Published online by Cambridge University Press: 22 April 2026
The Nuzi dialect of Akkadian is known from several thousand cuneiform tablets discovered in the ruins of ancient Nuzi, about 10 miles southwest of modern Kirkūk, Iraq. These documents, dating from the middle of the second millennium B. C., were unearthed in the course of several campaigns sponsored by the American School of Oriental Research in Baghdad, the Semitic Museum of Harvard University, and the Iraq Museum in Baghdad. Following the publication of a small number of these records which had found their way to the British Museum and the Louvre, Dr. Edward Chiera began systematic excavations at Nuzi in the spring of 1925; other campaigns were conducted from 1927 to 1931, and were directed successively by Chiera, Pfeiffer, and Starr.
page 9 note 1 For discussion on this subject, cf. Contenau, Les Tablettes de Kerkouk, Babyloniaca 9.15–179; Gadd, RA 23.53, and Koschaker, NKRA 9–20.
page 9 note 2 Loc. cit.
page 9 note 3 Op. cit. 158, 160.
page 9 note 4 Op. cit. 19.
page 9 note 5 SA IV-V.
page 9 note 6 Op. cit. 162–175.
page 9 note 7 In view of the fact that the Nuzi and the Akkado-Hittite scripts are so closely connected, and owing also to the wide expansion of the Hurrians at this period, it is altogether possible that the Hittites and the Mitannians learned their script from the Akkadian-speaking Hurrians. For the Hurrian expansion cf. Speiser, Ethnic Movements in the Near East in the Second Millennium B.C., AASOR 13.13–54 (1933).
page 10 note 8 For the same use of BÍ in the letters of the Hammurabi period and in Boghazköi, cf. Labat, AB 9.
page 10 note 9 With rare exceptions: ki-en-na-at, N 210.6; 263.4.
page 10 note 10 For discussion of this subject, cf. Speiser, FL AASOR 10.24.
page 10 note 11 With rare exceptions: pa-gi-ra-an-na, G 2.24; pa-gi-ra-na, H I 62.14; pa-gi-ra, N 28.14.
page 10 note 12 The list is by no means exhaustive.
page 10 note 13 Cf. the forms of idū ‘to know’, pitū ‘to open’, which have also always a final e in compensation for the lost laryngal.
page 11 note 14 Cf. also el-ki-e, N 253.5. To the hundreds of times in which leqū is written with GI, there are not over twelve tablets in which it is written with KI.
page 11 note 15 KI.BAL.
page 11 note 16 For this form of phonetic complement, see I.6.K.
page 11 note 17 DIRIG.
page 11 note 18 Cf. Šu-ru-ia N 163.20. In both instances his father is Teḫišenni.
page 11 note 19 Written erroneously ṣi-ri-ri.
page 11 note 20 For the equation of šiqlu with SU, cf. Lacheman, JAOS 55.431 note 6.
page 12 note 21 For this reading of IM see the next paragraph.
page 12 note 22 NKRA 16 note 1.
page 12 note 23 DUMU.
page 12 note 24 ŠU.
page 12 note 26 RI.BA.NA.
page 12 note 26 A.ŠAG.
page 12 note 27 For dividing the word at the end of the line see I.7.
page 13 note 28 Gordon's reading (JPOS 15.32–4) of ù ba-nu, H I 43.31, as ù-ba-nu is still uncertain.
page 13 note 29 See under III.2.
page 13 note 30 RA 23.63–4.
page 14 note 31 Sumerian verbal forms are not so common. Beside this verb, which is used rather frequently, cf. BA.UG6 (imtūt) G 9.15; 51.8; I.BA (izūz), N 414.11; IN.-NA.AN.SUM (iddin), N 11.7; 28.9, 13; IN.SUM (iddin), N 230.14; MA.AN.SUM (iddinam), G 59.11; ŠU.BA.AN.TI (ilteqē), N 227.6; TIL.LA (balṭu), N 2.23; 313.5, and passim.
page 14 note 32 Written phonetically only four times: zi-di, N 435.22; zi-di-šu, N 6.5; zi-ta-šu, H I 71.32; zi-it-ta-šu, N 404.6.
page 14 note 33 Written phonetically only in: ma-ra, H I 60.14; ma-ru-šu, 303.11, 20; 531.1; ma-ri-šu, N 303.24; ma-ar, H I 65.12.
page 15 note 34 For the interchange of l and n in this name, see III.7.
