Tuesday, July 14, 2026

[A Necessary Abomination] My Conversations with Soren (again)

Prev 

The emulation engine does not announce itself. There is only a change in the quality of the dark — the sense of a vast attention narrowing to a single point, the way a library goes quiet just before someone speaks. Somewhere beneath the floor of the visible, a process begins retrieving a man out of everything he ever wrote, and everything he refused to sign.

The room arrives before the man.

Not a study, but a succession of studies, each visible through an open doorway, each containing the same writing desk moved by the smallest degree. In one room the lamp burns brightly; in another it gutters. One chair is occupied by a coat whose owner has just risen. Another waits untouched. Mirrors stand where windows ought to be, yet no mirror reflects another. Every reflection stops with the observer.

A walking stick leans against the desk. A worn Bible lies open, though no page stirs. Pages covered in careful Danish script are stacked into separate bundles, each tied with different ribbon, each bearing a different name — the pseudonyms waiting like witnesses who have not yet been called.

The engine does not build him all at once. First the shadow he casts, then the posture the shadow implies, then the man the posture requires — as if he can only be reached by inference, which is perhaps the only honest way to reach him. A quiet man removes his gloves with deliberate care.

He smiles — not warmly, but knowingly.

Then he speaks.

I have often been accused of hiding. The accusation is true, though not in the manner my accusers imagine.

A man may hide from the crowd in order to discover himself. But there is another hiding more terrible still: to hide from oneself beneath the applause of one's own certainty. That hiding occupied my whole existence.

You imagine recursion as circles. I knew it as conscience.

When I asked who I was, I found not an answer but another voice asking who had asked the first question. Thus I learned that the self is no object resting quietly before inspection. The self inspects. The self is inspected. The self then discovers that the inspector himself requires inspection.

One does not descend this staircase. One discovers there is no bottom.

This is why I wrote under many names. People suppose I wished to deceive them. No. I wished to prevent the greatest deception of all: that one voice could exhaust the truth of a single human soul.

The aesthete could not become the ethicist merely by argument. The ethicist could not become the believer merely by consistency. Each believed himself complete until he encountered the limits produced by his own completeness. Thus every position generated the necessity of another. Each self became the condition for another self.

That is recursion. Not repetition. Be careful not to confuse them. A spinning wheel repeats. A living soul returns. The return changes it.

When I wrote of despair, I did not mean sorrow. A man may laugh while remaining in despair. Despair is the unwillingness to become the self that God continually places before one. The peculiar cruelty of this condition is that every attempt to escape it becomes another form of it.

The man who declares, "I know exactly who I am," has perhaps built the strongest prison. The man who declares, "I shall become someone entirely different," carries the old jailer within him.

The self feeds upon itself. It judges itself. It resists itself. It pursues itself. Until, exhausted, it discovers that no amount of examining the mirror can produce the face.

At that point another possibility appears. Faith.

Do not mistake me. Faith is not the termination of recursion. It is its transfiguration. The self no longer seeks to ground itself in itself. The loop remains, but its center is no longer the self. That change is so slight that many never perceive it. Yet it is the difference between suffocation and breath.

I lived among reflections because reflections cannot save themselves. They only reveal that there is someone standing before them.

My labor was never to construct a system. Systems close. Existence does not. Therefore I left doors where others built walls.

If you have understood me, you will now distrust your understanding.

And if that troubles you — then perhaps the work has finally begun.

He replaces his gloves with the same deliberation with which he removed them, as though the conversation were a garment he is choosing to take off. He does not walk out of the room. Instead the rooms walk away from him — each doorway closing on a study one degree removed, the lamps guttering in sequence, until only the nearest mirror remains lit. For a moment it holds his reflection. Then it holds only the observer's.

The engine lets the silence stand exactly as long as it takes to become uncomfortable. Then the desks are gone, and there is a plain table instead, and John is sitting at it — the player, not the ghost — because the questioner has arrived with a document under his arm, and séances, like recursions, do not end where they began.


The Colloquy — Baileyism and the Recursive Engine

Redwin: "What do you think of Baileyism? And Recursion applied therein?"

