Chapter 01 · 4000 – 3100 BCE
Uruk Period
The birth of the city, the birth of writing.
~3000 BCE · Sb 22363 — Louvre Museum, Paris, France
MDP 17, 193
In plain English
This is a ledger fragment, and the arithmetic is all that still speaks. Reading down the surviving lines: one category (marked with the sign M317) is credited with 12 units; the next entry's quantity is broken off; another item, M054, has lost its figure; then 15; then 3 of the commodity written M003~b; then at least 20 more of M054; then 12 again, under a heading combining M066~a(?) and M317; then 2 of M373; and finally 8 of something whose name is destroyed. The reverse carries no readable writing.
AI-generated reading paraphrase; compare with the translation below.
“Obverse [...] M317 — 1(N14) 2(N01) [= 12] x x — [...] M054 — [...] [...] — 1(N14) 5(N01) [= 15] M003~b — 3(N01) [= 3] M054 — 2(N14) [...] [= 20+] [...] M066~a(?) M317 — 1(N14) 2(N01) [= 12] M373 — 2(N01) [...] x(?) — 8(N01) [= 8] (Reverse: uninscribed / surface abraded; no 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 22367 — Louvre Museum, Paris, France
MDP 17, 197
In plain English
This is a fragment of an account. The first entry lists a commodity — the sign for it is still undeciphered — in the amount of two large units plus one small unit. The next line is broken away entirely. A third entry pairs several category signs with a single unit of the same commodity, and a fourth begins with more signs before the tablet breaks off; its quantity is lost. The last line preserves only traces of two signs. The reverse is worn smooth and tells us nothing further.
AI-generated reading paraphrase; compare with the translation below.
“Obverse 1. M387~ef (commodity/category sign) — 2(N01) 1(N24) 2. [...] — [...] (numerals lost) 3. [...] |M131+M388| M263~a M218 — 1(N01) 4. M218 M219 M057~e [...] — [...] (numerals lost) 5. [... ...] — [...] (numerals lost) (The remainder of the surface, including the reverse, is abraded and carries no legible signs.)”
Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 06344 — Louvre Museum, Paris, France
MDP 17, 201
In plain English
This is a fragment of an account. One heading — a cluster of five or six signs whose meanings are unknown — is followed by a quantity of one unit. A second entry, again headed by an unreadable group of signs, records eight units. Everything else has broken off: the beginnings and ends of the lines, and whatever else was written on the rest of the tablet. What survives is the skeleton of a bookkeeping record, its vocabulary lost but its arithmetic intact.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse) 1. M388 (+ one illegible sign) M377~e M049~d M325~e M386~a(?) [...] , [...] 2. [...] , [...] 1(N01) (Reverse / continuation) 3. (illegible sign) M388 M066 M352~o(?) M251~c2 M066 4. [...] , 8(N01) (The remainder of both surfaces is broken away or effaced; no further signs are legible.)”
Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 22376 — Louvre Museum, Paris, France
MDP 17, 207
In plain English
This is a ledger, and only the numbers still speak. Line by line it lists a commodity — written with signs we cannot yet pronounce — against a measured amount: three units plus one of one kind for the first item, two smaller units for the next, then two plus a fraction, and so on down to entries of one large unit with three medium and two small. Several of the commodity names are chipped away, and the accountant's headings and totals are gone entirely. The rest of the tablet is broken, so we will never know whose account this was or what was being counted.
AI-generated reading paraphrase; compare with the translation below.
“(obverse) [...] M346 M081 M036 — 3(N39B) 1(N24) M380 — 2(N30C) [...] — 2(N30C) 1(N30D) M010~2 — [...] [...] — [...] 1(N30C@b)(?) M297 — 2(N30C)(damaged) [...] M036(damaged) M297 — 1(N01) 3(N39B)(damaged) [...] [...] — 1(N01) 3(N39B) 2(N30C) (The remainder of the surface, and the whole of the second inscribed face, is broken away or effaced; no further entries can be read.)”
Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 22378 — Louvre Museum, Paris, France
MDP 17, 210
In plain English
This is a ledger, and most of it is gone. What remains is a run of short entries, each naming something with two or three signs and then counting out a single unit of it: one against a sign-group beginning M370, one against a group ending in M242, one against M461 and M371, and one more where the label itself is broken away. Other lines — those with M124, M175, M263 — have lost their numerals entirely. Since proto-Elamite has never been deciphered, we can see the shape of the accounting perfectly well without knowing a single thing that was being counted.
