The Bauhütte Standard: reality before loyalty

Jun 21, 2026

Young man, there’s a place you can go, I said

The private origin of this standard was a bad midterm. I went to Marianna Csörnyei’s office looking, I suppose, for rescue, exception, or absolution. She was not unkind; that was part of the force of it. She asked, quite pleasantly, why I was there. There are two kinds of people, she said quietly — or at least this is how the sentence lodged in me. There are the lazy smart people, who can figure things out on the fly, and the less gifted diligent ones, who do the hard work. I seemed, however, to have contrived a third and least useful category: too stupid to solve the harder problems by improvisation, too lazy to memorize the easier proofs. Why was I there? I had no answer, and that was the only embarrassment in the exchange: not that she had seen me clearly, but that the question had arrived before I had done the work of asking it myself. I thanked her and left, not humiliated but illuminated. The remark did not diminish me. It located me.

I do not know whether she meant the sentence as taxonomy, diagnosis, or mathematical mercy; in any case its moral has improved with age. The point was not that talent does not matter. The point was that talent, whatever its amount, is not an exemption from the duller virtues. One may prefer to be the first type; one had better acquire the disciplines of the second. Maturity may be no more glamorous than this: the moment one stops treating diligence as an insult to intelligence.

The lesson eventually escaped mathematics. Work is not merely the arena in which competence is displayed; it is the discipline by which competence is produced. A serious institution should therefore do more than consume talent. It should train judgment, expose evasion, and make the duller virtues impossible to despise.

It should also preserve the right kind of friction. Work has to grind against reality; otherwise it remains a gesture, a polish, an agreeable motion above the surface of things. Friction—the tooth, the catch, the refusal to slide—is what allows power to be transmitted. When work is made too easy, too smooth, too carefully insulated from resistance, the whole machine merely skates. It may produce noise; it may even produce elegance. But it does not move the weight. The object is unchanged. Reality has been flattered, not altered.

This is not a defense of pain, still less of mindless grind. Quite the opposite. Mindless grind is not friction; it is waste. Punishment does not make work serious. Reality does. The useful resistance is the kind that teaches: the stone that cracks, the proof that fails, the client who will not invest, the sentence that will not carry the thought, the machine that will not run. A real object supplies the loss function. It gives error a shape. It tells you not merely that you have failed, but how. It measures the discrepancy between intention and result, between the thing one meant to build and the thing that actually stands. Without that discrepancy, there is no learning gradient. There is only motion, confidence, and the pleasant continuation of one’s existing form.

More importantly, the appetite for frictionlessness is itself diagnostic. This was the lesson Csörnyei had taught me: if resistance appears only as an obstacle to be removed, rather than as the medium through which competence is formed, improvement has not really begun. One is not yet working better; one is arranging the conditions under which one can continue unchanged—or decline without noticing.

This is the alarming part. Credentials can outlive competence. Every résumé should probably carry the disclaimer printed on an investment prospectus: past performance is no guarantee of future returns. A degree, a title, a prestigious employer, a prize — these are not competence engines. They are records of an earlier contact with reality: evidence that, under some earlier conditions, a person cleared a particular bar. Whether that capacity remains alive depends on what the intervening years have asked of it. A résumé records previous contact with reality. It does not prove continuing contact.

It saddens me, rather than amuses me, to see a quant Ph.D. from a top school unable, in middle age, to do basic mathematics with any force or speed. The point is not mockery. It is warning. The diploma may remain accurate as biography long after it has ceased to describe a living capacity. The spreadsheet calculates, the assistant checks, the model summarizes, the meeting rewards fluency, the institution rewards confidence, and the unused faculty quietly withdraws. The brain is not a trophy case. It does not preserve old victories in glass. It is more like muscle: what is not exercised atrophies; what is not corrected becomes theatrical; what is not tested becomes opinion. That is how decline happens — not usually by collapse, but by insulation.

The academic version is especially treacherous because recognition can become rent. This is the academic cousin of the value-extractor in Good companies: the title keeps extracting authority after living contact with the work has thinned. A prize, chair, named professorship, or founding label — father of this, mother of that — can become a rent-bearing title detached from continuing work. I have seen professors spend decades preaching the result of their early thirties, long after the result had become peripheral, non-generative, or preserved chiefly by reputation. The old work becomes an annuity. The room pays interest to the name. Students are invited to spend their careers approaching a monument whose spring may already have dried. The academic value-extractor is not always a fraud. He may once have met the thing. The danger is that the chair, prize, and founding label keep meeting it on his behalf.

