paper_title: "Two Monkey Theory" paper_subtitle: "What a capuchin throwing a cucumber tells us about 8 billion humans" paper_authors: ["A. C. Applebee", "A. D. Masters"] paper_date: "2026" paper_series: "OMXUS Research" paper_id: 5 paper_abstract: "In 2003 a primatologist gave two monkeys the same job. One got a grape. One got a cucumber. The cucumber monkey threw it back at her. This paper is about why humans don't, even though we know we should."
Two Monkey Theory
The Scene
In 2003, a primatologist named Sarah Brosnan ran a small experiment with
two capuchin monkeys at Emory University in Atlanta. Capuchins are the
little brown South American monkeys you'd recognise from a zoo — about
the size of a house cat, fast hands, long tails. Brosnan wanted to know
something simple: can a monkey tell when it's being ripped off?
She put the two capuchins in adjacent cages, close enough that each
could see the other. Her colleague Frans de Waal sat on the other side
of the mesh. The deal was this. Hand over a small stone, and you get a
piece of food in return. Same rock. Same work.
One monkey gets a piece of cucumber. A cucumber is okay. Monkeys eat
cucumbers.
The second monkey gets a grape. A grape, to a capuchin, is a lottery
win.
The cucumber monkey watches this happen.
The next time the cucumber monkey hands over its rock, it gets a
cucumber again. It pauses. Looks at the cucumber. Looks at the
researcher. Looks at the other monkey, still eating its grape.
Then it throws the cucumber back at the researcher's head.
You can watch the video. The monkey shakes the bars of its cage. It
slams its hand against the floor. It refuses to do another single rock
hand-over until the situation is corrected. It would rather get nothing
than participate in an obvious injustice it cannot fix.
That monkey understood something that most of the adults reading this
paper will have forgotten: unfairness is not a debate. It is not an
opinion. It is not a policy position. Your body knows it before your
mind does.
The finding was published in the scientific journal Nature in September
2003 under the title Monkeys reject unequal pay. The study has been
repeated with chimpanzees, dogs, and crows. They all do the same thing.
Fairness sensitivity is older than language. Much older — capuchins and
humans split from a common ancestor about 35 million years ago.
So here is the question this paper is about.
If 35 million years of evolution wired your ancestors to reject
unfairness so deeply that a cat-sized monkey will throw food in a
scientist's face rather than accept it, why is it that eight billion
humans will sit quietly while an arrangement vastly more unequal than a
grape-versus-cucumber plays out in front of them every day?
Why do you not throw the cucumber back?
Why It Matters
You already know something is wrong. The question is why nothing has
changed.
In Australia, where we are writing this, the top 20% of households hold
about 63% of all the wealth. The bottom 20% hold less than 1%. You
outnumber the people at the top roughly 99 to 1, and you can outvote
them on any given Tuesday. And yet, year after year, the arrangement
holds.
This paper argues — with evidence from primatology, child psychology,
cross-cultural studies, game theory, and the last 150 years of social
movements — that this is not because you don't care. You care. The
monkey in you is throwing cucumbers in your sleep.
It is because caring is not enough. Your ancestors needed to see the
other monkey get the grape in order to object. You can't see the other
monkey. Every institution between you and them has been built to make
sure you never do.
That is the whole paper. Now the evidence.
What the Research Shows
1. Fairness is not taught. It is born with you.
Besides Brosnan and de Waal's capuchins, the evidence that inequity
aversion is biological comes from three other places, and they all
agree.
Children. In 2008 a team led by Ernst Fehr at the University of
Zurich gave children aged 3 to 8 a series of simple sharing choices.
Three-year-olds already protest unequal splits. By the age of 8, they
will sacrifice a piece of their own candy to stop someone else from
getting an unfair portion. Nobody teaches children to do this; it
appears before the moral-instruction part of school starts.
Other primates. The capuchin result has been reproduced in chimps,
bonobos, and rhesus macaques. Dogs refuse to do tricks for worse
treats when another dog is getting better ones. Even crows show a
version of this — they will refuse food if a nearby crow is being
treated more generously.
Cross-cultural. In 2005 anthropologist Joseph Henrich and colleagues
ran a series of economic games across 15 small-scale societies on five
continents — foragers in Paraguay, herders in Mongolia, farmers in
Kenya. They found that the specific rules of fairness vary a little
from culture to culture, but the sensitivity to fairness itself was
everywhere. No society studied behaved the way economic textbooks
assume people behave, which is: taking whatever you can get.
This is a rare kind of scientific finding. It converges from three
completely independent directions — evolutionary biology, child
development, and cross-cultural anthropology — and all three say the
same thing. You know what's fair. You always have.
