No CrossRef data available.
Article contents
Some Observations on Nuzu Akkadian
Published online by Cambridge University Press: 22 April 2026
Extract
The Nuzu tablets, excavated by an American expedition, have consumed a good deal of the labor of American Assyriologists in recent years and will continue doing so for some time to come. After the publication of the material has far advanced, the grammar of the new dialect is in progress of reconstruction. The efforts made in this direction have so far been of a merely descriptive kind. It seems time now to apply also the linguistic line of approach to Nuzu Akkadian. Though the registration of the philological facts is the necessary basis for such work, it will in turn help to clarify difficulties which the descriptive grammar is unable to remove.
Information
- Type
- Research Article
- Information
- Copyright
- Copyright © 1938 Linguistic Society of America
References
1 The following abbreviations will be used. N = E. Chiera, Joint Expedition with the Iraq Museum at Nuzi (5 vols.); H V = E. Chiera, Excavations at Nuzi I; H IX = R. H. Pfeiffer, Excavations at Nuzi II; Berkooz = M. Berkooz, The Nuzi Dialect of Akkadian (Language Dissertations 23).
la w is rare at the beginning of words, e.g. wa-al-du 39:6, 12; wa-la-az-zu 53:7. In most cases w in this position has disappeared (as in the other dialects); example as-bu 99:7 etc.; aš-bá-ak 10:19; aš-bá-ku-mi 60:25 follow the Babylonian practice, but ur-ka-az-zu 30:12, 42:14 the Assyrian.
2 See Berkooz 50 f. It is worth noting that the glide w in Šawuška = Šauška (cf. Friedrich, Anal. Or. 12 129, fn. 5) and in cases like Ninuwari = Ninu(a)-ari may be spelled with m. Such an inorganic m must not be regarded as formantic, as L. Oppenheim, WZKM 44.197, does.
3 The orthodox view asserts on the contrary that intervocalic w had become m (Ungnad, Bab. Ass. Gram. 2, 14). This necessitates the further assumption that this secondary m (together with etymological m) later on became w again. Since to- in initial position in Bab. and Ass. had disappeared at roughly the same time, there was no longer need for the sign wa.
4 The view expressed in the text seems to be shared by Fr. Thureau-Dangin, see his Syllabaire Accadien 34.
5 See the collection of C. H. Gordon, BASOR 64 23-8.
6 ti is spelled ‘ti’, 'di' or 'te' in the same word saṭir'. With ‘ti’: 22:40; 28:14; 32:43, 54:23; 58:47; 61:32; 66:49; 67:31; with ‘di’: 63:35; 65:28, 50; with ‘te‘ 55:48.
7 For tu there can be quoted with 'tu' (rare): a-na ba-al-ṭú-ti-im-ma 8:14; with 'du': i-pa-aṭ-ṭù-ur 27:12; il-ṭù-ru 59:16; 65:47; with ‘tum’: bal-ṭu 4 23:11; 30:14. For ṭa for which a special sign has never been developed ‘ta‘ is used: ḫu-ṭd-ar-ri 3:60; ṭá-a-ti 8:26, 28, 67; iṭ-ṭá-ba-aḫ-šu-nu-ti 6:43.
8 As for qi, the observation of Berkooz (10 f.) can be referred to, that 'ki' serves for qi (as for ki and gi) and 'gi' for qe (as for ke and ge). It should be added that the same distribution is valid for the Mitanni letter (F. Bork, Mitanni-sprache 17).
9 qu is expressed by ‘ku‘ : thus in the frequent forms of leqū, e.g. il-qú-ú 47:4, etc.; 48:36; te-el-qú-ú 70:16, 24; ni-il-qú-ú-ma 74:19; i-liq-qú-ú 51:18, etc.; li-qú-ú-ma 84:4; il-te-qú-ú 7:47; 21:19; 22:13; 54:7; 74:8; il-te-qú-ma 8:7; furthermore ir-te-qú-šu-ma 6:51; šu-qú-ul-ta-šu 97:2; il-tar-qú 1:47.
