Prologue. The expedition
In 1931 the Soviet psychologist Alexander Luria set out on an expedition through the kishlaks of Uzbekistan. With him he carried problems that, in the psychometrics of the time, counted as the baseline test of formal logic. The subjects — local peasants, many of whom had never been to school and could not read — were presented with a strict syllogism:
“Far to the north all bears are white. Novaya Zemlya lies far to the north. What colour are the bears there?”
The typical answer, as recorded in the protocols, sounds startling and sobering to a modern city-dweller: “I have not been there, I do not know.” The researcher tried to lead his interlocutor to a logical inference and pressed: “But from my own words, does it not follow? What follows from what I have said?” The peasant simply stood his ground: “Your truth may be true somewhere, but I do not know.”
At first the expedition recorded a defect — the primitiveness of thinking unable to master the operation of formal inference. But if one re-reads those protocols carefully and without academic arrogance, it becomes clear that the subject is not making a logical error at all. He does not say “the bears are black” or “the bears are grey.” He consciously and principally refuses to apply formal logic where he has no direct personal experience. In dozens of Luria’s Uzbek records the same hard-won life formula repeats: about what a person has not lived through himself, he does not speak.
When these materials at last saw the light as a separate book — nearly half a century later, in 1974 — a second, far deeper side of this experience opened up. The Uzbek peasant was showing not an intellectual limitation but a hard epistemic discipline: an impermeable boundary between the personally lived and the reported on hearsay. His thinking had a narrow but absolutely intact and undamaged domain of operation: any inference was built exclusively from personal experience, and beyond that experience the person honestly and plainly declared his ignorance. Call this epistemic humility — a person knows exactly where his calibration contour ends.
The man in the first paragraph — the armchair expert or the remote manager — is built in exactly the opposite way. His domain of judgment is unbounded: he reasons easily, confidently and peremptorily about anything. And the problem here is not the breadth of his horizon. His feedback contour is not merely narrow — it has been fatally swapped out.
1. The individual: the blind spot of self-check
In 1999 the psychologists Justin Kruger and David Dunning ran four series of tests — on humour, knowledge of grammar and formal logic — and came up against a fundamental property of human thinking. To solve a complex task well, a person needs competence. But to adequately assess the quality of a decision already made, the very same competence is required, applied a second time — recursively, to oneself. A subject deprived of grammatical competence physically could not tell a wretched sentence from a model one — including the one he had just set down with his own hand.
From this experimental fact a construction follows that is directly opposite to the banal cliché (“dumb people are simply too highly opinioned of themselves”). It is not a matter of two different mental skills — the ability to do the work and the ability to judge it. It is one and the same skill, run through two loops: first at the external material, and then recursively at oneself. Remove the competence and what disappears from the system is not merely the correct result — the metric of quality disappears from it. The blind spot forms exactly where, under normal conditions, self-check should fire. The man is not merely “over-confident” — at a deep level he is deprived of the very calibrating signal that should have been damping his confidence.
The mechanics of the effect and regression to the mean
The Dunning–Kruger curve needs an honest disenchanted caveat: in strict mathematics the effect reproduces even on completely random numbers. If you model the level of real ability and the level of self-assessment with an ordinary generator of random variables, the familiar classic picture — “overestimation among the weak and underestimation among the strong” — appears automatically, out of statistical regression to the mean. Fresh statistical analyses show: the arced chart of Kruger and Dunning owes its geometry to properties of the self-assessment scales, not to the psychology of the subjects. But that does not mean the blinding effect does not exist in nature. It means it must be looked for not in people’s “bad character” or “stupidity”, but in the objective environmental conditions under which the calibrating feedback becomes physically unavailable to the person.
This state has an exact term from classical cognitivistics: the symbol grounding problem. Any formal system is able to manipulate signs and words flawlessly, without the slightest direct contact with what those signs denote in the physical world.
A person whose calibration against reality is blocked is built exactly the same way: the test passed with top marks, the report written in flawless officialese — inside its own symbolic system everything is virtuoso and correct. The catastrophe arrives only where blunt reality collects the bill for these signs: in a concrete action, on concrete material, in a world that does not read neatly formatted reports.
The very same diagnosis is now made of modern large language models: a symbolic system without grounding inevitably begins to hallucinate. And this is not a poetic analogy but direct proof of a single mechanism: this is how any system of signs behaves when it has lost the quickening response of the physical environment.