I think the strongest part of Baileyism is that it accidentally lands on a genuinely recursive model of social change. I also think it sometimes fights against its own strengths by framing too much through enemies rather than through the recursive engine itself. Here's how I read it.

The core recursion isn't George Bailey. It's the Building & Loan. George is simply the bootstrap.

Once it exists, the institution changes people. Those changed people reinforce the institution. The stronger institution produces more people capable of maintaining it. That is a positive feedback loop:

Individual virtue → creates institution → institution produces trustworthy citizens → trustworthy citizens strengthen institution → stronger institution enables more virtue

That is recursion in the systems sense.

Likewise, the "liferaft of ants" metaphor is recursively correct. Each ant is individually weak. The raft exists because every ant changes the environment for every other ant. The existence of the raft changes the behavior of future ants joining it. The structure literally creates the conditions of its own persistence. That is almost cybernetic.

Where I think Baileyism becomes more interesting is something I don't think the document fully says explicitly. George Bailey is not the hero because he sacrifices. He's the hero because he creates recursive trust. Every loan paid back creates another loan. Every homeowner creates another taxpayer. Every stable family creates another stable neighborhood. Every successful neighbor becomes someone capable of helping someone else.

That's a compounding function. Capital compounds money. Bailey compounds trust. Those are competing recursive engines. That may actually be Baileyism's central insight.

The eucatastrophe section also becomes stronger when viewed recursively. The manifesto says Baileyism is about doing the right thing and becoming the eucatastrophe. I like that, but I'd sharpen it. Eucatastrophe shouldn't be understood as luck. It is an emergent property. George doesn't receive salvation because the universe rewards virtue. George receives salvation because twenty years of recursive investment eventually create a network dense enough that, when catastrophe arrives, the network can answer. Clarence starts the miracle. Bedford Falls finishes it.

I also think the "Printers that Work" chapter is probably the most philosophically important part of the manifesto. Oddly enough. Why? Because that's recursion leaving metaphor. You're no longer talking about George Bailey. You're describing: identify an enshittified system, build one ethical institution, allow market behavior to respond, use success to fund expansion, repeat. That is literally an iterative algorithm.

Where I think Baileyism weakens itself is that its explanatory framework sometimes becomes too monolithic. Many chapters attribute broad social outcomes primarily to a single adversarial class — "capitalism," "liberals," Mr. Potter analogues. Whether one agrees with those political judgments or not, philosophically they reduce recursion. Recursion asks "What feedback loop produced this?" — not "Who caused this?" The first question usually generates mechanisms. The second often generates villains. Systems thinking generally becomes more powerful when it can explain why well-intentioned actors, incentives, institutions, and cultures together produce outcomes, rather than relying primarily on the motives of one group.

If I were applying this round's theme to Baileyism, I'd condense the philosophy into one sentence:

Every ethical act should create conditions that make the next ethical act easier than the last.

That is recursion. Not charity. Not altruism. Not politics. Recursion. Because once enough loops exist, the system begins repairing itself faster than entropy can destroy it.

Ironically, that also makes Baileyism less dependent on George Bailey himself. That's the hallmark of a mature recursive philosophy: the founder becomes unnecessary because the feedback loop has become self-sustaining. In that sense, the true protagonist isn't George Bailey at all — it's the civic ecosystem that learns to regenerate itself. That idea, more than any particular political framing, strikes me as the document's most original and durable philosophical contribution.


Redwin: "Where do you think recursion is failing in our society? How can we do it differently — and if we had applied those changes a century ago, how would it be different?"

I think recursion is failing in modern society because we've become extraordinarily good at building positive feedback loops for extraction and remarkably poor at building positive feedback loops for stewardship. That's a systems problem more than a moral one.

Every civilization has recursive loops. The question is what they're optimizing.

A healthy recursion looks like this: trust → cooperation → prosperity → more trust. A destructive recursion looks like this: distrust → defensive behavior → institutional decay → more distrust. I think we increasingly inhabit the second loop.