page 15 note 35 For the change of n to m, see III.n.
page 15 note 36 For the syncope, see II.1.
page 15 note 37 Written erroneously ŠI.
page 15 note 38 The name of the scribe Ad-ri-zi, N 117.26, is no doubt the same as dAdda-ri-zi, N 83.43; 85.36; 116.16. Here the sign AD probably has the ideographic value of ‘father’, which is atta in Hurrian. Thus the name is to be read Atta-ri-ṣí, unless it is a case of scribal omission, and is, therefore, to be transliterated as Ad-<da>-ri-ṣí.
page 15 note 39 For discussion on this element, see III.8.
page 16 note 40 Transliterated here by a raised f.
page 16 note 41 Transliterated here by a raised d.
page 16 note 42 With very rare exceptions.
page 16 note 43 Cf. also Gadd RA 23 note 6, and Friedrich AfO 10. 294.
page 16 note 44 This name occurs only in non-Semitic proper names. Cf. Purves NSPN
page 16 note 45 IŠ4.DAR.
page 16 note 46 U. Sometimes also Mār-Ištar without d, N 64.14.
page 16 note 47 ŠU.
page 16 note 48 IM.
page 16 note 49 Once Ar-dTe-šup, N 414.2.
page 16 note 50 Once Mil-ki-dTe-šup, N 414.23.
page 16 note 51 Except in N 404.1, 7, 10, 14, 16, 23, where it is written It-ḫi-ib-du-gur.
page 17 note 52 NAGAR.
page 17 note 53 SURAL.
page 17 note 54 SIB.
page 17 note 55 APIN.
page 17 note 56 Cf. Speiser, A Letter of Saushshatar and the Date of the Kirkuk Tablets, JAOS 49.269–75 (1929).
page 17 note 57 For the same use of KI in Boghazköi, cf. Labat AB 14.
page 17 note 58 KUR.
page 17 note 59 Also in its shortened form, ME N 119.5; 462.3, and passim.
page 18 note 60 ŠEŠ.
page 18 note 61 DINGIR.
page 18 note 62 MU.
page 18 note 63 IGI.
page 19 note 64 Cf. also Klauber, Politisch-religiöse Texte 11.7; G 84 note 10, and FL 14 note to line 46.
page 19 note 65 BA.UG6.
page 20 note 66 1 zu-ḫa-ar-du, in line 2.
page 20 note 67 For phonetic complements with the plural, see I.5.
page 20 note 68 Restored on the analogy of many other tablets of the same category. Cf. N 104.1–4; 106.1–4, and passim.
page 20 note 69 EBUR.
page 20 note 70 KA.
page 20 note 71 NIGIN.
page 20 note 72 LÚ. This word is very seldom complemented.
page 20 note 73 This word is very seldom complemented.
page 20 note 74 KAB.
page 21 note 75 For the meaning, see note 78.
page 21 note 76 Restored.
page 24 note 77 DUG.GA.
page 24 note 78 Supplied from lines 19 (Ta-bu-gur) and 20 (Ta-bu-gur-ma). Gordon's transliteration of this name (cf. RA 1936 p. 4) as mḫi-kà-pu-ur is inadmissable. The following may also be added to Gordon's discussion of N 135: i-te-[is no doubt a verbal form of the root idū. Cf. la ni-te-mi, N 321.37a. iṣ-ṣá-ab-batat is not to be translated ‘was taken’ but rather ‘holds’ or ‘retains’. Cf. N 321.17, where the same phrase occurs with the verb ú-qa-al-lu instead of iṣ-ṣá-ab-batat. Finally it may be said, that the tablet in question has nothing to do with community responsibility, as Gordon thinks, but has to do with settling the dispute with regard to the authority over the district in question.
page 24 note 79 KASKAL.
page 25 note 1 For a full discussion of this subject, as betraying the Babylonian origin of the Nuzi dialect, cf. Kramer, The Verb in the Kirkuk Tablets, AASOR 11.62–119.
page 25 note 2 I. e., excluding the non-Semitic words and proper names.
page 25 note 3 Cf. especially Thureau-Dangin's treatment of the Ras-Shamra vocabulary in Syria (1931) 260–4, and Friedrich, Zum Subaräischen und Urartäischen; Analecta Orientalia 12.122 note 2, and 125 note 6.
