Chapter 01 · 4000 – 3100 BCE

Uruk Period

The birth of the city, the birth of writing.

Illustration of a scribe recording grain beside a canal, reed boats and mudbrick buildings in ancient Uruk
Uruk, around 3300 BCE.

~3000 BCE · Sb 15279 — Louvre Museum, Paris, France

MDP 26S, 4770

In plain English

A short ledger, headed by a single classifying sign. Five entries follow, each pairing a category or supplier with a quantity: one unit, six, two, two more, and nine. At the bottom the scribe drew the account together — fifteen units plus two measures of one capacity type and one of another, closed off with a further large-unit mark. What the goods actually were is lost with the script itself: the numbers survive, the nouns do not.

AI-generated reading paraphrase; compare with the translation below.

“(Obverse) M157 (heading sign) [...] — M206~d: 1 (unit) M153+M320+M153 — M206~d: 6 (units) M195+M038~a — M206~d: 2 (units) [... ...]: 2 (units) M206~d: 9 (units) (Summary / reverse) M288: 1(N14) 5(N01) 2(N39B) 1(N24) 1(N34)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15280 — Louvre Museum, Paris, France

MDP 26S, 4771

In plain English

This is a short ledger. Under a heading sign, the scribe lists seven entries, each naming a commodity or account and giving its amount: the first, a compound entry whose middle signs are now rubbed away, is credited with 3 units; the next receives 1 unit plus two smaller measures; three further accounts get 1 unit each; one entry is too damaged to read; and the last is the largest — 2 units together with four subdivisions of decreasing size. A single figure stands at the bottom, apparently the total for the whole account. What the goods actually were, we cannot say: the script is still undeciphered.

AI-generated reading paraphrase; compare with the translation below.

“(obverse) M401 (heading / label sign) M393~a M340 [x x x] M054 : 3(N01) M288 : 1(N01) 2(N39B) 1(N24) M054 : 1(N01) M265 : 1(N01) M354 : 1(N01) M286(?) — sign damaged M297~b : 2(N01) 2(N39B) 1(N24) 1(N30D) 1(N39C) (total / final entry) 1(N34) Rendered as a ledger, so far as the numerical notations can be read: Heading: M401. Entry: M393~a M340 [three illegible signs] M054 — 3 units (N01). Entry: M288 — 1 unit (N01), 2 (N39B), 1 (N24). Entry: M054 — 1 unit (N01). Entry: M265 — 1 unit (N01). Entry: M354 — 1 unit (N01). Entry: M286 (broken, reading uncertain). Entry: M297~b — 2 units (N01), 2 (N39B), 1 (N24), 1 (N30D), 1 (N39C). Closing figure: 1 (N34).”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15285 — Louvre Museum, Paris, France

MDP 26S, 4776

In plain English

A short tally sheet, badly cracked. Under a single heading, one entry after another is credited with the same amount — two units each: two for the first item, two for the second, two for the third, and so on down the list, with several entries broken away in the middle. The last preserved line simply reads 'two.' The back of the tablet is blank. What exactly was being counted is not written on any surviving part of the tablet, and the signs that would have named the recipients or the goods cannot be read.

AI-generated reading paraphrase; compare with the translation below.

“(Obverse) 1. M157 — (heading) 2. M005~a M213~b M346 — 2 (units) 3. M005~a (x) [...] — [2 (units)] 4. [...] — 2 (units) 5. M249~c M001 — 2 (units) 6. M131 M388 [...] — 2 (units) 7. [...] M388 M260~1 M260~1 — 2 (units) 8. M075~ab [...] (remainder of line lost) 9. [M346] — [...] 2 (units) 10. 2 (units) (Reverse: uninscribed / surface blank and broken) Note: the counted commodity is not named in any surviving line; all preserved quantities are written with the numeral sign N01, twice in each entry, in the decimal-type system used for discrete counted objects.”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15289 — Louvre Museum, Paris, France

MDP 26S, 4782

In plain English

A short ledger, opening with a heading sign. Three near-identical lines follow, each pairing the same two commodity signs with the same quantity: one large measure plus one small one. A fourth entry, on a damaged patch, lists a single unit against another sign-plus-numeral combination, and a fifth records three large measures and one smaller unit. The last line gives the total: one unit of the largest denomination. The reverse was left blank. What the goods actually were, and who handled them, cannot be recovered — proto-Elamite words are still unread.

AI-generated reading paraphrase; compare with the translation below.

“(Obverse) (heading) M157~a M387~a M036 — 1(N39B) 1(N30C)(?) M387 M036 — 1(N39B) 1(N30C) M387~ef M036 — 1(N39B) 1(N30C) |M260~1 + 1(N24)| — 1(N01) M288 — 3(N39B) 1(N24) (total) 1(N34) (Reverse: uninscribed / surface blank)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15292 — Louvre Museum, Paris, France

MDP 26S, 4786

In plain English

Under a single heading, the scribe itemises eight or nine allotments, each naming a recipient or commodity class and setting a figure beside it: four units against the first entry, one large unit plus two small against the second, one measure of a different kind against the third, single units for two further named items, one more for an entry now broken away, and so on down to a partly damaged line worth one unit plus one measure. The last line is the tally: two units, four measures, and several fractional amounts, gathered under the same classifying sign that runs through the list. The reverse and edges carry only further impressed numerals, and several readings are damaged.

