Imagine travelling back 10,000 years.
You arrive in a landscape full of wild plants. There are forests, grasslands, rivers, nuts, fruits, tubers, seeds and animals. You meet a community of highly knowledgeable hunter-gatherers. They know which plants are poisonous, which fruits ripen when, where animals congregate, which roots can be dug up after rain and which seeds can be stored through winter.
Now ask a seemingly obvious question:
Why don't they simply become farmers?
After all, farming eventually transformed human history. It allowed populations to become denser, settlements to become permanent, food surpluses to accumulate and, eventually, states, armies, writing and large-scale technological systems to emerge.
So why didn't everybody become a farmer at roughly the same time?
This is the puzzle Jared Diamond tackles in Chapter 8 of Guns, Germs, and Steel, provocatively titled “Apples or Indians.”
And the title captures the central question perfectly:
Was the problem with the Indians—or with the apples?
Diamond's answer is one of the most important ideas in the book.
But it is also an answer that becomes considerably more interesting when viewed through the lens of modern archaeology.
The great agricultural puzzle
Diamond begins with a geographical mystery.
Some regions of the world seem perfectly capable of supporting agriculture, yet did not independently develop intensive food production at the same time as the Fertile Crescent.
Why not?
California was biologically rich.
Eastern Australia was biologically rich.
Large parts of Africa were biologically rich.
The eastern United States was biologically productive.
New Guinea was extraordinarily rich in plant diversity.
If humans everywhere were equally intelligent and equally capable of experimenting with plants, why did agriculture arise independently in only certain places?
Diamond proposes two broad possibilities.
Possibility 1: The people were different
Perhaps some human societies were simply more innovative.
Maybe they were more willing to experiment.
Perhaps they had cultural traditions that encouraged agriculture.
Or perhaps some populations were somehow inherently better at recognizing the possibilities offered by their environment.
Diamond is deeply suspicious of this explanation.
And for good reason.
It easily turns into an argument about supposedly superior and inferior peoples.
Possibility 2: The plants were different
Perhaps the crucial variable was not human intelligence but the biological raw material available to humans.
And this is where Diamond places his bet.
His argument is essentially:
Agriculture can emerge only if a sufficiently useful package of plants is available for domestication.
Not just one edible plant.
Not just a plant that can be cultivated.
A successful agricultural system needs a collection of species capable of providing enough calories, protein, carbohydrates, oils and other nutrients to sustain a growing population.
That turns out to be a much harder biological requirement than it first appears.
There are 200,000 flowering plants. So why aren't we eating more of them?
This is one of Diamond's most powerful observations.
The planet contains an enormous number of plant species.
But almost all of them are useless as crops.
Some are poisonous.
Some produce too little edible tissue.
Some are difficult to harvest.
Some grow too slowly.
Some have seeds that are too small.
Some have terrible nutritional profiles.
Some cannot reproduce conveniently.
Some are impossible to store.
And some simply aren't worth the enormous labour required to cultivate them.
Even today, despite modern genetics, biotechnology and enormous economic incentives, humanity has domesticated very few entirely new major crops.
Our ancestors therefore faced an astonishingly restricted menu.
The real question wasn't:
“What plants grow here?”
It was:
“What plants grow here that are biologically suited to becoming crops?”
That distinction is enormous.
The Fertile Crescent won the botanical lottery
Diamond's favourite example is the Fertile Crescent.
The region contained several plants with unusually favourable properties.
Wheat and barley, for example, had large, nutritious seeds.
They were annual plants.
They could be harvested in large quantities.
Their seeds could be stored.
They responded strongly to human selection.
And several of the important crops had reproductive characteristics that made their cultivation relatively straightforward.
The result was not one magical plant.
It was a package.
Grains supplied carbohydrates.
Legumes supplied protein.
Flax supplied fibre and oil.
And several domesticable animals supplied meat, milk, hides and traction.
This combination was extraordinarily powerful.
The Fertile Crescent therefore wasn't merely a place where someone happened to discover farming.
It was a place where the biological inventory happened to contain an unusually good starter kit for agriculture.
Diamond's broader point is that this biological lottery occurred before recorded history—and therefore long before anyone could have consciously chosen the geopolitical consequences.
And then there is the apple
Why call the chapter “Apples or Indians”?
Because the apple provides a beautiful thought experiment.
North America contained wild apples.
Native Americans knew their environment extraordinarily well.
So why didn't they domesticate apples?
Was there something about Native American societies that prevented them from becoming apple farmers?
Diamond's answer is: not really.
The problem was largely the apple.
Apples are relatively difficult plants to domesticate.
Unlike wheat, they cannot simply be planted from seed and reliably reproduce the desirable characteristics of the parent tree.
Producing particular apple varieties requires vegetative propagation, especially grafting.
