by Eckart Frahm
(SAA 17 2: 17–22)
On the other hand, clear proof of the official recognition of the Aramaic language comes from a set of bilingual lion‐weights of Shalmaneser V and other Assyrian kings; these objects carry short inscriptions both in Akkadian and in Aramaic. However, the recognition seems to be based on a “double standard,” since the Akkadian inscriptions mostly say “n mina‐unit(s) of the king,” whereas the Aramaic ones read “n mina‐unit(s) of the land.” Late Assyrian palace reliefs from Tiglath‐pileser III’s reign until Assurbanipal’s also illustrate the increasing need for writing in Aramaic: Assyrian and Aramaic scribes were regularly represented side by side, with the former writing Assyrian cuneiform on their clay tablets, while the latter wrote onto scrolls, most likely using the Aramaic alphabet (Figure 17.2). The textual evidence for Aramaic scribes at Assyria’s court confirms this development.
Figure 17.2 Assyrian and Aramaic scribes as depicted on a Neo‐Assyrian relief.
Source: Reproduced with permission of T. Lipasti.
As already pointed out, most Aramaic inscriptions from the Neo‐Assyrian period are unfortunately lost because they were written on perishable materials (papyrus and leather). But Aramaic alphabetic epigraphs scratched on clay tablets of a mainly legal nature have been found throughout the Neo‐Assyrian Empire (Fales 1986, 2007: 99–102; Figure 17.3). Aramaic writing is also preserved on some stone stelae, seals, metal objects, ivory, and pot shards (ostraca), but Aramaic characters painted in ink survive only rarely. Truly bilingual Akkadian‐Aramaic celebrative inscriptions on stone, which display more or less the same text in two languages, seem to be attested only during the early part of the Neo‐Assyrian period (Fales 2007: 106).
Figure 17.3 Assyrian‐Aramaic triangular corn‐loan docket (after Fales 1986, fig. 3).
Other languages
During their long history, the Assyrians came in contact with speakers of many different languages, several of which are known almost exclusively from personal names in Assyrian sources. In the Old Assyrian period, when Assyrian merchants were active in Anatolia, their foreign contacts are reflected in the hundreds of non‐Assyrian names that are mentioned in the extant documents, the names originating from the Neshite (or Proto‐Hittite), Luwian, Hurrian, and (Proto‐)Hattic languages. The Assyrian king Šamši‐Adad I was himself of Amorite origin. During the second half of the second millennium, Babylonian was the lingua franca of the Middle East, while a Kassite dynasty ruled over Babylonia; Assyria annexed large Hurrian‐speaking areas and was in regular contact with Indo‐European Hittites and Luwians living in atti. In the Neo‐Assyrian period, Assyria confronted – and interacted with – Šubria and Urartu in the north, Mannea in the east, Elam in the southeast, Babylonia in the south, and Syro‐Palestine (where West Semitic languages such as Ugaritic, Phoenician, Moabite, and Edomite were spoken), Egypt, and the Eastern Mediterranean in the west. All these regions had their own distinct languages.
Because of this multilingual environment, many loan words entered Assyrian, and, conversely, many Assyrian words were absorbed by foreign languages. Surprisingly, however, the word targumānu(m) “interpreter, translator, dragoman” is rarely mentioned in Assyrian sources (von Soden 1989: 353–4), although interpreters were certainly necessary and enjoyed official recognition. The problem of understanding the languages of distant countries was sometimes treated in official sources, as in an (unfortunately broken) account from Assurbanipal’s annals concerning a messenger of the Lydian king Gyges arriving in Nineveh, where no one could understand him (Borger 1996: 182, 218). Interpreters are never mentioned in contexts in which Assyrians were communicating with speakers of other Semitic languages, i.e., with Babylonians or West Semites; they are only attested when the Assyrians were dealing with people of Anatolian, Hurrian, Šubrian, Urartian, or Mannaean origin. Assyrian libraries have yielded large numbers of Sumero‐Akkadian bilingual texts, reflecting the important role Sumerian played in religion and scholarship, but there are also a few multilingual tablets that include words from other languages (Elamite, Lullubean, etc.).
Writing Systems
Cuneiform, probably invented by Sumerians in the mid‐fourth millennium BCE, was the writing system par excellence in the Near East for millennia. Cuneiform signs were produced by impressing a reed stylus on damp clay to make distinctive intelligible patterns, which the educated could read. Cuneiform writing was widely borrowed and adapted by speakers of the various languages in the region. Not restricted to clay, cuneiform was also written on stone, metal, wax‐covered writing boards, and other materials. The most prominent use of cuneiform writing for a language other than Sumerian was for Akkadian, which comprises both Assyrian and Babylonian dialects.