Incompetence is not an excess of arrogance. It is the absence of the signal “you are wrong.”
2. The brain: the knowledge is there — the tether is not
On 13 September 1848 an accident on a railway line entered every neurology textbook forever. An explosive hurled an iron ramrod more than a metre long into the skull of the construction foreman Phineas Gage, destroying the ventromedial prefrontal cortex (vmPFC). Gage survived, fully retaining his fluent speech, memory, intelligence and the ability to calculate in his head. But he permanently lost the ability to carry his life plans to completion: every decision he made after the injury wrecked his career, his finances and his reputation with a frightening, mechanical regularity.
A century and a half later the neurobiologist Antonio Damasio, assembling the case histories of patients with similar frontal-lobe damage, formulated exactly what is broken in them. What broke was not the intellect — what broke was the connection between knowledge and action.
The key instrument of this connection is the somatic marker. These are anticipatory bodily alarm signals that in a healthy person fire before the rational mind has time to formulate a thought and make a decision. Before a doubtful or risky step the skin already knows the price: the palms grow damp, the heart stumbles out of rhythm, a cold feeling of dread rises in the gut — and the hand stops before the final words can fly out of the mouth.
The experiment with the losing decks (Iowa Gambling Task)
Damasio and his colleagues studied the anatomy of this process with a card task. In front of the subject lie four decks of cards and a starting capital. Two decks bring a modest but stable win. The other two pay large quick money but hide rare, catastrophic penalties that lead to total bankruptcy.
In healthy participants the skin conductance reaction (the micro-sweating of the skin that registers dread) to the “dangerous” decks appeared long before they could consciously produce a logical explanation of their choice. The body understood the trap first. Patients with ventromedial cortex damage, by contrast, reasoned about the losing decks verbally with absolute correctness and not a single error — and in the same second kept drawing cards from the losing deck, politely agreeing that it was bankrupting them.
Medicine calls this state “acquired sociopathy” — not because the person becomes a villain, but because his anticipatory brake loop has been physically cut out. He is capable of drawing an ideal, flawless picture of the negative consequences — and failing to connect it with the choice of “here and now.” The knowledge is in the head; the tether to action is not.
It is important here to separate the two severings clearly, so that the mechanisms are not confused. In chapter 1 the calibrating signal itself was entirely absent: an incompetent mind does not see its own error. Here, by contrast, the signal is present in the body, the body emits it — but the link between signal and deed is surgically cut. The one thing the two severings share: this is not a deficit of mind. Reason and computation are preserved in full. It is the feedback loop itself that is broken.
Intelligence is not only computation. It is the brake that fires before the signature.
3. The body: the negative feedback you cannot fake
The founders of cybernetics, Norbert Wiener and Ross Ashby, divided all feedback processes into two fundamentally different signs.
Negative feedback is correcting. Its signal is always directed against the deviation of the system from the norm. It continually returns the system to equilibrium: that is how a home thermostat switches off the heating as the temperature rises, and how a hand that has touched a hot kettle is instantly drawn back.
Positive feedback is amplifying. Its signal maintains, feeds and deliberately winds up the deviation: that is how a howling screech arises when a microphone is brought up to a speaker. The first system calibrates and holds stability; the second winds up and destroys.
The human brain is calibrated against reality exclusively through the physical body, and this calibration is hard-negative. It has concrete neural carriers, long since measured on the brain’s structure:
- Handwriting. In children who learned letters in their own handwriting the visual neural reading contour forms more fully and switches on faster than in peers who trained only key-pressing: the fine-motor load leaves a deep structural record of the symbol in the brain. High-density EEG experiments in adults confirmed it: neural connectivity during handwriting is substantially higher and broader than during keyboard typing, where any press is an identical click.
- Physical load. In a randomized study on 120 older participants a single year of regular aerobic training grew the anterior hippocampus by two percent — effectively rewinding one to two years of the brain’s inevitable age atrophy.
- Skills and neuroplasticity. MRI studies of jugglers showed pure plasticity in real time: three months of intensive training and the grey-matter volume grows; three months without practice and the grown structures fall back to their starting level.
A separate, most powerful class of these calibrating carriers is craft and direct interaction with material.
Physical material has a fundamental property unavailable to any social signal: it responds instantly, objectively and without personal ill-will. A turned part does not fit into the socket not because the physical world is looking at you sourly or judging your views, but simply because it does not fit.