And that is a sophisticated technique.
Interestingly, even in Eurasia, large-scale cultivation of apples occurred much later than the initial rise of agriculture.
So the fact that Native Americans had wild apples does not mean that they had an obvious agricultural opportunity waiting to be exploited.
The apple was there.
But the agricultural apple was not.
That distinction captures Diamond's argument beautifully.
But here's the really important point: agriculture isn't simply “planting crops”
This is where the chapter becomes more interesting than a simple geography-versus-culture argument.
We tend to imagine agriculture as a binary:
hunter-gatherer → farmer
But the archaeological record doesn't look like that.
There is a huge middle ground.
People can:
- collect wild plants;
- protect useful plants;
- clear competing vegetation;
- burn landscapes;
- transplant plants;
- scatter seeds;
- irrigate particular areas;
- harvest selectively;
- encourage particular species;
- return repeatedly to productive locations;
- cultivate plants without genetically domesticating them;
- and eventually select plants whose characteristics have been altered by human harvesting.
In other words:
domestication and agriculture are not the same thing.
A plant can be cultivated without being genetically domesticated.
And humans can substantially modify landscapes without becoming conventional farmers.
This distinction has become increasingly important in archaeology.
New Guinea complicates Diamond's story
Consider New Guinea.
Diamond uses New Guinea partly as an example of a region where agriculture developed under significant constraints.
But archaeological research has revealed something remarkable at Kuk Swamp.
People were manipulating plants and wetlands there thousands of years ago.
Evidence indicates plant exploitation around 10,000 years ago, followed by clear cultivation using mounded systems roughly 7,000–6,400 years ago.
Taro was being used early, and bananas were being intensively cultivated by this period.
This was agriculture—but it didn't look like the wheat fields of the Fertile Crescent.
There were no vast fields of wheat.
There was no single agricultural template.
Instead, people developed agricultural systems suited to New Guinea's own plants, soils, climate and landscapes.
This is an important lesson:
There isn't one evolutionary pathway to agriculture.
There are many.
Eastern North America delivers an even bigger surprise
Diamond discusses the eastern United States as an example of an area where agriculture developed later and with a smaller suite of domesticates.
But the archaeological record shows that Indigenous peoples were not simply sitting around waiting for maize to arrive.
They independently domesticated several plants.
These included:
- squash;
- sunflower;
- marshelder;
- chenopod.
Evidence places the domestication of these plants roughly between 5,000 and 3,700 years ago.
Genetic and archaeological evidence has subsequently strengthened the case that eastern North America was indeed an independent centre of plant domestication.
And something even more interesting happened.
These early agriculturalists did not necessarily abandon hunting and gathering.
Instead, domestication appears to have been incorporated into an already successful mixed economy.
At sites such as Riverton, people were cultivating multiple domesticates while continuing to exploit a rich variety of wild resources. Archaeological evidence does not suggest a simple collapse of the old hunter-gatherer system followed by an abrupt conversion to farming.
This is crucial.
The transition wasn't necessarily:
“Hunting failed, therefore farming began.”
It could instead be:
“Life was already working reasonably well, and people gradually added new ways of obtaining food.”
Maybe farming wasn't inevitable—or even desirable
This is perhaps the most important part of the story that gets lost in simplistic accounts of the agricultural revolution.
From our modern perspective, agriculture looks obviously superior.
But for a hunter-gatherer, becoming a farmer could mean:
- harder physical labour;
- more repetitive work;
- dependence on fewer species;
- greater vulnerability to crop failure;
- poorer nutrition in some contexts;
- infectious disease associated with dense settlements;
- greater exposure to famine;
- and eventually greater social inequality.
If wild foods are abundant, why would you voluntarily spend enormous amounts of time cultivating a field?
The answer may be:
You wouldn't.
At least not immediately.
Archaeologist Bruce Smith's work on eastern North America is particularly revealing. Early domestication occurred in resource-rich river valleys where there is little evidence of resource exhaustion or population pressure forcing people into agriculture. Instead, cultivation appears to have been integrated into existing hunting-and-gathering economies.
That changes the question.
Instead of asking:
“Why didn't these people become farmers?”
we should sometimes ask:
“Why would they?”
The environment matters—but so does the human response to it
This is where I think Diamond's argument is simultaneously brilliant and incomplete.
He is absolutely right to insist that biology and geography matter.
You cannot domesticate a plant that does not exist.
And you cannot build a cereal-based agricultural civilization without suitable cereal plants.
But the presence of a potentially domesticable species does not automatically produce agriculture.
Humans have to notice it, value it, manage it, transport it, select it and incorporate it into their social lives.
And those processes are cultural.
