The first writing was not poetry, scripture, or law. It was accounting. Sumerian temple administrators at Uruk, around 3200 BCE, needed to track sheep deliveries, beer rations, and barley quotas across an economy far too large to hold in any one head. They began, centuries earlier, by sealing little clay tokens — a cone for grain, a sphere for oil — inside hollow clay envelopes; then they pressed the tokens' shapes onto the outside so the contents could be read without breaking the seal; then the impressed marks broke free of the tokens altogether, became abstract signs scratched with a reed stylus, and could stand for any noun, then for the sounds of words, then for anything sayable. By 2500 BCE this cuneiform could record a love poem, a schoolboy's complaint, a lawsuit against a fraudulent copper merchant, or the Epic of Gilgamesh.
The deeper consequence of writing was not literature; it was time. A purely oral culture lives within the memory of its oldest member — perhaps eighty years, and faltering at the end. A literate culture remembers as far back as its oldest legible tablet, and Sumerian tablets are still being deciphered more than four thousand years on. The change runs deeper than longevity, though. Speech is bound to a speaker and a moment; writing detaches the message from both, letting a command travel across distance and a promise across generations without its author present. Law could now be fixed in a public stele rather than re-negotiated by each generation's elders — the impulse that runs through Ur-Nammu's code and, later, Hammurabi's. Debts could outlive the death of either party, binding heirs to bargains they never struck. A king could issue commands to subjects he would never meet and to successors not yet born; an administrator could audit a granary he had not personally counted, trusting the mark on the tablet over the word of the man before him. Bureaucracy in the strict sense — rule by the written file — begins here, and with it the possibility of an institution that outlasts every person inside it: the temple, the treasury, the state. Writing also armed memory against power, because a record can be checked: what a king once decreed could be produced against him, and the scribe who kept the tablets quietly became indispensable to the throne that depended on them.
Every database row, every cryptographic signature, every blockchain block is a lineal descendant of those clay tablets — marks on a durable surface that let one person commit a fact and a stranger, later, rely on it. The ability to bind the future to the past through inscription is one of the half-dozen technologies that made civilization possible; we have only changed the surface from wet clay to spinning silicon. The same anxieties recur, too: who controls the record, whether it can be forged, how long it will outlast the machines that read it.