Akkadian is a dead language of the Semitic family, itself a part of the Afro-Asiatic language family. Within Semitic, there were only two members of the Eastern branch, Eblaite and Akkadian. Some of the oldest texts in existence are written in Akkadian, and it continued as a liturgical language until 100 A.D. is certain contexts. It existed in a very small, tight sprachbund with Sumerian, a language isolate. The writing system of Sumerian came first and was used for the Akkadian language as well, despite being severely ill-adapted for that use. It is assumed that scribe and scholars were all fluent in both languages, making it much less cumbersome originally. Akkadian borrows heavily from Sumerian and is therefore quite different from other Semitic languages. For example, it is SOV, where most Semitic languages are VSO. Akkadian uses triconsonantal roots, a three case system, singular-dual-plural numbers, two genders, and polypersonal agreement. Nouns also occur in the construct state.
The vowels of Akkadian are a, e, i, and u. They all occur in short and long varieties, long being written with a macron. Long vowels which arise from contraction are written with a circumflex.
The consonants of Akkadian which are the same as their IPA symbols are b, d, g, h, k, l, m, n, p, r, s, t, w, and z. š is sh, ṣ is ts, y is /j/. ṭ is ejective /t/, q is ejective /k/. ḫ is /x/. Some systems of romanization use ˒ as a symbol for the glottal stop, some omit it.
Syllables are labeled a light, heavy, or ultra heavy. Light syllables are CV, ending in a short vowel. Heavy syllables are CVC, or CVV (long vowel). Ultra heavy syllables are CVVC. The rules of accent are
- if the last syllable is ultra heavy, it bears the stress
- otherwise, stress falls on the last non-final heavy or ultra heavy syllable
- without any of the above, stress falls on the first syllable.
Because of the Summerian influence, there are many times where a kind of vowel harmony exists. It takes the form of as and es being generally incompatible within the same word. Fortunately, most inflectional morphology is not subject to these changes.
The nominative (N), genitive (G), and accusative (A) case endings in the singular are -(t)um, -(t)im, and -(t)am. The optional t is for feminine words, typically. A situation where all three cases are distinguishable is called tripartite. Often, the genitive and accusative are inflected with the same morpheme. These situations are called dipartite. The dual endings are dipartite, -(t)ān, and -(t)īn. The plural endings are also dipartite, -ū, and -ī in the masculine, or -ātum, and -ātim in the feminine. Nouns in the construct lack case endings.
- mār šarrim ṣeḫram amḫaṣ - I struck the king's young son.
There are no articles.
Prepositions take the genitive form of their noun phrase.
Independent personal pronouns occur singular and plural, but with gender in the 2nd and 3rd person. (c stands for "common".)
- 1cs anāku, 1cp nīnu
- 2ms atta, 2mp attunu, 2fs atti, 2fp attina
- 3ms šū, 3mp šunu, 3fs šī, 3fp šina
Adjectives agree with their head noun in gender, but only inflect for singular-plural. About half the adjectives inflect in the feminine plural as -ūtum, -ūtim, instead of -ātum, -ātim.
The demonstrative adjectives are annûm - "this", and šū - "that". Unfortunately, šū is also the 3rd person personal pronoun, which can make this confusing at times.
In the Semitic system, word-groups have as their base a tri-consonantal form, written C1C2C3. Conjugations and word derivations are formed by infixing vowels, appending suffixes and prefixes, and sometimes doubling consonants. For example, the 3cs durative G-stem of P-R-S is iparras, which can be generalized as iC1aC2C2aC3. Verbs must agree with their subjects in person, number, and gender. Complicating these patterns is a rule of vowel syncope, where two or more short vowels cannot occur in a row. Typically, the second of two short vowels is syncopated, unless
- it is the end of a word
- it is part a vowel hiatus
- often, before r
- occasionally, before l
- before pronominal suffixes
- in Sumerian loanwords
Further complicated the inflectional paradigm, the consonants n, w, y and ˒ are typically "weak", and fail to conjugate as might be expected.
The consonant n is often assimilated in conjugation and inflection into a doubling of the following consonant. For example, the verbal adjective from Š-K-N šikin- has fs šakittum, not *šakintum.
Verbless clauses take the place of "to be" clauses.
Word order is SOXV, where X stands for adverbial phrases. Akkadian is fairly strict about being verb final.
Common prepositions are
- ana - to, for, at
- ištu - from, out of
- ina - in, among, with, by, from
- itti - with, in the company of
- eli - on, upon, over, above
- ša - of (the)
- kīma - like, as, according to, instead of
- ul(a) - not. Must appear immediately before the verb
- u - "and" for nouns phrases and sentences
- -ma - "and" for verbs (which causes nasal assimilation)
- -ū (lū) - "or"
However, note that most successive causes follow without a conjunction (called, asyndeton).
The writing system of Akkadian is among the most difficult in the world. Imagine writing Chinese with a stylus in clay and letting it evolve and then be simplified for 1,000 years, and use it to write Russian, and you will have some idea. As typically evolves first, Akkadian cuneiform glyphs are syllabograms, little pictures that used to represent something physical and how it was pronounced in Summerian, but have since been requisitioned for merely their sound. However, the original Sumerian meaning is still used, since the scribes who wrote this way were all fluent in both languages (at least originally). Because Summerian phonology was markedly different from Akkaidians, very difficult hacks were needed. Long vowels and emphatic consonants were seemingly absent from Sumerian, so Akkadian īnum might be written i-nu-um in the cuneiform.
Take, for example, the sign 𒄑.
- It may represent the sound *iṣ, part of the spelling of iṣ-ba-at for iṣbat or ki-iṣ-rum for kiṣrum
- It may represent the word iṣum, leaving you to fill in the blanks as to the case ending or lack thereof
- It may represent the determiner TREE telling you that the following is the name of a tree or type of tree, which you might not have otherwise known
Sumerian had a lot of glyphs for the same sound. Also, each glyph typically had one than one sound it could represent. The most common glyph for a given syllable is written without any diacritical marks. The second most frequent with ´, the third with `, and then subscript numbers. Sumerian syllables were V, CV, VC, or CVC. Akkadian long vowels are often not indicated at all in the Sumerian spelling, though usually they are at the end of word. Consonantal gemination -- so important in Semitic languages -- is also not spelled in the cuneiform either. CeC also typically stands in for CiC too.
Of course, as the oldest texts in the world, there are often gaps and damages, making reading all the more difficult. For the learner, there are also differences in the style of writing over the 3,000 year history of the language, ranging from Old Babylonian to Neo-Akkadian, where the characters were simplified and cleaned up.
Unicode does a passable job at containing all possible symbols, but there is a growing list of errors. They are due for updates and corrections soon.
If you are thinking about learning cuneiform, you will never remember it simply by looking. I highly recommend whittling some square chopsticks into styli and buying a few kits of Playdoh. Without tactile re-enforcement, none of this will stick in your brain at all.