Jump to ContentJump to Main Navigation
A Linguistic History of Arabic$

Jonathan Owens

Print publication date: 2006

Print ISBN-13: 9780199290826

Published to Oxford Scholarship Online: September 2007

DOI: 10.1093/acprof:oso/9780199290826.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 24 February 2017

(p.276) Appendix 2. Summary of Variables, Mesopotamia, Western Sudanic Region, Uzbekistan, Shukriyya

(p.276) Appendix 2. Summary of Variables, Mesopotamia, Western Sudanic Region, Uzbekistan, Shukriyya

A Linguistic History of Arabic
Oxford University Press

1. These variables are referred to in the text under Ap 2.1 for phonology or Ap 2.2 for morphology plus the number in brackets, which is the number by which the variable is identified in the data bank. /k/for instance is sect. 2.1.2. The numbering sometimes has gaps in the sequence, these being left for related features.

Unless otherwise specified, the suffixal elements in the data do not themselves have another suffix. The addition of further suffixes often induces allomorphy of different types, which is not accounted for here.

2.1. Phonology

  1. (1) qaaf (reflex of SA /q/); 1. q, baqar ‘cattle’: Mardin 2. g, bagar: Kirenawa. (2) /k/1. k: kalib: CB, 2. c [t∫]: calib: MB

In those dialects with a k ∼ c alternation, c basically occurs in the environment of front vowels, or high front vowels but not back ones. The precise conditions vary from area to area, however.
  1. (3, 4) /ħ//ʕ/; 1. laħam, ‘meat’, baaʕat, ‘she sold’, 2. laham, baaʔat: Kirenawa. (5)*ha/ʕa; 1. laħam, gaʕad ‘he stayed’, 2. *leħem ‘meat’, *geʕed: Bagirmi

In the Bagirmi area proto *ħa/ʕa raise to *ħe/ʕe. Subsequent to this raising of /a/to /e/, (3, 4) above occur.
  1. (6) /ɤ/; 1. ɤ: ɤasal ‘wash’: MB, 2. q: qasal: Kirenawa, 3. x: xasal: Abbeche, 4. G (=[ɠ] (pre-glottalized voiced velar stop or injective) Gasal: Awlad Eli (Cameroon/ Chad). (7) /ð/; 1. ð, dabaħ ‘slaughter’: Hiit, 2. z, zahab ‘go’: Daragözü, 3. d, daba(h) ‘slaughter’: Abbeche, 4. d̥, d̥aba(h): Kirenawa, 5. v, vahab ‘go’: Siirt. (8) /θ/; 1. θ, θlaaθa ‘three’: Hiit, 2. t, tal̥aata: Kirenawa, 3. s, sal̥aasa: Aajiri, 4. f, faafe: Siirt

This correspondence applies to basic vocabulary.
  1. (9) /ð̥/; 1. ð̥, ð̥all ‘remain’: Hiit, 2. d̥: d̥ahar ‘back’: Kirenawa, 3. z̥: z̥ahar: Daragözü, 4. v̥: v̥ahar; Siirt. (10) t̥; 1. t̥: t̥aaɤ ‘it flew’: JB, 2. ɗ̥: ɗ̥aar: Kirenawa, 3. t: taar: Abbeche. (11) Realization of ‘jiim’ 1. j, jimal ‘camel’: (MB), 2. y, yimal: (southern Mesopotamia). (12) Word-internal imala 1. No imala, kaatib ‘has written’: Kirenawa, 2. Imala, keetib: CB

(p.277) The imala in the Mesopotamian area has two degrees, with some dialects (e.g. JB) having a further raising to ii in some forms (e.g. nominal but not participial, kliib ‘dogs’).
  1. (13) r; 1. r, baarid ‘cold’: Kirenawa, 2. ɤ, beeɤid: CB. (14) CCVC-V (verb); 1. CCVC-V, bu-ktub-u ‘they write’: Kirenawa, 2. CəCC-V ykitb-uun: Hiit, 3. CCC-V: yiktb-o Daragözü.