10 This sign, used for zé in Old Babylonian, is almost entirely absent; see Berkooz 11. In this volume the name Pa-i-zé-ni 95:21. For ṣi there serves ‘zi’: ú-uṣ-şí 23:15; 30:18; 36:7; 58:40; ú-še-iṣ-sí 23:18; 63:12; uṣ-te-ṣí 45:8; i-iṣ-ṣí-du 88:10; ma-ḫa-ṣi 1:34, mi-iḫ-ṣí 72:14, 20; ṣí-di-ta 8:7, 11; ṣí-mi-it-tu 100:13 etc.; ṣí-mi-it-ti 8:25; iṣ-ṣí-el 75:14; iṣ-ṣí-e-lu-ma 75:3.
11 ṣu is represented by ‘zu’: ú-uṣ?-ṣú-ú 34:25; ú-še-iṣ-ṣú-ú 61:28; ú-še-iṣ-ṣú 60:18; it-ta-ṣú-ú 8:11; ú-ṣú-ur-ti 73:12; pè-ṣú-ú 100:16; it-tar-ṣú 72:12.
12 Examples, Kramer 68; Gordon, Nouns 9.
13 Since F. Bork's statement in Die Mitannisprache (1909) 15 f. this has been uncritically repeated time and again; most recently, M. Berkooz 39 ff.
14 Fr. Thureau-Dangin, Le Syllabaire Accadien v.
15 Even for Ḫurrian the existence of different sets of stops has been proved by the Ras Shamra documents, cf. J. Friedrich, Anal. Or. 12.130 f.
16 'sa': i-SA-as-sí 67:27; 96:10 (usually i-ša-as-sí with the variant i-sà-as-sí, e.g. 29:12) of šasū 'call'; SA-ṭi-ir 58:47 (usually ša-ṭi-ir or šá-ṭi-ir) of šaṭāru ‘write’; SA-dì-in-ni-e 1:38, etc. seems to be a special case, cf. Sum. s u. u d. d i n. gigir ŠL 354, 335. ‘si’: SI-id-dì 15:9 (instead of ši-id-dì). ‘su’: customary in sutān (Berkooz 11), but even here replaced by šu (e.g. 67:8).
17 'za' for sà: it-ta-sà-aḫ-ma 10:21; sà-a-mu 98:4; ri-ik-sà 23:18; ir-pí-sà-an-ni-mi 72:7; ir-ta-ap-sà-an-ni 3:55, 60; 8:32. 'zi' for sí: si-ḫi 73:6 etc.; sí-i-bit-e 49:26; sí-is-sí-ik-ta-šu 32:8; ra-pa-sí 3:32; ri-ik-sí 44:1; 55:1; 57:11; various forms of šasū 'call', such as i-ša-as-sí (pass.) and al-ta-sí-iš 4:12. ‘zu‘ for sú: ḫu-us-sú-sí 69:18; i-pár-ra-sú 69:17; ir-ku-sú(-u) 44:3; 55:3; ir-ta-ak-sú 31:14; i-ša-as-sú-ú 16:16; 51:29; i-ša-sú-ú 59:7; il-ta-sú-uš 71:15.
18 Instead of ša also šá sporadically occurs: šá-ṭì-ir 61:32; šá-te-ir 53:48; ú-ri-ḫul-šá 60:22; 61:17.
19 The use of 'az' and 'uz' for as and us is noteworthy; for classical Old Babylonian employs 'áš' and 'uš' with these values. 'az' for as: ar-ta-ka 4-as H IX 145:21; i-ša-as-sí (pass.), 'uz' for us: ḫu-us-sú-sí 69:18.
20 The rare occurrences of ‘áš‘ must be mentioned here. I find the sign only in one genuine Akkadian word, namely in na-ÁŠ-qú 95:6; the corresponding feminine na-sí-iq-tu 4 H IX 17:6 shows that na-ás-qú must be read as in Old Babylonian. Is this an archaism? The sign, possibly denoting some variety of a sibilant, occurs also in Ḫurrian words, e.g. in ka 4-ás-ka/i (cf. Gordon, Nouns 80 f.), interchangeably with aš (e.g. ka 4-aš-ka/u 64:6, 13). Cf. also the city a-ka 4-ás 48:17 (notice a-ka-a-a-še meš H IX 34:4).