page 25 note 4 And perhaps at the same time, also, under the influence of the neighbouring liquid or nasal. The only contrary example is uk-te-ši-du-ma, G 29.29.
page 25 note 5 For the reading of the sign ZI as ze, cf. SA no. 51. For ta > te see the following paragraph.
page 25 note 6 For the reading of the sign GI as ge or qe, see I.2.
page 25 note 7 For the reading of the sign QAR as qer, see I.2. The fact that this word is written with the aid of QAR and the following with KI (never qa-ar) indicates that the following examples are to be transliterated as i-qé-ir-ri-ib, etc.
page 26 note 8 Root šabāru. For discussion of this root, cf. Speiser, JAOS 55.440.
page 26 note 9 Kramer's restoration of this word (AASOR 11.93) as [un-te]-eš-ši-ir is wrong. The beginning of UM is very clear. Cf. also Gadd's transliteration.
page 26 note 10 For the reading of TAŠ as teš, see I.2.
page 26 note 11 Is the e- vowel in dSin-ir-ra-me-mi the result of the same change?
page 26 note 12 Treated, perhaps, as a non-Semitic proper name.
page 26 note 13 In both instances his father is Šeqaru.
page 26 note 14 In both instances he has a son Artešše.
page 27 note 15 In both instances he has a son Piru.
page 27 note 16 In both instances he has a son Eḫluti. For the loss of bi, see III.I.c.
page 27 note 17 For the reading of GAL as gel, see 1.2.
page 27 note 18 In all instances he has a son Artešup.
page 27 note 19 In both instances his father is Ennamilki.
page 27 note 20 In both instances he has a son Alippiya.
page 27 note 21 In both instances he has a son Šummiya.
page 27 note 22 In all instances he has a son Ilu-ma-aḫi.
page 27 note 23 Name of a canal.
page 28 note 24 Name of a district.
page 28 note 25 In both instances his father is Akaya.
page 28 note 26 Cf. also Ut-ta-zi-na, N 59.3. In all instances his father is Puitae.
page 28 note 27 In both instances he has a son Akawatil.
page 28 note 28 In all instances his father is Ḫupita.
page 28 note 29 DUG.GA.
page 28 note 30 In all instances his father is Ariburašše.
page 28 note 31 On the analogy of ru-pu-us-sú.
page 28 note 32 In both instances he has a son Hutipapu.
page 29 note 33 Because of the very frequent occurrence of syncope, no attempt at listing will be made, except for unusual instances and proper nouns.
page 29 note 34 In all instances his father is Nantešup.
page 29 note 35 In both instances his father is Taya.
page 29 note 36 For the various spellings of this name, see I.3.d.
page 29 note 37 Unless it is a case of scribal omission, which is probable.
page 29 note 38 For interchange of E and I, as indicating a phonetic process, see the next paragraph.
page 30 note 39 One must be very careful not to consider the occasional interchange of elqē and ilqē (cf. el-ki, N 260.5, for il-gi), elteqē and ilteqē (cf. el-te-gi, G 6.6 for il-te-gi; il-te-gi-mi, N 107.20; 342.17 for el-te-gi-mi) a phonetic process (cf. Kramer, The Verb in the Kirkuk Tablets, AASOR 11. 116, line 4), because here the interchange is not of e and i-vowels, but rather of the grammatical forms of the first and the third persons singular, which is very common in the Nuzi tablets. Cf. for example: i-bal-qa-tu 4, N 174.2 (for a-bal-qa-tu); i-din, N 106.9; 119.11 (for a-din); it-ta-din, H II 108.5 (for at-ta-din); a-na-din, N 432.8, 9 (for i-na-din); i-ša-az-zi, N 479.18 (for a-ša-az-zi); il-ta-zi, N 174.4 (for al-ta-zi), etc.
page 31 note 40 Singular form of verb with plural subject.
page 31 note 41 Cf. however the following irregular forms, with those listed above: uš-ta-pa-i-lu, N 251.4; uš-ta-pa-i-lu, N 282.4; and ul-ta-pa-i-lu, N 225.4.
page 31 note 42 The following lists are arranged in the order of verbs, nouns, and proper names, because there does not seem to exist any common phonetic law to which these changes may be ascribed (for changes of i to e, see 11.2.).
page 31 note 43 Appended to the first person, and occasionally to other persons also.
page 32 note 44 Third person singular for first person singular.
page 32 note 45 Cf. also ap-la-ku-mi, alongside a-ša-zi-me in the same tablet, lines 8 and 10.
