1Part Two, in four paragraphs
So this part stands on its own, even if you have not read the last one.
Part Two answered the question Part One asked. Six or more kinds of human, and only one left. Why us? Harari's answer: around 70,000 years ago something rewired the inside of the Sapiens brain. Not a bigger brain — a differently wired one. He calls it the Cognitive Revolution, and right after it the evidence explodes: boats, needles, jewellery, cave paintings, long-distance trade, burials with objects placed in them. What it gave us was a new kind of language. Its first job was probably gossip — keeping track of who can be trusted, which is the only thing that makes cooperation safe. But gossip runs out at around a hundred and fifty people, because that is roughly where personal knowledge runs out. The thing that breaks the ceiling is fiction. We can talk about things that do not exist — gods, nations, money, companies, laws — and get millions of strangers to act as though they are real. That is why five hundred Sapiens beat fifty Neanderthals who were individually stronger. It was never a fight about intelligence. It was a fight about numbers, and stories are how you get numbers.
Two chapters, and they do very different jobs.
Chapter 3 asks what our lives were actually like for the two and a half million years before farming. That is the overwhelming majority of our existence, and it is the world your body still thinks it lives in.
Chapter 4 asks what we did the moment we arrived somewhere new. It is the shortest chapter in this part and the one that changed how I see our species.
One is about who we were. The other is about what we cost.
2The arithmetic that should rattle you
Almost everything you think of as normal human life is younger than your great-great-grandmother's family tree.
Humans have existed for about two and a half million years. For roughly two million four hundred and ninety thousand of those years, every single human being alive got food by hunting animals and gathering plants. That was the job. There was no other job. Farming is about twelve thousand years old. Cities and writing, about five thousand. Industry, about two hundred and fifty. Computers, about seventy. The phone in your pocket, about twenty.
Forager: somebody who lives by gathering wild plants and hunting wild animals, rather than growing food or keeping livestock.
You will also see hunter-gatherer, which means the same thing. Harari prefers forager partly because the older term puts hunting first, and for most groups the gathering was doing more of the feeding.
Why it matters: this is not a stage we passed through on the way to being properly human. This is what being human was, for more than ninety-nine per cent of our existence.
That is the fact underneath the entire chapter. Your body and your mind were shaped for a world that no longer exists. Every instinct you have was tuned for a small band of people walking through open country looking for food, and then the world changed under you in a few hundred generations while the instincts stayed exactly where they were.
3Why you want the sugar
The most useful idea in the chapter, and the one that made me stop blaming myself for something.
There is a field that studies this mismatch. It has a heavy name and a simple job.
Evolutionary psychology: the study of how instincts that were useful in our ancient environment still steer our behaviour today, even when the environment has completely changed.
The core move is always the same. Take a modern habit that looks irrational. Ask what problem it would have solved for a forager. Usually there is an answer.
Why it matters: it converts a lot of things we treat as personal failings into design mismatches. That is not an excuse. It is a diagnosis, and diagnoses are more useful than blame.
Harari's example is the fridge, and it is the best short explanation of a craving I have read.
Why do you want sugar and fat? Not because you are weak. Because for two and a half million years, sugar and fat were rare, precious and completely unreliable.
Picture your ancestor walking into a clearing and finding a fig tree loaded with ripe fruit. What is the correct behaviour?
Eat as many figs as you physically can, right now. Not because you are greedy. Because tomorrow the baboons will have found the tree, or the birds will, or it will rain, or you will have moved on. There is no fridge. There is no tomorrow-fruit. The only storage available is your own body.
An ancestor with excellent self-control who ate three figs and saved the rest for later had children who inherited excellent self-control, and those children starved slightly more often than everybody else's. Restraint was a fatal personality trait.
That instinct is still in you, completely intact. But you do not live near a fig tree. You live near a shop that is open until eleven at night and sells sugar by the kilogram, and there is a machine in your building that will bring more of it to your door.
Same instinct. Different world. And now the instinct is eating you.