21 Fr. Thureau-Dangin, RA 30 93 ff.; cf. A. Goetze, Orient. 6.16 f.
22 It would be a promising undertaking to investigate the syllabary of the Akkadian texts from the Ḫurri/Mitannian territory.
23 F. Weissbach, Die Achaemenideninschriften zweiter Art 29.
24 On the nominal cases which sometimes seem confused see below p. 139 f.
25 As in Old Babylonian.
26 A new treatment of this phenomenon is badly needed; the last one was given by P. Haupt, AJP 8.265-91 (1887).
27 The specific conditions remain to be established. Here belong the numerous present and pi'el forms like tezerreb, iqebber, iqerreb, uḫemmeṣ, umeššer, etc. It may be added that all this has nothing to do with the Assyrian vowel harmony which is entirely absent (the cases quoted by L. Oppenheim, WZKM 45.43 are to be explained otherwise).
28 For gg, Berkooz 44, for dd, ibid. 45, for bb, ibid. 49. It may be asked whether i-na-an-ZA-ar-šu (present of naṣāru) testifies to the nasalization of the peculiar spirant (A. Goetze, Or. 6. 12 ff.) contained in this root.
29 See Berkooz 62 f.
30 For this section cf. C. H. Gordon, AJSL 51.1-21.
31 W. von Soden, ZA NF 6.181 ff. gives the best information about the forms extant in the older dialects of Akkadian.
32 For this latter (C. H. Gordon 3, fn. 3) an-na-aš-ni 5:4 is a new corroboratory example. Cf. furthermore an-na-ši-in KBo I 23:10. The an-na-aš, attested in the epic dialect (W. von Soden, ZA NF 6.184) seems closely akin.
33 J. Lewy, Stud. zu den altass. Texten aus Kappadokien 11.
34 I noted for the second person i-na-an-din-na-ku-nu-ši H IX 5:11; for the first iz-za-ak-ra-an-na-ši N 324:14 and it-ta-ad-nu-na-si N 360:20.
35 Thus after the -ū of the 3rd plural: it-tab-lu-uš 8:14; uš-te-ši-bu-uš 6:53; idukkū-uš 74:26; 75:31; ú-ka 4-lu 4-uš 24:6; ú-me-eš-ši-ru-uš 8:5; id-du-uš 68:5; it-ta-du-uš 38:34; 41:21; 71:24, 32; 73:33; it-ta-an-nu-uš 40:13; iš-ta-lu-uš 70:11, 16; 73:9, 18; il-ta-sú-uš 71:15. Also after -i of tertiae infirmae: al-ta-sí-iš 4:12. The protection of an additional element after the verbal form preserves the u of the suffix: e.g. id-du-šu-mi 71:21; ni-il-ta-sí-šu-ma 4:6.
36 See W. von Soden, ZA NF 6.176 ff.
37 mannu-mē ‘whoever’, minum-mē ‘whatever’, immati-mē ‘whenever’ pass.; cf. also i-na a-i-im-me-e u 4-mi ‘on whatever day’ 91:10; a-i-ka 4-am-me-e ‘wherever’ N 328:16; 365:7.
38 Not mentioned by Gordon. attu-ya 1:18; N 338:29; 342:15; 359:9; 370:17; attu-ka N 321:41. On ša at-ti-ḫu N 321:37 see J. Friedrich, ZDMG 91.212.
39 C. H. Gordon 19 f.
40 K. F. Müller, Das assyrische Ritual 1.42.
41 C. H. Gordon 21. Add now 7:49 ff.: ‘Ḫašip-abu has collected clothing for the ubaru people; the clothing of the others he has returned; the garments of mine alone he retained’; 56:12 ff.: ‘the other sons of mine are grown up (and) have taken wives; but thou didst not take a wife.‘
42 For the meaning notice the contrast with allūti (see preceding note) and e.g. 10:13 ff.: ‘Two sheep Peškilišu has taken; the one belonging to the soldiers they have released; mine alone they have retained’.