Akkadian cuneiform is a logo‐syllabic script based on its Sumerian predecessor and written from left to right. A word sign (or logogram) is a sign that represents an entire word; it is usually transliterated in modern scholarship in non‐italicized, small capital letters. A syllabic sign is a sign that stands for a syllable of the type V, CV, VC, or CVC (with V designating a vowel and C a consonant); these are normally transliterated in italics. Akkadian cuneiform is both homophonic and polyphonic, i.e., different signs can have the same phonetic value (distinguished in transliteration with an accent or an index number, e.g., il, íl = il2, ìl = il3, il4, il5), and a sign can have multiple logographic and/or syllabic readings (e.g., the sign can have the logographic values DINGIR = ilu(m) “god” and AN = OA šamāʾum, MA šamāʾu, NA šamê (pl.) “heaven,” as well as the syllabic values an and ìl). The use of polyvalent signs introduces the possibility of ambiguity, making the correct reading of each sign dependent on the context. Modern sign lists register cuneiform signs with both their respective logographic and syllabic readings.
Fortunately, for a particular period and area, the homophony and polyphony of cuneiform signs is limited. Moreover, the Akkadian writing system includes mechanisms, such as determinatives and phonetic complements, that help modern scholars choose the intended reading. The determinatives originate from logograms and indicate the class or category to which the following or preceding noun belongs. In Neo‐Assyrian, the most common determinatives are LÚ before a profession or gentilic, the so‐called Personenkeil m before male personal names, and the determinative d before divine names. Phonetic complements, on the other hand, are syllabic signs that usually repeat the last (or more rarely the first) syllable of a word rendered by a logogram, thus directing the reader to the word’s correct reading (e.g., the sign mu in DImu tells us that DI has to be read as šulmu). Determinatives and phonetic complements are optional devices that are dependent on the time of writing (the Personenkeil is rare in Old Assyrian, but is used systematically in Neo‐Assyrian) or the whims of the scribe.
During Assyria’s history, the use of cuneiform underwent drastic changes. Old Assyrian cuneiform, mainly known from documents owned by private individuals, is simple: the number of signs is small and the use of logograms and determinatives limited. Literacy was probably high during this period. Middle Assyrian and, especially, Neo‐Assyrian cuneiform are more complex. The scholarly tablets, in particular, which were written by a small group of highly educated specialists, used an extensive repertoire of cuneiform signs and values, so that lesser‐educated scribes might have struggled to interpret these learned texts.
Similar to cuneiform in terms of difficulty and learnedness are the hieroglyphic scripts that were used in the ancient Near East, not only for Egyptian, but also for Luwian. Many foreigners familiar with hieroglyphic writing lived in the Assyrian heartland in the Neo‐Assyrian period. Especially the knowledge of Egyptian scribes and ritual experts was highly valued at Nineveh’s court in the seventh century. Even though hieroglyphic texts are rarely discovered in Assyria, because their mostly perishable writing materials are now lost, a certain Assyrian fascination with hieroglyphs can be gauged from the so‐called Assyrian hieroglyphs, which are iconographic representations of Sargon’s and Esarhaddo
n’s royal titles (Finkel and Reade 1996).
In contrast to the complicated writing systems of Akkadian and Egyptian, Aramaic was rarely written in cuneiform or hieroglyphs; it was, instead, recorded in a consonantal alphabetic script. Following the invention of the alphabet in Canaan in the second millennium BCE, various alphabetic scripts appeared in the Levant, such as Phoenician and Ugaritic. These scripts had one character for each consonant of the language, requiring fewer than thirty signs. Yet despite their economy, alphabetic scripts spread slowly in the ancient Near East – until Aramaic gained prominence. In the 11th or 10th century, the Aramaeans adopted and further developed the Phoenician alphabet with its linear letter forms and right‐to‐left direction. With only twenty‐two signs, the Old Aramaic script is simple; however, the Aramaic language had more than twenty‐two consonantal sounds and, hence, some signs had to represent more than one sound. Moreover, the Aramaic alphabet, like the Phoenician and Ugaritic ones, only represents consonants. Vowels are not systematically indicated, “restricting knowledge of patterns of vowels and stress to some extent” (Millard 2007: 86) and causing interpretive challenges for the modern translator. Regardless, with the increasing Aramaization of the Neo‐Assyrian Empire, the Aramaic script became more and more common and probably displaced cuneiform as the dominant writing system shortly before the empire collapsed.
