Eleanor Crane: I want to start with a confession — I've been mildly irritated by a thing for the past few days, and I think it's actually the whole episode.
Ben Okonkwo: Okay, I'm in — what's the irritant?
Eleanor Crane: The way we talk about writing as this great leveler. Knowledge gets free, the priest loses his grip, the ordinary person gains access. It's a very clean arc. And I keep thinking — wait, is any of that actually what happened? Because Egyptian priests spent decades mastering hieroglyphics. Decades. And that was not a bug, that was how you held power.
Ben Okonkwo: Hm — so the complexity of the script was itself a gatekeeping mechanism.
Eleanor Crane: Exactly that. And then you go back further — back to Uruk, 3400 BCE, the oldest documented writing system we have — and what you find is grain receipts. Livestock tallies. No one sat down to write a poem. They sat down to count sheep, literally, because a state needed records at scale.
Ben Okonkwo: So writing starts as an administrative technology, not a liberating one.
Eleanor Crane: Right — and that one fact, I think, reframes everything. The core idea here is actually simple: writing didn't democratize knowledge, it shattered one monopoly on truth and replaced it with many competing gatekeepers. Before writing, you needed the elder, the memory-keeper, the person who held the law in their head. After writing — well, you needed the scribe. You needed the priest who could read the temple wall. Different lock, same cabinet.
Ben Okonkwo: And those competing gatekeepers — that fragmentation — is actually what makes bureaucratic state power possible. Not despite the complexity, because of it.
Eleanor Crane: That's exactly the thread. And Jack Goody and Ian Watt were pulling at this back in 1963 — this argument that literacy doesn't spread knowledge evenly, it creates a cognitive class. People who can read and write don't just know more. They occupy a structurally different position in relation to what counts as true.
Ben Okonkwo: Which means the question isn't just who controls the text — it's who gets to decide what the text means. And whether that's actually more stable or less stable than the elder with the memory.
Eleanor Crane: And stable is the wrong question, I think — because the oral system wasn't about stability. It was about control through indispensability. The elder wasn't just a library. The elder was the only door.
Ben Okonkwo: Right — but here's where I want to push, because 'distributed' sounds like it means accessible. Like, every village has its own elder, so knowledge is spread out, decentralized. That sounds almost democratic.
Eleanor Crane: Mm. Is that the objection you think the listener is sitting with?
Ben Okonkwo: It's exactly what I'd think if no one corrected me. Distributed feels like — spread the wealth, right? But a monopoly isn't about where the knowledge lives. It's about whether you can reach it without going through a specific person. Those are completely different things.
Eleanor Crane: Give me a scene.
Ben Okonkwo: Okay — Sumerian village, 2500 BCE. Harvest comes in. A family says they owe thirty units of barley. The collector says sixty. There is no tablet, no seal, no document. The only court of appeal is the elder's memory. And the elder's memory — I mean, whether it's accurate or whether it favors whoever the elder owes a favor to — those two things are, functionally, indistinguishable. There's no mechanism to tell them apart.
Eleanor Crane: That's not just inconvenient. That's a complete power structure dressed up as a knowledge system.
Ben Okonkwo: And it's nearly impossible to coordinate across distance, which is actually — wait, that's the other half of it. Oral knowledge is hard to verify even in the room where it's held. Transmit it across twenty miles through a messenger's memory and you have a completely different problem.
Eleanor Crane: The messenger whose entire job was to carry knowledge in their body. They weren't just a courier — they were the record.
Ben Okonkwo: Which is exactly what Jack Goody was tracing — him and Ian Watt, the 1963 paper. Goody's argument is that oral cultures aren't just pre-literate, they're organized differently. He calls what writing produces 'logographic' culture — a cognitive class system where literate elites don't just know more, they inhabit a structurally different relationship to what counts as true. The elder's authority and the scribe's authority are not the same kind of authority.
Eleanor Crane: And Goody is saying that gap — that structural difference — is baked into the technology itself, not just into whoever happens to be holding it.
Ben Okonkwo: That's the load-bearing claim, yeah. Oral knowledge required the person. Writing — eventually — requires the reader. And those are not the same constraint. One pins authority to a body. The other pins it to a credential.
Eleanor Crane: And a credential you can fail to hold. Which is — I think that's actually the thing that doesn't get said enough. Because the earliest writers weren't philosophers. They were accountants. Uruk, 3400 BCE — the tablets we find are livestock tallies. Someone counted goats and pressed the count into clay. That's the birth of writing.
Ben Okonkwo: The first writers were accountants. I mean — when you say it that way, it's almost funny.
Eleanor Crane: It should be. We've spent centuries treating writing as the technology of revelation — and it started as a receipt.
Ben Okonkwo: Okay but — wait, actually, this is where I want to slow down, because the receipt itself is doing something more specific than it looks. Picture a temple administrator in Uruk, 3200 BCE. She has grain coming in from six villages. She can't be in six villages. And the messengers' counts don't match. So the tablet isn't just a record — it's a way of making her authority present without her body being present. That's the rupture, right there.
Eleanor Crane: Which is exactly what the cylinder seals are solving for, a few decades later — 3200 down to 2750 BCE, roughly. Someone realized you could press a carved stone into soft clay, and the impression proved the transaction was authorized, even when the authorizing person was five miles away and three days gone.
Ben Okonkwo: Hold on — so verification before literacy, almost. You don't need to read the tablet. You need to recognize the seal.