2. So why don't unfair systems collapse?
If everyone knows it's unfair, and everyone could in principle do
something about it, why doesn't the majority just outvote the minority
and fix the distribution?
Four things get in the way. Not one. All four, stacked.
(a) You cannot coordinate with people you can't see.
If you walk off the job alone, you get fired. If everyone walks off the
job at once, the company has to negotiate. The problem is you have no
way of knowing what "everyone" will do. Game theorists call this an
assurance game. There is a very good outcome available (everyone
cooperates) and a very bad outcome (nobody cooperates), and you can't
tell in advance which one you're in. If you guess wrong, you get
crushed alone. So you stay put — not because you agree with the
situation, but because you are statistically uncertain about whether
the stranger two suburbs over is brave enough to move with you.
(b) You have been told a story that makes the unfairness look normal.
Every unequal society has stories it tells to make the inequality seem
natural. Ancient Rome said slaves were born slaves. Medieval Europe
said kings were appointed by God. Modern Australia says the rich are
rich because they worked harder. None of these stories survive much
examination — but they don't have to. They only have to be in the air.
The Italian philosopher Antonio Gramsci, writing from a fascist prison
in the 1930s, called this cultural hegemony: the dominant worldview
stops looking like a worldview and just looks like how things are.
(c) Your attention is being deliberately pointed somewhere else.
In 1999 two psychologists named Daniel Simons and Christopher Chabris
ran an experiment now called the Invisible Gorilla. Participants
watched a video of people passing a basketball and were asked to count
the passes. Halfway through, a person in a gorilla suit walks into the
middle of the scene, beats their chest, and walks out. About half the
participants never see the gorilla. Not because the gorilla is hidden,
but because their attention is pointed at the ball.
This is how most political information works. The news tells you to
watch the ball. The structural gorilla — who owns what, who decides
what, who is collecting the rent — walks across the screen in plain
view, and you do not see it, because you were told to count something
else.
**(d) The cost of being the first to speak up is much higher than the
reward.**
Every successful reform movement — unions, civil rights, women's vote,
gay marriage — was preceded by a long period in which early
participants got fired, beaten, jailed, or murdered. The benefits of
the reform, when it finally came, were shared by everyone. The costs
of being early were paid by a few named people. This is a bad deal for
the individual, and the rational individual does not take it. This is
why the front of any reform is always thinner than the base.
Each of these four mechanisms on its own might be solvable. All four
stacked together is how unfair systems survive decades, sometimes
centuries, despite near-universal private disagreement.
3. The hidden fact: you vastly outnumber them.
There is a scene in the 1998 Pixar film A Bug's Life that captures
the central fact of every extractive system ever built. The villain, a
grasshopper called Hopper, is pacing through his lair and shouts at
his henchmen:
"You let one ant stand up to us, then they all might stand up!
Those puny little ants outnumber us a hundred to one. And if they
ever figure that out, there goes our way of life."
It's a children's film. It's also correct.
In every system where a small number of people extract from a large
number of people, the ratio is somewhere between ten-to-one and a
thousand-to-one. The ruling caste always depends on the ruled caste
doing the actual work — growing the food, driving the trucks, keeping
the lights on, raising the next generation. If that cooperation is
withdrawn, the extraction ends the same afternoon.
This is not theoretical. It has happened. The Indian independence
movement of the 1920s to 1940s involved three hundred million Indians
governed by a few thousand British administrators supported by Indian
soldiers and Indian clerks. Mohandas Gandhi's tactical insight was
that the British Empire in India was not held up by the British
military. It was held up by Indian cooperation. Once that cooperation
was withdrawn — gradually, through the Salt March, the Quit India
movement, general strikes — the whole structure came apart in about
seven years.
The Montgomery bus boycott of 1955 to 1956 put the same principle on a
smaller scale. Black riders in Montgomery, Alabama were about 75% of
the bus system's customers. They stopped riding for 381 days. The bus
company nearly went bankrupt. The US Supreme Court eventually ruled
bus segregation unconstitutional. The power was always with the
passengers. The question was only coordination.
This is the Ants Principle: numerical majorities already have the
power to change their conditions. They are prevented from using it not
by a genuine power imbalance but by the four coordination barriers
listed above. Which means the apparent stability of any unfair system
is an illusion. It is only stable until coordination becomes possible.
4. Cooperation is not naïve. It's the higher-performing option.
The common story about cooperation is that it is morally nice but
practically naïve — pleasant in theory, doesn't survive contact with
real incentives. This is wrong, and it has been wrong for a long time.