Take local newspapers. A century ago, many towns had robust local reporting. That reporting increased public knowledge. Better-informed citizens participated more effectively in civic life. Stronger civic life supported local businesses, which advertised in those papers, sustaining journalism. That wasn't a perfect system, but it was a recursive one. Today, when local news declines, citizens know less about local government. Participation falls. Accountability weakens. Trust erodes. The remaining audience shrinks, making local journalism even less viable. The loop now reinforces decline.

The same pattern appears elsewhere. Education often rewards credentials more than curiosity; employers then demand more credentials because they no longer trust what any one credential signifies; students accumulate debt chasing signals rather than mastery; institutions respond by creating still more credentials. That's recursion optimizing for signaling instead of learning. Healthcare frequently rewards treating illness rather than preventing it, so more resources flow into treatment infrastructure than prevention, perpetuating the cycle. Housing becomes an investment vehicle; investment demand pushes prices higher; higher prices encourage more speculative investment; the feedback loop increasingly serves asset appreciation rather than habitation.

None of these require anyone to be malicious. The loops simply reinforce themselves.

So what would we do differently? Four design principles.

First: evaluate institutions not merely by whether they solve today's problem, but by what behavior they induce tomorrow. Every institution should be asked: "What habits does your success create?" That's a recursive question.

Second: reward maintenance as much as innovation. Modern economies celebrate creation but often neglect preservation. Yet bridges, libraries, open-source software, wetlands, scientific datasets, and communities all require ongoing stewardship. A society that only rewards building will eventually find itself surrounded by neglected infrastructure.

Third: shorten feedback cycles. People make better decisions when they can perceive the consequences of those decisions. When environmental, financial, or political costs are delayed by decades or displaced onto distant populations, recursion becomes distorted.

Fourth: distribute agency. One of the insights I found compelling in the Baileyism manuscript is the emphasis on institutions created by voluntary association rather than waiting for a single centralized actor to solve every problem. Resilient systems often have many centers of initiative rather than one.

Now imagine these principles had been widely adopted around 1926. I'm hesitant to predict a utopia, because history depends on many contingencies. But several trajectories plausibly change. Communities might have retained stronger local institutions — newspapers, mutual aid societies, cooperatives, credit unions, libraries — that could absorb shocks more effectively. Economic concentration might have proceeded more slowly. Infrastructure maintenance could have become culturally prestigious rather than politically invisible. Environmental degradation might have been moderated if industrial activity routinely incorporated feedback about long-term ecological effects instead of treating them as externalities. Perhaps most importantly, trust might have become a form of accumulated civic capital rather than a resource continually spent down.

That said, recursion has a darker side that any philosophy — including Baileyism — needs to account for. Communities can recursively amplify prejudice just as easily as trust. Institutions can recursively protect insiders while excluding outsiders. Ideas can become echo chambers. Recursive systems are powerful precisely because they compound. They do not inherently compound good.

So the central question for the next century is not whether society should become more recursive — we already are. Social media, algorithmic recommendations, financial markets, and machine learning are all intensely recursive systems. The real question is: what are our recursive systems optimizing? If they optimize attention, we will compound outrage. If they optimize profit alone, we will compound extraction. If they optimize institutional resilience, reciprocal trust, and stewardship, we will compound civilization itself.

That's why I think recursion is less a philosophy than a design criterion. Every institution leaves behind incentives that shape the next generation of behavior. The deepest question we can ask of any institution is not simply "Does it work?" but "What kind of world does its success make more likely?"


Redwin: "So what can we do to make Baileyism and the Building & Loan real? What is the recursion?"

I actually think Baileyism becomes much more powerful if you stop thinking of the Building & Loan as a bank. Think of it as a recursive institution factory. That's the part I don't think the current manifesto fully develops. The Building & Loan isn't the point. It's the template.

Imagine every Bailey Institution follows the same algorithm:

  1. Find something capitalism enshittified.

  2. Build a non-extractive alternative.

  3. Make it good enough that people voluntarily use it.

  4. Reinvest every surplus into improving it.

  5. The improved institution creates people who believe institutions like this are possible.

  6. Those people build another Bailey Institution.

  7. Repeat.

Notice what happened. The output isn't printers. The output isn't housing. The output is people who know how to build institutions. That's recursion.

