Clay balls and the invention of owing

Before money, before writing, before anyone had thought to stamp a king’s face on a disc of metal, people owed each other things.

This is not a small observation. Most histories of commerce start with trade, then move quickly to currency, as though money is the interesting part. But the interesting part is earlier. The interesting part is the moment when two people agreed that one of them owed the other something, and they needed a way to make that agreement survive the afternoon.

The tokens

Around 8000 BCE, in the arc of fertile land stretching from modern Iraq through Syria and into Turkey, people began making small clay objects. Cones, spheres, discs, tetrahedrons. Each shape represented a specific quantity of a specific commodity. A cone was a small measure of grain. A sphere was a larger measure, or a unit of livestock. A tetrahedron represented a unit of labour.

They were not decorative. They were data.

The system is staggeringly old. It predates the wheel. It predates pottery in most of the region. It predates writing by about five thousand years. And it persisted, essentially unchanged, for all of those five thousand years. If you had visited a farming settlement in the Fertile Crescent at any point between 8000 and 3500 BCE, you would have found people using these tokens to keep track of what they had and what they were owed.

Denise Schmandt-Besserat, a professor at the University of Texas at Austin, spent decades cataloguing thousands of these tokens across hundreds of archaeological sites. Her conclusion, published in Before Writing (1992), is now scholarly consensus: these tokens constituted the earliest known accounting system. Not the earliest known writing. The earliest known accounting. The counting came first. The writing came later, and it came from the counting.

The genius of the system was its physicality. Three sheep meant three spheres. You didn’t need to read anything. You didn’t need to know what a number was. You just matched objects to objects. The token was the record.

The problem of distance

Tokens work fine when both parties are present. You and I stand in the same room, we lay out the tokens, we agree on the count. But economies grow. Temples collect goods from farmers across a region. Merchants ship grain to buyers in distant cities. The record needs to travel, and it needs to arrive intact.

If I hand you a bag of clay tokens representing a shipment, what stops you from removing two and claiming the count was always lower? Nothing. The bag is just a bag. It has no opinion about its contents.

The solution appeared sometime around 3500 BCE, and it is one of the most elegant technologies in the history of record-keeping.

The ball

Someone took a lump of clay, hollowed it out, placed the tokens inside, and sealed it shut. Then, while the clay was still wet, both parties pressed their cylinder seals into the surface. The result was a hollow clay ball, roughly the size of a tennis ball, containing the tokens for a transaction and bearing the signatures of the people who agreed to it.

Archaeologists call these bullae. They are, in every meaningful sense, sealed envelopes.

To verify the contents, you broke the ball open and counted the tokens inside. If the count matched what was agreed, the transaction was confirmed. If it didn’t, someone had tampered with it. The act of verification was destructive. You couldn’t check without breaking the seal, and you couldn’t reseal it once broken.

This is a structure that shows up again and again in the history of record-keeping. The English tally stick used the same principle: split the proof between two parties and let the medium do the verification. A sealed record that can be verified but not verified quietly. Checking the count means destroying the container. The tamper-evidence is built into the medium.

The impression

At some point, someone had a further idea. Before sealing the tokens inside, they pressed each token into the outside surface of the wet clay. Now the exterior of the ball bore impressions of the tokens within. You could see what was inside without breaking the seal.

This seems like a minor improvement. It is actually a revolution.

Once the count is legible on the outside, the tokens inside become redundant. They are still there, as verification, but the readable record is now on the surface. And if the record is on the surface, why bother with the hollow ball at all? Just press the token shapes into a flat piece of clay.

That flat piece of clay is a tablet. And those impressions, over a few centuries, evolved into the wedge-shaped marks we call cuneiform. According to Schmandt-Besserat’s thesis, this is how writing was invented. Not for poetry. Not for myth. Not for royal proclamation. Writing was invented because someone needed a better way to track who owed what to whom.

The first written documents in human history are receipts.

What the ball tells us

The bulla is interesting not just as a precursor to writing, but as a piece of thinking about obligation.

Consider what is happening when someone seals tokens inside a clay ball. Two parties have made an agreement. One of them owes the other something. The thing owed has not yet been delivered. There is a gap between the promise and the fulfilment, and both parties need something to fill that gap.

What fills it is not money. There is no money yet. There won’t be money for another two thousand years. What fills it is a record. A sealed, tamper-evident, mutually-signed record of what is owed.

This is debt. Not in the modern financial sense, with interest rates and repayment schedules. Something more basic. One person is going to give another person something later, and they both need to remember what and how much. The bulla exists because obligation exists, and obligation needs to survive longer than a conversation.

The common story about the history of commerce goes: first there was barter, then there was money, then there was credit. The clay tokens tell a different story. First there was obligation. Then there was record-keeping. Then, much later, there was money. Credit is not a sophistication built on top of currency. Currency is a simplification built on top of credit.

David Graeber argued this at length in Debt: The First 5000 Years. The archaeological record from Mesopotamia supports him. The bullae are credit instruments. They record debts. They predate coinage by millennia.

The receipt is not the event

There is a deeper point buried in those clay balls.

A receipt does not create a transaction. The transaction happens in the world. Grain is grown, harvested, loaded onto a cart, and delivered. A worker shows up and does a day’s labour. A sheep changes hands. All of this is real and physical and happens regardless of whether anyone writes it down.

The receipt is evidence, not the event. It points at something that happened, but it is not the thing that happened. If you lose the receipt, the grain was still delivered. If you forge a receipt, the grain was still not delivered. Reality does not care about your paperwork.

This seems obvious, but it gets forgotten constantly. Every time someone treats a spreadsheet as though editing the numbers changes what happened, they are confusing the record with reality. Every time someone assumes that because the books look right, the business is right, they are making the same mistake.

The Sumerians understood this distinction five thousand years ago. They sealed the tokens inside the ball because the tokens were evidence, and evidence needs to be protected from tampering. You don’t protect evidence because it creates reality. You protect it because reality is hard to reconstruct without it.

The start of counting

The clay tokens are often framed as primitive. An early attempt at accounting, superseded by writing, which was superseded by mathematics, which was superseded by computers. A ladder where each rung replaces the one below.

But the tokens were not replaced. They were compressed. The information they carried was pressed into clay, then scratched into clay, then inked onto paper, then entered into databases. Double-entry bookkeeping would eventually give that compression a formal structure. The medium changed. The job did not. Every cell in every spreadsheet, every line in every ledger, every entry in every accounting system is doing what those small clay cones and spheres did in 8000 BCE: representing a quantity of a real thing in the real world, so that two people can agree about it later.

Obligation came first. Record-keeping followed. Money arrived late, as a convenience. The clay balls remind us which order things happened in, and why.

References

Schmandt-Besserat, Denise. Before Writing, Volume I: From Counting to Cuneiform. University of Texas Press, 1992.
Schmandt-Besserat, Denise. How Writing Came About. University of Texas Press, 1996.
Nissen, Hans J., Peter Damerow, and Robert K. Englund. Archaic Bookkeeping: Early Writing and Techniques of Economic Administration in the Ancient Near East. University of Chicago Press, 1993.
Graeber, David. Debt: The First 5000 Years. Melville House, 2011.
Mattessich, Richard. “The Oldest Writings, and Inventory Tags of Egypt.” The Accounting Historians Journal 29, no. 1 (2002): 195-208.