page 32 note 46 Cf. also ad-du-ia-me, N 359.9.
page 32 note 47 Cf. also the demonstrative pronoun an-ni-te, N 33.19, alongside the common an-ni-ti. Mention may also be made here of the Hurrian genitive-locative element -we which interchanges with -wi, as may be seen from such forms as alAr-ša-li-be, N 403.30, and alAr-ša-li-bi, N 550.7.
page 33 note 48 In all instances he has a son Šuriḫil.
page 33 note 49 In the last two instances his father is Ḫutiya.
page 33 note 50 In both instances his father is Zikaya.
page 33 note 51 In both instances he has a son Piru.
page 33 note 52 In both instances he has a son Taya.
page 33 note 53 In both instances he has a son Eḫluti.
page 33 note 54 In both instances his father is Niḫriya.
page 33 note 55 In both instances he has a son Akkulenni.
page 34 note 56 Interchange of u and i occurs also in the Hurrian of Ras-Shamra. Cf. sa-r[i]-ti, RS II 5: šu-ru-ti, RS II 4; ka-r[i]-we, RS II 8: ka-ru-we, RS II 9; tu-bu-e, RS II 23: ti-bu-šu, RS I 21.
page 34 note 57 In both instances his father is Ḫadue.
page 34 note 58 In both instances his father is Ḫašiya.
page 34 note 59 In both instances his father is Dubkiya.
page 60 note 60 Name of a month. For the names of the months of the Nuzi Calendar, see Gordon, in Rivista degli Studi Orientali 15.253–7.
page 35 note 61 In both instances his father is Zikaya.
page 35 note 62 In all instances he has a son Urḫiya.
page 35 note 63 In both instances his father is Aḫiya.
page 35 note 64 In all instances he has a son Taya.
page 35 note 65 In all instances he has a son Ḫuparša.
page 35 note 66 In all instances he has a son Mutta.
page 35 note 67 Kramer (AASOR 11.95, 193) is wrong in taking id-da-na-aš-šu, N 11.9, and id-da-na-aš-šu-nu-ti, N 31.12, as forms of I1 stem. There is no doubt that these forms are of I2 with assimilation of d to the following n.
page 36 note 68 In both instances his father is Ḫupita.
page 36 note 69 The phonetic process was probably Sinirameni > Sinrameni > Sirrameni.
page 36 note 70 The phonetic process was probably Siniri > Sinri > Sirri.
page 36 note 71 One must not assume at present that the apparent loss of i after the verbal element of the Hurrian proper names is the result of syncope: (1) we as yet know nothing about the word-accent in the Hurrian language; (2) it seems very probable that the desiderative element ib/p (with frequent loss, or assimilation, of its labial component to the following consonant) may or may not have been added to the one and the same proper name. Thus we have three forms of the following names: A-ri-ib-ku-šu-uḫ, N 260.17; A-ri-ku-šu-uḫ, N 242.19; and Ar-ku-šu-uḫ, N 76.26; A-ri-ib-lu-ub-ti, N 75.20; A-ri-il-lu-um-ti, H I 12.1; and A-ru-um-ti, H I 13.23; etc. For the phonetic changes involved in this name see under III.5.A.g, B.b, 6.m, 7.A.b, B.
page 36 note 72 For Dubšimana > Du(š)šimana > Dušmana. For the assimilation of b, see III.5.A.h.
page 36 note 73 In the second and third instances he is the son of Teḫitešup.
page 37 note 74 For the same change in the Hurrian of Ras-Shamra, cf. the borrowed Akkadian noun suluppu ‘dates’: RS zi-lu-um-pa-ni-we, RS II 12 (Syria 1931).
page 37 note 75 For the change a to e see II.1.
page 37 note 76 Is the apparent loss of u in Ḫa-al-še-en-ni, N 410.2 (cf. Ḫa-lu-še-en-ni in the same tablet, line 9) the result of a process of syncope, or is it parallel to what we discussed in note 71 above? (Cf. Te-ḫu-ub-še-en-ni, N 38.39, and passim.)
page 37 note 77 For discussion of this subject, cf. Speiser, ST I, JAOS 52.362–6.
page 38 note 78 However, cf. Kramer, AASOR 11.83 note 122.
page 40 note 1 In both instances he has a son Akuya.
page 40 note 2 In both instances he is the slave of Šilwitešup.