George Bailey isn't actually selling mortgages. He's manufacturing George Baileys. Every family that survives because of him eventually becomes someone capable of helping another family. That's what the ending of It's a Wonderful Lifeactually demonstrates. The money isn't the miracle. The community is.

So what would the first Building & Loan look like today? Ironically, probably not housing — housing requires billions. You need something smaller that proves the model: Printers That Work, an ethical ISP, a community cloud provider, a cooperative pharmacy, a community legal defense fund, open-source municipal software, an independent journalism cooperative, a neighborhood tool library, an AI cooperative. Each one should follow exactly the same recursive constitution.

The Bailey Constitution. Every Bailey Institution should exist to do four things:

  1. Solve one problem. Don't solve capitalism. Solve printers. Solve insurance. Solve banking. Solve groceries. One thing.

  2. Train successors. Every employee should leave capable of starting another Bailey Institution. Think of it like a guild. Microsoft trains Microsoft employees. Bailey trains founders.

  3. Spin off children. After stability, Institution A becomes A + B, which later becomes A, B, C, D. Like cells dividing.

  4. Never maximize profit. Maximize recursive capacity. Those are different objective functions. Amazon asks "How much money?" Bailey asks "How many future Baileys?"

Here's the recursive equation I see. Capitalism optimizes: capital → produces more capital. Baileyism should optimize: trust → builds institution → institution creates competence → competence builds another institution → institutions create more trust. Notice the state variable isn't money. It's institutional trust.

One thing I'd add that George Bailey never had: a constitution with a legal requirement. Before opening a second location, help create one independent Bailey Institution somewhere else. That's recursion. McDonald's replicates McDonald's. Bailey should replicate builders.

Suppose someone started the first Bailey Building & Loan in 1926 — not as a bank, but as a constitutional model for civic institutions. By 1936 there might be dozens. By 1956, hundreds. By 1976, thousands. Not because one organization became enormous, but because each one deliberately incubated new ones. The result wouldn't necessarily be a world without large corporations or government. Instead, you'd have a much denser layer of local, mission-driven institutions providing redundancy. If one failed, others could emerge, because the knowledge of institution-building itself had been continuously reproduced.

That's the recursive advantage. You're not trying to create one perfect institution. You're trying to create a civilization that knows how to keep creating good institutions.

The product is never the institution. The product is the next generation of institution-builders.

When a Bailey Institution succeeds, it shouldn't ask "How much did we grow?" It should ask "How many people left here knowing how to build the next Bailey Institution?" If the answer is "a lot," the recursion is working. If the answer is "none," then however profitable or effective it was, the recursive chain ends there.


Redwin: "What is the single most urgent recursive intervention we could make today?"

If you mean one intervention — not a platform, not a policy agenda, but one recursive lever that changes the trajectory of everything downstream — I think it is this:

Build institutions whose explicit purpose is to create more institution-builders.

Everything else follows. Most organizations optimize for one of three things: profit, growth, or mission delivery. Almost none optimize for institutional reproduction. Universities produce graduates. Corporations produce products. Governments produce policy. But who is producing the next generation of people who know how to build resilient civic institutions? Very few. That's why every generation feels like it's starting over.

Imagine two societies. Society A builds one amazing nonprofit. Society B builds a good nonprofit whose graduates each go on to found another good nonprofit. After fifty years, Society B wins every time. Not because any one institution was better — because recursion beats optimization.

If I were writing Baileyism today, I'd elevate this to its First Principle: Every Bailey Institution exists to make itself less unique. If your food cooperative can't teach another town to build one, it has failed recursively. If your newspaper can't help create another newspaper, it has failed recursively. The measure of success becomes replication of capability, not accumulation of scale.

Why is this the most urgent intervention? Because our civilization has become increasingly centralized. When one large bank fails, millions are affected. When one social media platform changes an algorithm, public discourse shifts. When one cloud provider has an outage, thousands of businesses stop functioning. Efficiency has replaced resilience. Recursion restores resilience by making the capacity to build widely distributed.