Main Features of Assyrian Grammar
In this section, we focus on the main features of the Assyrian dialect and some of the changes that occurred between Old, Middle, and Neo‐Assyrian. It should be noted that our knowledge of Assyrian will improve in the near future as grammars of Old Assyrian, Middle Assyrian, and Neo‐Assyrian are presently in preparation. Recent overviews of the related Babylonian dialects are listed in the section on Further Reading.
Phonology
Since Assyrian is an extinct language, we can only approximately reconstruct how the native Assyrians actually spoke. For Akkadian in general, the estimated pronunciation is based on the cuneiform orthography and on the knowledge of cognate languages such as Aramaic, Arabic, and Biblical Hebrew. In comparison to the original twenty‐nine proto‐Semitic consonants, the number of consonants in Akkadian is greatly reduced, probably due to the influence of Sumerian. In alphabetical order, the twenty consonants of Akkadian are: ʾ, b, d, g, ḫ, j (y), k, l, m, n, p, q, r, s, , š, t, ṭ, w, z. The approximated pronunciation of ḫ is “ch” as in Scottish “loch,” and the emphatic consonants q (qoph), (ṣade), and ṭ (ṭeth) were glottalized (as in Ethiopian and South Arabic; Kouwenberg 2003). In the Akkadian cuneiform system, signs ending in a labial, dental or velar stop or in a sibilant (except/š/) did not distinguish between voiced, voiceless, and emphatic. The aleph (ʾ) represents the glottal stop. The consonants w and j (y) are still pronounced in OA, but in MA, word‐initial w mostly disappears (although initial wa‐ becomes u‐); intervocalic w within a word appears mostly as b (awātum > abutu). By the Neo‐Assyrian period, the independent phonemes w and j cease to exist in Assyrian.
As with other Semitic languages, Akkadian has the three basic vowels a, i, and u, and the secondary vowel e. The vowels can be short or long, and are typically transcribed either unmarked (short vowels), with a macron ā, , ī, ū (long vowels), or with a circumflex â, ê, î, û (always stressed, resulting from vowel contraction). According to the Assyrian vowel assimilation, a short a in an open, unaccented syllable is assimilated to the vowel of the following syllable, e.g., MA abutu (nom.), abata (acc.), abete (gen.) “word, matter.” This feature is not found in the related Babylonian dialects but rather is particular to Assyrian, although several other phonological rules operate in both Assyrian and Babylonian Akkadian, for instance Geers’ law of dissimilation, according to which one of the emphatic consonants of a Semitic root with two emphatic consonants dissimilates (e.g., Semitic *abāṭu “to seize” became Akkadian abātu). Another rule applicable to both dialects is that the initial m in nouns dissimilates to n if the noun contains a labial b, p, or m (e.g., *markabtu “chariot” became narkabtu).
Morphology
Nominal and verbal roots
As a Semitic language, Akkadian has a consonantal root system. The root usually consists of three consonants (or radicals) as in k‐r‐b “bless,” although some biconsonantal or quadriliteral roots are also attested. Combined with various prefixes, infixes, and suffixes, the roots are converted into words with specific meanings according to the resulting consonant‐vowel pattern, e.g., karābu “to bless,” ikribu(m) “blessing,” and kāribu(m) “one who blesses.” The weak radicals ʾ, j (y), n and w cause assimilation.
Pronouns
Akkadian has independent personal pronouns – possessive, demonstrative/relative, reflexive, and interrogative – and pronominal suffixes. The Assyrian independent personal pronouns are shown in Table 17.1.
Table 17.1 Independent personal pronouns
OA/MA NA OA MA NA
nom. obl. acc./gen. dat. obl.
Singular 1c. I anāku yāti *yāti yāši (a)yāši
2m. you atta ku(w)āti *kuāti kuāša kāša
2f. you atti kuāši *kāši
3m. he šūt šû, šūtu šu(w)āti šuāti/u/e šuāšu šāšu
3f. she šīt šî, šīti šiāti šiāti/u šuāša šāša
Plural 1c. we nnu annu/i niāti *niāti nāši nāši
2m. you attunu kunūti *kunātunu kunāšunu kanāšunu, kāšunu
2f. you attina (only attested in OA) kināti *kinātina *kināšina *ka/ināšina
3m. they šunu šunūti šunātunu *šunāšunu šunāšunu, šāšunu
3f. they šina šināti šinātina *šināšina šināšina
Whereas OA has only one form for the oblique case of independent personal pronouns, MA has separate forms for the accusative/genitive and the dative. By the Neo‐Assyrian period, the archaic accusative/genitive forms disappear.
Independent possessive pronouns are frequent in OA but less so in MA/NA. They may be used as attributes or independently and are declined as adjectives (Table 17.2).
Table 17.2 Independent possessive pronouns
OA MA NA
Singular 1 mine yāʾum yāʾu iyû
2 yours ku(w)āʾum kuʾāʾu ikkû
3 his/hers/its šu(w)āʾum (m.),
šiāʾum (f.) *šuʾāʾu *iššû
Plural 1 ours niʾāʾum *niʾāʾu innû
2 yours kunūʾum kunāʾu ikkanû
3 theirs šunūʾum *šunāʾu iššanû
Yet even in OA, possession is mostly denoted by pronominal suffixes attached to nouns. Added to verbs, pronominal suffixes may indicate direct or indirect objects (Table 17.3).
Table 17.3 Pronominal suffixes
Nouns Verbs
Possessive Direct object Indirect object Direct & indirect object
OA MA NA OA MA OA MA NA
Singular 1c. ‐ī/, ‐ya ‐ī, ‐ya ‐i, ‐(y)a ‐ī/, ‐ni ‐ni ‐(a)m,
‐nim ‐a(m),
‐ne(m) ‐(an)ni
2m. ‐ka ‐ka ‐ka ‐ka ‐ka ‐kum ‐ku ‐(ak)ka
2f. ‐ki ‐ki ‐ki ‐ki ‐ki ‐kim ‐ki ‐(ak)ki
3m. ‐šu ‐šu ‐šu ‐š(u) ‐š(u) ‐šum ‐šu(m) ‐(aš)šu
3f. ‐ša ‐ša ‐ša ‐š(i) ‐š(i) ‐šim ‐ši/e(m) ‐(aš)ši
Plural 1c. ‐ni ‐ni ‐ni ‐niʾāti ‐nâši(n) ‐niʾāti ‐nâši(n) ‐(an)nāši
2m. ‐k(u)nu ‐kunu ‐kunu ‐kunu ‐kunu ‐kunūti ‐kunu ‐(ak)kunu
2f. ‐kina ‐kina ‐kina ‐kina ‐kināti ‐kina ‐(ak)kina
3m. ‐š(u)nu ‐šunu ‐šunu ‐šunu ‐šunu ‐šunūti ‐šunu ‐(aš)šunu
3f. ‐š(i)na ‐šina ‐šina ‐šina ‐šina ‐šināti ‐šina ‐(aš)šina
The demonstrative pronoun “this” is annium (OA), anniu (MA), and (h)anniu (NA); “that” is ammium (OA) and ammiu (MA/NA); OA also has allium “that” (for the distinctive uses, see Kouwenberg 2012). Their declension follows that of adjectives.
The d
eterminative pronoun is ša. It is generally indeclinable in Assyrian. The determinative ša is followed by a genitive and often used as “the one of” to form compound words (e.g., ša‐pān‐kalli, “palace supervisor,” lit. “the one before the palace”), as “of” in periphrases of the construct state (e.g., maartu ša šarri “the guard of the king”), or adverbially (e.g., ša‐šrāti “in the morning”). ša also functions as the relative pronoun “who(m), which, that, what,” introducing relative clauses.
Mannu(m) “who?” and mīnu(m) “what?” are the interrogative pronouns. They are declined for case, but do not have feminine or plural forms. In addition, the interrogative adjective ayyum (OA), ayyu (MA/NA) “which?, what?” agrees in case, gender, and number with the word to which it refers. The indefinite pronouns are mamman(a) (OA), mamma (MA), memmni (NA) “anybody,” and mimma (OA/MA), memmni (NA) “anything;” when negated these stand for “nobody” and “nothing.” The indefinite collective pronoun is (am)mala (OA), ammar (MA), (am)mar (ša) (NA) “all that; everyone who.” The noun ramunu “(one)self” with pronominal suffixes is used as a reflexive pronoun, e.g., ana raminīšu “for himself.”
Nouns
The Akkadian nouns are marked for gender, number, and case (Table 17.4).
Table 17.4 Noun declension
Masculine Feminine
OA MA NA OA MA NA
Singular
nom. ‐um ‐u ‐u ‐u ‐(u)tum ‐(u)tu ‐(u)tu‐(u)tu
Acc. ‐am ‐a ‐(a)tam ‐(a)ta
Gen. ‐im ‐e/i ‐i ‐(i)tim ‐(e/i)te ‐(i)ti
Plural
nom. ‐ū ‐ū ‐āni, ‐āti, ‐ūti, ‐ī/ ‐ātum ‐ātu ‐āti