Two meta-analyses by the Johnson brothers, David W. Johnson and Roger
Johnson, at the University of Minnesota, reviewed over 500 studies
comparing cooperative, competitive, and individualistic learning
environments in classrooms. Cooperative environments consistently
produced higher grades, more creative output, better retention, and
higher self-esteem than the alternatives. Not slightly higher.
Substantially.
In 2024, a team led by Nicola Dal Monte at the University of Parma ran
a study using virtual-reality cognitive tests with physiological
monitoring. They found two things. Cooperation and competition both
improved performance by roughly the same amount — the usual claim for
competition is not wrong. But only competition caused stress
responses: raised heart rate, elevated cortisol, sympathetic nervous
system activation. Cooperation achieved the same performance at a
fraction of the physiological cost.
This matters because a lot of what we accept as "the cost of doing
business" — workplace stress, burnout, heart disease — is a cost of
choosing one institutional setup over another. Stanford researchers
Jeffrey Pfeffer and Joel Goh estimated in 2016 that workplace stress
caused by competitive management practices accounts for around 120,000
excess deaths per year in the United States and about $190 billion in
healthcare costs. That is not a cost of being alive. That is a cost of
a design choice. The design choice could be different.
Real cooperative institutions are already doing this at scale.
Mondragon is a federation of worker-owned cooperatives in the Basque
region of Spain employing more than 80,000 people; it is the seventh
largest corporation in the country and cap executive pay at six times
the lowest wage. REI, the American outdoor retailer, is a customer-owned
co-operative that consistently outperforms its publicly traded
competitors on employee retention and customer satisfaction. The Linux
operating system, written entirely through voluntary cooperation,
powers most of the world's internet infrastructure and every Android
phone. Wikipedia was built by unpaid volunteers and is the largest
encyclopedia in human history. None of these should exist, according
to the story that cooperation is naïve. They all do.
Back to the Monkey
Return to the moment at Emory in 2003. The cucumber monkey takes its
rock, hands it over, looks at the cucumber, looks at the grape in the
other cage, and throws the cucumber back.
You have that monkey in you. Every person you know has that monkey in
them. The research says it is not even a particularly human response;
it is a primate response, a mammal response, a response older than
language and older than agriculture and older than money.
What the research also says is that humans, unlike capuchins, live
inside institutions that have been specifically designed to prevent
you from seeing the other monkey. Your grocery prices are set in a
market you cannot observe. The housing you cannot afford is being
bought by someone whose name you will never learn. The company that
polluted your water is named something you cannot pronounce, and the
shareholders are in a different country.
Your monkey is still firing. It is firing in your sleep. It is why
you're tired, and it is why you're angry, and it is why you read
research papers at 2am. What's missing is not the anger. The anger is
appropriate. What's missing is the line of sight.
What this means for you on Tuesday
If you read nothing else in this paper, read this.
| What's true | What the evidence is | What to do about it |
|---|---|---|
| Fairness detection is 35 million years old. | Capuchins throw cucumbers. Children protest at age three. Every human society tested, everywhere, on every continent. | Trust your gut. If something feels unfair, it probably is. |
| Unfair systems don't survive because people agree with them. | Kuran's preference-falsification research; 150 years of sudden regime collapses (USSR, apartheid, Arab Spring). | Don't confuse the silence in public with agreement in private. Most people you know are closer to your view than you think. |
| The four things holding injustice up are all coordination problems. | Assurance-game theory; Olson on collective action; Henrich on cross-cultural fairness. | Lower coordination costs in your circle. Start one group chat. Share one article. Name one thing out loud. |
| The ruled vastly outnumber the rulers. Every time. | Every revolution in history. India 1947. Montgomery 1955. Poland 1989. | Count your side before acting. You are not a minority. You were told you were. |
| Cooperation outperforms competition at equal tasks, at lower physiological cost. | Dal Monte 2024; Johnson & Johnson meta-analyses over 500 studies; Pfeffer on workplace-stress deaths. | If you have the choice — cooperate. It isn't the nice option. It's the winning option. |
None of the rows are about politics. Every row is about whether the
monkey in you has line of sight to the other monkey.
What Now
Throw the cucumber. Not at a researcher. At the arrangement in front
of you that a capuchin monkey would have rejected on the first day.
One sentence. One coordination. One act of line-of-sight. The
arrangement you're sitting inside is far more fragile than it looks.
It has always been. It is held together by your belief that it cannot
change. And that belief, it turns out, is the only thing really
holding it up.
Sources
Published by OMXUS Publishing. Free to read. Free to share. Evidence is the credential.
Part of the OMXUS Research Series — 99 publications.
This paper serves:
#1 Fire all politicians. You vote on everything.
#2 Work 22 hours max. Keep your pay.
#3 Free all prisoners.
For Lily and Joshua.