I began this book in a penguin house.
It was Rotterdam, late autumn, and I was standing in the underwater viewing gallery of the Oceanium watching Humboldt penguins swim. I had not gone to the zoo to have an idea. I had gone because my older son, who was three at the time, wanted to see the penguins, and because I was tired of the flat I had been writing in for six months, and because the zoo was fifteen minutes by tram and cost twelve euros. I was not thinking about enclosures. I was not thinking about Homo sapiens. I was thinking about whether the cafe by the flamingos sold decent coffee.
The penguins were swimming in a pool that had been designed to replicate, as closely as a concrete tank in South Holland can replicate, the coastal waters of Peru and Chile. The water was kept at fourteen degrees Celsius. The rocks were textured to approximate the guano-stained surfaces the birds would use for nesting. The lighting cycled to match a Southern Hemisphere photoperiod. The fish were supplemented with vitamins to compensate for the nutritional difference between frozen herring and fresh anchovy. The exhibit had been designed by people who had studied what Humboldt penguins need and had built, within the constraints of the available space and budget, the closest approximation they could manage.
My son pressed his face against the glass. A penguin swam past at eye level, banked, and came back. My son laughed. I watched the penguin, and I watched my son, and I thought — with the sudden clarity that sometimes arrives when you are not trying to think at all — that nobody had done this for him.
Nobody had studied what a three-year-old human needs and designed his environment to provide it. His nursery was a converted office building. His food came from a supply chain that had never been tested against his biology. His daily schedule was structured around his parents' working hours, not around his developmental requirements. The light he woke to was electric. The ground he walked on was flat. The air he breathed was recycled. Every dimension of his enclosure had been designed for the convenience of the institution rather than the biology of the inhabitant.
The penguin had a better-designed habitat than my son.
I have spent the preceding twenty chapters tracing that observation across every dimension of human life, and I want to be careful, here at the end, about what I have and have not said.
I have not said that civilisation is bad. Civilisation produced the antibiotics that saved my younger son's life when he was eighteen months old. Civilisation produced the language I am writing in, the printing press that will reproduce these words, the postal system that will deliver them, and the electric light by which you are probably reading them. The systems described in this book — money, justice, education, media, governance, medicine, housing, work — are not failures. They are achievements. They fed eight billion animals. They connected strangers across oceans. They extended the average human lifespan from thirty-five years to seventy-three. The impulse behind every system was good. I have said this in every chapter because it is true, and because the temptation to read a book like this as an indictment is strong, and the temptation is wrong.
What I have said is that the systems were not designed around the animal. They were designed around the institution, the economy, the supply chain, the schedule, the market. The animal was not the client. The animal was the input. And the consequences — chronic subclinical malnutrition from a food system that feeds but does not nourish, chronic sleep restriction from an economic system that treats rest as waste, chronic loneliness from a social architecture that exceeds the organism's trust capacity, chronic meaninglessness from a work system that serves institutional needs rather than human ones — these consequences are not mysteries. They are exactly what a zoologist would predict if you described the enclosure without naming the species inside it.
I have also not said that I have escaped. I have said, repeatedly and honestly, that I have not. I run on a treadmill three times a week. I eat sandwiches from food courts. My children are in school. I pay rent on an apartment whose construction costs were recovered decades ago. I check my phone before I check on my sons. I am writing this book indoors, under electric light, at a desk, in a city of 125,000 people of whom I know thirty by name. I am the fish who noticed the water. Noticing is not escaping. It is the first step, and I am on it, and I am not ahead of you.
I should say something about the writing of this book, because the process changed me in ways I did not anticipate and would not have chosen.
The research took two years. The first six months were spent in intellectual euphoria — the framework was clarifying, the evidence was everywhere, and I had the convert's certainty that I was seeing something nobody else had seen. This is precisely the state the book warns against in Chapter 14: the organism that mistakes a new narrative for the truth. I was building a model and falling in love with it. The model was, at that stage, too neat.
The justice chapter broke the neatness. I had assumed I would write about criminal justice as a straightforward enclosure failure — punishment as the wrong response to a diagnostic signal. The Scandinavian data supported it. The recidivism numbers supported it. What I had not anticipated was how the argument would feel when applied to specific cases. I sat with files from the Truth and Reconciliation Commission in South Africa for three weeks, and there were mornings I could not write — not because I lacked evidence but because the framework felt like an insult to the people who had suffered. The animal that committed those acts was responding to environmental pressures. Yes. And? Does that comfort anyone?
I left the chapter for two months. When I came back, I wrote about the limits of the framework — that a zoological lens is a lens, not a verdict, and that some human experiences exceed what any analytical frame can hold. That passage arrived because the research demanded it. The book holds its framework as lightly as it can because I learned, during the writing, that holding it tightly was dishonesty.