Prizes sharpen the problem because they do not merely remember. They institutionalize praise. This is where the Adlerian warning belongs. In The Courage to Be Disliked, the Adlerian point is that praise and rebuke both create a vertical relation: one person stands above another as evaluator. The sugar and the vinegar differ; the hierarchy is the same. Praise and rebuke are not opposites. They are administrative siblings. One says, “I approve”; the other says, “I condemn”; both presume the authority to grade another person from above. An award is that structure raised to institutional scale: praise with a seal, a dinner, a purse, a citation, and a public memory attached.

That vertical form is dangerous because the prize claims jurisdiction. To award is to say not only, “this is good,” but “this is the one worth remembering here.” That claim is almost impossible to make honestly. If the judgment is not exhaustive, it is parochial; if it is not superlative, it is arbitrary; if the committee lacks the courage to withhold the prize when no clear superiority exists, it is not judging excellence but feeding ceremony. Most awards fail all three tests. They must select because the institution must appear capable of selection. The ceremony needs a head.

There is also a sense in which prizes are the field bribing itself. The bribe is not paid to genius but to mediocrity: a ceremonial payment that lets ordinary colleagues preserve their self-respect. We are not small; you are merely great. We were not slow; you were exceptional. We are not provincial; you are a rare genius. The award converts indictment into anomaly. It takes work that might have exposed the inadequacy of the existing judges and turns it into a harmless exception that leaves the judges intact.

The old Einstein myth — that general relativity was understood by only a few people — makes the absurdity plain. If one can certify Einstein, one must in some sense stand above Einstein; and if one stands above Einstein, why is one not receiving the prize? Committees launder this embarrassment. They diffuse partial understanding, reputation, politics, hindsight, and institutional vanity into authority. No individual judge has to claim the impossible height from which the highest work can be ranked. The institution puffs itself up until it appears tall enough to bless the work that once exceeded it.

This is why the recipient’s posture matters. To accept an award is already to accept, for the purposes of the ceremony, the limited authority of the awarding institution. The honest recipient says: I need your signal — not because it settles rank, but because it may carry the work farther than it could travel alone. Refuse the prize, and say the work does not require the ceremony. Accept it, and admit the dependence. The corrupt recipient wants the benefit of both positions at once: dependence without confession, contingency without modesty, visibility without debt. He wants to murmur “how humbling” while spending the prize as if the hierarchy had been settled.

Napoleon understood the problem more honestly than most prizewinners. If no authority stands above you, then you crown yourself; the gesture is outrageous, but at least it does not pretend humility before a power it has already displaced. Ordinary awards cannot do that. They must occupy the middle position or become ridiculous. They are useful only when they say: here is work that should be seen. They become corrupt when they imply: here is the final order of merit. The award may preserve public memory, but it also protects the institution from the accusation greatness would otherwise make against it.

The rule should be stricter still: an award should not come with anything of marketable value. Prize money makes the bribery literal. If the purpose is recognition, money contaminates the signal by turning honor into transfer. If the purpose is support, call it a grant, fellowship, commission, subsidy, laboratory, salary, or patronage, and drop the ceremony of superiority. Do not pretend that a check proves excellence. Do not pretend that a salary, chair, office, residency, or guaranteed platform is merely symbolic. These things have market value; they change incentives, careers, and bargaining power. They are not innocent ornaments on judgment. They are the judgment becoming property.

No wonder billionaires love to endow prizes in mathematics and the natural sciences. The act of crowning crowns the crowner. A patron who cannot stand inside the work can still purchase proximity to its height: summon the committee, fund the purse, photograph the laureate, and let reflected genius wash over the donor’s name. The laureate receives money; the patron receives borrowed altitude. This is not always ignoble. Civilization has always needed patrons. But patronage is honest only when it admits what it is doing: giving resources to work, not buying a place near the altar.

If the money is to have social use, it should be spent prospectively rather than ceremonially. Give it to the young, the strange, the under-seen, the not-yet-protected — the person whose future work might fail without time, tools, shelter, or institutional permission. Fund the dangerous interval before reputation has arrived. Buy freedom from rent, equipment, teaching load, visa anxiety, medical precarity, committee life, and premature market pressure. That is patronage. Stacking money onto an already established name is something else: reputation paying interest to itself.

The only defensible public use of a prize is illumination. Awards should discover the overlooked, not decorate the established. A prize should be a flare, not a laurel. Once a person is already visible, endowed, chaired, translated, invited, and protected by reputation, the marginal utility of another crown is close to zero. At that point the award no longer discovers talent; it compounds authority. It turns recognition into rent. This is also a credentialed form of cognitive resignation: the institution remembers and rewards an old contact with difficulty while the faculty that produced it may no longer be under strain.

