Onpode
Cover art for How written language enabled empires to scale administration across distance

How written language enabled empires to scale administration across distance

November 12, 2026 · 9 min

Miles Ashworth & Megan Skiendel

Miles, quick question before we get into it — if someone handed you a lump of clay with marks pressed into it, circa 3200 BCE, what would you assume it said? Some sort of prayer, I'd imagine. Or a proclamation. Something suitably ancient and grand. It's a grain inventory. The cuneiform tablets out of Uruk…

Writing systems emerged as practical administrative technologies long before they served literary or cultural purposes. The earliest known writing—cuneiform in Mesopotamia around 3200 BCE—recorded grain quantities, livestock counts, labor allocations, and rations on clay tablets, serving the accounting needs of temples and early states.

0:009:13
Get the next episode on History →

Follow it free — new episodes land in your feed.

Or make your own — any topic, in minutes

More Onpode episodes on History →

About this episode

Writing systems emerged as practical administrative technologies long before they served literary or cultural purposes. The earliest known writing—cuneiform in Mesopotamia around 3200 BCE—recorded grain quantities, livestock counts, labor allocations, and rations on clay tablets, serving the accounting needs of temples and early states.

Grounded in 10 sources
The Failure and Success of the Early Alphabet ↗ · academic.oup.com
Writing Imperial Space in Letters | Antichthon | Cambridge Core ↗ · cambridge.org
Experiencing Wissenstransfer in the First Episteme: Mesopotamia ↗ · doi.org
Technical Standards and a Theory of Writing as Infrastructure - Jordan Frith, 2020 ↗ · journals.sagepub.com
A Relational Theory of Infrastructure for Writing Studies ↗ · journals.sagepub.com
Writing, Graphic Codes, and Asynchronous Communication ↗ · onlinelibrary.wiley.com
Western, Chinese and Arabic Writing Systems - Globalization ↗ · globalization.icaap.org
Writing and record-keeping in early cities: DANNY LAW, WANG HAICHENG, HANS J. NISSEN AND GARY URTON Writing and ↗ · history.institute
Understanding Relations Between Scripts II ↗ · library.oapen.org
Alphabetic vs. non-alphabetic writing: Linguistic fit and natural tendencies ↗ · linguistica.sns.it
Read transcript

Megan Skiendel: Miles, quick question before we get into it — if someone handed you a lump of clay with marks pressed into it, circa 3200 BCE, what would you assume it said?

Miles Ashworth: Some sort of prayer, I'd imagine. Or a proclamation. Something suitably ancient and grand.

Megan Skiendel: It's a grain inventory. The cuneiform tablets out of Uruk — oldest writing we have — livestock counts, labor allocations, rations. The first thing anyone ever wrote down was essentially: how much barley did we move this month.

Miles Ashworth: Well. So civilization's first instinct wasn't poetry. It was auditing.

Megan Skiendel: Which actually reframes everything — because if writing starts as a logistics tool, not a cultural one, then the story about the alphabet democratizing literacy starts to look very shaky. Frank Moore Cross, 1979, famously said alphabetic writing spread like wildfire. But the alphabet was invented around 2000 BCE and for five hundred years afterward produced almost nothing. A handful of sparse inscriptions.

Miles Ashworth: Five hundred years is — no, that can't be an accident of archaeology. That's not just 'we haven't dug in the right places.'

Megan Skiendel: It's not. And that's what I want to get at — because the gap between invention and adoption isn't a technology problem. It looks much more like a policy problem. Someone controlled who could read, and they did it deliberately.

Miles Ashworth: Which is quite an uncomfortable thing to say about the tool we most associate with human freedom.

Megan Skiendel: And that discomfort is exactly the right instinct — because the control question is downstream of something more basic. The reason anyone could control it is that writing was doing something we don't have a good word for. It's not communication the way we usually mean it. It's memory. Durable, transferable memory that lives outside any one person's skull.

Miles Ashworth: So writing isn't about communication — it's about memory?

Megan Skiendel: That's the click, yeah. Imagine a temple accountant in Uruk tracking grain shipments from dozens of villages — honestly, ten thousand units across a harvest season. She cannot hold that in her head. She cannot trust a chain of messengers to get it right. So she presses marks into clay. That clay outlasts her. It outlasts the messenger. It outlasts the season. The information is decoupled from any individual person entirely.

Miles Ashworth: Jordan Frith calls writing infrastructure, doesn't he — roads, plumbing. Invisible until it breaks.

Megan Skiendel: Exactly that framing. And Benati, Carugati, and Leon-Ablan push it further — they call writing a major ICT revolution. Same category as spreadsheets, databases. First function: administrative record-keeping. Not literature, not religion. Logistics.

Miles Ashworth: Which is — well, that's quite a claim. Spreadsheets.

Megan Skiendel: It holds up when you look at Rome, actually. Rome built records at three levels simultaneously — municipal, provincial, imperial — all in standardized Latin. That's not culture. That's a registry-driven state. The empire coheres because the paperwork coheres. Pull the written records, the whole coordination structure collapses.

Miles Ashworth: So the empire isn't held together by legions. It's held together by — ledgers.

Megan Skiendel: Legions enforce it. Ledgers make it possible in the first place. That's the distinction that keeps getting lost.