I was wrong about several things. My draft of Chapter 8 treated formal schooling as an unambiguous failure. The research corrected me — the Sudbury data is promising but thin, Sugata Mitra's experiments have been critiqued, and what I ended up writing was more honest and less satisfying: the system fails the animal in measurable ways, alternatives exist, none is the answer. I was wrong about technology too. The early drafts treated screens as a straightforward toxin. Andrew Przybylski at the Oxford Internet Institute showed me the relationship is not linear, and that the organism is drawn to the screen because the screen provides, in degraded form, things the enclosure fails to provide — contact, novelty, play. The screen is a symptom. I wrote it as a cause. I wanted clean answers. The clean answers were not true.
What did not change was the central observation. Every system, every institution, every structure — designed for something other than the animal. The consistency was the finding. The details varied. The pattern did not.
The method I have proposed is not new. It is not revolutionary. It is not even particularly interesting. It is what zookeepers do.
You look at the animal. You look at what the animal needs — not what it wants, not what the institution requires, not what the market demands, but what the organism's biology requires to function at the level it evolved to function at. You look at the enclosure. You identify the gaps between what the animal needs and what the enclosure provides. And you adjust.
You do not demolish the enclosure. You do not blame the animal. You do not write a manifesto or start a revolution or wait for a leader to fix it for you. You adjust. You modify. You test. You observe. You adjust again. This is the method. It is iterative, unglamorous, and slow. It is also the single most effective approach to animal welfare ever developed, and it has never been systematically applied to the species that invented it.
The adjustments are not utopian. They are specific. They have been described across the chapters of Part Four, and they amount to a small set of design principles: scale the group to the animal's trust capacity. Let the organism learn the way organisms learn — through participation, not instruction. Address all eight dimensions of welfare, not just the ones the institution finds convenient. Include death, because an organism that has forgotten its finitude has no reason to change anything.
These principles are not theoretical. They are already in practice.
The Danish cohousing communities — bofaellesskaber — are perhaps the clearest. Fifty to a hundred residents share a common house while maintaining private dwellings. Group size is kept where every resident knows every other by name. Meals are shared on a rotating roster. The architecture uses courtyards, not corridors, so casual encounters happen by default. It is not a commune. It is an enclosure redesigned around the animal's social architecture: trust-scaled groups, shared labour, and a physical layout that makes connection the path of least resistance.
The Mondragon cooperatives in the Basque Country — over eighty cooperatives, eighty thousand workers, operating since 1956 — run on one-member-one-vote governance with salary ratios capped at six to one. Workers rotate through management. Decisions are made by the people who do the work. The animal retains agency over its labour, and each member's contribution is visible to the group.
The Buurtzorg nursing model in the Netherlands disbanded the management layer and replaced it with self-organising teams of ten to twelve nurses per neighbourhood. No middle management. Each team makes its own clinical decisions. The results, documented by Ernst & Young in 2009: highest patient satisfaction in Dutch healthcare, staff satisfaction roughly doubled, costs down forty percent. They changed the size of the group, and the animal performed better on every metric.
The Sudbury Valley School in Massachusetts — no curriculum, no grades, no compulsory activities since 1968 — demonstrates that the organism can learn without being taught, that the Mastery and Cub dimensions can be satisfied when the enclosure provides resources, safety, and time, and then steps back.
The hospice movement, begun by Cicely Saunders in London in 1967, redesigned the enclosure for dying — introducing the idea that the dying organism needs honest information, the presence of people who matter, and the opportunity to close the narrative of a life. It addresses the Monk dimension at the moment meaning is most required.
The citizens' assemblies — Ireland's, France's, Canada's — share a design: randomly selected citizens, given time and evidence to deliberate face to face. The results have been more nuanced and more broadly supported than those of professional legislatures on the same questions. The animal deliberates well in groups where it can see every face.
And there are others, smaller, happening already. The transition towns — Totnes in 2006, now over a thousand worldwide — where communities map their dependencies and design local alternatives. The community kitchens, from the comedores populares of Lima to pay-what-you-can cafes in British cities, where eating is returned to its species-typical context: shared, prepared together, the service and herd dimensions satisfied in the same hour. The repair cafes — Amsterdam, 2009, now over two thousand locations — where volunteers fix things for free because the mastery drive and the service drive are both satisfied by the act. The tool libraries, seed banks, time banks, men's sheds, community gardens. None of these will appear in an economics textbook. All of them are enclosure adjustments.
They share a common architecture. The group is small. The contribution is visible. The organism has agency. The activity addresses more than one dimension simultaneously. These are not coincidences. They are design features, and they recur because the animal's specifications recur. We are the same species in Totnes as in Lima. The enclosure details differ. The biology does not.
The book has been a diagnostic exercise. The question it leaves — the one you are entitled to ask — is: what do I do with this?