Material is an uncompromising generator of negative feedback whose forbearance cannot be bought or a discount begged for. In the dialogue between a craftsman and material the other side holds an absolute, irreplaceable veto. Social approval can be imitated or earned through connections; the response of physical material can only be earned through skill.
This is the key formula on which the whole further argument rests: body and material are the most incorruptible and honest calibration line of the human mind of all that exist on Earth.
Material does not hand out likes. It simply does not assemble.
4. Tools: domain-by-domain atrophy
From the mechanics described above a systematic law follows naturally: any function handed to technology inevitably loses its natural training loop. And this function atrophies not as some abstract “general decay of the mind” but locally and specifically — domain by domain, skill by skill.
The action of this law is easiest to test where the price of incompetence is not measured in likes and views but in hundreds of human lives.
Aviation: the price of abandoning the yoke
On 12 February 2009 the flight liner Colgan Air 3407 crashed near Buffalo (50 dead). The official investigation revealed the fatal loss of manual-flying skill in the crew. The US Congress demanded a systematic inquiry, and in 2013 an FAA expert panel issued the devastating report “Operational Use of Flight Path Management Systems”: 28 of the gravest findings and 18 urgent recommendations. The main conclusion: manual flight had degenerated into passive monitoring of the display, and the skill of direct control of the aircraft is degrading year by year.
Just a few months after Buffalo, on 1 June 2009, Air France 447 crashed into the Atlantic. The final report of the French BEA bureau recorded it: the pilots had not practiced manual control at cruise altitude for years; the sudden failure of the automation caused bewilderment, and the missing motor skills finished off the catastrophe.
Four years later in San Francisco Asiana 214 struck the seawall on a visual approach to landing. In its final report the NTSB plainly recommended that airlines break their existing automation policy — and forcibly return pilots to manual flying in training and on regular routes.
Hundreds of passengers killed — that is the price of feedback arriving into the system only after the calibrating loop has been handed over entirely to the autopilot. The skill fades quietly and unnoticed — right up to the moment when the automation works perfectly.
Navigation: the shrinking hippocampus
Véronique Bohbot’s laboratory tested the action of the same law on ordinary drivers and GPS navigators. The research design would stand up to any scientific criticism: 50 participants, a detailed accounting of experience. It turned out that long experience of driving by navigator reliably predicts a deterioration of spatial memory on independent trips.
The hypothesis that “navigators are simply used by those who are poor at orientation to begin with” was checked and ruled out. A three-year observation recorded an accelerated decline of the mental maps precisely among the active gadget users.
The control group: London taxi drivers
The existence of this law is confirmed by the control group too — where the function was, in principle, handed to no one.
The famous London taxi drivers spend three to four years memorizing the most difficult map of the capital for the “Knowledge” exam. Research by Eleanor Maguire and Kathleen Woollett showed: in those who pass the exam, the posterior part of the hippocampus physically grows, at the level of grey matter. In those who fail the test, or in bus drivers running a strictly fixed route, no structural changes in the brain occur.
The law works in both directions with pharmacy-grade precision: what you load grows, what you hand to the algorithm shrinks. This makes our diagnosis not a grim verdict but an engineering task: the loop is restored if its physical load is returned to it.
The other domains demonstrate the same mechanism:
- Search. A person’s memory restructures from remembering the essence to remembering the address — “where to find it” instead of “what this actually is.”
- Medicine. When there is a ready answer from a diagnostic algorithm the doctor’s vigilance objectively falls, and a full check of the diagnosis is replaced by a subconscious search for confirmations.
| Tool | What atrophies | Facet of the law |
|---|---|---|
| Autopilot | Manual flying | Skill fades unnoticed; the price comes as a catastrophe |
| GPS | Spatial memory | The neural substrate shrinks without load |
| Taxi drivers (contrast) | On the contrary: the hippocampus grows | The law is two-way |
| Search | Memory of content | The function moves to the cloud |
| Diagnostic algorithm | Operator vigilance | Monitoring degrades into a search for confirmations |
5. The smartphone and the environment: a polarity inversion
The navigator in the car or the autopilot in the cabin is a local episode: step out of the car and the natural calibration loop returns. The smartphone radically changes the very condition of our lives: every offloading channel — navigation, memory, arithmetic, grammar, other people’s judgments — is gathered into a single device that stays at the body twenty-four hours a day. The episode has become the climate of habitation. This is the difference between a single bout of flu and an incurable change of the atmosphere.