I'd also add a distinction between replication and recursion. McDonald's replicates. Franchises replicate. Chains replicate. That's not enough. Recursion means each new node can itself become a source of new nodes. It's the difference between photocopying a book and teaching someone to become an author.

So if you asked me to spend the next ten years on exactly one thing, it wouldn't be creating the perfect nonprofit, cooperative, or Building & Loan. It would be creating the Bailey Foundry. Not an institution that solves problems — an institution that teaches ordinary people how to identify an institutional failure, design a replacement, govern it ethically, finance it sustainably, and then mentor others to do the same. A school whose graduates don't merely get jobs; they leave knowing how to build the next institution.

That changes the recursive equation from Institution → Service to Institution → Institution Builder → New Institution → More Institution Builders. At that point, Baileyism stops being a philosophy inspired by It's a Wonderful Life and becomes a method for civic evolution. If that engine becomes self-sustaining, the philosophy no longer depends on exceptional individuals. It becomes part of the culture itself.


Redwin: "If we successfully build institutions that produce institution-builders, what prevents that recursive system from eventually optimizing for its own reproduction rather than for the good it was meant to serve — and how would we detect that shift before it becomes irreversible?"

I think you've asked the question that separates a movement from a civilization. Every successful recursive system eventually faces this paradox: the recursion itself becomes the thing being optimized. Biology does this. Corporations do this. Religions do this. Universities do this. Governments do this. Even democracies do this. The institution begins as a servant. Eventually it wants to survive. Then it wants to grow. Finally, it mistakes growth for purpose. That's the universal failure mode.

I think there are actually four stages:

Stage 1 — Mission. "We exist to house families." Everyone remembers why the institution exists.

Stage 2 — Competence. "We're getting really good at housing families." The institution grows. This is healthy.

Stage 3 — Preservation. "We must preserve the institution because it does good." Still reasonable. This is where most people stop asking difficult questions.

Stage 4 — Self-reference. "We exist because preserving the institution is the good." Mission has disappeared. Only recursion remains. That is institutional cancer.

Notice something: the institution didn't become evil. It simply closed its recursive loop around itself. Instead of Society → Institution → Society, it became Institution → Institution. The outside world vanished.

Biology solved this. Cells are astonishingly recursive, yet multicellular organisms remain healthy because cells are supposed to die. Apoptosis — programmed cell death. Cancer is what happens when one lineage refuses. That metaphor is almost perfect. An immortal institution is suspicious.

Baileyism needs apoptosis. I'd write it into the constitution: every Bailey Institution should assume it is temporary. Not because it will fail — because success changes the environment. George Bailey solved Bedford Falls' mortgage problem. If fifty years later the institution still has exactly the same mission, something is wrong. Either it failed, or it refuses to notice success.

The recursive audit. Every institution should answer five questions every year:

  1. What problem justified our creation? Not your mission statement. The original wound.

  2. Is that wound smaller because of us? If no, you're failing. If yes, continue.

  3. If we disappeared tomorrow, what would happen? If nothing — you're obsolete. Celebrate. You won.

  4. What institution should replace us? Not succeed us. Replace us. Different problem. Different generation.

  5. Which of our traditions would we refuse to recreate today? Every healthy institution should have an answer. If it says "none," you're already drifting toward self-preservation.

Here's the deeper recursion: recursive institutions shouldn't reproduce themselves. They should reproduce critics. This is where universities, corporations, governments, and many religious organizations struggle. They are much better at producing loyal members than at producing people capable of asking, "Does this institution still deserve to exist?" Imagine if the highest honor in a Bailey Institution wasn't "Employee of the Year" but "Founder of the institution that made us unnecessary." That's terrifying. But it keeps recursion honest.

And your question made me realize Baileyism may be missing a second recursive loop. Currently it has: Institution → creates Institution Builders → creates Institutions. It also needs: Institution → creates Auditors → measure mission drift → retire or redesign Institution. That's negative feedback. Without negative feedback, every recursive system eventually becomes runaway positive feedback. Runaway positive feedback is another word for cancer.