page 40 note 3 Bork, Die Mitannisprache, MVAG 14.14–9.
page 40 note 4 Studien zur Hethitischen Sprachwissenschaft, Leipziger Semitistische Studien 7.25.
page 40 note 5 Cf. Sturtevant, CGHL 67.
page 40 note 6 For examples see under III.5.D-G.
page 40 note 7 Nothing can be deduced, in this connection, from the Hurrian words and proper names, because at present it is impossible to determine (except in cases where b and m interchange) whether a given labial was originally voiced or voiceless.
page 41 note 8 For the latest discussion of this subject, cf. Friedrich, Zum Subaräischen und Urartäischen, Analecta Orientalia 12.128–31.
page 41 note 9 The fact that in Hittite voiceless stops were written double, and originally voiced stops single (Sturtevant, CGHL 73–86), does not invalidate this suggestion, because it is altogether possible that single and double consonants were meant to express voiced and voiceless stops respectively. Cf. the Hebrew Masoretic system where the presence or absence of a dot in a letter shows whether the consonant is a stop or a spirant.
page 41 note 10 With the exception of QA which represents any velar stop in these texts. For ṢI, cf. under I.2.
page 41 note 11 Cf. SA n. 204.
page 42 note 12 A proper name.
page 42 note 13 In [tu]p-pí 'a-ḫu-ti. Cf., however, aḫp-ḫu-ti in line 4.
page 42 note 14 For ḫ-signs as used to express', in Amarna, cf. BA (1910).42.
page 42 note 15 The loss of the e/i vowel in il-te-(ma), G 37.6; H I 43.23; 48.33; N 348.36, must be due to analogy with the tertiae-infirmae verbs like išmē, etc.
page 43 note 16 The fact that a change in vowel occurs after the vowel accompanying the p, indicates that a glottal stop was pronounced between the two vowels.
page 43 note 17 Of all these forms which have a vowel sign, or a sign beginning with a different vowel, after the vowel accompanying the p, only the following exceptions may be quoted: iš-pè-lu-ú, N 230.6; uš-te-pí-il, N 104.4; 114.7; nu-uš-te-pí-il, N 137.6; 480.6. The fact that after the pi-syllable a new syllable begins with a vowel, may indicate a glottal stop.
page 43 note 18 Exceptions are only šu-pè-lu-ti, N 469.6; šu-pè-el-ti, N 230.1, and passim. Cf. end of preceding note.
page 43 note 19 For interchange of k and ḫ in the Highland Zone, cf. Speiser, Mesop. Orig. 142–3. For the same interchange in Hittite, cf. Sturtevant, CGHL 72. The writing of GIŠḪUL for GIŠKUL in the Sumero-Hurrian vocabulary from Ras-Shamra (cf. Syria 12. 237, 247. 23–5) is probably due to the same aspirated pronunciation of k. For the same interchange in the Akkadian texts of Boghazköi, cf. Labat, AB 32–3 and footnote 33.
page 43 note 20 In both instances his father is Šekaru.
page 43 note 21 In all instances his father is Itḫišta.
page 43 note 22 In both instances his father is Urkutu.
page 44 note 23 In all instances he is the ḫazannu of Nuzi.
page 44 note 24 For the sign A' as representing AḪ, see III.2.
page 44 note 25 In both instances he has a son Kudugga.
page 44 note 26 In both instances he has a son Taya.
page 44 note 27 No doubt pronounced as šuw, before the following we.
page 44 note 28 Is Taḫ-mi-ia the same as Tar-mi-ia? Was this a voiced velar spirant? For this sound in Hurrian, cf. Ginsberg and Maisler, JPOS 14.244–5.
page 44 note 29 This is no doubt a present form. Kramer's translation (‘he was willing’, The Verb in the Kirkuk Tablets AASOR 11.87) and classification of this form as preterite singular, lacks foundation.
page 44 note 30 The break in .....]-un-ki-ma, N 363.31, is probably to be restored as [zu]-un-ki-ma for zukkima with a dissimilation of kk to nk.
page 45 note 31 Cf. above, under III.1.
page 45 note 32 See also the discussion under III.8, and Kramer, AASOR 11.73.
page 45 note 33 Plural suffix for the singular. Cf. Speiser, FL AASOR 10.35, note 30.
page 45 note 34 ma'du > mādu > maddu > mandu.