Have you ever caught yourself unable to drive to a familiar house without the navigator switched on, although you have lived in that district for three years? The navigator did not let you get lost at a turn — and at the same time it did not let your brain learn the map. It did not cut the loop out by force. It simply did not let it switch on.
But the main step of this section is not the gadget’s permanence. The tools of the previous chapter were severing the calibration loops one at a time and at random. The social feed and the algorithms do something categorically different: they invert the sign of the feedback itself.
Return to cybernetics. For calibration the human mind needs negative feedback — a clear “you are wrong” signal, directed against the deviation that has arisen. The feed does not simply switch that signal off — it replaces it with its opposite in sign.
Instead of the sobering “you are making a mistake” there arrives “you got a like”: positive feedback that instantly feeds and grows the very error that has already arisen in the head. The loop is not merely cut out — it has been re-soldered anew and started on wind-up. The system receives the most powerful stimulative signal in the wrong direction: the conditions of the blindness from the first chapter are reproduced industrially, in billions of people at once.
A function handed off to a device
│
▼
The training loop fades, no error signal arrives
│
▼
Confidence separates from competence
│
▼
The feed rewards confidence ──► positive feedback
│
▼
Offloading deepens ──► (back to the top of the cycle)A strict caveat
Do not build the argumentation on cheap psychological alarmism. The brain-drain effect (“the mere presence of a smartphone on the desk lowers cognitive capacity”) turned out, in strict repeat replications, to be far weaker than the original loud publications. Filter bubbles, conventionally spoken of as the chief machines of polarization, give a far more modest effect according to large recent studies than was assumed.
But the argument of this article does not rest on those shaky metrics — it stands on the architecture and the economics. Polarization may be exaggerated; the inversion of the calibration sign is not.
This architecture is paid for directly by the attention economy. The only business model of a digital platform is engagement: time at the screen, response, emotional return. The uncalibrated user is profitable to the IT giants not because anyone is deliberately scheming his stupidity, but because his uncalibrated confidence reacts instantly, loudly and cheaply. Doubt is friction, and friction lowers conversion. Optimizing engagement systematically selects for the behaviour whose link to physical reality is the thinnest.
Exactly the same operation the environment performs on the far side of the screen — in business. Companies that wish to reach the user buy not an advertising slot but a seat at the automatic auctions. By 2021 Google had completely moved its ad auctions to the first-price model: you pay exactly as much as you yourself bid.
Industry measurements show a mad picture: the cost of acquiring a customer (CAC) on these auctions is growing faster than any product metric — plus 222 percent over eight years. Meanwhile the average loss per new buyer in e-commerce shot up from 9 to 29 dollars.
The marketing department — seemingly the most instrumented with figures and sensors organ of modern business — is calibrated not on product quality or customer satisfaction. It is calibrated on survival in the auction. Its error signal comes not from a living human being but from the auctioneer.
The loop that, for the ordinary user, has been re-soldered to the like, for the commercial director has been re-soldered to the ad bid. Both sides are calibrated to a synthetic signal whose full owner is a third party.
A like does not say “you are wrong”. It says “more.”
The counter: the break is already visible
Until now we have been dissecting the inner anatomy of this mechanism — from a single neuron to the international ad auctions. But the argument has a population-level check too: the global counters measuring humanity as a whole have already registered the break in dry figures.
The first counter: the reversal of the Flynn effect. Throughout the twentieth century the average IQ of humanity grew continuously. But in recent decades, in the developed countries, this curve has broken and turned downward. The researchers Bratsberg and Rogeberg worked through a body of Norwegian data and convincingly showed, in within-family and cross-generational analysis, that both the rise and the reversal are caused by changes in the environment, not by genetics or by shifts in the composition of the samples. Population intelligence grew while the environment created training loads, and went down the moment the environment changed to a load-relieving one.
The second counter: the adult-skills studies (PIAAC). The OECD program for assessing literacy, mathematical skill and problem-solving ran its first measurement in 2013 and its repeat in 2023. The result: for the first time in the history of the measurement, a direct decline in adult literacy has been recorded in most of the participating countries.