The modern picture is therefore less like a one-way causal chain:
Environment → domesticates → agriculture → civilization
and more like a feedback loop:
Environment ↔ people ↔ plants ↔ technology ↔ institutions ↔ landscape
Humans change plants.
Plants change human societies.
Human societies change landscapes.
Those landscapes change which plants thrive.
And the process continues.
Domestication may have been an accident
This is one of the most fascinating insights from modern domestication research.
We often imagine ancient farmers deliberately breeding plants:
“This wheat has bigger seeds. Let's plant it.”
“This plant doesn't shatter its seeds. Let's select it.”
Sometimes selection probably was intentional.
But much of domestication may have been unintended evolution.
Imagine prehistoric people repeatedly harvesting the largest seeds and preferentially transporting them.
Next season, the largest seeds are disproportionately represented in the plants growing around human settlements.
Repeat this for hundreds of generations.
Eventually, the plant population itself changes.
Humans have become an evolutionary force.
The plant has, in effect, domesticated itself in response to the human environment.
Modern reviews of plant domestication emphasize precisely this protracted co-evolution: adaptations of crops often emerged as unintended consequences of human economies rather than as the result of conscious breeding programmes.
This makes domestication look less like invention and more like evolutionary entanglement.
And not every domesticated plant became a global crop
This is another place where Diamond's framework can be misleading if taken too literally.
The successful crop species we see today are survivors of a gigantic evolutionary and cultural experiment.
Many plants were cultivated and then abandoned.
Some were domesticated but disappeared.
Some remained locally important.
Some never became globally important because they were replaced by crops introduced from elsewhere.
In eastern North America, for example, marshelder was once cultivated but disappeared as a major crop.
Why?
Not necessarily because it was biologically inferior.
History intervened.
Once crops such as maize entered new regions, they could dramatically reorganize existing agricultural systems.
So we need to distinguish:
domestication
from
successful agriculture
from
long-term crop survival
from
global economic importance.
These are four different evolutionary filters.
The Amazon makes the distinction even harder
The Amazon provides an especially fascinating challenge to a simple farmer-versus-forager narrative.
Many Amazonian societies cultivated plants extensively, yet numerous important plants remained somewhere between wild and fully domesticated.
Trees and tubers, in particular, can exist along a continuum between wild populations and strongly domesticated crops.
Indigenous peoples also transformed landscapes through practices such as selective enrichment, burning and forest management.
Thus a forest that looks “wild” to an outsider may actually be partly a human-created ecological system.
In other words:
A landscape doesn't have to look like a wheat field to be agricultural.
Research on ancient Amazonia increasingly emphasizes this continuum of plant management, cultivation and domestication.
So what really determines whether agriculture emerges?
The best answer today is probably not “apples” or “Indians.”
It is:
Apples + Indians + landscapes + climate + technology + culture + time + chance
Consider the variables.
1. The biological inventory
Were there plants with:
- large edible seeds?
- high nutritional value?
- predictable reproduction?
- short generation times?
- favourable responses to selection?
- good storage properties?
This is Diamond's strongest point.
2. The animal inventory
Plants weren't the whole story.
Large domesticable mammals could provide:
- meat;
- milk;
- traction;
- transport;
- manure;
- hides;
- wool;
- and, eventually, pathogens that profoundly affected human history.
This is why Diamond immediately follows “Apples or Indians” with “Zebras, Unhappy Marriages, and the Anna Karenina Principle.”
The plant and animal inventories together created radically different possibilities.
3. Climate and seasonality
A plant might be perfectly nutritious but useless as a staple if its growing season, rainfall requirements or storage characteristics are unsuitable.
Climate also determines whether agriculture can be stable from year to year.
4. Human population density
Population pressure may sometimes have encouraged cultivation.
But it clearly isn't a universal explanation.
Some domestication occurred in resource-rich landscapes without obvious signs of population crisis.
5. Existing abundance
This is the paradox.
The better hunting and gathering is, the less attractive farming may be.
A landscape overflowing with fish, shellfish, nuts, tubers and game may delay agricultural intensification.
Agriculture can therefore arise not because the environment is poor, but because particular crops make cultivation sufficiently rewarding.
6. Technology
Agriculture depends on technology.
Not just ploughs.
Knowledge of:
- fire;
- storage;
- irrigation;
- soil management;
- plant propagation;
- harvesting;
- processing;
- grinding;
- cooking;
- fermentation;
- and preservation
can radically alter the usefulness of a plant.
The same species can be almost useless to one society and extremely valuable to another.
7. Cultural preferences
People don't maximize calories.
They eat foods because they are tasty.
Because they are culturally meaningful.
Because they are easy to prepare.
Because they are associated with identity.
Because particular foods are exchanged socially.
Because they are used in ceremonies.
Or simply because people like them.
This sounds trivial.
