Reading track · 53,971 tablets

The Ledger

Reading Mesopotamia as a book of accounts — the axis on which writing itself was invented.

Writing was not invented to record poetry, prayer, or the names of kings. It was invented, on the evidence we have, to count things: sheep, grain, beer, textiles, workers, days. The oldest readable documents in human history — the archaic tablets of Uruk, from around 3300 BCE — are receipts and allocation records, and on one influential account (Denise Schmandt-Besserat's), even they descend from something older still: small clay tokens sealed inside clay envelopes, each token standing for a commodity, until someone realized the marks pressed on the envelope's surface made the tokens inside redundant.

That origin never stopped shaping the corpus. Of the roughly half-million cuneiform documents recovered to date, the overwhelming majority are administrative — by most estimates well over three quarters. This is not a defect of preservation; it is what these societies used writing for. A temple in Early Dynastic Girsu, a royal depot under the Third Dynasty of Ur, a merchant family in Old Assyrian Kanesh, and the Eanna temple of Neo-Babylonian Uruk all kept books, and the books survive where the buildings do not.

Reading the economic corpus in order is therefore the closest thing we have to watching an economy invent itself: measurement standardized, labor priced in barley and silver, credit extended against harvests, interest calculated, audits performed, deficits carried forward from one accounting year to the next. Most of these texts were written to be read once and discarded. That is precisely their value — no genre of tablet lies less.

Anchor tablets below are selected automatically from the corpus — the richest readable witnesses of this subject in each era — and new ones surface as the translation engine works through the backlog. Every translation is labeled with its source; engine translations carry their confidence level on the tablet page.

4000 – 3100 BCE

Uruk Period

The very beginning: numerical tablets and proto-cuneiform allocation records from Uruk itself, many still only partly legible. The famous signature-like sign group read as Kushim — possibly the earliest personal name in recorded history, an accountant — belongs to this world.

~3300 BCE · VAT 14942 — Vorderasiatisches Museum, Berlin, Germany

ATU 5, pl. 013, W 6710,a

“1. 1(N14) 3(N01), [commodity-sign ZATU678] — male workers (ERIM~a) 2. 1(N14), [commodity-sign ZATU678] — [sign ZATU765] 3. 5(N01), [sign ZATU718?] 4. 5(N01), woman/female (SAL) 5. 5(N01), PAP~a (supervisory/summary marker) 6. 1(N14), copper(?) (URUDU~a?) 7. 4(N14) 8(N01), [total, no commodity given]”

Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved)

Read the full tablet entry

~3300 BCE·Uruk PeriodOur engine

ATU 5, pl. 071, W 9579,cc

Obverse: 2(N01), AB2 NUN~a ZATU639 LAM~b NIMGIR GU4 (2 units: cow [qualified by] NUN~a, ZATU639, LAM~b, herald, ox) 1(N01), AB2 X BA# [...] (1 unit: cow [qualified by] the unread sign X; disbursed(?) — remainder broken) 1(N01), AB2 SZE~b UNUG~a (1 unit: cow [qualified by] barley, Uruk) 4(N01), AB2# GU4 ZATU639 UNUG~a (4 units: cow(?), ox, ZATU639, Uruk)

ATU 5, pl. 071, W 9579,cc

ATU 5, pl. 077, W 9579,dv

ATU 5, pl. 089, W 9656,ab

FMB 27

Nisaba 25, 01

Nisaba 25, 02

AAICAB 1/3, pl. 245, Bod S 295

AAICAB 1/4, Bod S 340

AAS 009

AAICAB 1/1, pl. 059, 1924-0553

AAICAB 1/2, pl. 093, 1935-541

AAICAB 1/1, pl. 055, 1923-427

TLB 05, 10

Abi-sare 2005 / CDLI Seals 012800 (CDLI Seals 012800 (composite))

Šamši-Adad I 2003

Sargon I 1 / CDLI Seals 003474

SAA 15 072. Adda-rami and Horses (CT 53 606)

SAA 15 105. Horses and Recruitment Officers of Calah (ABL 0127)

SAA 17 083. Pardoning the Offences of Borsippa (ABL 1076)

CUSAS 15, 184