The institutional form I have in mind is the Bauhütte: the builders’ lodge gathered around a serious work. Not a workshop in the modern corporate sense — not a breakout room, not a training session, not a culture exercise with pastries — but a house of work: cathedral, stone, scaffold, plans, apprentices, masters, inherited techniques, and corrections supplied by the material itself. The stone cracks or holds. The arch stands or falls. Gravity, that severe auditor, does not care who spoke most persuasively in the meeting. For its opposite, see Good companies: the value-extractor manages the signs of work; the Bauhütte returns judgment to the object.

Once people are fairly paid, the most important — and perhaps the most humane — thing an institution can give a talented person is not another title, prize, or increment of compensation, but a Bauhütte: a place where their powers can engage a real problem with enough force to matter. Talented people do not wither only from neglect or underpayment. They also wither from idling. A mind made for pressure becomes strange when left to spin in air — clever, restless, decorative, self-consuming. Give it a hard object and the same energy acquires traction. The gears catch. Effort enters the world. Something moves.

In a company, that engagement must eventually earn its keep; otherwise the house cannot endure. But commercial success is healthiest when it records contact with reality rather than substitutes for it. It is evidence that the gear has caught, not the source of the work’s human meaning. Before the title, the bonus, or the market verdict, the work has already done something serious: it has given the worker necessity, and necessity is the one honor no title can counterfeit.

The converse is equally important: a role with no resistant object is not neutral. It trains too. I have seen too many people spend years in lucrative but ornamental roles and discover that the habits of those roles had followed them home. The role is rarely the whole cause, but it is a daily rehearsal. If a person spends eight hours a day learning that promises may remain vague, consequences may be passed elsewhere, appearances may substitute for obligations, and omissions need not be repaired, those habits do not remain neatly at the office. Unaccountability at work can become undependability in private life.

“How you do anything is how you do everything” is false as a law, but exact as a warning. Repeated conduct does not merely reveal character; it rehearses character into being. This is the seed in the storehouse consciousness: insight may name the pattern; only repeated practice retrains it. A decorative job can therefore be lucrative and still exact a private cost. It may strengthen the résumé while weakening the habits by which a person becomes trustworthy. A Bauhütte is humane partly because it reverses that rehearsal. It gives responsibility, consequence, and dependability a real object on which to practice.

None of this makes the work easy. A real problem is not an employee-wellness program; it is infinite, indifferent, and often soul-crushingly daunting. It bites. It humiliates. It refuses the convenient self-image of the talented person as already sufficient. Some people will turn away from that pressure, and they should. That is part of the selection the Bauhütte performs. Those who came for status, ease, optics, or managerial warmth will discover that the object is rude and ungrateful. Those who remain are the people one wanted anyway: not because they enjoy suffering, but because they find home where work has teeth.

Years of formal feedback, private counsel, failed collaborations, durable partnerships, and unflattering self-examination have made two things clearer to me: the conditions under which I do my best work, and the conditions under which my strengths need deliberate design before they become useful to other people.

The name I want for that design is the Bauhütte Standard.

I do my best work with people who are answerable to something real. I do my worst when work becomes theater For the full pathology, see Good companies: the value-extractor manages visible signs of work while leaving delivery’s slow obligations to others.—when impressions outrun obligations, when agreement quietly replaces dissent, when warmth is used to excuse vagueness, and when correction is received as injury rather than information.

The Bauhütte Standard is not a personality framework. It is a way of asking whether work can remain answerable to reality after people, incentives, fear, charm, fatigue, hierarchy, and good manners have all begun their quiet campaign against it. It asks who can stay with the object on the bench: the client, the market, the system, the model, the sentence, the risk, the promise, the machine. It asks who can be corrected by that object without converting correction into grievance, status injury, or political leverage. It asks what interfaces are required so that high standards do not become brutality, and warmth does not become fog.

Reality before loyalty is not a rejection of loyalty to people. It is the only form of loyalty that does not quietly betray them. A room protected from reality may feel kind for an afternoon; over time it lets the world deliver the correction more violently than a colleague needed to. The Bauhütte exists to make truth arrive early enough, and humanely enough, that it can still be used.

What follows is a taxonomy, but its purpose is not to sort people into the worthy and the defective. It is to map interfaces: where my native mode produces speed, precision, and useful pressure; where it turns too severe, too fast, or too poorly translated; and what kinds of people, processes, and counterweights make the work truer without making the room needlessly brutal. The standard is neither comfort nor severity. The standard is disciplined, frictional contact with reality.

