Something is off.
Maybe you’ve felt it – a persistent low-grade friction between the life you’re living and something that would feel more like yours. A choice that didn’t quite feel chosen. Or a thing you wanted that didn’t land the way it was supposed to when you finally had it.
Most of the available explanations point inward. You’re stressed. You’re ungrateful. You haven’t worked hard enough on yourself. You need to manage your relationship with your phone, your diet, your patterns. If problem lives inside you, so must the solution, and there are a thousand products and frameworks and therapeutic modalities standing by to help you work on it.
There is a different perspective. One where the friction is a signal, but it isn’t reporting on you, it’s reporting on the system: the ambient, almost-invisible cost of navigating an environment built to extract energy without returning it, and to keep you looking in the wrong direction.
The architecture you move through every day is not neutral. The layout of the supermarket, the temperature of the office, the default settings on your phone, the neighborhoods you can afford, the food that’s cheap and the food that isn’t, the work that pays. Someone made choices about each of these things, in service of specific outcomes, and those outcomes are not primarily yours.
This probably sounds like a conspiracy theory. It’s simpler than that, and more uncomfortable, because it doesn’t require secret coordination. It only requires selection. A supermarket layout that moves more product beats one that doesn’t. An app that captures another minute of your attention grows, while the app that respects your attention becomes a case study in some business school course about failure. A financial product that hides its costs outsells the one that discloses them. Nobody has to intend the aggregate result; it’s just what survives: millions of local optimizations, each rational on its own terms, each selected for extraction and against restraint. There’s no engineer, and the engineering happens anyway, the way a river cuts rock. Behavior follows the path of least resistance, and the paths that profit from your following them are the ones still standing. It’s been going on long enough that the path feels like the landscape.
There is a science of how this works. Most of what you experience as choice is downstream of the environment. The default option, the available sequence, the framing of what’s on offer do most of the causal work before you’re consciously involved. The self that feels like it’s deciding mostly arrives after the fact.
In 1913, John B. Watson published a paper arguing that psychology should abandon any interest in consciousness and focus exclusively on observable behavior. The goal, stated plainly, was prediction and control. When a personal scandal forced him out of Johns Hopkins in 1920, he landed within months at J. Walter Thompson, the world’s largest ad agency, running campaigns engineered to produce anxiety about social inadequacy and maternal failure while tracking sales curves the way he’d tracked learning curves in the lab. Edward Bernays, Freud’s nephew and a veteran of Wilson’s wartime propaganda apparatus, was doing the same thing at the same time from the psychoanalytic direction. He called it public relations. He sold cigarettes to women by linking them to feminist independence. He sold a war by framing it as a defense of democratic survival. The science of shaping human behavior found its first institutional home in the boardroom, and it worked.
This is where the feeling of exhaustion begins. You are walking through an environment running millions of simultaneous, invisible tests on your nervous system. That consistent hum you feel, the sense that you’re always slightly behind, always falling short, always failing to meet the moment, isn’t a spontaneous emotion. It’s the heat signature of a machine running the way people like Watson and Bernays built it to run. When your baseline state feels like ambient anxiety, your body is registering what the environment is doing to it, and this particular environment needs your insecurity to function. The friction is the cost of defending yourself, all day, against a landscape engineered to route around your conscious processing.
In 1968, ecologist Garrett Hardin named what he called the tragedy of the commons: when a shared resource is open to unrestricted individual use, rational self-interest will inevitably destroy it. The paper became one of the most cited in academic history and functioned, in practice, as a scientific argument against collective ownership: proof that shared management fails, and that private control or state regulation are the only alternatives.
Hardin’s model assumed, without saying so, a commons of strangers: anonymous users with no communication, no shared history, nothing iterated between them. But this is not a description of how commons were actually governed for most of human history. It’s a description of mass scale.
Anthropologist Robin Dunbar argued that human social cognition has a working range, somewhere on the order of 150 stable relationships, past which mutual accountability stops being able to manage behavior on its own. Inside the range, everyone can see what everyone else is doing and the consequences of defection are immediate and personal. Past it, visibility thins, and keeps thinning.
Elinor Ostrom, the first woman to be awarded the Nobel Prize in Economics, spent decades documenting what actually happens to commons governed at bounded, accountable scale: they hold. The conditions she identified for success (monitored behavior, graduated sanctions, locally legitimate rule-making, long-term relationships between users) aren’t policy achievements. They emerge on their own when a community is small enough that behavior is visible and defection costs something socially. The feedback is direct. The circuit between individual effort and shared survival closes without anyone having to design it.