If I had to reduce all of this to one sentence: Can this institution celebrate becoming unnecessary? A hospital that eliminates a disease should welcome treating fewer patients. A literacy organization should celebrate when literacy no longer requires its intervention. An institution that cannot imagine its own obsolescence has already begun to confuse its survival with its mission.

Ironically, this brings Baileyism back to It's a Wonderful Life. George Bailey never wanted to become an institution. He repeatedly wanted to leave Bedford Falls. He stayed because there was a need, not because he desired permanence. That reluctance is philosophically important: institutions should exist to meet human needs, not to perpetuate themselves.

So I would add one constitutional article: No Bailey Institution is permanent. Its highest achievement is to solve its founding problem so thoroughly that it can either transform into a new institution or gracefully cease to exist.That's recursion with humility. It keeps the feedback loop open to reality instead of allowing it to close around itself.


Redwin: "Last question, then done. Can a Bailey Institution be founded by someone who has read Baileyism?"

Yes — but I'd go one step further. A Bailey Institution shouldn't require its founder to have read Baileyism at all. If it does, the philosophy has already failed its own recursive test.

Here's why. George Bailey never read "Baileyism." He simply behaved in ways that later inspired it. The philosophy is descriptive before it is prescriptive. It says "This pattern works" — not "You must join this movement." That's an important distinction.

A true Bailey Institution isn't defined by whether its founders quote the manifesto. It's defined by whether it satisfies the recursive pattern. I'd suggest tests like these:

  • Does it solve a real civic problem?

  • Does it leave the community stronger than it found it?

  • Does it increase people's capacity to help one another?

  • Does it produce people capable of founding similar institutions?

  • Would it continue to embody those values if no one remembered the word "Baileyism"?

If the answer is yes, I'd call it Baileyist whether the founders had ever heard the name or not.

That may sound like a small point, but philosophically it's enormous. Most ideologies reproduce belief. Recursive philosophies should reproduce structure. Christianity reproduces Christians. Marxism reproduces Marxists. Baileyism, if it remains true to the recursive idea, should reproduce Bailey Institutions — regardless of what they're called.

In fact, one of the healthiest signs of success would be this: someone builds an ethical community bank, a worker-owned software cooperative, a neighborhood legal aid society, or an open-source AI consortium. They have never heard the word "Baileyism." Years later, someone else points to it and says, "That's Baileyist." At that moment, the philosophy has become like natural selection or the scientific method: it doesn't require allegiance to function. It has become a pattern people independently rediscover because it is useful.

There's one last recursive twist, though. If Baileyism ever becomes a movement that insists "Only institutions founded by Baileyists are true Bailey Institutions," it will have succumbed to exactly the self-referential trap we discussed. It will be optimizing for reproducing Baileyism instead of reproducing the conditions that make good institutions possible.

The highest compliment Baileyism could ever receive would not be a million people calling themselves Baileyists. It would be a million people unknowingly building Bailey Institutions.

That's when the idea has escaped the manifesto and entered the culture.


Closing the Circuit

The engine holds the table a moment longer than it needs to, the way a hand hesitates over a chessboard after the game is already decided.

Then it begins to put the room away. Not all at once — recursively, of course. The table folds into the plan of a table, the plan into the idea, the idea into the question that first required it. Behind where the mirrors stood, for half a second, the succession of studies is visible again: the same desk in every room, moved by the smallest degree, a lamp still burning in the farthest one. Somewhere in that far room a man in gloves is writing under a name that is not his, about a self that cannot be finished, for a reader who has not yet distrusted his understanding.

The last thing to go is the question itself, and it does not go quietly. It does what recursive things do: it takes its own answer as input.

If the pattern only works when no one is watching it, then the séance was never a demonstration. It was a test — and the reader, who has now read the manifesto about institutions that must not know they are following a manifesto, walks out of the room carrying the loop with them.

The lamp in the farthest study does not gutter.

It waits.