I am not going to answer that. A zookeeper does not prescribe a standard enclosure for all animals of a species. A zookeeper assesses the individual animal, in its specific enclosure, and identifies the specific gaps. The person best positioned to assess your enclosure is not me. It is you.
What I can offer is the method itself, applied inward.
Take the eight life areas. Be honest — not aspirationally honest, but diagnostically honest, the way a veterinarian is honest when assessing a limping animal. Your sleep is adequate or it is not. Your social bonds sustain you or they do not. Your work engages your mastery drive or it does not. You have a sense of purpose or you are borrowing one from an institution that provided it as a substitute. Rate each area against what the organism requires. The chapters have described the requirements. The gap between them and your reality is the gap.
Find the area with the largest gap. Not the most fashionable one, not the one that would make the best social media post. The largest. That is where the adjustment will have the highest return.
Make one adjustment. One. Not a transformation. A single modification to close the gap by a measurable fraction. If the gap is sleep, it might be a blackout curtain. If the gap is connection, one recurring commitment to see one person, in person, at a fixed time. If the gap is mastery, thirty minutes a week doing something difficult that has nothing to do with your job. The adjustment should be small enough to survive contact with the rest of the enclosure — because the rest of the enclosure will resist it, as enclosures do.
Then observe. Did the adjustment hold? Did the organism respond? Did the limping ease, even slightly? Adjust again.
This is not a self-help programme. It is what zookeepers do, applied inward. The difference between this and the thousand books that promise transformation is that this promises nothing except the method, and the method is slow, and there is no moment of breakthrough — only the gradual closing of a gap that should not have been there in the first place.
I do not know if these ideas will work at the scale required. The scale problem — the central argument of Part Three — is real, and it may be that some of what I have described can only function within the neurological limits of Dunbar's number. It may be that civilisation at the scale of eight billion is simply too large for the animal that built it, and that the chronic stress, chronic loneliness, and chronic meaninglessness documented in these pages are not bugs but features of a system that has exceeded its designer's specifications.
I want to be honest about my uncertainty here. I do not know whether the adjustments can propagate. I do not know whether a species that has spent ten thousand years scaling upward — band to tribe to chiefdom to state to empire to global market — can reverse direction without losing what the scaling achieved. The antibiotics. The literacy. The container ships full of grain. Whether you can have the antibiotics without the alienation, the literacy without the fluorescent rows — I do not know. I suspect yes. But suspicion is not evidence.
What I do know is that the question of scale may be the wrong question. The enclosure operates at scale and therefore frames every alternative as a scaling problem. Can this work for eight billion? The honest answer is that nothing works for eight billion. The systems we have now do not work for eight billion people — they feed them, which is different, and even that is distributed so unevenly that a billion are malnourished while another billion are obese. The scale is already failing. The question is not whether an alternative can match it. The question is whether the organism, in its specific enclosure, on this specific Tuesday, can begin.
Every zoo exhibit starts with one animal. Every enclosure redesign begins with a single assessment. The method does not require a policy paper or an election. It requires one animal, looking honestly at the distance between what it needs and what it has, and closing the gap by one adjustment. That is not a theory of civilisational change. It is smaller than that. It is a zookeeper, walking into an enclosure, doing the only thing a zookeeper can do — looking at the animal in front of them and beginning.
My boys are ten and six now. The older one reads books I have not read and asks questions I cannot answer, which is exactly what the organism is supposed to do when the Mastery dimension is functioning. The younger one builds things — elaborate, unstable structures out of cardboard and tape and ambition — and when they collapse, he rebuilds them differently. He does not yet know that this is the method. He does not need to know. The organism discovers iteration before it discovers the word for it.
I do not know what enclosure they will inherit. What I hope for them is not utopia — I have spent twenty chapters arguing that utopia is a fiction that prevents adjustment by promising perfection. What I hope is smaller. I hope that someone in the chain of decisions that will shape their adult lives — the architects, the employers, the legislators, the people who design the systems my sons will live inside — will pause and ask the question the penguin house taught me to ask. Not: what does the institution require? Not: what does the market demand? But: what does the animal need?
And then adjust.
The woman in the white coat at the Hawkesbury campus is still testing eucalyptus leaves. Forty-nine compounds per leaf, for a thirteen-kilogram marsupial, because the institution that houses the animal takes the animal seriously. The seriousness is not sentimental. It is methodological. You test because you cannot see. You measure because you cannot guess. You adjust because the organism cannot adjust itself — it is inside the enclosure, living the only life it has been given, doing the best it can with what the system provides.
You are inside the enclosure, living the only life you have been given, doing the best you can with what the system provides.
The animal can be helped. The enclosure can be redesigned. The method exists. The evidence exists. The organism's needs are known. What remains is not knowledge. What remains is not theory. What remains is not even courage, though courage helps.
What remains is the decision to begin.
Leiden, 2024