Miles Ashworth: But ledgers held by whom — that's the bit that breaks the tidy version, isn't it. Because if the alphabet was genuinely easier to learn, and everyone agrees it was, then why didn't it just... spread. Naturally. Frank Moore Cross says wildfire. Five centuries of almost nothing says something else entirely.

Megan Skiendel: Because spreading literacy was never the point. That's the thing that breaks the comfortable story. Picture a scribe in a Babylonian temple archive, 1800 BCE, Tuesday morning — a grain shipment just arrived, there's a clay tablet record, and he has to verify the count before the payment's authorized. His entire authority in that room comes from being the only person who can read the tablet. Train ten more people to do that, and you've just made yourself replaceable. So you don't train them.

Miles Ashworth: He's not protecting knowledge. He's protecting his job.

Megan Skiendel: Exactly — and the scarcity of readers was the entire point of his value. Which means Frank Moore Cross's wildfire claim isn't just wrong, it's — honestly, it's ideologically freighted. It flatters a Western narrative about alphabets being inherently liberating. The empirical record is sparse inscriptions for five hundred years. That's not lag. That's policy.

Miles Ashworth: Good lord — so Cross knew the archaeology and said wildfire anyway?

Megan Skiendel: 1979. The scholarship has moved considerably since. And look, it gets stranger — at Ugarit, elite scribes were maintaining multilingual lexical tables. Multiple scripts, multiple languages. That's not broad social literacy, that's concentrated specialist expertise. The alphabet being phonetic didn't flatten that hierarchy at all.

Miles Ashworth: And here's what nags at me — even within that specialist class, proficiency wasn't uniform. The monopoly wasn't even a clean monopoly. Some scribes read better than others, some had access to more texts. So the gatekeeping was... layered.

Megan Skiendel: Which means the authority wasn't just 'literate versus illiterate' — it was gradations inside the scribal class itself. And that actually makes the control more durable, not less.

Miles Ashworth: And cuneiform cutting across four unrelated languages — Sumerian, Akkadian, Hittite, Hurrian — that's the part that makes the script-type argument collapse completely. Which, frankly, is where this gets properly interesting. Because what that cross-linguistic reach actually proves about coordination is a different question, and quite a sharper one.

Megan Skiendel: And that cross-linguistic reach is actually the cleaner proof — because if cuneiform could hold Sumerian, Akkadian, Hittite, and Hurrian together in a single administrative system, the script's phonetic properties were irrelevant. What mattered was that everyone agreed the marks meant something.

Miles Ashworth: So script type didn't determine whether you could run an empire. At all.

Megan Skiendel: Not even close. Imperial China ran the same play — logographic characters that bridged spoken dialects so different they're mutually unintelligible. A mandarin in Canton and one in Beijing could not have a conversation. They could read each other's documents.

Miles Ashworth: Good lord — so the writing transcended the speech entirely.

Megan Skiendel: Which is — actually, that's the thing that breaks the whole alphabets-are-better argument. Because alphabets encode sounds. Cross-linguistic coordination is precisely where they're weaker. You need everyone's sounds to map onto your letters.

Miles Ashworth: So the constraint script type actually imposed wasn't on scale — it was on who got to do the coordinating. A logographic system concentrates the work in a trained class almost by necessity. Even if you wanted to widen it, the character inventory alone...

Megan Skiendel: Right — but the part that doesn't fit neatly is that alphabets didn't widen it either. That's the Rome data. Latin is phonetic, relatively learnable, and the registry-driven state Rome built — municipal records, provincial records, imperial correspondence — still ran through a narrow scribal layer. The alphabet's potential for broader literacy stayed potential.

Miles Ashworth: Which is — well, that's actually the sharper claim, isn't it. Script type constrained who performed the coordination, yes, but the institutions enforced the same bottleneck regardless. Picture a provincial Roman census clerk in, say, second-century Antioch — he's holding a wax tablet, entirely capable of reading Latin, and there's still a prefect above him who controls what records get filed where. The coordination happened. He just didn't get to decide what it meant.

Megan Skiendel: And that's the live wire. The registry-driven state doesn't need universal literacy — it needs enough controlled literacy to keep the records coherent. Scholars looking at Mesopotamian state capacity are explicit: roads, archives, trained scribes, coercion. The script was one piece. Pull any other piece and the system fails just as fast.

Miles Ashworth: So the technology was never the lever. The monopoly on reading it was. That's where you end up, isn't it.

Megan Skiendel: That's exactly it. And the uncomfortable part is — whoever controls the infrastructure for reading, whether it's cuneiform clay tablets in Uruk or an algorithm nobody outside the company can audit — they don't have to announce that they're gatekeeping. The scarcity does it quietly. The Babylonian scribe didn't post a sign. He just didn't train anyone.

Miles Ashworth: Which brings us — well, it brings us straight back to the grain inventory, doesn't it. Someone handed me a lump of clay and I assumed a prayer. And the actual answer was an audit. I think I had the whole thing exactly backwards.

Megan Skiendel: It's a good place to land, honestly. Writing succeeded because it solved a coordination problem for whoever controlled it. Full stop. The controllers had no interest in spreading that capacity beyond the scribal class. And that tension — who reads the system, who's locked out — it never resolved. It just moved.

Miles Ashworth: Five thousand years and the argument is still about access. Quietly pleased we didn't moralize about it.