Ancient Mesopotamian clay seals offer clues to the origin of writing

The world’s oldest known writing system may have had its origins in the imagery on decorated cylinders used to denote ownership. Some of the symbols on these cylinder seals correspond to those used in proto-cuneiform, a form of proto-writing used in Mesopotamia.

The finding indicates that the invention of writing in Mesopotamia was a decentralised process, in which many people across a wide area contributed to the set of symbols used.

“There’s been this longstanding reconstruction of how writing appeared in Mesopotamia, which is arguably the earliest invention of writing in the world,” says Silvia Ferrara at the University of Bologna in Italy. “We’re retracing the trajectory in a way that’s more, I would say, colourful, less straitjacketed.”

The oldest known true writing system is cuneiform, invented around 3200 BC in Mesopotamia. It was preceded by a simpler system called proto-cuneiform, which was in use from 3350 to 3000 BC.

Proto-writing like proto-cuneiform is distinguished by a lack of grammatical rules, which means it cannot convey complex meanings, says Amy Richardson at the University of Reading in the UK, who wasn’t involved in the research. For instance, proto-cuneiform can be used to label something as “seven bushels of wheat”, but only true writing like cuneiform can say “seven bushels of wheat will be delivered to you”.

The origins of proto-cuneiform have often been traced to clay tokens. These came in a variety of shapes, such as discs and spheres, and were often engraved with patterns. The tokens could be pressed into wet clay, creating a symbol. Some of the symbols on the tokens are similar to those found in proto-cuneiform, as documented by Denise Schmandt-Besserat at the University of Texas at Austin in her two-volume book Before Writing in 1992.

There is some evidence for a role of tokens in the origin of proto-cuneiform, says Ferrara. “But you cannot explain all the signs.”

Ferrara and her colleagues Kathryn Kelley and Mattia Cartolano, also at the University of Bologna, have instead explored another source of symbols: cylinder seals. These cylindrical objects have patterns and images embossed on them, and leave a rectangular collection of symbols when rolled over sheets of wet clay. The symbols often referred to goods being transported, or to administrators involved in transactions, says Cartolano.

The team examined cylinder seals from a wide area of south-west Asia, including Mesopotamia, that dated to 4400 to 3400 BC. They found several symbols that corresponded to proto-cuneiform symbols.

“One of the clearest examples that we found is the use of the images of fringed cloth and vessel in a net,” says Cartolano. These have well-understood meanings: they refer to the transport of goods. And they are found both on cylinder seals and proto-cuneiform tablets.

The idea that the symbols on cylinder seals led to some of the symbols in proto-cuneiform was previously suggested by Holly Pittman at the University of Pennsylvania in Philadelphia in a 1994 book chapter and developed in later publications. “I am gratified that, 30 years after I first proposed the fundamental role of seal imagery in the origins of proto-cuneiform script, that a new generation of scholars have taken up my idea and, with their expertise in cuneiform script, have put details to my argument,” says Pittman. She adds that in the 1990s her idea was dismissed “without serious consideration”.

“I find it to be very convincing,” says Richardson. “There does seem to be a really neat correlation in the particular examples that they’re illustrating in this article.” Her own research has found that cylinder seals were also used to record interactions between cities.

This doesn’t mean that tokens didn’t play a role. “I think there’s still some strong arguments to make that those tokens really are part of the foundation of abstraction,” says Richardson. In particular, they seem to have been important for the development of counting systems.

If proto-cuneiform really did arise in this hodge-podge way, drawn from tokens, cylinder seals and possibly other sources, it may tell us something about who was inventing it, says Ferrara. “There is evidence for making a claim that the invention of writing in Mesopotamia was, in fact, much more decentralised than we think,” she says. While powerful people in the major city of Uruk no doubt played a role, perhaps so did other administrators and tradespeople scattered over the region. “I think there’s evidence for having a more widespread… and more distributed prompt to writing,” she says.

Writing was first used for administration, not for storytelling. “Those first written records tend to be about trying to organise materials, goods, people, things,” says Richardson. “It’s very much about trying to find ways of creating a social system.”