopher Derek Parfit, and it is a genuine puzzle, not a debating trick.

For you to exist, one specific sperm had to meet one specific egg, which required your parents to meet at one specific moment, which required a chain of billions of such moments going back through every ancestor you have. Change anything, anywhere, and a different person exists instead of you.

The best Harari response is that the victims were the actual people in 8500 BC, not us. That works — but it also shrinks the claim considerably. It becomes “a specific set of dead people had worse lives than their grandparents,” which is true of countless generations in history and would not normally earn the title “history’s biggest fraud.”

5.6Axiom 1 is a philosophical position, not a fact — and it is contested

Harari treats “judge the individual, not the species” as obvious. In moral philosophy it is one side of a live, unresolved dispute:

• Average view: what matters is how good the typical life is. Harari’s view. A world of 8 million happy foragers beats 8 billion stressed farmers.
• Total view: what matters is the sum of all wellbeing. If each of 8 billion lives is even slightly worth living, that world wins by a distance.

Neither is obviously correct, and both lead to conclusions people hate. Parfit showed that the total view leads to the Repugnant Conclusion: a gigantic population of lives barely worth living beats a small population of wonderful lives. But the average view has monsters of its own — taken strictly, it says the world improves when unhappy people cease to exist, and it can prefer a single ecstatic person to billions of contented ones.

The honest summary: Harari’s foundation stone is not a discovery about the past. It is a choice about values, presented as though it were a finding.

5.7“Fraud” implies a road not taken, and there was none

A fraud means you would have been better off refusing. But nobody could refuse. Once one neighbouring band farmed, the choice was gone. A “bad deal” with no available alternative is not a fraud — it is a condition, like gravity or ageing.

Harari knows this; he says so himself when he explains that once one band settled, agriculture became irresistible. But that admission quietly cancels the word he chose for the chapter title. You cannot be defrauded by the only option on the table.

5.8A claim that cannot be tested is not a scientific claim

“We did not domesticate wheat; wheat domesticated us.” What observation could show this to be false? None. There is no measurement that comes out differently depending on whether the sentence is true.

That is fine for literature. It is not fine when the sentence is offered as a historical finding, and readers cannot tell which mode they are in. Sapiens switches between the two constantly and rarely signals the change.

5.9The “iron law” problem

Harari calls the luxury trap “one of history’s few iron laws.” Professional historians are, as a group, deeply hostile to iron laws — the twentieth century was substantially wrecked by people who thought they had found one. It is also internally odd: a book whose main argument is that history is driven by invented, changeable stories reaches for an unchangeable law when it wants to make its most memorable point.

The anthropologist C. R. Hallpike, reviewing Sapiens, put the general complaint bluntly: whenever Harari’s facts are broadly correct they are not new, and whenever he strikes out on his own he often gets things wrong. That review is harsh and not universally accepted, but it names something real — the book’s most quotable lines are usually its least supported ones.

C — The hidden side that both camps ignore

5.10There was no “average farmer”

This is the most important omission in the popular version of the argument. Harari compares the average forager to the average farmer. But farming did not make everyone worse off equally. It split humanity in two.

Agriculture produced, for the first time, a small group of people whose lives were better than any forager who ever lived — a Sumerian priest, an Egyptian scribe, a landowner. It simultaneously produced a large group whose lives were worse than any forager who ever lived — a chained field slave, a bonded labourer, a woman grinding grain for twelve hours.

5.11The plough, and the fact that it is still in your society today

Here is a piece of evidence that supports Harari far more strongly than anything in his chapter — and he does not use it.

Economists Alberto Alesina, Paola Giuliano and Nathan Nunn, writing in the Quarterly Journal of Economics in 2013, asked why some societies have much more rigid gender roles than others. Their answer traced to the plough. Hoe agriculture uses upper-body and lower-body labour in ways women and men can share. Plough agriculture requires sustained upper-body strength and control of a large animal, which pushed men into the fields and women into the home.

Their finding: descendants of societies that historically used the plough — measured today, across countries and across migrant communities within countries — still show lower female participation in work, politics and business. A farming implement from thousands of years ago is still visible in employment statistics in 2026.

This is Axiom 10 proven with modern data. It is also a warning: the consequences of the wheat bargain did not stay in the Neolithic.