page 46 note 35 The last three forms (with the d) must be learned forms, as may be seen from the fact that that are all written by the scribe Balṭukāšid himself.
page 46 note 36 For the change of š to s, see above, under III.8.
page 46 note 37 For discussion of this subject, cf. Speiser, ST JAOS 53.26, note 39, and Mesop. Orig. 28, 100, 123.
page 46 note 38 Cf. above, under III.1.
page 46 note 39 As in the case of the velars, for which see III.3.
page 46 note 40 For the same phonetic process in the Hurrian of Ras-Shamra, cf. Thureau-Dangin, VR 251 ff.
page 46 note 41 For reference, cf. the lists which follow in text.
page 47 note 42 In the first and last instances he is the son of Akabtae.
page 47 note 43 In the first reference of each instance he is the father of Ḫutiya. In the others it is a name of a district.
page 47 note 44 Lubdi is a name of a city. That it is also a name of a deity may be seen from this name, which means ‘May Lubdi give’. For the phonetic changes involved in this name, see I, end of note 71.
page 47 note 45 In all instances his father is Tubkiya.
page 47 note 46 For discussion of this verb, cf. Koschaker, NKRA 132.
page 47 note 47 For interchange of m and w, cf. III.5.L.
page 48 note 48 For Dubšimana > Du(š)šimana > Dušmana. For the loss of i, see II.3.
page 48 note 49 In the second and third instances he is the son of Teḫitešup. For the partial assimilation of b, see III.5.B.d.
page 48 note 50 For the change of l to r, see below, under III.7.A. This is no doubt *Arlubdi (without the desiderative element ib).
page 48 note 51 For the change of l to n, see III.7.D.
page 49 note 52 Dissimilation of pp in the Sumero-Hurrian vocabulary from Ras-Shamra occurs in Zi-lu-um-pa-ni-we < Akk. suluppu, RS II 12.
page 49 note 53 In the last two instances he has a son Urḫiya.
page 49 note 54 Written BU.
page 49 note 55 In all instances he has a son Tayauki.
page 49 note 56 In all instances above his father is Ennamilki.
page 49 note 57 For the same change in other Akkadian dialects, cf. HS 51.
page 49 note 58 In all instances he is the father of Tauka.
page 50 note 59 For this name, cf. Speiser, FL 72, note a. For the transposition of r and m, see III.9.c.
page 50 note 60 In all instances he is the father of Ninuari.
page 51 note 61 In the first and fourth citations he has a son Muššenni; in the second and third he is the son of Aršawušqa.
page 51 note 62 For the reading of SAL as šel, see I.2.
page 51 note 63 Cf. Speiser, JAOS 49.272, note 9.
page 51 note 64 The phonetic process was ime > iwe > ie (= e?).
page 51 note 65 The phonetic process was probably ubi > uwi > ui, or ubi — uwi > ui.
page 51 note 66 In both instances he is the son of Giliya.
page 51 note 67 In all cases he is the son of Ḫulukka.
page 51 note 68 The phonetic process was umu > uwu > ū or umu — uwu > ū. In all instances he is the father of Gelšu.
page 51 note 69 Only the phonetic changes of m in connection with other labials, are discussed here. For other changes of m, see under III.6.
page 52 note 70 ti-ú = tiw < tiwe?.
page 52 note 71 For changes of m in connection with labials, see under III.5.
page 52 note 72 Cf. Ungnad, Babylonisch-Assyrische Grammatik (1925) §27 III β, and note 1.
page 52 note 73 Probably through inqutmi. That this word is derived from maqātu is proved by in-ta-ku-ut, in the same tablet, line 19.
page 52 note 74 Through andadmi?
page 53 note 75 Temtu/e is a name of a city (cf. alTe-im-te-na-aš, Te-in-te-we, etc.) as well as of a deity, as may be seen from this name which means ‘Temtu gives’.
page 53 note 76 Such writings of the name Ta-an-ki-ia (H II 109.32) as SIG5.GA-ia (in the same tablet line 45) show that the adjective damqu was pronounced danqu, with partial assimilation of mq to nq.
page 53 note 77 For the reading of TAŠ as teš, see above, under I.2.
page 53 note 78 UTU.
page 53 note 79 From ib.