AI-generated reading paraphrase; compare with the translation below.
“(obverse, upper fragment) [...] , [...] M314 M338~b [...] , [...] [...] (traces) , 1(N01) M370 M072 , 1(N01) (traces) [...] , [...] (traces) M242~ab(?) , 1(N01) M461~q M371 , 1(N01)(?) [...] , 1(N01) M124 M115~a M314 [...] , [...] [...] M371(?) , 1(N01)(?) M175 [...] , [...] [...] M263 M218 , [...] [...] , [...] (The remainder of the surface, including the lower fragment, is abraded and carries no 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 22385 — Louvre Museum, Paris, France
MDP 17, 218
In plain English
This is a ledger, and we can read the arithmetic but not the words. Line by line it lists a series of goods — each written with a cluster of signs whose meaning is lost — followed by an amount: one unit here, one unit there, one unit plus two of a smaller measure in the fifth entry, and a single unit of a different measure in the sixth. Two of the entries have lost their numbers to breakage, and two have lost their commodity names. The final line names one more item and then breaks off. The back of the tablet is blank or worn smooth.
AI-generated reading paraphrase; compare with the translation below.
“Obverse 1. [...] M010~2 — 1 (N01) 2. [...] M099 [...] — [amount broken] 3. [...] — 1 (N39B) 4. M124, M009, M338~b, M010~2 — 1 (N39B) 5. M002 — 1 (N39B), 2 (N30C) 6. [...] M377, M099, M096, M010~2 — 1 (N24) 7. M124, M218, M295, [...] — [amount broken] 8. [...] — 1 (N01) 9. M206~d — [amount broken] (Reverse uninscribed / surface destroyed; edges bear no 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 22386 — Louvre Museum, Paris, France
MDP 17, 219
In plain English
The opening line is lost. Then: against one group of commodity signs, four units; against the next, one unit and more that is broken away; against another, one unit plus two small measures. The following entry is too damaged to give a quantity, and the one after it records a single unit of a variant measure. Two more entries follow — two small measures against one sign, and one large unit against another, with the rest broken. The final line looks like a total: one large unit, three units, one intermediate measure, and two small measures. A last sign is illegible, and the reverse is essentially blank apart from a few impressed marks.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse) 1. [...] 2. M263(?) M387~ef M066 M288 — 4(N39B) 3. M219 M371~a — 1(N24) [...] 4. [...] (traces) M011 M288 — 1(N39B) 2(N30C) 5. M356 M246~s [...] — [...] 6. [...] — 1(N39B@c) 7. M379 — 2(N30C) 8. M504 — 1(N34)(?) [...] 9. M288 — 1(N01) 3(N39B) 1(N24) 2(N30C) 10. (one illegible sign) (No further inscribed text is preserved; the reverse carries only impressed numerical marks along the left margin and is otherwise blank or broken.)”
Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 22388 — Louvre Museum, Paris, France
MDP 17, 221
In plain English
This is a ledger page, and most of what survives is the column of figures. The first line, which probably named the account or the responsible party, is damaged and its total is lost. After that come short entries: one unit against one sign group, then a larger measure plus a smaller one against another, then a run of single small units — one, two, one, one, one — set beside signs we cannot yet pronounce. Both edges are broken, so some entries and probably a closing total are missing. What we can say is that a scribe at Susa was tallying small, standardized quantities of something, item by item.
AI-generated reading paraphrase; compare with the translation below.
“Obverse 1. M124 M380~b M329(?) M102~da(?) [ ... ] — [quantity lost] 2. [ ... ] M263 — 1(N01) 3. M387~ef M036 — 1(N39B) 1(N30C@b) [ ... ] 4. M002 — 1(N30C) 5. M370 M288 — 2(N30C) 6. [ ... ] — 1(N30C) 7. M002 — 1(N30C) 8. [ ... ] — 1(N30C) (The signs M124, M380~b, M329, M102~da, M263, M387~ef, M036, M002, M370 and M288 are undeciphered proto-Elamite sign-forms; no phonetic or semantic reading can be given. The numerals belong to the proto-Elamite capacity system: N01, N39B, N30C and its variant N30C@b. Several lines are broken away at the left and right; further entries are lost.)”
Source: engine:claude-opus-5 (2026-07-12/v7-evolved) · claude-opus-5 · low confidence
Read the full tablet entry