I have read a lot of writing about food and willpower and most of it is either cruel or useless.
This is neither. It does not tell you the craving is imaginary and it does not tell you to try harder. It tells you the craving is two and a half million years old, that it was correct for almost all of that time, and that the thing which changed is not you.
That reframing does not make anybody thin. But it moves the problem out of the category of character and into the category of environment, and problems in the second category are much easier to actually work on.
4The curtain of silence
Harari does something here that most popular writers will not, and it raised my opinion of him considerably.
Now the hard question. What were their lives actually like? And here he stops and tells you how little anybody knows. Then he tells you again. And again.
Problem one: everything they made has rotted
Foragers built almost everything out of wood, bamboo, leather, reed, fibre and bone. Every single one of those materials rots. What survives is stone. So when we dig, we find stone tools, and stone tools, and more stone tools — and we named the entire era after them.
Calling it "the Stone Age" is like a future archaeologist digging through our rubbish, finding only the plastic, and calling us the Plastic Age.
They would have our bottle caps and our packaging. They would not have a single book, a single song, a single photograph, a single conversation, or any idea what we cared about.
Harari's suggestion is that "the Wood Age" would be closer to the truth. The stone is what lasted. It is not what mattered.
This is a small point that quietly destroys a lot of confident writing about prehistory, including some of Harari's own later in the chapter.
Problem two: there was no such thing as "the forager lifestyle"
There were thousands of separate bands, spread across every climate on Earth, eating completely different food, with different neighbours and different histories. They would have differed as much as modern countries do. Some may have been gentle and some brutal. Some obsessed with ancestors, some indifferent. Some rigidly hierarchical, some flat. Any sentence that begins "ancient hunter-gatherers believed…" should make you narrow your eyes. Which ones? Where? When?
Problem three: the curtain of silence
This is the big one, and it is the phrase from this chapter I have kept. We have essentially no idea what they thought. Not their politics. Not their gods. Not their rules about who could marry whom. Not their arguments, their music, or their jokes. Tens of thousands of years of complete human lives — love affairs, betrayals, revolutions in belief, whole religions rising and dying — and almost every trace of it is gone. Not misplaced. Gone. There is no archive. There never was one. This is by far the largest blank space in human history, and it covers most of human history.
Everybody in the argument about forager life — Harari, his sources, his critics, me — assumes that "work" is a thing you can count.
The famous claim is that foragers worked a thirty-five to forty-five hour week. To produce that number, somebody had to decide what counts.
Does walking to the next valley count as work, or as travel? Does carrying water count? Does processing food count — the pounding, the soaking, the hours of preparation? Does making a tool count? Does teaching a child to track count? Does gossip count, when gossip is the maintenance of the only insurance system you have?
For a forager there was no line between working and living. The category "work" — a bounded activity you do for someone, in exchange for something, during specific hours — is an invention of a wage economy. Measuring a forager's working week is applying a ruler that was manufactured ten thousand years later.
The number is not wrong exactly. It is an answer to a question those people did not have.
5What the evidence does suggest
With all those warnings standing, here is the picture — and one finding that genuinely unsettled me.
They probably worked less than you do
Studies of forager societies that survived into the twentieth century — including groups in the Kalahari, living on genuinely poor land — suggest something in the range of thirty-five to forty-five hours a week of food-getting and related work. Finding food took a few hours a day. The rest was walking, talking, making things, resting.
Count honestly. Your job. Your commute, both ways. The work you take home. Cooking, cleaning, shopping, bills, repairs, paperwork. The messages you answer at ten at night because the phone is right there.
For a great many people that comes to well over fifty hours, and a fair amount of it is not restful even when it is not officially work.
The comparison is unfair in both directions — they had no medicine, no dentistry, and a real chance of dying from an infected cut. But the specific claim that we bought ourselves leisure is not obviously true, and that is the claim worth examining.