5.12We did adapt — the mismatch story is only half true

A popular version of Harari’s argument says our bodies are “Stone Age bodies in a modern world.” Genetics complicates this considerably.

• Lactase persistence. Most mammals stop digesting milk after infancy. Several human populations independently evolved the ability to keep digesting it, in the 10,000 to 6,000 year window after animals were domesticated. That is evolution keeping pace with farming.
• Amylase. Amylase is the enzyme that breaks down starch. Populations with starch-heavy farming diets carry more copies of the amylase gene. Recent work found that haplotypes carrying duplicated copies increased roughly sevenfold in the past 12,000 years in West Eurasians — one of the clearest signals of recent selection in our genome.
• Malaria and sickle cell. Forest clearance for farming created standing water and mosquito habitat, which drove malaria, which drove the spread of the sickle-cell trait. That is a genetic scar and a genetic adaptation in the same gene — a perfect miniature of the whole bargain.

So the honest picture is not “wheat beat us.” It is co-evolution: two species reshaping each other, with costs and adaptations on both sides. That is less quotable and more accurate.

5.13The trap did eventually open — after 11,800 years

Human height only recovered to, and then exceeded, pre-agricultural levels in the last century or so, and only in some places. Child mortality only collapsed in the last hundred and fifty years. Famine only stopped being routine in most of the world in living memory.

Two conclusions follow, and both are uncomfortable for somebody. For Harari’s critics: the bad deal genuinely lasted for about 99% of the time since it was struck. That is a very long time to call it a temporary adjustment. For Harari: it did end. The trap had an exit. It took eleven millennia and required industrial agriculture, sanitation, vaccines and fossil fuels — but the door was not permanently welded shut, which is exactly what his framing implies.

5.14The mourner is standing on the thing being mourned

You can only grieve for the foragers from inside the world that replaced them. Writing, printing, universities, archaeology, radiocarbon dating, publishing, literacy, and the leisure to read a book about it are all products of the surplus Harari is condemning. No forager band could have produced the critique.

This does not make the argument wrong. Plenty of true criticisms are made from inside the thing criticised. But it should reduce everyone’s certainty, because it means the entire debate is conducted by the winners of the process, using tools the process created, judging by values the process invented.

5.15The idea has a political tail, and it is worth knowing about

The “civilisation was a mistake” idea has a history outside the seminar room. In gentle forms it produces environmentalism, food reform and a healthy scepticism about technology. In extreme forms it has produced anti-modern movements, and in a few cases has been used to justify violence against people working in science and technology.

Harari holds none of these positions and argues against nostalgia explicitly. But readers should know the idea has a slope, and that “we should never have started” has been used as a premise for conclusions that most people who quote the chapter would reject. Recognising a trap is useful. Concluding that the people inside it are the enemy is not.

Part 06What every field of study says about it

One sentence, eighteen departments. Each one has looked at this question with its own tools and reached its own verdict. Read together, they are more informative than any single field — and they do not agree.

6.1Archaeology

Verdict: the “revolution” framing is dead. Nobody in the field now believes in a sudden switch. Domestication of wheat took perhaps two to three thousand years of unconscious selection. Sites like Göbekli Tepe show monumental building before farming. Ohalo II in Israel shows people processing wild cereals over 20,000 years ago. The story is long, patchy, and reversible for most of its length.

6.2Bioarchaeology and paleopathology

Verdict: agrees with Harari on direction, disputes his confidence. The decline in stature and rise in dental and infectious markers is real and repeated across continents. But the osteological paradox, differential burial and preservation, and small sample sizes make the size of the effect much less certain than popular accounts suggest.

6.3Genetics and evolutionary biology

Verdict: complicates the story in Harari’s favour and against him at once. Ancient DNA confirms early European farmers were shorter than their genes predicted — strong evidence of environmental stress. But lactase persistence, amylase duplication and malaria-related adaptations show humans changed genetically in response to farming within a few thousand years. This was a partnership, not a conquest.

6.4Epidemiology and medicine

Verdict: strongly supports Harari. This is the least controversial part of the case. Crowd diseases require crowds. Zoonotic diseases require animals living beside us. Both are agricultural products. The single deadliest consequence of the Agricultural Revolution is probably not hunger or overwork; it is the invention of the epidemic.

6.5Nutrition science

Verdict: supports Harari on variety, corrects him on calories. Diet diversity collapsed and deficiency diseases appeared. But the total calorie supply per unit of land rose enormously, which is why populations grew. “Worse food, much more of it” is the accurate summary.