43 Cf. most recently A. Schott, OLZ 1937 col. 362, fn. 1.
44 For this section cf. C. H. Gordon, Bab. 16.1-153.
45 šu-ti-iq 31:13; ši-ti-i-iq 57:9 (both for šūtuq); mi-du-ud 92:8 (for muddud). Cf. Berkooz 36 f.; L. Oppenheim, WZKM 45.43. Such alleged genitives are kaspipí 9:13; tup-pí 54:22; 61:30; ú-ri-ḫul-ši 62:18. The form of a nominative is encountered in it-ti X IḪa-bi-ru-↑ 40:2; ama-bu-ul-ta-nu 60:47.
46 ‘tum’: a-na pa-ni ši-bu-ti an-nu-TUM 54:2f.; 95:2 (cf. from Old Assyrian I. J. Gelb, AJSL 53 43 f.). ‘rum’: qāt (or aban) X tup-šar-RUM pass. (with ri e.g. N 493:26; 530:24).
47 Examples are abundant; only a few can be quoted: 1 en-zu IḪu-ti-ia a-na ia-ši iddinana 10:1 f.; ù kirū an-nu-um-ma IHu-i-ba-a-bu ú-za-ak-ka 4 21:11 f.; ù 1 amtu SIG5.GA na-ás-qú ... IAr-ša-wa a-na IE-ḫé-el-te-šup i-na-an-din 95:6 ff. With circumflexed-ú, e.g.: ùkirū an-nu-ú ITul-pu-un-na-a-a ... ú-ka 4-al 21:14 f.; ù ša-nu-ú UDU Iki-bi-ia il-te-qè 6:53 f. With the first person of the verb e.g. 6 šqil kaspu şa-ar-pu a-na IPeš-ki-ili-šu at-ta-din 9:2 ff.; 3 imēr ku-nu-šu meš ... a-šar IPu-ḫi-še-en-ni ... el-te-qè-mi 62:24 ff. With the third person plural of the verb : ù tup-pu an-nu-ú i-na 3 šanāti meš ù i-na 4 šanāti meš a-na pa-ni-šu-nu i-ša-as-sú-ú 51:26 ff. For contradistinction compare the consistent use of the cases in the frequent phrase šumma eqlu paqirāna irtaši, or for instance ù IKi-pa-a-a am. mu-ru l na-am-zi-ta a-šar IŠúk-ri-ia il-te-qè 6:16 f.; ù IIn-ni-ku ù IA-kap-še-en-ni ṭup-pa an-ni-ta il-ṭù-ru 59:15 f.
48 See L. Oppenheim, Af O 11.60. In the new volume, e.g. ša al-lu-ti TÚG.ṭI.-A.MEŠ ut-te-ir-ru ù ud-du-ia lu-ba-ri ik-ta-la 7:51 f.; i-na pa-na-ni-im-ma IA-ka 4-am-mu-uš-ni a-na aš-šu-ti iddin ù 40 šiqil kaspi-ia a-šar mu-ti-ia il-te-qú-ú 54:5 ff.
49 X Y Z 3ameluti ma-an-za-ad-du-uḫ-li-e ša IḪa-na-te 43:10 ff.; amēlu ana amēli ma-ḫi-iṣ pu-ta H IX 74:11 f., and frequently in the texts there following.
50 Cf. C. H. Gordon, Bab. 16.17 f.; also Berkooz 17 f. The formation in -ānu/i occurs only sporadically; DINGIR.MEŠ-ni (i.e. ilāni N 89:10; also in the new volume 47:3?) is offset by DINGIR.MEŠ-e (i.e. ilē N 352:26); še-ir-a-ni meš N 551:14 by UZU.MEŠ-ra N 360:20, 25.