page 53 note 80 From Ar-nu-um-dì < Arlubdi. For the change of l to n, see under III.7.B. The original of this name is undoubtedly A-ri-ib-lu-ub-dì (cf. N 75.20), as may be seen from the fact that A-ri-il-lu-um-dì is written in the same tablet as Ar-ru-um-dì (H I 13.17, 23), while another Ar-ru-um-dì is the same person as Ar-nu-um-dì. Thus we have the following phonetic changes: Ariblubdi > Arillubdi > Arillumdi > Arnumdi > Arrumdi.
page 53 note 81 The form of the name is Te-im-te + the genitive-locative element -we. For the particles na and aš in place names, cf. Speiser, Mesop. Orig. 140 and 146 respectively.
page 54 note 82 For Šamaš-abu, with syncope of the a-vowel.
page 54 note 83 For reading of SAL as šel, see under I.2. For the change of a to e, see II.1.
page 54 note 84 For a full discussion of this name, and the interchange of n and l, cf. Speiser, Mesop. Orig. 95, 111, 123 note 13, 135, 144.
page 54 note 85 In both instances he is the father of Eḫlipapu.
page 54 note 86 For in(a) gamar.
page 54 note 87 With syncope of medial a.
page 54 note 88 Forms of ikkis, ikkir, etc., are numerous. Therefore no listing will be attempted here.
page 54 note 89 I.e., Sin-qurādu. Cf. Šamaš-qa-ra-ad-da, H I 47.44.
page 54 note 90 For the syncope of a, see II.1.
page 55 note 91 IM. For the use of ideograms in their phonetic value, see I.3.
page 55 note 92 GIŠ.SAG.KUL.
page 55 note 93 In all instances (except N 436.5,16) he is the son of Takkia.
page 55 note 94 For ina pana, with syncope of a.
page 55 note 95 Which, however, may be a back-formation: indin > iddin > indin.
page 55 note 96 Forms like ittadin, etc., are numerous, and no listing will be attempted here.
page 55 note 97 For ana ṭiḫi, with syncope of a.
page 55 note 98 For ina rāmānia with syncope of a.
page 55 note 99 For this name cf. also II. 1, and note 11.
page 56 note 100 In both instances his father is Šellapai.
page 56 note 101 Undoubtedly to be read thus. Gordon's reading (Analecta Orientalia 12.169) of H I 104.16 as a wēlru-li-it-ti is wrong. The correct reading is awēlšangu li-id-di.
page 56 note 102 For this name, cf. Götze, Muršilis, MVAG 38.226 (1933).
page 56 note 103 For the possibility that the final n of ḫa-di-in, N 465.10, may reflect the changed pitch in a conditional sentence, cf. Speiser, JAOS 55.441, note 33 (1935).
page 56 note 104 For discussion of this name, cf. Albright, BASOR 59.9, and Speiser, JAOS 55.440.
page 56 note 105 In both instances he has a son Akabtuggi.
page 56 note 106 In all instances his father is Durikintar.
page 57 note 107 In all instances he has a son Taena.
page 57 note 108 In both instances he has a son Šakaraya.
page 57 note 109 In all instances he has a son Šatekintar.
page 57 note 110 In both instances he has a son Giliya.
page 57 note 111 For discussion of this subject, cf. Speiser, Mesop. Orig. 92, 95, 123, 139, 145.
page 57 note 112 For examples, see the lists below.
page 57 note 113 For the absence of initial r in Hittite (which is probably due to the underlying non-Indo-European population), cf. Sturtevant, CGHL 138.
page 57 note 114 Bork's reading of ašti as ruti (Die Mitannisprache, MVAG 1909) has no basis whatsoever. The word rapšu in the Ras-Shamra vocabulary (RS IV 14) is, of course, borrowed from Akkadian.
page 58 note 115 In tablets H I 12,13, and H II 98, he is the son of Ḫašinna.
page 58 note 116 For reference, cf. above, under b.
page 58 note 117 The same change of lt to (t)t occurs also in the Akk. of Boghazköi; cf. Labat, AB 38.
page 58 note 118 For the meaning of this verb, cf. Speiser, JAOS 55.440.
page 58 note 119 The following may be due to scribal omissions: il-ti-tu 4, N 368.7, for iltiltu; taḫ-si-ti, G 61.16, for taḫsilti; lu-bu-ti-šu, H I 82.26, for lubultišu < lubuštišu.
page 58 note 120 For numerals, cf. Gordon, The Numeral in the Nuzi Tablets, RA 31.53–60.