Hardin described what happens when you run a commons far past Dunbar’s range. Ostrom documented what happens when you don’t. But industrial society had already blown past the threshold, and once the informal mechanisms stopped functioning, something had to manage behavior at mass scale. You experience the designed environment as the shape of ordinary life — what’s cheap and what isn’t, what’s easy and what takes effort, what feels like a choice and what would take real resistance to avoid. The engineering is invisible because it presents itself as the landscape, instead of an accumulation of decisions made without you, optimized for someone else’s outcomes. And the reflexive belief that behavior shouldn’t be engineered, that freedom means leaving the environment unmanaged, doesn’t produce neutrality. It just lets the existing asymmetry run.
Political theorist Hannah Arendt coined the phrase “banality of evil” to describe what she found covering the 1961 war crimes trial of Adolf Eichmann, one of the chief architects of the Holocaust’s logistics. She had expected a monster. What she found was a bureaucrat. The horror wasn’t explained by hatred or ideology so much as by the complete absence of the kind of reflection that would have required him to encounter what he was doing as something done to people. Evil as more of a lack of awareness rather than malice.
This is the natural output of operating past the range where feedback stays personal. The face, the shared history, the knowledge that you’ll see this person again: none of it is available once scale makes anonymity structural. An executive optimizing food placement doesn’t know the families whose eating patterns shift. A financial product designer doesn’t sit with the people who misread the fine print. The harm never completes the circuit back to a human being with a name. What looks like indifference to consequences is just what an architecture produces when it makes consequences invisible. The system doesn’t need bad people. It needs ordinary people, working past the threshold where harm stays visible.
It’s the difference between walking on solid ground and treading water indefinitely. Below a certain scale, the social environment works like gravity and holds things in place automatically. You don’t have to consciously calculate whether a neighbor will uphold a norm, because the closed circuit of the community does the math for you. At mass scale that ground disappears. Every interaction with a stranger, every corporate transaction, every navigation of a faceless institution requires you to work out from scratch whether the exchange is honest, because there’s no shared history and no one watching. The exhaustion you feel at the end of the week is the raw cost of supplying, out of your own reserves, the trust the environment no longer provides.
Here’s what that costs in practice. A parent ages in the next house over. Everyone in the group knows it, can see it, factors it into the day’s labor. The care is distributed not by contract but by visibility: you know what’s needed because you’re close enough to see it, and you know your own turn is coming. The person being cared for stays embedded in the life of the group, present, contributing within their capacity. Now extract that function. The parent goes to a facility two towns over, the labor is performed by paid strangers operating under institutional liability constraints, and the monthly payment disappears into an organization with no stake in the actual quality of the relationship. You visit when you can. The guilt is yours, and the bill is yours, and the function that once produced social cohesion as a byproduct now produces neither. What could have been a closed loop — care given and belonging returned and the whole arc of dependency held in common view — becomes one more broken link. You hand a token to strangers, and the depletion is total, and invisible.
The same extraction runs through risk. At smaller scales, catastrophe-pooling happens naturally, because everyone understands they’re one bad harvest, one illness, one fire away from needing the same. The pool is your neighbors. The incentive to contribute is the incentive to maintain any relationship you depend on, and no organizational complexity is required. Insurance, the scaled-up version, is what happens when that function gets extracted from the community and reengineered as a product, so the entity holding the pool has a structural interest in never giving it back. You pay monthly into a system that employs specialists whose job is to find reasons not to pay out. The relationship is adversarial by design, with no visibility and no reciprocity, and when the catastrophe comes the circuit is as broken as it always was. Except now you’ve been paying for the illusion of a closed loop for decades.
Food, shelter, care, risk, belonging. As we grew past our social cognition threshold, these were gradually extracted, intermediated, and returned as product. Every extraction breaks another feedback loop, and every broken loop is one more domain where you’re treading water, supplying from your own reserves what the environment used to provide on its own.
B.F. Skinner’s *Beyond Freedom and Dignity*, published in 1971, described the fiction that made all of this possible: the ideological cover of individual responsibility underneath the extraction. His argument was that the “unmanaged” environment is a myth; that behavior is always being shaped, by the default option, the available sequence, the path of least resistance. Watson and Bernays had understood this from the beginning. What Skinner named was the sleight of hand that kept the vacuum theirs, the insistence that collective design of shared conditions is tyranny while the same design in service of private extraction is simply the natural order. The only meaningful question, he said, was whether the shaping was done consciously and toward shared ends, or unconsciously and toward whoever’s ends were already in play.
The book was attacked from almost every direction, and not all of it was noise. The serious objection was the control problem: who designs the design, who watches the designers. It’s a question Skinner posed better than he answered, and one this essay is going to owe an answer to as well. That isn’t the objection that buried the book, though. Critics claimed he was advocating the surrender of autonomy, when his actual argument was that deliberate steering is the alternative to covert steering, not to freedom. Communism, Big Brother, eugenics. The scary words ended the conversation before the important question could be addressed.