51 2 ka 4-lu-mu meš NITA ḫu-ra-pu 66:12 f.; UDU.MEŠ an-nu-ú 66:30; 10 SAL. MEŠ nu-ul-lu-a-ú 32:15; cf. also (with -i instead of -u) 4 DINGIR.MEŠ GAL. MEŠ an-ni-i 50:19.
52 See C. H. Gordon, Bab. 16.20. It must however be emphasized that the grammatical agreement becomes much more consistent than Gordon realizes, as soon as the rules established in the text are taken into consideration. In many cases a more correct transliteration of ideograms furnishes better grammatical agreement.
53 For this section cf. C. H. Gordon, RA 31.53-60.
54 W. von Soden, ZA NF 7.131.
55 A. David, RA 25.99 f.
56 For šulušū, ḫumušāu see Gordon, l.c.
57 D. Cross, Movable Property 27, fn. 43, is apparently mistaken when she separates tiši'u from this group; cf. H IX 104:12.
58 For this section cf. S. N. Kramer, AASOR 11.63-119.
59 Pronoun and adjective exhibit the same feature; it seems that the native language of Nuzu had no specific form for the feminine.
60 ni-il-qú-ú-ma 74:19; ni-ik-ki-ru-ma 74:21; ni-it-tam-ga 5-ru-mi TCL IX 12:20.
61 aš-bá-ak 10:19; ti-te -en-na-ak 29:4. But ap-la-ku-mi 62:29; 94:12 and also 67:22; aš-bá-ku-mi 60:25; ḫu-ub-bu-la-ku-mi 73:5; 96:5; na-ta-ku-mi 73:8; š-ba-bu-mi 73:19.
62 See A. Goetze, JAOS 56.297-334.
63 Compare, e.g., attadin 1:19 with la addin 1:20; elteqē-mi 6:58 with la dqē-mi 6:56. Cf. JAOS 56.313 ff.
64 lu-ul-lik-mi 1st sg. 70:7; li-iq-bi-šu-nu-ti 3rd sg. 76:13; li-id-di 3rd sg. H V 104:16; 106:14; li-pu-uš-ma 3rd sg. 23:11. The exception lu-ša-ka 4-al 3rd sg. 23:11 may be due to the interchange of u and i (above fn. 45).
65 The treatment of this question by Kramer, AASOR 11.79 ff., is not altogether satisfactory. The prefix of the 3rd person is i- already in Old Akkadian, so that no a at all is involved in this process. It has nothing to do with the Babylonian umlaut nor with the Assyrian vowel harmony which, in their turn, must be strictly kept apart.
66 Cf. W. von Soden, ZA NF 7.133 ff.
67 There is no proof whatsoever that the preposition lit is a contraction of liwit (so also Berkooz 51). Already C. J. Gadd (RA 23.104) has correctly referred to P. Jensen, KB VI 2.5*. The preposition is also Middle Babylonian, see H. Waschow, Bab. Briefe aus der Kassitenzeit 17.
68 untu also in the Amarna tablets (letters from Assyria and Mitanni) and in Boğazköy (see R. Labat, L'Akkadien de Bogh. 226). It seems likely that untu is also contained in šu-un-tu 4 65:48; possibly ‘the year he wrote the tablet (is) the one when he brings the ašūḫu wood from the country of Ḫanigalbat’.
69 As regularly in Amarna and Boğazköy this conjunction is frequently construed with the indicative of the verb.
70 E.g. la a-wa-tù-mi la a-ni-ik-šu-mi 4:10; la ni-ku-ul-mi 5:13; la ik-ki-ru-šu-nu-ti 74:9; la iq-bu-ú-ma 75:4; la za-ku 89:2.
71 Examples for ù in almost every text. The ù is frequently put after -ma: ni-i-nu ni-il-qú-ú-ma ù i-na-an-na ni-ik-ki-ru-ma 74:19 ff.; ù ṭup-pa-šu šạ IA-bu-uk-ka a-na pa-ni daiānē meš uš-te-li-ma ù il-ta-sú-uš ... 71:13 ff.
- 3
- Cited by