6.6Ecology and earth science

Verdict: supports and extends the argument. Salinisation of Mesopotamian fields, deforestation across Eurasia, soil erosion, and the reshaping of whole biomes into monocultures. William Ruddiman has even argued that early farming — rice paddies producing methane, forest clearance producing carbon dioxide — began altering global climate thousands of years before industry. On that view the Anthropocene starts with the hoe, not the steam engine.

6.7Economics

Verdict: has a name for the trap and a disagreement about the cause. The Malthusian trap describes exactly Harari’s mechanism: gains in productivity are consumed by population growth, so income per person returns to subsistence. Gregory Clark’s work argues this held until about 1800, meaning the average person was no better off than a forager for the entire agricultural era. But Ester Boserup reverses the causation, and economists remain split on which way the arrow runs.

6.8Political science and state theory

Verdict: strongly supports Harari, via James C. Scott. Grain is the only crop with all the properties a tax system needs. Every early state ran on grain. Early states were also small, fragile, and frequently escaped from — Scott argues that “barbarians” outside state control often had better lives than subjects inside it, and that people fled states far more often than the standard story admits.

6.9Sociology and gender studies

Verdict: provides the best modern proof of a Neolithic consequence. The plough hypothesis links ancient farming technology to measurable gender gaps in employment and politics today, across countries and among migrant communities. Few claims about the deep past have this quality of evidence.

6.10Demography

Verdict: confirms the mechanism precisely. The Neolithic Demographic Transition shows a sharp rise in birth rates at the onset of farming, visible in the age structure of cemeteries. Settled life shortened birth spacing from about four years to about one or two. More births and more child deaths at the same time — which is exactly the shape of Harari’s trap.

6.11Psychology

Verdict: supports the luxury trap as a mechanism of mind. Hedonic adaptation is the well-documented tendency to return to a baseline level of satisfaction after any improvement. The Easterlin paradox describes how rising national income does not straightforwardly raise reported happiness. Both say the same thing Harari says: improvements get absorbed, then become the new floor, and the floor is never enough.

6.12Philosophy

Verdict: the argument’s weakest joint. Population ethics has no settled answer about comparing worlds of different size. The non-identity problem makes it unclear who was harmed. And “fraud” requires consent, alternatives and agency, none of which were available. Philosophy does not refute Harari; it shows that the moral part of his claim is asserted rather than argued.

6.13Religion and theology

Verdict: got there first, by about 2,500 years. Genesis chapter 3 describes humans expelled from a garden where food simply grew, and condemned to “eat bread by the sweat of your brow,” with painful childbirth added for good measure. Harari himself remarks that farmers ate bread by the sweat of their brows. Chapter 4 then has Cain, a tiller of the soil, murder Abel, a keeper of sheep. Whether or not that is a folk memory of the Neolithic, it is the same story with the same moral, and it survived because it described something people recognised.

6.14Law

Verdict: traces its own origin to the same event. Land ownership, boundaries, inheritance, contracts, debt, and the concept of theft as a serious crime all require durable, storable, divisible property. Foragers had possessions; they did not have real estate. Almost every legal category that structures modern life is a Neolithic invention.

6.15Linguistics

Verdict: adds a consequence Harari does not mention. The farming/language dispersal hypothesis, associated with Colin Renfrew and Peter Bellwood, argues that today’s great language families spread because farming populations expanded and absorbed or displaced forager populations. If it holds even partly, then most of the languages spoken on Earth today — including the one you are reading — are downstream of the wheat bargain.

6.16Military history

Verdict: supports the mechanism. Storable surplus creates something worth raiding, which creates walls, which create garrisons, which create standing armies, which require taxation, which requires more surplus. Jericho’s wall and tower go back roughly 10,000 years — among the earliest large defensive structures known, built within a thousand years of the first farming villages.

6.17Computer science and systems thinking

Verdict: recognises the pattern immediately and has better words for it. What Harari describes is path dependence and lock-in: an early choice constrains all later choices, and the cost of switching grows until switching becomes impossible even when the alternative is better. Engineers call the accumulated cost technical debt. The QWERTY keyboard is the textbook example — designed for mechanical typewriters that no longer exist, unchangeable because everyone already learned it. Farming was humanity’s first and largest instance of lock-in.