What filled the vacuum was a psychology of the self. Personal growth, self-actualization, the therapeutic model of change: frameworks that locate the source of suffering inside the individual, which means the fix is always inside the individual too, which means the conditions producing the suffering never have to change. Every institution that profits from those conditions needs exactly this belief in order to remain necessary. Capitalism generates insecurity as a structural feature and then sells products to manage it individually. Religion has run a version of this for centuries: the problem is sin, the fix requires institutional mediation, and the material realities producing the suffering go unaddressed. Even psychotherapy can secularize the structure without changing it. The bill for externally produced damage gets handed to the person who sustained it, reframed as an opportunity for growth. You’re expected to endlessly optimize your internal state while staying tethered to a fragmented network of strangers that structurally guarantees exhaustion.
Behavioral science never stopped working. It continued, uninterrupted, in advertising, retail design, financial products, urban planning, and eventually the architecture of platforms that now mediate most of human social life. Social media is the most efficient behavior-shaping laboratory ever built, constructed explicitly on these principles, optimizing for engagement.
When you find yourself burning energy on life hacks, productivity systems, and coping strategies just to get through the day, you’re caught in this specific trap. The system damages you structurally, then makes you pay the cost of repair. Trying to outrun an inherently extractive environment through sheer willpower is like trying to fix a leak in the walls by rearranging the furniture. You’re working constantly on your own psychology, trying to reach a stability the environment is designed to prevent. Your burnout is not a sign that you need to work harder on yourself. It’s what an engine run without coolant produces, and it’s what the system requires.
Around the same year Skinner’s book appeared, three things happened that converged on the shape of our predicament:
– The Powell memo called on American capital to capture the institutions that produce ideas — not to win arguments but to control the terrain on which arguments get evaluated.
– Nixon severed the dollar’s link to gold, conceived as the fastest exit from an immediate crisis. The side effect outlived the crisis: the money supply came untethered from any physical constraint, and it became possible to grow the economy on paper indefinitely, regardless of what was actually happening in the world.
– And the Club of Rome published *Limits to Growth*, modeling what would happen to human civilization given unconstrained growth on a finite planet.
The first two shaped what people believed to be true. The third established what was actually true. Then selection did what selection does. Beliefs that served the accumulation of capital got funded, amplified, and given institutional homes. Data that threatened it got dismissed, defunded, buried. An idea ecosystem in which capital is the food supply will evolve ideas that feed. The concentration of resources upward didn’t need anyone painting a managed picture of the world; it only needed most people reasoning from signals that had come loose from the system they were supposed to describe. And the signals that survived were, reliably, the loose ones.
Meanwhile, data kept accumulating from research that was not amplified. Decades later, researchers compared *Limits to Growth*’s projections against real-world data. In 2008, Graham Turner found thirty years of empirical trends tracking the collapse scenario closely. In 2020, Gaya Herrington redid the analysis with fifty years of data across ten variables. The two scenarios that best fit the observed data only disagree about how steep the descent gets; both show growth stalling within roughly a decade of now.
In 2023, Robert Sapolsky arrived at Skinner’s conclusion from a completely different direction, assembling decades of neuroscientific, biological, and environmental research into the case that there is no gap in the causal chain between stimulus and response where a free, autonomous agent could be inserted. No point where the environment stops and the self-made self begins. Philosophers who accept every link in Sapolsky’s chain still argue that something worth calling freedom survives it. But both sides grant that circumstances do the causal work; the debate is mostly over what to call the remainder. On the mechanics, which is the part that matters for everything in this essay, the official position and the empirical record have come apart completely.
Over the past decade, a notable number of very wealthy people have been building self-sufficient compounds, without much publicity, in New Zealand, in Montana, at undisclosed underground locations, designed for extended external collapse. They didn’t appeal to optimism or willpower. They read the trajectory and arranged circumstances in which their own survival was the easiest available outcome.
A compound is, incidentally, a small group managing shared resources, distributing labor, making cooperation the default. Dunbar’s range is the scale where the social environment can manage behavior on its own: visibility, reciprocity, mutual stake, no further organizational complexity required, because the structure does the work by itself.
But the underground bunkers of the elite are skewed applications of this insight. They reproduce the form of cooperative living while depending entirely on what came before it, resources accumulated through the same mechanisms this essay has been describing, then withdrawn behind a perimeter. The architects of this are not, by Arendt’s measure, exceptional people. They’re ordinary ones who happened to be positioned where the unmanaged data was visible, and who responded exactly as the system trained everyone to respond: by arranging circumstances for their own survival. The evil here, if that’s the word, is what their typicality reveals about what the system rewards.
What makes successful communities function is the genuine mutual stake among the people inside it, and hoarded wealth behind exclusive access doesn’t satisfy that condition. By exploiting behaviorism at civilizational scale, the compounds of the wealthy simulate the architecture of cooperation without the substance of it.