This is not the banal “the young generation has got worse at studying.” This is an adult function, stripped of its constant calibrating loop, shrinking before our eyes — exactly as the grey matter of a juggler’s brain falls away after three months without practice.
The laboratory microscope and the population counter converge on a single point: the environment sets the direction. The question “why are there more uncalibrated people” is no longer an emotional rhetoric. In our hands there is a chart where this “more” is documented.
6. The class: knowledge without skin
In the mid-twentieth century the economy of the developed world formed a whole class whose daily practical activity has not the slightest physical contact with the object it manages — James Burnham and William Whyte called this the managerial revolution. The profession of these people consists in “managing the managing.” But to grasp the scale of the disaster one need not lean on essayism: the question “why are there more of them” has a mathematically exact quantitative selection mechanism.
Selection: how the system reproduces unfitness
The model of the economists Francesco Caselli and Massimo Morelli proves this: a “bad manager” is not a random anomaly and not a corruption of morals — it is a stable systemic equilibrium. Bad managers generate huge negative externalities for the good ones, so that the benefit of the office rises precisely when the average quality of the staff is low. The system with inevitability selects and reproduces its own composition.
The famous Peter Principle (“a man rises to the level of his incompetence”), which had seemed a half-joking anecdotal hypothesis, has turned into a strictly measurable fact. On data for 53,000 salespeople from 214 American firms, the researchers Benson, Lee and Shue showed: promotion to a managerial post systematically goes to those who are best at their current sales. Yet a person’s earlier results do not at all predict his quality in the role of boss. The firm with an admirable constancy loses a strong executor and gains a weak manager.
Density: the spread of the bureaucratic layer
By the counts of Gary Hamel and Michele Zanini, the ratio of managers to workers in US federal agencies rose from 1:2.2 in 1998 to roughly 1:1.3, and the US bureaucratic class exceeded 24 million people.
Here the cybernetic law of requisite variety of Ross Ashby enters: a regulator whose own variety is bounded by the frame of a dashboard is physically incapable of governing a system whose variety exceeds its dashboard. This is exactly why the reporting multiplies in geometric progression: there was catastrophically not enough of it even before. The tracker RegData records the continuous accumulation of regulatory constraints in the law. Bart Jansen called this process the juridification of business ethics: living human judgment is replaced by a formal rule, and real responsibility by bureaucratic compliance. David Graeber added to this the concept of “bullshit jobs” — the existence of a vast class of vacancies whose uselessness their own practitioners openly acknowledge.
The bridge: the corporate prefrontal cortex
Here the anatomical homology with the brain damage from the second chapter receives its missing link.
The somatic marker is the biological form of what Nassim Taleb called “skin in the game.” An organism that bears personal physical or financial losses from its own decisions stops before the deed is done, because the anticipatory pain is registered by the body.
An institution or corporation in which the taking of a decision is detached from the losses is a whole organism with the ventromedial cortex surgically removed. The one who signs the document, whose destructive consequences will arrive in another office, in another financial year and in other human lives, keeps drawing cards from the losing deck with a polite smile. The pain is simply not registered by the system until the moment of the total collapse.
The asymmetry: the transfer of risk downward
This institutional construction has two ends, and both work with a tilt.
At the top floor the “skin in the game” is absent entirely. The transfer of risk from top to bottom is described by the concept of the “moral crumple zone” of Madeleine Elish: all the blame for a complex system’s failure automatically flows to the person nearest the accident who acted on instructions, while the designers and beneficiaries of the system stay clear. And where a learning algorithm makes the decision, systemic “responsibility gaps” arise — for the catastrophe no one answers at all.
At the bottom floor, by contrast, the Peltzman effect of Sam Peltzman works: when an external compensation system lowers the price of error as the person perceives it, his natural vigilance rationally falls — not from any ill intent, but by pure arithmetic. Though the Peltzman effect is context-dependent and disputed in its details, the terminal state is described by the legal system with one exact word: negligence — the inadequate performance of a duty through careless conduct that caused harm. The term strikes not the character of a particular person but the organizational mechanism: it is the role itself that is so constructed.