Eleanor Crane: You need to recognize the seal. Which means what's being transmitted isn't knowledge exactly — it's institutional identity. 'This came from someone with the authority to make this claim.' That's the ancestor of the notary stamp, the watermark, the digital certificate.
Ben Okonkwo: And it fractures the monopoly — not because more people can read, but because a single keeper can no longer control whether anyone verifies what they say. The check exists outside the body. Hm. Plato hated this.
Eleanor Crane: Plato genuinely hated this.
Ben Okonkwo: No, seriously — his argument in the Phaedrus is that writing would erode natural memory. His phrase is something like: it produces the appearance of knowledge without its substance. You think you know because you can point to a text. But you haven't held it. And Eric Havelock, building on that later, argues writing didn't just change what Greeks knew — it reorganized how Greek cognitive culture worked. The whole claim structure of a literate mind is different from an oral one.
Eleanor Crane: Now, Plato's not wrong about what was lost — but I'd want to ask: lost for whom? Because the elder whose authority lived in memory — that person is facing obsolescence. The scribe has a new monopoly. Writing didn't remove the cognitive class, it issued new credentials to a different class.
Ben Okonkwo: Which is Goody and Watt's whole architecture. Literate elites don't just know more — they control what counts as knowable. And the part that comes next, what the state actually gained from this — the standardization, what happens when you can reproduce the same disagreement in writing across three provinces simultaneously — it gets genuinely strange.
Eleanor Crane: Different lock. Much bigger cabinet.
Ben Okonkwo: And that cabinet gets a lot bigger once you can standardize. Like — same tax levy, same troop conscription order, transmitted without distortion across, I don't know, three hundred miles. That's not just convenient. That's a different kind of state.
Eleanor Crane: A king who has never met you can govern you. Because the text travels for him.
Ben Okonkwo: Right — and that's the mechanical basis of bureaucratic state power. Knowledge standardization across distance. The same law, word for word, arrives in a village the king has never visited.
Eleanor Crane: And this is where I want to hold two positions honestly, because there's a real argument on both sides. Writing democratized knowledge — by making the same information independently verifiable, it shifted authority from persons to texts. The priest no longer has exclusive custody of what the gods supposedly said. Multiple parties can now hold and contest it. That's the emancipation argument.
Ben Okonkwo: But then — actually, no, walk me through the counter first. Because I can feel the 'and yet' coming.
Eleanor Crane: The counter is that literacy itself became the new restricted credential. Standardization served state power as much as it served any individual freedom. You still couldn't read the tablet without training. And that training — I mean, who controlled access to it?
Ben Okonkwo: The scribal class.
Eleanor Crane: So Walter Ong's point in 'Orality and Literacy' is that there's no clean replacement. Orality and literacy stay entangled — the two modes keep operating simultaneously, which is why the honest answer to 'did writing democratize or control' is neither. What it actually created was reproducible disagreement.
Ben Okonkwo: Reproducible disagreement — say more on that.
Eleanor Crane: Multiple parties can hold the same text. Contest it. Cite the same source and reach opposite conclusions. That's not emancipation and it's not pure control — it's fragmented authority. Which is, I'd argue, stranger and more consequential than either.
Ben Okonkwo: Okay — and this is where I want to deploy what I think is the most interesting example of exactly that. Hangul.
Ben Okonkwo: 15th century Korea. The Chinese writing system is the prestige system — dense, requires years of elite education, functions as a credential gate. So the Korean court creates Hangul, a vernacular script, deliberately engineered to be easy to learn. Not evolved over centuries — invented. Specifically to break the Chinese-educated elite's grip on textual authority.
Eleanor Crane: Wait — so writing was used as a weapon against an existing scribal monopoly. By a state.
Ben Okonkwo: By a state. Which means writing's effects aren't determined by writing itself — they're determined by who designs it, and what existing monopoly they're trying to fracture. Hangul is the state using literacy as political resistance.
Eleanor Crane: So the technology is genuinely neutral — and I mean that in the uncomfortable sense, not the flattering one. It can serve the tax collector and the village pushing back against him. The same clay, different hands.
Ben Okonkwo: Which is the thing Goody and Watt's framework almost misses — they're so focused on what literate elites do with writing that the case where writing dismantles a literate elite barely fits the model. Hangul shouldn't exist, in their framework. And it does.
Eleanor Crane: And I think that's actually where the whole episode lands for me — not that writing freed knowledge or controlled it, but that it made knowledge portable. Reproducible. And the moment you can reproduce something, you can reproduce the disagreement about it. Which means no single gatekeeper can hold it anymore. You get many gatekeepers, competing.
Ben Okonkwo: Fragmented, not dissolved.
Eleanor Crane: Fragmented, not dissolved. And we've been running that dynamic ever since Uruk, 3400 BCE — every time a technology makes knowledge more portable, more reproducible, we re-run the same competition. Different players, same structure.
Ben Okonkwo: Which is — I mean, I don't want to overstate it, but that does make the question about large language models feel less like a new problem and more like... the same clay, again. Do they create a new monopoly over truth-generation, or do they just accelerate the fragmentation that started at Uruk? I genuinely don't know the answer. But I think the framing is right.
Eleanor Crane: The framing is exactly right. And that's actually enough — that's the modest thing I'll take away. Not a verdict, just a shape. The first clay tablet pressed in southern Mesopotamia wasn't liberation and it wasn't capture. It was the opening move in a competition that never closed.
Ben Okonkwo: Good place to press stop, I think. This one kept me honest.