AI-generated reading paraphrase; compare with the translation below.

“(Heading) M157~a M038~i(?) M141 M086 |M036+1(N24)| — 4(N01) |M036+1(N30C)| — 1(N14) 2(N01) M140~a M288 — 1(N39B) M262~b — 1(N01) M262~ba — 1(N01)# [...] — 1(N01) M248~a M002 — 1(N30D) |M343~h+1(N30C)| M037 M288 — 1(N01) 1(N39B)# (Total) M288 — 2(N01) 4(N39B) 1(N24) 2(N30C) 1(N30D)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15294 — Louvre Museum, Paris, France

MDP 26S, 4787

In plain English

This is a damaged page from an accountant's ledger. Several entries survive, each naming a commodity or account holder and then giving a figure: one line records 9 large units plus 2 small ones with a further high-order mark; another lists 2 large units, 4 small units and grain measures; a third, headed by the sign M203~c, gives 6 large units before the tablet breaks. Further entries follow — one with the repeated sign M387 totalling 4 large units and grain measures, one with only 1 unit left readable, one giving 3 units, and a final line with 4 large units before the surface is lost. The sign-groups that identify what is being counted cannot be read; the numbers are all that still speak.

AI-generated reading paraphrase; compare with the translation below.

“(obverse, column entries) [...] , [...] — 1(N45) 9(N14) 2(N01) […] [...] — 2(N14) 4(N01) 3(N39B) 1(N30C) M203~c M288 — 6(N14) [...] [...] M387~i M387~i M054 M288 — 4(N14) 3(N39B) 1(N30C) M340 [...] — [...] [...] — [...] 1(N01) M054 […] M288 — 3(N01) M371(?) […] […] M048~c M288 — 4(N14) [...] (The remaining surfaces — the four edges and the reverse — are uninscribed or too abraded to bear legible signs.)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15295 — Louvre Museum, Paris, France

MDP 26S, 4788

In plain English

A short ledger, mostly numbers. The first entry records 40 units; then 12 of one commodity, 13 of another, and 10 of a third item listed under a longer string of signs. Two further lines are broken and their amounts are lost. Near the end come 3 units, then an entry combining 3 units with a measure of grain or liquid, and finally a larger closing figure — a big unit plus 30 plus 4 — which looks like a total. The commodity names themselves are written in the proto-Elamite script and cannot be read. The reverse of the tablet is blank apart from modern museum numbering.

AI-generated reading paraphrase; compare with the translation below.

“Obverse 1. [...] — 4(N14) [40] 2. [...] M288 — 1(N14) 2(N01) [12] 3. [...] M066(?) M288 — 1(N14) 3(N01) [13] 4. M387~a M111~i M242~b [...] — [numeral lost] 5. [...] M066 M352~o M218 M259(?) M218 M288 — 1(N14) [10] 6. M038~h [...] — [numeral lost] 7. [...] M288 — 3(N01) [3] 8. M047 [...] M288 — 3(N01)(?) 1(N39B)(?) 9. M288 — 1(N45) 3(N14) 4(N01) (Reverse uninscribed; only modern museum markings — "4788" and "Sb 15295" — are written on it.)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry

~3000 BCE · Sb 15297 — Louvre Museum, Paris, France

MDP 26S, 4790

In plain English

The tablet opens with a heading sign, then runs down a column of entries: one large unit against the first item, single units against the next two, and then three separate lines each recording five units of the same commodity-and-measure combination. Several entries in between have lost their numbers where the surface broke away. The last itemized line carries a compound quantity — one unit plus a string of smaller capacity measures. The reverse gives the grand total as a single large numeral. What each of the commodities actually was, we cannot say: the script has never been deciphered.

AI-generated reading paraphrase; compare with the translation below.

“(obverse) 1. M157~a — (heading; no quantity) 2. M305, (illegible sign) — 1(N14) 3. M096 M112 M136~c M263~b1 — 1(N01) 4. M305 [...] M260 — 1(N01) 5. |M036 + 1(N30D)| — 5(N01) 6. M111~b M388 [...] — [quantity lost] 7. |M036 + 1(N30D)| — 5(N01) 8. (illegible sign) [...] — [quantity lost] 9. |M036 + 1(N30D)| — 5(N01) 10. [...] — [quantity lost] 11. (illegible sign) |M036 + 1(N30D)| — 1(N01)(?) 12. (illegible sign) — 1(N01) 13. M297 — 1(N01) 2(N39B) 2(N30C) 1(N30D) 1(N39C) (reverse / total) 14. 1(N34)”

Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence

Read the full tablet entry