Chapter 01 · 4000 – 3100 BCE
Uruk Period
The birth of the city, the birth of writing.
~3000 BCE · Sb 15099 — Louvre Museum, Paris, France
MDP 06, 247
In plain English
This broken clay tablet is a simple tally sheet. At the top is a single sign that may serve as a heading or category label for the whole record. Below it, the first entry lists a string of symbols followed by a quantity: one large unit, one middle unit, and four small units. A second, damaged entry follows with a different quantity — two middle units and two small units — but the rest of that line, and whatever object or person it refers to, is broken away. A third entry survives only as a number, six small units, with everything else lost. In short, this is an accounting record — someone kept a count of something (goods, rations, or labor) in three or more lines — but because the proto-Elamite script itself has not yet been deciphered, we can read the numbers with confidence but not the words that explain what was being counted.
AI-generated reading paraphrase; compare with the translation below.
“Heading: M157 Entry 1: |M195+M057|(?) |M218+M288| M001 M288 — 1(N34) 1(N45) 4(N14) Entry 2: [...] M195~d M288 — 2(N45) 2(N14) [...] Entry 3: [...] — [...] 6(N14)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15100 — Louvre Museum, Paris, France
MDP 06, 251
In plain English
This tablet is a bookkeeping note. It lists several items, one per line, each followed by a number: an entry marked with one sign gets a count of 1, another gets 2, a cluster of several signs together gets roughly 1 (the exact figure is broken), a further entry gets 5, and a final entry is tallied with a mixed count — 4 of one unit, 2 of another, and 1 more of a third, though that last figure is damaged. Because the proto-Elamite signs themselves have never been deciphered, we can recognize that this is a tally of goods or produce and reconstruct the numbers with confidence, but we cannot say in modern terms exactly what was being counted.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse) Line 1: M157 — Line 2: M383(~j, damaged) M260(~1, uncertain), — quantity: 1 (unit N01) Line 3: M036 — quantity: 2 (unit N01) Line 4: M111(uncertain) M388 M057 M240 M377(~e) M347 M371 — quantity: <1> (unit N01) Line 5: M036 — quantity: 5 (unit N01) Line 6: M297 — quantity: 4 (unit N39B), 2 (unit N30C), 1 (unit N30D, damaged)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 15101 — Louvre Museum, Paris, France
MDP 06, 252
In plain English
This is an ancient bookkeeping tablet, not a letter or a story — think of it as a scrap of an inventory sheet. At the top is some kind of heading or document-type marker. Below it, several lines each pair a recurring sign (perhaps a category label or a commodity marker) with a number: one line records a single unit of a compound item; another lists several different quantities — 2, 1, 1, then a gap where the surface is damaged, then 5, 2, and 1 more of various units; further lines repeat similar short entries, some of them again broken by damage. The tablet ends with what looks like a summary or total line, but it is too damaged to read fully — only a couple of numbers ('2' and '2 more') survive. Because the script used here (proto-Elamite) has never been fully deciphered, we can transcribe the numbers and repeated signs faithfully, but we cannot say for certain what goods or transactions this record actually describes.
AI-generated reading paraphrase; compare with the translation below.
“Heading: sign M157. Entry: compound sign |M106~2+M288| — quantity: 1 (N02-unit) [sign damaged, marked with #] Entry: sign M288 — quantities: 2 (N34), 1 (N45), 1 (N14), [break in surface], 5 (N01), 2 (N39B), 1 (N24) Entry: compound sign |M106+X| [damaged, marked with #] — quantity: 1 (N08-unit) Entry: sign M288 — quantities: 1 (N34, uncertain reading), 9 (N14), 1 (N01) Entry: compound sign |M106~2+M288| — quantity: 1 (N02-unit) Entry: sign M288 — quantities: 1 (N45), 4 (N14) Final/totaling line (heavily damaged): [broken], sign M288-related notation, 2 (N34) [marked with #], [break], 2 (N01), 2 (N39B), [break]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15102 — Louvre Museum, Paris, France
MDP 06, 253
In plain English
This is a bookkeeping tablet in a script we still cannot read. It lists what look like four separate entries — quantities of five, two, one, and two units of something, each marked with its own combination of unidentified signs — and then closes with a single line recording one unit in a bigger measure, probably a running total of everything above it. The back of the tablet was left blank, or nothing survives on it. Because the proto-Elamite writing system has never been fully deciphered, we can describe the tablet's structure and numbers with confidence, but not say what commodity or transaction it actually records.
AI-generated reading paraphrase; compare with the translation below.