6.18Animal ethics

Verdict: the strongest support Harari has, and the least discussed. If you count non-human suffering at all, the ledger is not close, and no amount of human welfare data changes it. This is the one part of the argument that has never been seriously answered — mostly because it has rarely been seriously engaged.

Part 07The argument on Indian ground — Mehrgarh to the Green Revolution

Most discussion of this chapter stays in the Middle East. That is a mistake. South Asia contains both the strongest counter-example to Harari’s argument and the clearest live demonstration of it — the second one happening right now, in Punjab, on land you can drive to.

7.1Mehrgarh, around 7000 BC

South Asia did not wait for anyone. At Mehrgarh, at the foot of the Bolan Pass in Baluchistan, people were farming and herding by about 7000 BC. The dominant crop was barley — well-developed six-row barley — supplemented by wheat. Four kinds of wheat appear: einkorn, emmer, durum and bread wheat.

The wheats almost certainly arrived from South-West Asia. But the grains found at Mehrgarh are small and rounded, which indicates that local varieties were being bred for local conditions. And the animals tell an independent story: the humped Indian cattle, Bos indicus — the zebu, ancestor of every Indian cow you have ever seen — was domesticated in the subcontinent itself.

7.2The Indus Valley: the best counter-example in the world

Axiom 10 says grain surplus produces kings, priests, armies and steep hierarchy. The Indus Valley Civilisation is the hardest case that claim has to face, and it sits in South Asia.

Between roughly 2600 and 1900 BC, the Indus cities — Mohenjo-daro, Harappa, Dholavira, Rakhigarhi — were among the largest and most sophisticated on Earth. Grid-planned streets. Covered drains and household bathrooms. Standardised brick sizes across hundreds of kilometres. A weight system so uniform it is still startling. Long-distance trade. Writing.

And then the absences. No identified palaces. No royal tombs stuffed with treasure. No monuments showing a ruler’s face. No clear evidence of a standing army. No temples of the Mesopotamian kind. This is a huge agricultural surplus without the pyramid of visible tyranny that the theory predicts.

7.3The Green Revolution: the wheat bargain, replayed in one lifetime

Now the part that matters most. If you want to see Harari’s mechanism working — not inferred from bones, but documented in government reports, bank records and satellite data — you do not need a time machine. You need a train to Ludhiana.

What happened, in order

The problem. In the early 1960s India could not feed itself. Food came by ship under American aid. Serious forecasters predicted mass famine.

The solution. Semi-dwarf, high-yielding varieties of wheat and rice, developed through the work of Norman Borlaug and adapted for India under M. S. Swaminathan, were introduced from the mid-1960s. Punjab was the testing ground: good soil, existing canals, and farmers willing to try.

The result. It worked, spectacularly. India went from importing grain to exporting it. The predicted famine did not happen. Hundreds of millions of people did not starve. This must be stated plainly and kept in view for the rest of this section: the Green Revolution is one of the great humanitarian achievements of the twentieth century.

And then the terms of the deal appeared

The new varieties are not magic seeds. They are seeds that convert inputs into grain. They need reliable fertiliser, reliable pesticide, and above all reliable water — far more than the old varieties, and on a fixed schedule that canals could not always meet.

So farmers dug tubewells. To run tubewells you need electricity, so electricity for agriculture was made free or heavily subsidised. Free power means no reason to pump less. As the water table fell, farmers dug deeper and bought stronger pumps. Each of those steps was individually rational and individually beneficial — exactly the shape described in Axiom 6.

The lock: why nobody can stop first

Punjab grows wheat and paddy in rotation on the great majority of its cropped land. Paddy is a water-hungry crop being grown in a state that is running out of water. Everyone involved knows this. It has been in the newspapers for thirty years. And still it continues, for a reason Harari would recognise instantly:

• Guaranteed purchase. Minimum Support Price procurement is dependable mainly for wheat and paddy. A farmer who switches to maize, millet or pulses gives up a guaranteed buyer for an uncertain market.
• Debt. Average landholdings are small — around two hectares. Input costs for seed, fertiliser, diesel, pesticide and labour rise every year. Formal credit falls short, so many farmers borrow informally at punishing rates. A person servicing debt cannot afford one experimental season.
• Sunk investment. The tubewell, the pump, the laser leveller, the combine — all bought for this crop cycle. Switching crops means writing them off.
• No coordination. If every farmer in a district switched together, the risk would be shared. There is no mechanism to make that happen.

