Cuneiform 103 · Chapter 00
Warm-up · 5 quick retrievals from earlier chapters
A script crosses a border
Everything you have read so far was Sumerian — a language with no known relatives, written by the people who invented the script for it. This course reads Akkadian: a Semitic language, cousin to Hebrew and Arabic, with different sounds, different word-shapes, and a completely different grammar.1 And yet you will read it in exactly the wedges you already know, because around the middle of the third millennium BCE the scribes of Akkad and Babylon did not invent a script — they borrowed one whole.
What survives a borrowing like that, and what changes? The signs survive: their shapes, their names, their order of strokes. What changes is the work they do. Sumerian leaned on word-signs — 𒈗 was the word “king.” Akkadian’s words inflect and shift too much for that, so its scribes leaned the other way: they took each sign’s sound and used it as a bare syllable, spelling their own words piece by piece. The king of Babylon is šarrum — written ša-ru-um, three signs, three syllables, no “king”-sign in sight.
A different kind of language
The deeper difference is not the sounds but the machinery. Sumerian, as 102 taught you, builds sentences by stacking unchanging pieces: words keep their shape, grammar arrives as little tags on the end of a phrase (-e, -ra, -še3) and as a chain of prefixes marching toward the verb. Akkadian works from the inside out. Most of its words are a consonantal root — a skeleton of (usually) three consonants — dressed in changing vowel patterns that do the grammar:
| One skeleton: n-d-n, "give" | Says |
|---|---|
| nadānum | "to give" — the dictionary shape |
| iddin | "he gave" — one word, subject included |
| inaddin | "he shall give" — the laws' favorite verdict |
The consonants stay; the vowels flex. Hebrew and Arabic, its cousins, work the same way — this is the Semitic family trait. Three more differences will organize everything ahead:
- The noun carries its own case. Where Sumerian hangs a tag on the end of a whole phrase, an Akkadian noun inflects its own ending — -um, -am, -im (chapter 02).
- The ergative is gone. No more lugal-e for doers of transitive deeds: an Akkadian subject is simply the subject, whatever the verb.
- The verb is one dense word. Iddin already contains “he” — no chain of prefixes, but an inner pattern (chapter 05).
And one thing did NOT change: the verb still comes last. Semitic languages usually lead with the verb — Akkadian kept Sumerian’s verb-final habit after centuries of sharing the same cities, the same scribes, and the same clay. The two languages are unrelated, but they were neighbors so long they began to rhyme.
The signature sounds
Three signs open the course — the ones that carry sounds and words Akkadian reaches for constantly.
| Sign | Name | Key | Reads | Means | Notes |
|---|---|---|---|---|---|
| 𒊭 | ŠA | which | [ša] | the particle ša, "which, of" | the busiest word in Akkadian — glue for nouns and clauses |
| 𒅇 | U3 | and | [u] | the conjunction u, "and" | Sumerian mostly did without an "and"; Akkadian says it out loud |
| 𒌝 | UM | tether | [um] | — | the ending of every noun-subject: šarrum, awīlum |
Two layers of writing it down
From here on, examples come in two forms, and the course will always keep them straight. Transliteration is what the tablet actually spells, sign by sign, hyphenated: ša-me-e. Bound transcription is the reconstructed Akkadian word those signs spell, written joined-up with its vowel lengths marked: šamê, “of heaven.” The first is evidence; the second is interpretation — useful, standard, and always one step removed from the clay. A bound transcription never appears in this course without the spelling it came from.
The sounds Akkadian brings
A Semitic language arrives with a Semitic mouth, and several of its sounds simply did not exist in Sumerian — which means the borrowed script had no signs waiting for them. The scribes’ solution, every time, was to draft a nearby sign into the new job; this course meets each recruit at the moment a real text needs it.
Here is the complete delegation — every sound Akkadian brings that Sumerian’s script had never needed to write:
| Sound | What it is | You will meet it |
|---|---|---|
| ṣ | an emphatic s — pressed, with an edge | chapter 02, the stele's opening |
| ṭ | an emphatic t — the same pressure on t | chapter 07, the king's promise |
| q | an emphatic k — the third of the pressed set | in the laws' vocabulary: ḫulluqum, "to destroy" |
| w | plain w, which Sumerian never wrote | chapter 06, inside awīlum |
| y | the glide y — rare, and often melted into its vowels | in words like ayyum, "which one?" |
| ʾ | the glottal stop — a catch in the throat between vowels | chapter 03, unwritten but audible in šā'im, "who decrees" |
| ā ē ī ū, â | vowel LENGTH — long vowels that change meaning; the circumflex marks length born from two vowels contracting | everywhere in bound transcription; the tablet often hints at length by writing the vowel twice |
That is the whole list. Everything else in Akkadian’s mouth — b, d, g; p, t, k; s, z, š; the rasping ḫ (printed plain h in values, as the corpora do); m, n, l, r; and the four vowels a, e, i, u — Sumerian’s script already handled. Traffic ran the other way too: Akkadian has no ŋ — the velar nasal of saŋ and niŋ2 stays behind with Sumerian, and no Akkadian word will ever ask you for it. None of the new sounds needs memorizing now: each arrives attached to a sign, a word, and a line of the stele, exactly where it will stick.
What your Sumerian years already bought
Here is the remarkable part, measured: of every syllable written in the Old Babylonian documents this course draws on, more than half are spelled with signs you already know.2 Sumerian values like a, na, ma, šu, ta, ba, ra, bi, la pass into Akkadian unchanged — same sign, same sound, new language around it. Two of the very first words:
urn:nabu:cdli:p464358 · license: attribution.
Not every sign crosses the border with the same value it had, and some cross with new values — when a familiar sign returns with an Akkadian reading, its table row will say veteran, and its glyph will link to its Akkadian page. Sign lore for this course lives on its own shelf, the Akkadian Addenda, kept deliberately separate from the Sumerian shelf: one sign, two lives, two pages.
Where this course is going
The destination is a real monument. Around 1750 BCE, Hammurabi, king of Babylon, set up a stone stele carrying almost three hundred laws between a prologue and an epilogue — the Codex Hammurapi, the most famous Akkadian text in existence. Its opening words are i-nu an ṣi-ru-um, “when lofty An…” — and by chapter 04 you will read them in the original, every sign yours. By chapter 06 you will read its first law. This is not a specimen course: it is a straight march to the stone.
Drill before you go on
- Write 𒊭, 𒅇, 𒌝 from memory; say ša, u, um.
- Spell ša-me-e in signs without looking, then say the bound transcription it spells.
- Which of the two prepositions above means “in”? Write it.
- In one sentence: what did Akkadian scribes keep when they borrowed the script, and what did they change?
Next: the syllabary at work — spelling whole Akkadian words with signs you mostly already own.
-
The grammar this course teaches follows the standard teaching grammar of the field: John Huehnergard, A Grammar of Akkadian, 3rd ed. (Winona Lake: Eisenbrauns, 2011) — hereafter the course’s silent authority on Old Babylonian forms. ↩
-
Measured over this site’s own frequency count of the Old Babylonian slice of the CDLI corpus (204,820 lines): the primary values of the 77 signs taught in 101–102 cover 55% of all syllable-occurrences before this course teaches a single new sign. ↩