| Level | What is broken | Who pays | Who it is charged to |
|---|---|---|---|
| Brain (vmPFC) | Decision ↔ bodily signal | The organism | No one — a disease |
| Body | Skill ↔ training load | The person | No one — a climate |
| Tool | Action ↔ control | The operator | The operator (crumple zone) |
| Environment | Calibration ↔ error signal | Everyone’s attention | The user |
| Class | Decision ↔ loss | Executors, the environment | The lower floor |
The homology can be checked component by component. The corporate top manager and the political functionary have different objects of management and different owners of the platforms, but the architecture of the break is absolutely identical in them: each is calibrated not on the real object but on a synthetic proxy signal whose owner is a third party.
| Parameter | Corporate manager | Political manager |
|---|---|---|
| The real object | The product and the living customer | The social fabric and the infrastructure |
| The synthetic proxy | CAC, ROAS, reach | Opinion polls, media indices, virality |
| The owner of the auction | Google, Meta — a trade for attention | Social networks, media conglomerates — a trade for affect and anger |
| The price of the break | Burned margin, product decay | Burned institutional stability |
A corporation’s burned margin and decaying metrics are visible in the nearest quarterly report. The state’s feedback, by contrast, lags by decades: the wear of urban networks and the demographic collapse physically do not fit inside the short electoral period. This is precisely why, for the political manager, the somatic marker is not merely dulled — it is amputated whole: until the moment of the catastrophe the pain is not registered at all.
A report has no body.
Conclusion. Design with friction
The Uzbek peasant from the prologue refused to reason about what he had not personally lived through — and in that lay his hard calibration: do not issue inferences that have no support in personal experience. The modern managerial class does not refuse to judge — it simply does not notice its own ignorance. The error signal “you are wrong” physically does not reach him; what reaches him is a smooth report.
If the article’s argumentation is correct, the antidotes should treat not the character or morals of people but the feedback loops — at every level of our graph:
- Body. Handwriting and manual labour in education are not nostalgia for the past, but the protection of that very motor record that builds the dense neural loop of reading and thinking.
- Skills. Manual flying, restored into the regulations at the insistence of the NTSB, and periods of manual navigation without GPS: the law is two-way, and the hippocampus responds to load.
- Environment. The design of systems with deliberate engineering friction — two-key protocols, physical interlocks, forced pauses before an irreversible action. This is the standing engineering practice where the system has no body of its own to be afraid in time — an artificially created somatic marker.
- Institutions. A personal signature under the decision, metrics fastened dead to long-term physical consequences, and the legal right of the one who signs the order to see its consequences live.
Why there will be no ready playbook
Any antidote on this list the modern system instantly packages and turns into a commodity. The industry has already successfully sold us mindfulness, meditation, minimalism and digital detox. Friction bought as a subscription from a digital platform becomes its own new product — positive feedback, new edition.
The genuine calibrating loop is restored exclusively through direct contact with what is, in principle, impossible to buy: with physical material, with weight, with the body, with the incorruptible environment and with one’s own pain from a mistake.
Two main objections
The first objection: “People are different by nature; a natural dispersion of mind and abilities exists.”
It exists — and it is not the subject of this article. The point is about the systemic function of the environment, not about the genetic distribution within the population. An environment that cuts the calibrating loops equally stupefies the middling and the genially gifted. The population counters of IQ and PIAAC show plainly: the vector of movement was set by the environment, not by the genes.
The second objection: “People have been frightened this way in every age — even Socrates feared that the invention of writing would destroy human memory.”
They feared — and they did not fear in vain. The function really was handed to an external carrier, and writing really did carry away part of the natural memory. But the fundamental difference of today lies in two things:
- In the sign. The older technologies (writing, print) removed individual loops, leaving the physical world calibrating. The social feed changes the sign of the calibration itself: the error signal arrives, but inverted.
- In the economics. For the first time in human history there is a gigantic investor in this break who earns billions of dollars on its preservation and deepening. This is not the moral of the story — this is its dry economics.
The loop test
Answer yourself honestly: when was the last time you changed a deep conviction under the pressure of blunt physical reality, rather than under the influence of something you read in the feed? If you cannot remember this — your calibration loop has been re-soldered.
But it is repairable — right up to as long as physical material remains in our hands.
First what is lived through, then the decision. When the order is reversed, the decision is made by the report.
“About what he had not seen, the man did not speak” — that is the position of the illiterate Uzbek peasant in the prologue, whom the scientific expedition counted as backward. Our present-day environment is confident that it can judge about everything that is not even in the report. The difference is just one: for the peasant it was the deepest humility of the mind, while for our civilization it is a fatal error of the architecture.