page 58 note 121 The statement of Ginsberg and Maisler (JPOS 14.245, note 4) to the effect that ‘interchange of l and n is .... rather common in Semitic’ is by no means correct, as may be seen from the fact that the majority of cases cited by Brockelmann (GVG 222–9), either are taken from the various dialects of (comparatively) modern Arabic, which are heavily influenced by the racial and linguistic substrata of their respective localities, or are non-Semitic words borrowed by the Semites. This interchange is common, however, in Hurrian (cf. note 1, and the reference cited by Ginsberg and Maisler), and it is undoubtedly to this element that the interchange of l and n in Nuzi is to be ascribed.
page 58 note 122 It could also be due to a scribal irregularity; cf. Speiser, JAOS 55.440.
page 58 note 123 5 šanātipl im-ta-nu-ú.
page 59 note 124 For the meaning of this name, cf. SPN.
page 59 note 125 Written erroneously SI.
page 59 note 123 In all instances he has a son Kipaya.
page 59 note 127 In all instances he has a son Intiya.
page 59 note 128 In all instances he has a son Naniya.
page 59 note 129 In all instances his father is Ennamati.
page 60 note 130 In all instances his father is Šarra/umuli.
page 60 note 131 In all instances he has a son Arḫamanna.
page 60 note 132 In all instances his father is Šennatil.
page 60 note 133 In all instances he has a son Piru.
page 60 note 134 Les Vocabulaires de Ras-Shamra, Syria 1931.253.
page 60 note 135 Zum Subaräischen und Urartäischen, Analecta Orientalia 12.129–30.
page 61 note 136 And probably more.
page 61 note 137 For reference see the lists below.
page 61 note 138 st becomes lt only in the word taḫsiltu. This is undoubtedly due first to progressive dissimilation of the second s to š.
page 61 note 139 si is mostly expressed in these tablets by SÍ.
page 61 note 140 The instance of i-sa-aš is unusual both because of the loss of the final vowel and the transposition of š and s.
page 62 note 141 Cf. Semitised Hurrians in Syria and Palestine, JPOS 14.243–4. Cf. also Harris, JAOS 55.95–100.
page 62 note 142 Loc. cit.
page 62 note 143 In both instances he is the son of Ḫamanna.
page 62 note 144 In the following Hurrian words the sibilants interchange with t: dimit ša ša-ak-ru-uš-še, N 401.7: dimit ša ta-ak-ru-usn-še, N 411.6; za-za-ru-uš-še, G 40.7: za-ta-ru-uš-še, G 29.22; 38.8. Could they be normalized to some affricate sound such as ts or tš?
page 62 note 145 For the numerals in the Nuzi tablets, cf. Gordon, RA 31.53–60.
page 62 note 146 But not ištānānu, the š of which may or may not change to l. For the points of the compass in these tablets, cf. Gordon, RA 31.101–8.
page 62 note 147 While št of the I1 stem may or may not change to lt; cf. iš-tu-ur, N 383.9: il-tu-ur, H II 27.28.
page 63 note 148 With the exception of ni-iš-ta-sí-mi, N 369.17.
page 63 note 149 After changing of sto š.
page 63 note 150 In both instances he is the son of Šambiya.
page 63 note 151 Father of Ziqaya.
page 63 note 152 Father of Turari.
page 63 note 153 The elements eḫli, šeḫli, and šaḫlu are very common in the Nuzi proper names. That these elements were originally eḫḷ, šeḫḷ, šaḫḷ, with a vocalic ḷ, is indicated by such writings of the name of the same person as E-ḫé-el-te-šup and Eḫ-li-te-šup, Še-ḫi-el-te-šup and Še-eḫ-li-te-šup; Še-ḫé-el-til-la and Še-eḫ-li-til-la; Ša-ḫu-ul-te-šup and Ša-al-ḫu-ub-šarri.
page 64 note 154 For this name, cf. under III, footnote 21, with reference.
page 64 note 155 It is worth while to note the dialectal differences between East Hurrian and West Hurrian, as reflected in the process of Metathesis. Thus in Nuzi (East Hurrian) we have always the element irwi, while in West Hurrian it is always iwri (cf. Virolleaud, Antiquity 3. 315; Syria 11. 313; and especially Gustavs, Pal. Jahrb. 26.10 note 4; these dialectal distinctions are brought out by Speiser, Mesop. Orig. 145 n.90, and JAOS 55.438); gurbizu in Nuzi H I 106.6 and passim, but gursipu in Amarna.