Jastrow (1973: 27) considers the stem /a/in the Daragözü form təftaħe ‘you.F. open’ to be an epenthetic vowel. This creates an interpretive problem. In terms of underlying form this would be classified in category ‘3’, but in surface realization as ‘1’. I have classified it as ‘3’.
  1. (15) CC-C (verb) 1. CəC-C, jibət-hum ‘I brought them’: MB, 2. CC-C, jəbtna: ‘you brought us’: Khaweetna, 3. CCə-C, jibt-u-hum: Kirenawa

  2. (17) CaCii; 1. CaCii, kabiir: Kirenawa, 2. CCii, kbiir: Hiit, 3. CiCii, cibiir: MB

Sasse (1971: 238) notes that both (1) and (2) occur in Mardin, nað̥iif ‘clean’, gbiir ‘big’, on a lexically (i.e. irregular) governed basis.
  1. (18) CaCa(C); 1. CaCa(C), katab ‘he wrote’, bagar ‘cattle’: Abbeche, 2. CiCa(C), citab, bugar: MB. (19) CaCaC-V (verb); 1. CaCaC-V, katab-at ‘she wrote’: Kirenawa, 2. CiCC-V, kitb-at: MB, 3. CaCC-V, katb-it: JB. (20) CaCaC-C(V) (verb); 1. CaCaC-C, katab-tu ‘I wrote’: Mardin, 2. CCaC-C, ktab-it: Hiit, 3. CiCaC-C, kitab-t: MB (or ktab-t). (21) CiCaaC; 1. CiCaaC, kilaab ‘dogs’: Kirenawa, 2. CCaaC, klaab: Hiit. (22) CVCC# (noun); 1. CVCC, əxt ‘sister’: JB, 2. CVCVC, uxut: MB. (23) CC-C (noun); 1. CəC-C, calib-hum ‘their dog’: MB, 2. CC-C: kalb-ki: ‘your.F dog’: Mardin, 3. CC-əC: kalb-uhum: ‘their dog’ Kirenawa

  2. (24) CgutCa gahwa (guttural = x, ɤ, q < ɤ, ʕ, ħ, h, ʔ); 1. CgutCa, gahwa: Mardin, 2. CgutaCa gahawa: Kirenawa. (26) CVCV(C) stress; 1. 'CVCVC, 'bagar ‘cattle’: Kirenawa, 2. CV'CVC, ba'gar: Aajiri. (27) Emphasis; 1. Emphatic consonants, t̥aaɤ ‘it flew’: CB, 2. No emphatic consonants, taar: Abbeche

2.2. Morphology

  1. (30) feminine plural; 1. yes, buktub-an ‘they F. write’: Kirenawa, 2. no, ykitb-u (common plural): MB. (31) two verb conjugations, low vs. high vowel (perfect verb); 1. yes, katab-tu /kbəɤ;-tu ‘they F. I wrote/I grew’: CB, 2. no,: ktab-it ‘I wrote’, lbas-it ‘I wore’: MB

The two conjugations are characterized by an opposition between a high stem vowel and a low stem vowel.
  1. (40) First person singular perfect suffix; 1. -tu, katab-tu ‘I wrote’: CB, 2. -t, ktab-it : MB

(p.278) I have not included the variant -eet which occurs in some southern Mesopotamian dialects (e.g. Shat̥t̥ al-Arab), kitb-eet ‘I wrote’.
  1. (41) 3 F.SG. perfect suffix; 1. -at, katab-at ‘she wrote’: Kirenawa, 2. -it, katab-it: Mardin. (42) 2 plural perfect suffix; 1. -tum, ktab-tum ‘you PL wrote’: MB, 2. -tun, katab-tun: Mardin, 3. -tu, katab-tu: Kirenawa, 4. -to, Daragözü

The variant -taw found in southern Mesopotamia is included under value ‘4’ here. In Daragözü -to is minimally contrastive with the 1SG perfect suffix -tu.
  1. (43) 3 masculine plural; (non-suffixed form) 1. -u, katab-u: ‘they wrote’: Khaweetna, 2. -o, katab-o: Kirenawa, 3. -aw, katb-aw: Hiit

In some qiltu dialects, before a pronominal suffix the variant -aw may appear, katab-o ∼ katab-aw-ha
  1. (44) 1SG perfect suffix; -t or (t); 1. always -t (or -tu), katab-tu: Mardin, 2. -t morphologically conditioned, ka'tab ‘I wrote’: Kirenawa

The conditioning factor for (2) is complicated. (see Owens 1993 b: 104).
  1. (45) verb C-aa-object suffix; 1. no, katab-ha ‘he wrote it.F’: Kirenawa, 2. yes, katab-aa-ha, saaf-aa-ha ‘he saw her’: Abbeche. (51) preformative vowel a, i or a/i; 1. a: yaktub ‘he writes’, 2. i, yiktub: Kirenawa