The Bauhütte, not the courtroom

Organizations tend to drift toward one of two failure modes. Some become courtrooms, where every exchange is implicitly a judgment of persons — who is competent, who is favored, who is at fault — and the actual object of the work recedes behind the management of standing. Others become theaters, where the visible performance of seriousness, alignment, and harmony stands in for contact with anything that can fail.

The Bauhütte is the alternative to both. In a Bauhütte, attention is fixed on the object on the bench. People are judged by their relationship to that object, not by their relationship to the boss. Correction is ordinary because the work is hard, not because anyone is on trial. Standards are high because the object is unforgiving, not because severity is a personality. Dissent is welcome because the object does not care about anyone’s feelings, and in the end neither can the work.

The distinction is where the friction goes. Friction with the object is traction; friction between egos is merely heat. The former moves reality. The latter warms the courtroom.

Building a Bauhütte-driven organization is mostly an interface problem. High standards are not enough; they have to be made usable — translated into sequence, structure, ritual, memory, and process — without being diluted into fantasy. The rest of this piece is an attempt to specify those interfaces.

The one-line

The sentence is personal, but the principle should not remain personal. Loyalty to reality is the highest form of loyalty to the people doing the work, because it prevents fantasy from becoming their downstream burden. The point is not to make the organization colder. It is to make its warmth less dishonest.

The simple diagnostic

The fuller diagnostic is still simple. No one lives entirely in one column, and everyone has bad weeks; what matters is the center of gravity.

These are not essences. They are collaboration states under pressure. The same person can occupy different positions depending on role, incentives, trust, fatigue, clarity, and design. The taxonomy is useful only if it produces a better interface; the moment it becomes a permanent sentence on a person, it has become the courtroom.

The three axes

The diagnostic resolves into three dimensions. Taken together, they give the framework its name — RIC: Reality, Interface, and Correction.

• Reality Contact asks whether a person submits to something outside opinion.
• Interface Intelligence asks whether they can make high standards usable by other people without falsifying them.
• Correction Metabolism asks what happens in them when truth arrives as friction — whether it becomes traction or heat.

The first and third are about whether the work can be trusted. The second is about whether it can be received — and it is the dimension that leaders with high standards most often neglect, because translation feels like someone else’s job. It is not.

What I do well, and where I need design

The taxonomy is only useful if it subjects me to the same discipline it applies to everyone else. Its purpose is not to explain why other people should adapt to my standards. Its purpose is to make my standards legible enough that they can be joined, challenged, improved, and carried by the organization without depending on my mood, stamina, or personal force.

My native strength is Reality Contact. I tend to notice when work has drifted away from the thing itself — when a client promise has outrun the machinery, when interface numbers have replaced true account state, when agreement has replaced decision, when polish has replaced ownership, when commercial warmth has become a substitute for saying what is and is not possible. At my best, I bring the room back to the resistant object: state, owner, sequence, evidence, obligation, risk, and consequence.

That instinct is valuable because serious work decays first as language. The organization still sounds responsible before it is responsible; still sounds aligned before anything has been chosen; still sounds client-centered while transferring pain downstream; still sounds rigorous while the ledger, model, system, or promise is telling a different story. I am useful when I make that evasion harder.

My second strength is compression. I can often see the structure beneath scattered symptoms: the repeated promise without an owner, the person managing credit rather than work, the system state being patched by human exception, the commercial story being paid for by operational debt. This helps in diagnosis, design, hiring, and triage. It also helps a small organization move quickly without pretending that speed is the same thing as clarity.

My third strength is formation. I care not only whether the artifact improves, but whether the person and the institution producing it become more accurate over time. A memo, model, system, portfolio, client interaction, or decision process should leave behind better judgment, not merely a completed output. The work should train the worker; See Deep persuasion. The short version: insight changes the sentence; practice changes the weights. Formation requires both artifact and producer to improve. the organization should train its own future judgment.

The improvement edge is Interface Intelligence. I can move too quickly from observation to classification, and from correction to verdict. What is meant as contact with the work can be received as judgment of the person. What is meant as a standard can arrive as contempt. What is meant as urgency can make the room too afraid to show unfinished reality. If that happens, the Bauhütte becomes the courtroom I am trying to avoid.