They probably ate better
A band might eat several dozen different plant and animal species across a year. Variety does two things. It gets you the full range of nutrients almost by accident, without anybody knowing what a nutrient is. And it means a bad year for one food is not a catastrophe — you eat the others. Later farmers ate wheat, or rice, or maize, or potatoes. When that one crop failed, there was nothing else, and people died. Famine is not an ancient constant. It is a farming invention, and Part Four goes into it properly.
They were probably healthier
Most of the great infectious diseases came from two things foragers did not have: domesticated animals living close by, and crowded permanent settlements sitting next to their own waste. Skeletons from the forager period tend to be taller and show fewer signs of malnutrition than skeletons from the early farming period that followed. That is a real, measurable finding, and it is one of the strongest planks in the argument coming in Part Four.
And this is the one that stings
A forager needed working knowledge of hundreds of plants — which ones feed you, which ones kill you, which ones only kill you at certain times of year. Animal behaviour. Weather. Tracking. Seasons. Tool- making. First aid. Navigation without instruments. And every person in the band. A modern person can survive knowing almost nothing, because the collective knows everything and we plug into it.
Not buy one. Make one, starting from nothing.
You would need to find and fell the right timber, cut and season it, locate graphite, mine it, mix it with clay in the right proportion, fire it, glue the halves, and — if you want the rubber and the metal band — find rubber and smelt metal.
No single human being on Earth can do all of it. There is a famous essay making exactly this point, and it is not a trick: our knowledge is collective and it is distributed, and almost none of it is in any one head.
Your ancestor could not build a pencil either. But they could keep themselves alive, alone, in a landscape, for as long as their body held out. Almost none of us could do that for a week.
And then Harari drops something I had to look up twice, because I did not believe it. Average human brain size appears to have declined since the forager period.
Not by a huge amount, and the cause is argued over. But the direction is real and it is measurable in the skulls.
As a species we know unimaginably more than we did. As individuals we may be dimmer than our ancestors, because being sharp is no longer a survival requirement. Nothing culls the person who cannot read weather any more. We built a society that carries its stupid people, which is a moral achievement, and one of its side effects appears to be visible in bone.
6A world with no gods above it
What they probably believed, and two questions Harari refuses to settle. I respect both refusals.
Most foragers were probably animists, and understanding what that means changes how you read the next several parts.
Animism: the belief that every place, animal, plant and natural event has a mind of its own — that a rock, a river or a fig tree can want things, and can be spoken to and bargained with.
Here is the part people miss. In an animist world there is no hierarchy. There is no separate realm of gods sitting above nature and issuing instructions. The bear is a person. The mountain is a person. You negotiate with them the way you negotiate with your neighbours, and sometimes you lose.
Humans are one kind of being among many. Not the management.
Why it matters: hold on to this. In Part Four, farming destroys it completely, and Part Eight shows you exactly how the gods got installed above us once we started owning animals instead of arguing with them.
Two questions he lays out and then declines to answer.
Did they live in families the way we do?
One view says the monogamous couple raising its own children is ancient and natural. Another argues for something much more communal, where sexual partnerships were fluid and children were raised by the band collectively — sometimes called the "ancient commune" theory. Harari gives both, points out that the evidence on each side is thin, and does not pick.
Were they peaceful?
The case for violent. At Jebel Sahaba in northern Sudan there is a cemetery somewhere between thirteen and fourteen thousand years old containing fifty-nine skeletons. Around forty per cent of them have arrowheads or spear points embedded in or lying against the bone, including women and children. That is not an accident and it is not a hunting injury. That is a massacre, or a long campaign of them. Ethnographic studies of some surviving forager societies have also recorded homicide rates well above those of modern states. The case for peaceful. Other forager sites from the same era show almost no signs of violent death at all. Small mobile bands have an option settled people do not have: when a conflict starts, you can simply leave. There is nothing to defend, no stored grain to take, and no territory worth dying for if the next valley is empty. And the ethnographic studies have a serious selection problem — the forager societies that survived into the twentieth century were mostly ones under pressure from expanding farming states, which is not a neutral condition.