The human cost is measurable. Punjab has recorded on the order of seven thousand farmer suicides since 2000, and researchers who have worked on the ground argue the true figure may be several times higher. Studies at Punjab Agricultural University have repeatedly identified debt and poverty as the leading causes.

7.4Why Punjab is the best test the argument has ever had

Go back to the Ratchet from Part 04 and check the four conditions against the Green Revolution:

All four conditions, satisfied, on the record, within living memory. This is not an inference from a skeleton. It is a documented case, with dated policy decisions, measured water tables and named consequences. If you want to argue that the luxury trap is a real mechanism rather than a literary device, Punjab is the evidence you should reach for.

7.5And what Punjab corrects in Harari

Here is the important amendment, and it is a hopeful one.

The trap in Punjab is not caused by the seeds. High-yielding varieties are not the villain; they fed a country. The trap is caused by the rules built around the seeds — free electricity that removes any reason to conserve, procurement guarantees that apply to only two crops, credit systems that make experimentation impossible, and the absence of any way for farmers to switch together.

Every one of those is a policy. Policies are written by people, and can be rewritten by people. Which means:

Part 08The luxury trap you are standing in right now

If the mechanism is real, it did not stop in the Neolithic. This part is a practical tool: how to tell whether something in your own life is a luxury trap, and what actually gets people out — because it is not willpower.

8.1The four tests

Take any technology, habit, product or working practice. Ask four questions. If the answer to all four is yes, you are looking at a luxury trap.

8.2Worked examples

8.3Why willpower never works

The single most useful thing in this document is this paragraph, so it is worth reading slowly.

A luxury trap is not a discipline problem. It is a coordination problem. The reason a Punjabi farmer keeps planting paddy is not weakness. The reason you keep answering messages at 11pm is not weakness. In both cases, the individual who stops first pays the full cost and receives none of the benefit, because the system carries on without them.

8.4What has actually worked

Historically, luxury traps have been opened by coordination, not by heroic individuals. Every one of these was once considered impossible:

• The eight-hour day and the weekend. No individual worker could refuse long hours. A law could.
• Seatbelts and helmets. Individually inconvenient; collectively enforced; tens of millions of lives saved.
• Leaded petrol. No consumer could solve it. A ban could.
• France’s “right to disconnect.” An attempt to make not-answering-after-hours a default rather than an act of courage.
• Phone bans in schools. Whatever you think of them, they are the correct shape of solution: change the default for everyone at once, so no child pays the price of going first.

Notice the pattern. In each case, the fix was to change what happens by default, for everyone, at the same time. That is the only known exit from a ratchet.

8.5Five questions worth sitting with

1. What did I adopt in the last five years because it would save me time — and do I now have more free time than before?
2. What could I not stop doing, even if I decided to, because the cost of stopping alone is too high?
3. What in my life started as a treat, became normal, and is now a fixed cost I cannot cut?
4. Which of my obligations did I actually agree to, and which arrived by accumulation while I was busy?
5. If I wanted to change one of these, who else would have to change with me for it to work?

The last question is the real one. If you cannot name the other people, you are not planning an escape. You are planning to sit down at a concert.

Part 09The scorecard, and the sentence rewritten

Eleven axioms went in. Here is what came out, with each one marked honestly, including the ones that did not survive.

The sentence, rewritten honestly

Harari wrote: “The Agricultural Revolution was history’s biggest fraud.” After all of the above, here is the version that survives every objection in this document:

That version is less quotable. It is also true, and it does the same work: it strips the comfort out of the word progress without replacing it with a fairy tale about the past.

The last thing worth saying

People argue about this chapter as though the question were whether the ancients were happy. It is not. The ancients are gone and their happiness is unrecoverable.

The question the chapter actually poses is about the shape of a decision. It says: be very careful with any choice that improves your position, requires constant maintenance, spreads because refusers lose, and cannot be undone once your dependents outnumber your options.

Farming was the first one. It was not the last. Punjab is living inside one right now, with the water table as the timer. And whatever bargains are being offered in the next twenty years — about work, attention, intelligence and judgement — will arrive in the same disguise the first one wore: as an obvious improvement, one small sensible step at a time, with no moment at which anyone is asked to agree.