So the central discipline is to separate correction from classification. Say, “This work does not yet meet the standard,” before saying, “This is the kind of person who produces this failure.” Say, “The promise outran the machinery,” before saying, “This is an optics problem.” Say, “The decision lacks owner, state, and consequence,” before reaching for a type. The type may be real, but if it arrives before the repair path, it becomes a sentence rather than an interface.

The second discipline is to reward early bad news. If people learn that early ambiguity gets punished, they will bring only the cleaned-up version of reality. That would be a complete failure of this framework. I do not want polished truth arriving late. I want rough truth arriving early enough to change the outcome. The measure of trust is not whether people avoid disappointing me; it is whether they can disappoint me before the world does.

The third discipline is calibrated recognition, not praise. The award problem appears again at the scale of daily management: praise and rebuke both try to place authority in the evaluator rather than in the work. A Bauhütte cannot run on the leader’s approval any more than it can run on the leader’s disapproval. The point is not whether I am pleased, whether I am disappointed, whether the leader’s face has warmed or cooled. The point is whether the work has become more answerable to reality.

So the signal must be converted into calibration. Not: good boy, bad boy. Not: I approve, I condemn. Rather: this sentence carried the thought; this model survived the data; this promise outran the machine; this decision lacked owner, state, and consequence. People need to know what worked and why it worked, but the authority should belong to the object on the bench. “This was strong because…” is not approval from above; it is training from the object. If correction names the boundary, calibrated recognition names the repeatable behavior. Without it, the organization learns only what summons judgment, not what compounds.

The fourth discipline is pacing. I often see the pattern before the other person has walked the path. That does not mean the path can be skipped. Some people need sequence before they can receive the standard: what happened, why it matters, what standard applies, what changes next, and what is not being judged. Speed without sequence feels like force. The bar should remain high; the staircase should be visible.

The fifth discipline is to give commercial force a legitimate, bounded place. Charm, warmth, recruiting, selling, and narrative are not corruptions by themselves. They become corrupt when they outrun memory and delivery. A serious organization needs people who can meet strangers and create trust before proof has fully arrived. My task is not to despise that talent, but to bind it to promise ledgers, owners, refusal scripts, and operating truth.

The sixth discipline is to build judgment into artifacts rather than keeping it in my head. If truth requires my presence, the organization is still orbiting me. Decision records, promise ledgers, definitions of done, written objections, pre-mortems, after-action reviews, ownership maps, and explicit refusal paths are how the standard becomes architecture. The goal is not that people become loyal to me. The goal is that the company becomes more loyal to reality when I am not in the room.

The practical commitment is therefore simple: keep the standard hard, but make the path to the standard clearer. Keep correction central, but remove unnecessary shrapnel. Keep distrust of theater, but do not mistake all warmth for theater. Keep demanding reality contact, but create enough trust that people bring reality before it has been cleaned up.

How to read the map

The figures below place every archetype in that three-dimensional space. The upper-right-back region — high on all three axes — is the Bauhütte core: the people and mechanisms that combine reality contact, correction, and translation. Distance from that corner is not a verdict. It is a specification — it tells you what machinery a given collaboration will require.

The first panel shows the three orthographic projections — each flattens one axis so the spread on the other two is unambiguous — alongside an isometric reference. The second is the full interactive score: drag to rotate, scroll to zoom, click any note to inspect it. The accidentals carry the key: natural fits, deliberate interfaces, counterweights.

People who tend to thrive in the Bauhütte

These are the naturals — people whose default relationship to the work already points the right way. They still have watch-outs; fit is not the same as finished.

People who require a deliberate interface

These are the interfaces — not bad people, but styles that, around a leader like me, need explicit design to stay honest. The friction is usually mutual, and usually fixable.

Counterweights the Bauhütte requires

These are the counterweights — the people, roles, and artifacts that make a high standard survivable and usable. A Bauhütte is not built from naturals alone; it is built from the counterweights that catch what a severe leader would otherwise miss, overread, or break.

The standard, restated

None of this is a character verdict, and none of it is an argument for being harder on people — it is closer to the opposite. These names describe failure modes, strengths, and collaboration states; they are not souls. The reason to name interfaces, build counterweights, and design paths for dissent is precisely so that high standards stop depending on anyone’s stamina, mood, or charisma — mine least of all. A Bauhütte run well does not need its master present for the truth to surface. If the master must be present, the standard is still a personality; if bad news waits for his mood, the work is still court politics; if approval and censure still orbit his face, the object has not yet taken authority.

The standard is not comfort. It is not severity. It is disciplined, frictional contact with reality — held steadily enough, and translated well enough, that other people can join it and the work can actually move the weight.