“(This tablet is written in the undeciphered proto-Elamite script; the great majority of signs have no established phonetic or semantic value. What follows is a structural rendering of the transliteration, not a semantic translation.) Obverse: Line 1: [M327+M342] (compound sign, damaged/uncertain reading; no numeral — likely a heading or classifier) Line 2: [M153+M106~a] M066 M242~b M096 M387~eh M004 M346 , 5(N01) — an entry of five units (N01) Line 3: M249~m M332~d(?) M057~a2 M346 , 2(N01) — an entry of two units (N01) Line 4: M370 M346 , 1(N01) — an entry of one unit (N01) Line 5: [M218~d+M288](?) x M218 M346 , 2(N01) — an entry of two units (N01) Line 6: M346 , 1(N14) — a closing entry of one unit in the larger N14 measure, likely a summary or total Reverse: no signs are preserved or visible; the surface appears blank/uninscribed.”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15103 — Louvre Museum, Paris, France
MDP 06, 254
In plain English
This clay tablet is an ancient bookkeeping record from Susa, kept in a script called proto-Elamite that nobody today can actually read. What survives looks like a list: several lines, each opening with one or more picture-like signs (perhaps naming a person, place, or type of goods) followed by a count — some number of units, marked with different-shaped number stamps for different sizes of quantity. A line near the top just reads '2 units' after its heading sign; later entries run to '5 units,' or compound counts made of several different-sized numeral marks together. A large piece is broken out of the middle of the tablet, so a number of entries in that section are simply gone. The surviving reverse continues in the same format — heading signs followed by numeral counts — before the tablet breaks off again near the bottom, its final total only partly preserved.
AI-generated reading paraphrase; compare with the translation below.
“Obverse: 1. [heading sign] M157 2. M374~c , 2(N01) 3. M124 M391 M218~b(#) M295~wa [...] , [...] 4(N01) 3(N39B) 1(N24)(#?) 4. [...] , 1(N14) 3(N01) 2(N39B) 5. [...] M066 , 1(N14) 3(N01) 3(N39B) 6. M218 M370 M219 M263 , 2(N01) 7. M263 M242~b M096 , 5(N01) 8. M386~a M242~b M096 , 1(N14) 2(N01) 2(N39B) 9. [...] M352~n |M218+X|(?) , [...] 10. M218(#) M295~ka M218 , 3(N01) Reverse: 11. M146(#) M380 M263 [...] , [...] 12. [...] M218 M295~ka M218 , 3(N01) 13. x , 14. [...] , 2(N45) 2(N14) 3(N39B) 1(N24)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15105 — Louvre Museum, Paris, France
MDP 06, 256
In plain English
This is a scrap of an ancient accounting record from Susa, written in the proto-Elamite script — one of humanity's oldest writing systems, and one we still cannot fully read. Each line follows the same pattern: a cluster of signs (naming some person, place, or commodity we can't identify) followed by a number. The surviving entries record quantities of one unit each, and a couple of two-unit groups, but the tablet is broken at both edges of nearly every line, so most of the context — what exactly was being tallied, and for whom — is lost. Two additional signs at the very end are too damaged even to attempt reading.
AI-generated reading paraphrase; compare with the translation below.
“(Line 1) [...] , [broken — no legible entry] (Line 2) [...] sign M376 , [1 unit (N08)?] (Line 3) signs M046 M254~b , 1 unit (N01) (Line 4) signs M218 M250~ba [...] , [...] (Line 5) [...] signs M024 M033 M376 , 1 unit (N08) (Line 6) signs M254~b1 M046 [...] , [...] (Line 7) [...] signs M380~b M295~y M263~1 M376 , [1 unit (N08)?] (Line 8) [...] sign M096 , 1 unit (N01)? (Line 9) [two unidentified signs] [...] , [...]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 15107 — Louvre Museum, Paris, France
MDP 06, 258
In plain English
This is an ancient accounting slip from Susa, roughly five thousand years old, written in a script — proto-Elamite — that experts still cannot fully read. Line by line it lists small quantities against unidentified categories of goods: 4 of one thing, 1 of another, 5, then 1 (in a different, larger counting unit), 1 more (in yet another unit), 4, 3, and 2. One entry is too broken to make out. The tablet closes with what looks like a summary or total line: 5 units of the item marked by a sign that recurs throughout the record, possibly a running tally or grand total for the whole account. What exactly was being counted — grain, animals, or something else — cannot currently be recovered from the signs themselves.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse, entries recorded as commodity-sign groups followed by numerical notations) 1. [...] M388(#) M057~b(?) M099(#) M371 M288 — 4(N01) [...] 2. [...] , [...] 1(N01) 3. [...] M004 |M218+M101| M371 — 5(N01) 4. M377~e |M106+M288| [...] , [...] 1(N24) 5. M380 M371 — 1(N14) 6. M295~u M220 — 4(N01) 7. M387~c x M387 M218 — 3(N01) 8. M301(#) |M296+M296| M057 — 2(N01) 9. M290(?) [...] , [broken] 10. (Total line) M288(#) — 5(N14)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15108 — Louvre Museum, Paris, France
MDP 06, 260
In plain English
What survives of this tablet is a short accounting entry broken into several pieces. One line records a quantity of 2 units next to an unidentified sign; another line lists a possible classifier sign followed by a gap; a third line pairs two occurrences of the same unknown sign with another sign and a count of roughly 5 plus 2 more units; a fourth line gives a count of 4 units next to an illegible sign; and the last readable line pairs two more undeciphered signs before the text breaks off. In short: this is a fragmentary tally of goods or produce, but because the proto-Elamite script is not yet decipherable, we can read the numbers reliably while the words for what was being counted remain unknown.
AI-generated reading paraphrase; compare with the translation below.
“(Line 1) [......] sign M302 (variant ~s?) , 2(N01) (Line 2) sign M288 , [......] (Line 3) [......] sign M262, sign M262, sign M096 , 5(N01) 2(N39B) [......] (Line 4) [......] x , 4(N01) (Line 5) sign M056 (variant ~e?), sign M288 , [......]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry