ASSYRIAN LANGUAGE. The Babylonians, though Semites, learnt the cuneiform (q.v.) script from the non-Semitic Sumerians, whom they found in southern Mesopotamia in the 3rd millennium B.C. ; and i,000 years afterwards the northern Assyrians adopted it. Sumerian had 600 signs, constituting thou sands of ideograms (signs depicting ideas). Some signs had only word-values, others both these and syllable-values. The Baby lonians primarily used only the syllable-values of these signs. But they freely borrowed their word-values, as a kind of shorthand, using in reading (as variants show) the Babylonian equivalent.
The use of Sumerian, when no longer spoken, in the temple services led the Babylonians and Assyrians to compile lists of signs and vocabularies, sometimes of more than 30o ideograms, giving the sign together with its name, Sumerian pronunciation and Semitic meaning, which they often glossed with synonyms. They also drew up lists of Sumerian paradigms and dialectical forms with Semitic renderings in parallel columns. Some lists contain foreign (e.g., Cossaean or Hittite) words. These and liturgical texts with interlinear translations have proved indispensable in the decipherment of cuneiform documents.
Assyro-Babylonian (Accadian), the oldest known member of the eastern group of the Semitic languages (see SEMITIC LAN GUAGES), had affinities with all the other groups, yet not enough to be called a mixed language. Thus with the central (Amorite Aramaic) it prefers a to a or o (cf. Bab.-Ass. tabu=Aram. tabh with Hebr. tobh "good") ; with the western (Canaanite-Hebrew Phoenician) group it shared much of its vocabulary (cf. Ass.-Bab. anaku = Can. anuki = Hebr. 'anoklii = Phoen. anech with Aram. 'and "I") ; and with the southern group it retained the case-endings (as in Arabic) and employed -k instead of -t for the first person singular in the permansive or perfect tense (as in Ethiopic). Babylonian and Assyrian differed only dialectically; e.g., Bab. gdtu=Ass. qatu "hand" and Bab. ashtur=Ass. altur "I wrote." During more than 2,00o years changes crept in; e.g., Bab. lawn or lames= Ass. lames, labes or Tapes "to surround." Each also exhibited variations within itself : thus s was sometimes preferred to sh in early Babylonian (as in sI for s/ies "he") and late (vulgar) Assyrian (as in Asdudu=Hebr. 'ashdodh "Ashdod"). Cappado cian, an Assyrian offshoot, also substituted s for sh before i and showed other peculiarities, such as the change of k to g before a and of t to d before u and the loss of the more strongly articu lated consonants.
Assyro-Babylonian stood out from the Semitic languages in sev eral ways. Its script showed the primitive vocalization and, par tially, the accentuation. Its connection with Sumerian brought in foreign words under a Semitized form. The gutturals ' ('dleph), li, li and ` (`ayin) were mostly weakened into vowels : e.g., Bab. Ass. aldku "to go" (cf. Hebr. halakh "went") and Bab.-Ass. Ulu (=Hebr. ba`al "lord"), although h always and ' and h sometimes were preserved, as in Bab.-Ass. ina'ddu "to be much" (cf. Hebr. rne'odh "very") ; so w and y generally disappeared, as in Bab. Ass. ides "to know" (cf. Hebr. yadha` "knew") . Assimilation of vowels (as in unikkir for unakkir "changed") was common and of consonants (as in nidittu for nidintu "gift") regular; final vowels were sometimes dropped, as in kashddk for kashddku "I am seized." Many uncontracted forms, lost in the cognate lan guages, were preserved in the older, though contracted in the later, speech; e.g., Capp. idu(w)ar=Bab.-Ass. ituar and itdr "re turns." It sometimes had sh where the other languages have h or ' ('alepli) ; namely, in certain pronouns, as in shes (cf. Hebr. lies) "he" and so on, and in the causative prefix, as in shuzkur (cf. Hebr. hizkir and Aram. 'adhkar) "mentioned." The pronominal suffixes with verbs alone expressed the direct, but with an affixed -m the indirect, object.
Certain nouns took the feminine ending in Assyro-Babylonian but not in the cognate languages, like Ass.-Bab. irsitu=Hebr. 'eyes "earth," or, though elsewhere masculine, were feminine in it, like Ass.-Bab. kabittu=Hebr. kabhedh "liver." Special vowels dis tinguished the cases (sing. nom. -21, acc. -a, gen. -i; du. nom. -a, acc.-gen. -e; plur. nom. u, acc.-gen. -i), to which various affixes might be added, though with little change in significance. These endings, however, soon sporadically and later generally, were used indiscriminately, and various transcriptions (cf. Gr. µeTEpB=Bab. Ass. mitertu "rain") show that they were usually disregarded in speaking. The dual, too, tended to fall into desuetude.
The verbs employed the usual Semitic inflections and derived themes, with modifications. There were three instead of two tenses : a "permansive" which expressed state, corresponding with the Semitic perfect (Bab.-Ass. zakir "was remembered"=Hebr. zakhar "remembered") ; a preterite corresponding with the im perfect (Bab.-Ass. izkur "remembered"=Hebr. yizkor "was re membering," "remembers," "will remember") ; and a present corresponding with the Ethiopic imperfect (Bab.-Ass. izdkar "re members" = Eth. yezdker "is remembering," "remembers," "will remember"). Subordinate verbs were marked by the termination -u or sometimes -a, which often in independent clauses had also the sense of the German her. The derived themes are nearly as numerous as those of Ethiopic, and one of these, or the perman sive, was used in place of a passive formed by modification of the vowels.
Syntactically, it showed no definite affinities. It shared several idioms with Hebrew and used the enclitic -ma for u "and," like the Hebrew "wdw-consecutive" but without its effect on the fol lowing tense, to connect each fresh verb in the narration of a series of events; unlike Hebrew, it often employed asyndeton and was rich in adverbs, co-ordinate and subordinate conjunctions, and other particles. Like Aramaic and Syriac, it often substituted a periphrasis with shd "of" for the "construct state," and it tended under Sumerian influence to throw the verb to the end of the sentence, as Syriac and Ethiopic frequently did. In prohibitions it used id "not" with the present or a "not" with a tense resem bling but hardly identical with the preterite : for example, id tazdkar or tazkur "do not remember." In prose the style was heavy and unadorned, often almost barbarous. In poetry there was little attempt at grace or charm, but the order of the words was more varied and the vocabulary somewhat fuller; the lines, which a caesura sometimes broke, were of roughly the same length, and there were occasional attempts at parallelism and rhythmic beat. (G. R. D.)