Chapter 01 · 4000 – 3100 BCE
Uruk Period
The birth of the city, the birth of writing.
~3000 BCE · Sb 15135 — Louvre Museum, Paris, France
MDP 06, 292
In plain English
This clay tablet from Susa is a bookkeeping record written in proto-Elamite, a script that predates readable writing systems and has never been fully deciphered. It lays out about nine or ten short entries — each a small cluster of signs, probably naming a commodity, person, or category — followed by a number, using two different counting systems, so it reads like a tally or inventory list. A single sign stands alone at the top, likely some kind of heading or document label. Several entries are damaged or partly illegible, the list breaks off before the end, and the back of the tablet appears to have been left blank. In short: it's clearly an accounting document, but exactly what was being counted remains unknown.
AI-generated reading paraphrase; compare with the translation below.
“Heading (isolated sign, position and function unclear): M157 (Obverse — entries, each a group of signs followed by a numeral notation) 1) M377 M217(var. e) M005(var. a, damaged) M124 M146(damaged) [illegible sign] M376(damaged) — quantity: 1 (N01, reading uncertain) 2) [illegible sign] M380 M371(damaged) M376 — quantity: 2 (N01) 3) M124 M049(var. n) M049(var. e) M057(var. a) M376 — quantity: 1 (N08A) 4) M391 M038(var. i) M096(var. 4) M376 — quantity: 1 (N01) 5) M146(var. 5) M222(uncertain) M218 M376 — quantity: 1 (N01) 6) M218(var. b, damaged) M332(var. d, damaged) M218 M376(damaged) — quantity: 1 (N01) 7) M371 M295(var. ka) M066(var. a) M376 — quantity: 1 (N08A) 8) M311(var. b) M388 M218 M386(var. a, damaged) [two illegible signs] [...] — quantity: 1 (N08A, damaged) 9) M391 M318(var. a) — quantity: 1 (N01) 10) [...] — [...] (text broken off) (Reverse: uninscribed, as far as visible on the photograph)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15138 — Louvre Museum, Paris, France
MDP 06, 297
In plain English
What survives of this clay record is a series of tally entries: numbers — three of one unit and one of another, then later two of one unit and one of another — set beside strings of pictographic signs that seem to mark categories of goods or institutions. Most of the tablet is broken away, so we only catch fragments of what was presumably a fuller inventory or accounts list. Because the proto-Elamite writing system used here has never been deciphered, we can recognize that this is a counting record without being able to say what exactly was being counted.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse, broken) [...] , [...] , [...] [...] sign M288 , 3(N01) 1(N39B) [...] [...] , [...] sign M124 sign M386~a(variant) sign M024 sign M371 [...] , [...] [...] , [...] [...] , [...] 2(N39B) 1(N24) sign M124(damaged) x [...] , [...] [...] , [...] sign |M317+M260|(damaged, uncertain) sign M054 sign M386~a(variant) sign M124 x [...] , [...] [...] , [...]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 15139 — Louvre Museum, Paris, France
MDP 06, 299
In plain English
This broken clay tablet is an ancient accounting record from Susa, written in the proto-Elamite script — a system that predates readable writing as we know it and is still not fully decoded. What survives are several lines, each beginning with a partly broken heading (possibly a name, category, or place, now lost) followed by a recurring sign, tentatively labeled 'M288,' whose exact meaning — commodity, classifier, or dividing mark — scholars haven't pinned down. After it come strings of numeral marks in different counting systems, presumably tallying quantities of goods, rations, or people. The tablet ends with a final line pairing that same sign with a single number, likely a summary total for everything listed above. Much of the tablet — including the very beginnings of most lines — is broken away, so the specific transaction being recorded cannot be recovered.
AI-generated reading paraphrase; compare with the translation below.
“(Column/line 1) [......], [......] 1(N24) 1(N30C) 1(N30D) 1(N39C) (Column/line 2) [sign] M288, 1(N45) 4(N14) [......] (Column/line 3) [......] M288, 9(N14) 4(N01) 3(N39B) [......] (Column/line 4) [......], [......] 2(N01) 3(N39B) 1(N24) 2(N30C) [......] (Column/line 5) [......], [......] 1(N30C) 1(N30D) 1(N39C) (Total line) M288, 3(N34)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 15140 — Louvre Museum, Paris, France
MDP 06, 300
In plain English
This is a fragmentary accounting tablet, its opening lines lost to damage. Where the text becomes legible, it lists a series of entries — each apparently a good or commodity identified by one or two undeciphered signs — followed by a small quantity: two units, then two units again, one unit, two units, and one unit. A final entry names another sign with no quantity recorded, possibly because it served as a heading, total, or the number is simply broken away. In short: a tally of small quantities of unnamed goods, kept by a scribe at Susa, but the specific items and purpose remain unreadable to us today.
AI-generated reading paraphrase; compare with the translation below.