It isn't.
A purely nutritional model of agriculture misses a huge part of human behaviour.
8. Social organization
Agriculture can create surplus.
But it can also require coordinated labour.
Irrigation, planting, harvesting and storage can favour new forms of cooperation.
And eventually, surplus can be appropriated.
Thus agriculture isn't merely a biological transformation.
It can become a social transformation.
9. Trade and diffusion
The independent invention of agriculture is only half the story.
Once agriculture exists somewhere, crops can move.
People move.
Technologies move.
Ideas move.
Genes move.
And sometimes an existing hunter-gatherer economy adopts a crop without independently domesticating it.
This means that the agricultural map we see today is the product of both independent invention and cultural diffusion.
10. Historical contingency
And finally, there is chance.
Two societies can have remarkably similar environments and yet take different historical paths.
A particular drought.
A migration.
A disease outbreak.
A new trade route.
A particularly useful mutation.
A technological discovery.
A cultural innovation.
Any of these can push a society onto a different trajectory.
Evolution doesn't produce a single inevitable answer.
Neither does human history.
The biggest lesson of “Apples or Indians”
The brilliance of Diamond's question is that it forces us to reject a very old assumption:
Differences in human history do not necessarily originate in differences between human beings.
Sometimes they originate in differences between environments.
That was an enormously important corrective to racist theories of history.
But there is a danger in swinging too far in the opposite direction.
If we say:
“Geography determined everything,”
we replace biological determinism with geographical determinism.
The modern evidence suggests something more interesting.
Human societies were not passive recipients of geography.
They were participants in ecological evolution.
They burned landscapes.
Moved plants.
Protected useful species.
Changed animal populations.
Constructed soils.
Built irrigation systems.
Created wetlands.
Selected seeds.
Transported crops.
And eventually reshaped entire ecosystems.
Humans didn't merely discover agriculture.
Humans and plants co-created agriculture.
Perhaps the real question isn't “Why did people become farmers?”
There is a deeper question hiding underneath Diamond's.
Instead of asking:
Why did some hunter-gatherers become farmers?
we might ask:
Under what circumstances did particular human–plant relationships become self-reinforcing?
Once cultivation began, cultivation could produce more food.
More food could support more people.
More people could mean more cultivation.
More cultivation could select plants for agricultural traits.
Better crops could make cultivation more attractive.
And eventually:
people became dependent on the plants they had themselves transformed.
That is a feedback loop.
Not a single invention.
Not a sudden “Agricultural Revolution.”
An evolutionary process.
From apples to civilizations
This is why the chapter matters far beyond apples.
The plants available to ancient humans helped determine which forms of agriculture were possible.
Agriculture affected population density.
Population density affected settlement.
Settlements affected disease.
Food surpluses supported specialists.
Specialists supported technologies.
Surpluses could support political elites.
States could organize armies.
Armies could conquer neighbouring societies.
And pathogens emerging from dense animal-human populations could become devastating weapons of history.
Diamond's larger argument is therefore not simply:
“Some people had better plants.”
It is:
Small differences in biological environments can, through thousands of years of feedback, become enormous differences in human history.
That is a powerful idea.
But the modern archaeological record adds an equally powerful qualification:
Plants had possibilities, not destinies.
Humans chose, experimented, managed, ignored, exchanged and transformed those possibilities.
And sometimes the plants transformed the humans right back.
The final answer: Apples or Indians?
So, was it the apples or the Indians?
Diamond's answer is:
Mostly the apples—or, more precisely, the entire package of plants and animals available to different human populations.
But the deeper modern answer is:
Neither alone.
The history of agriculture emerged from the interaction of:
biology × environment × climate × human behaviour × culture × technology × demography × social organization × diffusion × chance.
And perhaps that is the most interesting lesson of all.
The first farmer was probably not a person who suddenly looked at a wild plant and thought:
“I shall now invent agriculture.”
There probably wasn't a first farmer.
There were generations of people who gathered plants, returned to productive places, scattered seeds, cleared vegetation, burned landscapes, harvested selectively, experimented, exchanged knowledge—and slowly altered both the plants and themselves.
Agriculture was not invented in a moment.
It evolved.
And perhaps the greatest irony of the story is that the plants we now call domesticated were once wild.
The people who domesticated them were once hunter-gatherers.
And the landscapes we now think of as “natural” were sometimes already being shaped by humans.
The boundary between wild and domestic, nature and culture, farmer and forager turns out to be far blurrier than the conventional story suggests.
So the question is not really:
“Why did the Indians fail to domesticate the apples?”
It is:
“What happens when a species capable of culture begins to evolve together with the species it eats?”
That question takes us from apples—and Indians—to one of the most extraordinary evolutionary experiments in Earth's history:
the domestication of the planet.