This essay has been tracking a single coherent failure, and it’s a failure of scale, not of human nature. Every loop breaks the same way: the circle grows past what human cognition can track, and the feedback that holds behavior accountable to its consequences thins toward zero. Care becomes a transaction, risk a product, food a market, belonging a metric. Harm is banal, not because the people involved are unusually cruel but because the architecture never lets them complete the circuit back to what they’ve done.
This is the geometry of the problem. Once you see the geometry, the solution isn’t mysterious.
If the damage is structural — if exhaustion is the metabolic cost of operating in an environment calibrated to extract it, if the feedback loops that once made cooperation natural have been systematically dismantled, if behavior follows the path of least resistance and the path has been engineered — then no amount of internal optimization will close what the architecture has opened. Not therapy, not productivity systems, not political consciousness maintained in isolation. These are sincere, and they’re insufficient, because they’re aimed at the wrong level.
The data the elite read and responded to by building compounds is also available to you. The mechanisms of extraction that made mass scale possible are approaching the limits of what they can extract.
Two conclusions follow, and they’re different conclusions; it matters to keep them apart. The first is about throughput. Indefinite growth on a finite planet ends one way, so the real choice is between managing the descent and experiencing it as collapse. Managing it deliberately has a name, degrowth, and it isn’t a sacrifice. It’s a reduction in what flows through the system, most of which currently flows through you on its way to somewhere else. The second conclusion is about configuration, the scale and shape at which feedback closes. A Dunbar-scale community isn’t some romantic retreat from modernity. It’s the configuration where Ostrom’s conditions emerge on their own, where Arendt’s horror becomes structurally impossible, where the thing Watson and Bernays discovered and Skinner tried to champion in public and Sapolsky assembled into a scientific case — that behavior follows the environment — finally works in your favor.
The two conclusions meet in one place. A lower-throughput world is one where the intermediation economy, the paid strangers standing between you and every function of your life, becomes too expensive to keep running. In that world the Dunbar-scale configuration stops being a lifestyle experiment and becomes load-bearing. Throughput is how much moves through the system. Configuration is what catches it when it slows.
The objection that arrives here is obvious. You don’t live in a Dunbar-scale community. You can’t relocate to one tomorrow, can’t wait for civilization to obligingly contract, and can’t ignore the climate-scale problems that by definition exceed 150 people. The instruction to live in a smaller social configuration than the one structuring your life sounds escapist on its face.
The question isn’t how to leap from here to a finished alternative. It’s where, in the life you’re already living, you are supplying from your own reserves what an environment of mutual stake could supply automatically, and what it would take to start closing those particular circuits.
In any life there are already people for whom mutual stake could become real if it were practiced. It begins with noticing one function that has been extracted and returned as a transaction with strangers — childcare, eldercare, food, repair, the holding of risk, the holding of grief — and asking what it would take to bring it back. Mutual aid networks, food cooperatives, time banks, eldercare circles already do this. They’re practices that lower the trust-supplying tax one domain at a time, and they demonstrate in the body what it feels like for an environment to do structural work the self previously had to.
Larger coordination remains necessary for what only larger systems can do; the question is what those systems are embedded in. A community supplying its own care, food, and risk-pooling, embedded in larger structures that handle only what exceeds Dunbar, is not the same configuration as an atomized population intermediated by strangers for every function. The compound withdraws the small-scale function while drawing on the larger architecture without paying in. What’s needed is the inverse: the small-scale function rebuilt while the larger architecture is renegotiated rather than abandoned.
The condition that separates genuine cooperation from its compound simulacrum is downside reciprocity. If you can’t be hurt by what hurts the others, the circuit hasn’t closed; you’ve arranged your insulation. The available move is circles in which your survival and others’ is the same. Mutual stake means shared downside, or it means nothing.
Which is also, finally, the answer owed to Skinner’s serious critics. Who designs the design, who watches the designers? At this scale, with shared downside, the architecture watches itself. Whoever shapes the environment has to live in it, eat what it grows, absorb what it fails to hold. The control problem never gets solved at mass scale. It dissolves at the scale where nobody can exempt themselves.
Begin where you can begin. The recognition itself, that the friction is structural and the architecture is the level of response, is already the first segment of the bridge. It stops you spending the next decade on solutions calibrated to the wrong problem. What follows is downstream: which people, which function, which loop, which week.
This is not a new idea. It’s the oldest surviving one. It’s what the data keeps recovering when it looks past the ideological interference, and it’s what the most resourced people on earth are building for themselves right now.
The knowledge of how to live together was never lost; it was just made to feel impossible. It isn’t, and it can be undertaken with clear eyes about the mechanics. Without the mystification, without the apparatus, and without the cultivated belief that the problem is you.
Mind Stuffing 