Where it stands. Almost certainly, it varied enormously. Some bands lived peacefully for generations; some fought constantly. There is no single natural human level of violence waiting to be discovered, and the honest answer is a range rather than a number.
What would settle it. Many more well-dated cemeteries from many more regions. The current evidence is a handful of sites, and a handful of sites cannot describe a planet.
Why people care so much. Because the answer gets used. If humans are naturally peaceful, then war is a product of civilisation and could in principle be un-built. If humans are naturally violent, then the state and its police are what stand between us and each other. Both are political positions, and both sides have reached for the same fifty-nine skeletons.
And it was not gentle even at its best. A band on the move cannot carry people who cannot walk. Children who could not be fed, elders who could not keep up, and the badly injured were sometimes left behind or killed.
Harari does not hide this, and the chapter is better for it. There is no paradise here. There is a different set of problems.
7The Flood
The shortest chapter in this part, and the one that changed how I see my own species.
Around forty-five thousand years ago, Sapiens did something no human had ever done. We crossed open sea and landed in Australia. Sit with how strange that is. Every human before this had stayed inside the great landmass of Africa, Europe and Asia, where you can walk. Australia was cut off by open water. Reaching it required boats or rafts, provisions, planning, and the willingness to sail towards a place you could not see and had no evidence existed. Harari ranks it alongside Columbus crossing the Atlantic and Apollo 11 reaching the Moon — the first time our species stepped outside its own ecosystem entirely. I had never thought of it that way and I think he is right. And then something happened to Australia.
A continent of monsters, and then not
Australia had an extraordinary cast of large animals.
Diprotodon — a wombat the size of a small car, weighing around two and a half tonnes.•
Kangaroos two metres tall.•
Marsupial lions.•
A giant koala.•
Flightless birds twice the size of an ostrich.•
Snakes five metres long, and lizards weighing hundreds of kilograms.•
Megafauna: simply "big animals". In this context it usually means land animals with an adult body weight over about forty-five or fifty kilograms.
Why it matters: this is the group that keeps disappearing. Small animals mostly survived every one of these events. That pattern — big things die, small things do not — is itself a clue, and Section Seven explains why.
Within a few thousand years of humans landing, essentially all of them were gone. Of the twenty-four Australian animal species weighing fifty kilograms or more, twenty-three vanished. Around ninety per cent of the continent's large animals.
Australia has never recovered. The Australia we know today is the wreckage of a much stranger place.
Three defences, and why they do not hold
"It was climate change." These animals had already survived many climate swings. Australia went through repeated ice ages and warm periods over the preceding million years, and the Diprotodon shrugged all of them off. Then, on the single occasion when a climate shift happened to coincide with the arrival of a new two-legged predator, everything died. Harari's point is that this is a very suspicious coincidence, and I agree.
"It was fire, not spears." Partly true, and it makes things worse rather than better. Humans practised fire- stick farming — deliberately burning large areas of scrub and forest, because the regrowth attracts game and produces edible plants. Do that for a few thousand years and you do not merely kill animals, you convert the vegetation of an entire continent. Fire-tolerant eucalyptus took over. Everything that depended on the old forests had nowhere left to live.
"Humans were too few and too primitive." This is where the strongest answer comes, and it is in the next section.
Where it stands. A serious minority of researchers still put most of the weight on climate, and they are strongest in Eurasia, where humans and large animals lived alongside each other for a very long time before the extinctions. That Eurasian exception is the best argument against Harari and he passes over it quickly.
What would settle it. Tighter dating. The entire argument turns on whether the animals disappear before or after people arrive, and at several sites the two dates are close enough that the answer depends on the method.
8You do not need a massacre
The mechanism, and the reason this chapter should worry you about the present.
Here is the explanation that reframed the whole thing for me. Large animals breed slowly. A big mammal has a long pregnancy, produces one or two young at a time, and waits years between births. Its population can only replace itself at a crawl. Now suppose humans arrive and kill one extra Diprotodon every few months. Not a slaughter. Barely a hunting season.
Do that steadily for a few centuries and the deaths outrun the births, and the species quietly runs out. Extinction does not require a massacre. It requires being slightly too good at killing, for a long time. That reframing did something to me. It means you do not need villains. Ordinary people, doing an ordinary amount of hunting, at a modest and entirely reasonable rate, will empty a continent if you give them long enough. Nobody in that chain does anything wrong on any given afternoon. And these animals had no fear of us. Fear has to be learned across generations, by having your ancestors eaten. Australia's animals had never seen a Sapiens. A two-and-a-half-tonne Diprotodon would have watched a group of small, slow, clawless primates approach with mild curiosity, right up until the spears went in.
Then we did it again, on a bigger stage
Around sixteen thousand years ago, humans crossed from Siberia into Alaska. Sea levels were low during the ice age, so there was walkable land where the Bering Strait is now — a region called Beringia. From there people came down through the Americas and reached the southern tip of South America within roughly two thousand years. That is extraordinarily fast for people walking. And the American megafauna died. North America lost thirty-four of its forty-seven genera of large mammals. South America lost fifty out of sixty. Mammoths and mastodons. Sabre-toothed cats. Giant ground sloths weighing eight tonnes and standing six metres tall. Beavers the size of bears. And this one genuinely surprised me: horses and camels both evolved in the Americas, and both were wiped out there. The horses that Native American nations rode in the nineteenth century were descendants of animals brought back across the Atlantic by Europeans — a species returning to a homeland it had been exterminated from thousands of years earlier.
Add up the losses across every continent and Harari's summary is brutal: of the large land mammals alive before Sapiens spread out, roughly half were gone before anyone planted a single seed of wheat.
Not half the individuals. Half the kinds.
And it kept happening every time we reached somewhere new. Madagascar, around fifteen hundred years ago: the elephant bird, a three-metre flightless giant, and the giant lemurs, gone soon after people arrived. New Zealand, around eight hundred years ago: the moa, gone within a couple of centuries of the first Maori landing.
A last population of mammoths survived on Wrangel Island in the Arctic Ocean until roughly four thousand years ago — long after the Egyptian pyramids were finished — for one reason only. Nobody could reach them.
Then somebody reached them.
Both sides of this argument — overkill and climate — take it for granted that the story of an ecosystem is the story of its large animals.
That is where the evidence is, because big bones survive and small ones do not, and because a mammoth is a better sentence than a beetle.
But an ecosystem is mostly things you cannot see. Soil organisms. Insects. Seed dispersal networks. Fungal systems that connect whole forests underground. When the megafauna went, all of that reorganised too — and we have essentially no record of it, so it does not appear in the argument at all.
The consequence is that we are having a confident debate about the collapse of a system, using data from about one per cent of it, chosen because it fossilises well.
This is not an objection to Harari. It is a warning about the shape of the evidence, and it applies just as much to the extinctions happening around us now — which we are also measuring mostly in large, photogenic animals.
Three waves
THarari sets up a frame that runs through the rest of the book. There have been three waves of human-caused extinction. The first was foragers spreading across the world with stone tools and fire. The second came with farming, as we cleared land for fields. The third is industrial, it is happening right now, and it is the largest of the three. If you were waiting to find out when humans became dangerous to this planet, the answer is: on the first day.
9Where I think he is wrong in these two chapters
Chapter 4 is the strongest thing in the book so far. Chapter 3 is one of the shakiest.
1. The forager numbers come from a very small, very odd sample
he thirty-five-hour week, the varied diet, the good health — most of this rests on a small number of twentieth-century studies of specific groups, and it gets repeated as though it described all foragers everywhere. Worse, the groups available to study are not a fair sample. The forager societies that survived into the twentieth century did so in deserts, rainforests and Arctic edges, because farming states had taken every piece of good land. They also traded with farming neighbours, were pushed around by governments, and caught modern diseases. They are not time machines. They are the survivors of a long defeat, living on the worst land available, and using them to describe the Palaeolithic is a known problem in anthropology that Harari does not raise.
2. He warns against generalising, and then generalises
This is the tension I flagged in Part One and it is at its sharpest here. He spends pages telling you about the curtain of silence, the rotted evidence, the enormous variety between bands. Then he tells you fairly confidently what foragers ate, how long they worked, and what they believed about the world. I do not think this is dishonesty. A book consisting entirely of "we cannot know" would not be readable and nobody would publish it. But you should watch it happening, because those conclusions become load-bearing in Part Four.
3. The extinction figures are firmer in the prose than in the literature
"Twenty-three of twenty-four species over fifty kilograms" is a real and widely quoted figure. It is also sensitive to how you count. The underlying research generally counts genera rather than species — a genus being a group of related species — and different papers include different regions, different weight cut-offs and different dating standards. Change any of those and the number moves. The pattern is solid and I do not think it is in serious doubt. The decimal-point confidence is presentation.
4. Eurasia is the hole, and he steps over it
The overkill argument works beautifully where humans arrived suddenly into a naive ecosystem: Australia, the Americas, Madagascar, New Zealand. It works much less well in Africa and Eurasia, where humans and large animals evolved alongside each other over millions of years, the animals had time to learn fear, and the extinctions are patchier and harder to pin on arrival dates. That is the strongest argument the climate side has, and Harari gives it about a sentence. If I were arguing against him, that is where I would stand.
You are running two-and-a-half-million-year-old software in a world that is two hundred years old. The sugar craving was correct for almost all of that time. What changed is not you.
The biggest chunk of human history is unknowable. Wood rots, leather rots, and thoughts leave nothing. Anybody who describes prehistoric belief confidently — Harari included — is filling a gap with a guess.
Forager life was probably less laborious, better fed and healthier than the farming life that replaced it. That claim is contested at the edges, and it is the loaded gun that goes off in Part Four.
They were animists — a world of persons, some of whom are bears, with nobody in charge. Farming destroys this, and Part Eight shows you how.
And wherever we arrived for the first time, the big animals died. Not through massacre. Through being slightly too good at killing, for a few thousand years, at a rate nobody would have noticed. Half the planet's large land mammals were gone before farming even started.
10Words from this part
Everything I stopped to explain, gathered up.
| Word | What it means |
|---|---|
| Forager | Somebody who lives by gathering wild plants and hunting wild animals. Same thing as hunter-gatherer. |
| Evolutionary Psychology | The study of how ancient instincts still steer modern behaviour in an environment they no longer fit. |
| Curtainofsilence | Harari's phrase for the fact that we know almost nothing about what foragers thought, believed or argued about. |
| Animism | The belief that everything — animals, plants, rocks, rivers — has a mind. No gods above nature. Humans are not in charge. |
| Megafauna | Large land animals, usually meaning an adult weight over about fifty kilograms. The group that keeps disappearing. |
| Fire-Stickfarming | Deliberately burning scrub and forest so the regrowth attracts game and produces edible plants. |
| Beringia | The land bridge between Siberia and Alaska, walkable during the ice age when sea levels were low. |
| Overkillhypothesis | The view that human hunting was the main cause of megafauna extinction, rather than climate. |
| Genus | A group of closely related species. Most extinction counts are done in genera, not species — which is why the figures move about. |
11The timeline so far
Carried forward, with these two chapters added.
| When | What |
|---|---|
| 70,000 years ago | The Cognitive Revolution. Second exit from Africa. |
| 45,000 years ago | Sapiens cross open sea to Australia. Within a few thousand years, 23 of 24 large species are gone. |
| 16,000 years ago | Sapiens cross Beringia into the Americas. Reach the southern tip within about 2,000 years. |
| 14–13,000 years ago | American megafauna collapse. Jebel Sahaba cemetery in Sudan. |
| 12,000 years ago | Farming begins in the Middle East. Part 4 starts here. |
| 4,000 years ago | The last mammoths die on Wrangel Island, after the pyramids were built. |
| 1,500 years ago | Madagascar: the elephant bird and giant lemurs. |
| 800 years ago | New Zealand: the moa. |