“[...] x [...] , [...] [...] x , 2(N01) [...] M024#? M295~f , 2(N01) [M301] M338 [...] , [...] [...] M157? , 1(N01) x M145~a M371 , 2(N01) M145~a [...] , [...] x M110~b#? M380~b , 1(N01) M332~d , [...]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · low confidence
Read the full tablet entry~3000 BCE · Sb 15142 — Louvre Museum, Paris, France
MDP 06, 303
In plain English
This broken clay tablet, now in several pieces, is an ancient accounting record from Susa written in the still-undeciphered proto-Elamite script. It seems to list a series of entries, each combining an unknown sign — perhaps naming a commodity, animal, or category of goods — with a count marked by small impressed numerals. The first entry records a quantity of 1 unit; the middle lines are largely broken away, preserving only fragments of signs and numbers; and the final line, marked with a recurring sign that may function as a total or category heading, tallies several different numeral types together (1, 1, 2, and 1 of various denominations). Because the script cannot yet be read, we can describe the tablet's structure — a short list of counted items ending in a summary — but not translate its actual content.
AI-generated reading paraphrase; compare with the translation below.
“[...] M263~e(?) — 1(N01) x [...] — [...] [...] M269~a3(?)! — 1(N30C) 1(N30D)(?) [...] — [...] 1(N30C) M288(?) — 1(N01) 1(N24) 2(N30C) 1(N30D)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15143 — Louvre Museum, Paris, France
MDP 06, 304
In plain English
This is a fragment of an ancient accounting tablet from Susa, written in a script that nobody today can actually read. What survives looks like a list: line after line pairs an unidentified item or category — shown here only as sign codes like M101, M230, M024, M153 — with a small number, often '1' or '2,' recorded in the numeral system of the period. It's clearly some kind of tally or inventory, the sort of everyday bookkeeping a Susa official might have kept for goods, rations, or livestock, but because the proto-Elamite writing system has resisted decipherment, we cannot say what was actually being counted. Much of the tablet is broken, so several lines are only fragments of larger, now-lost entries.
AI-generated reading paraphrase; compare with the translation below.
“(Obverse — proto-Elamite numerical/administrative entries, script undeciphered) Line 1: [...] , 1(N14) 2(N01) Line 2: x [...] , [...] Line 3: [...] , 1(N14) Line 4: M101(variant b) , 2(N14) Line 5: [...] x M230, M024#, M033# , 1(N14) Line 6: M146# x , [...] Line 7: [...] , [...] 1(N14) 1(N01) Line 8: [...] M254 (variant a, partially preserved) , [...] Line 9: [...] , 1(N01)# Line 10: [...] M387 (variant a), M110# , [...] Line 11: [...] , 1(N01) Line 12: [...] M153, M059 (variant f, uncertain) , [...] Line 13: [...] M309#?, M314# [...]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 15144 — Louvre Museum, Paris, France
MDP 06, 305
In plain English
This is a small accounting tablet, broken into several pieces, written in the proto-Elamite script that nobody has yet learned to read. At the top is a single sign that may serve as a heading or label. Below it, a badly broken line pairs that same sign with another mark, followed by a line that opens with what looks like a section marker and two more unidentified signs, then breaks off. The next lines are clearer only in their numbers: one entry counts 4 units of one kind and 1 (damaged) of another; the next counts 1, 1, and 2 units under a third, uncertain sign; another surviving fragment shows just 1 unit before the text breaks again. The tablet ends with a line giving 3 units, which may be a summary or total. Because the proto-Elamite writing system itself is not deciphered, we can report the sign-shapes and the quantities faithfully, but not what commodity or transaction they actually record.
AI-generated reading paraphrase; compare with the translation below.
“Obverse (Heading, isolated sign at top of tablet): M157 [...] M157+M288 (compound sign) , [...] M328~b (section-opening sign) — M103~1 — M286 (damaged, reading uncertain) , [...] [...] , [...] — quantity: 4 (N39B)-units, 1 (N30C)-unit (damaged) M219 (sign uncertain) , quantity: 1 (N39B)-unit, 1 (N24)-unit, 2 (N30C)-units (damaged) [...] (illegible sign) , quantity: 1 (N39B)-unit (damaged) [...] [...] , quantity: 3 (N39B)-units [likely a summary/total entry]”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry~3000 BCE · Sb 06386 — Louvre Museum, Paris, France
MDP 06, 309
In plain English
This looks like an inventory or receipt: each line lists one or more unidentified commodity marks followed by a quantity. The first entry records something totaling 1 plus 2 units (in two different counting systems); the second, a single unit of a different measure; the third pairs two more unread signs with a count of 1, though the line breaks off; the fourth line strings together five unread signs with no legible number surviving; and the last line, mostly lost, ends with a count of 9. Because the commodity signs themselves haven't been deciphered, we can say numbers were recorded but not what was being counted.
AI-generated reading paraphrase; compare with the translation below.
“(Line 1) [...] sign M066, sign M288 — 1(N14) 2(N01) (Line 2) [...] — 1(N45) (Line 3) sign M254~a, sign M372~a, sign M288 — 1(N14) [...] (Line 4) sign M388, sign M066~1, sign M352~n, sign M370, sign M218, sign M288 — [...] (Line 5) [...] — [...] 9(N14)”
Source: engine:claude-sonnet-5 (2026-07-12/v7-evolved) · claude-sonnet-5 · experimental confidence
Read the full tablet entry