There are no dialects in this sample with only /a/(southern Borno Arabic does have such). If there are two values, a, i, the quality is usually determined by harmony with the stem vowel. Other distributions are possible, however. In Umm Hajar in Chad, for instance, i occurs in open syllables while a occurs in closed, t-i-biiʕ ‘she sells’, t-a-ktub ‘she writes’. Dialects with both i and a have a value of ‘1.5’.
  1. (52) imperfect suffixes (without object suffixes); 1. -uun, -iin, yikitb-uun ‘they write’: MB, 2. -u, -i, yiktub-u: Kirenawa

Siirt has a special stressed form, without -n, yənkəs'r̥uu ‘they were broken’. I have coded this as 1.5, since it lacks -n (coding = ‘2’), but in its special stress is differentiated from all other dialects which do not have -n. Alternatively, one could give it a separate coding of ‘3’. Before suffixes, in some dialects (many Anatolian qiltu) the n is deleted, Mardin yiktub-uu-ha ‘they write it.F’. As noted in sect. 8.6, I do not usually catalogue variants of suffixes with a further attached suffix.
  1. (53) 1 singular imperfect; 1. a-, a-ktub ‘I write’: Kirenawa, 2. n-, n-uktub: Abbeche

The 1SG prefix is a-. In the indicative it will often be prefixed with b-, ba-ktub.
  1. (54) 1 plural imperfect; 1. n-, n-uktub ‘we write’: Kirenawa, 2. n-…-u, n-uktub-u: Abbeche (55) Harmony-determined imperfect suffixes in strong verb 1. Only one form, -iin/uun: təkbaɤ-uun ‘you.PL grow’, MB, 2. Two forms, high or (p.279) low vowel -i, -e/-u, o, determined by stem vowel (see 6.2 (21)): tiktub-u ‘you.PL write’, tigɗ̥aʔ-o, ‘you.PL cut’, Kirenawa. (61) 1 singular object suffix on noun 1. -i (unstressed), 'beet-i ‘my house’: CB, 2. '-ii, bee't-i: Abbeche

Nigerian is distinguished from Chadian Arabic in that in Nigerian Arabic the verbal object suffix -ni is not stressed, whereas it is in Chadian. This difference is not catalogued.
  1. (62) 2 F.SG object suffix; 1. -ik, beet-ik ‘your house’: CB (after C-), 2. -ki, beet-ki: Kirenawa

A number of dialects (e.g. CB) have both variants on a conditioned basis, e.g. -ki after a long vowel, otherwise -ik.
  1. (63) 2 M.PL object suffix 1. -kum, beet-kum ‘your PL. house’: MB, 2. -kun, beeti-kun: Mardin, 3. -ku, beet-ku: Kirenawa. (64) 3 Pl. object suffix; 1. -hum, beet-hum ‘their house’: Hiit, 2. -hun, beet-hun: Mardin

I do not classify according to the vowel of this suffix, which can be distinctively front or back. In Hiit it is in fact [-him]; similarly, for the preceding 2M.PL suffix.
  1. (65) 3 M.SG. object suffix; 1. -u, beet-u ‘his house’: Hiit, 1.2. -u, VV-nu, katab-uu-nu ‘they wrote it’, CB, 1.3. -u, VV-hu, ijiib-uu-hu ‘they bring him’: Mardin, 2. -a, beet-a: Kirenawa

Bagirmi Arabic has -e after a front vowel (beet-e), otherwise -a. This variable is an exception to my practice of not treating post-suffix allomorphy. However, the allo morph u ∼ nu ∼ hu is deWned by any vowel, not only a suffix vowel, e.g. ʕa∫aa-nu ‘his supper’. Variants with the -nu ∼ hu post-vocalic allomorph are given a per cent classiWcation which is closer to ‘1’ than to ‘2’, since the component vowel -u is closer to -u than to -a. After a vowel, dialects classified here as ‘1’ or ‘2’ both have the variant ": (stress shift and length), e.g. ʕa'∫aa ‘his supper’ (both Hiit and Kirenawa, allowing for pharyngeal change).
  1. (66) H-deletion from object pronouns; 1. no h-deletion: beet-hum ‘their house’: Khaweetna, 2. h-deletion: beet-um: Kirenawa

There is a large amount of variability in the deletion of the /h/. For example in Daragözü in Turkey the /h/ has apparently completely disappeared, so that even after a vowel only the h-less form occurs, e.g. kəl-tuu-a ‘I ate it.F’. This variant has not been given a different coding.
  1. (67) Stress shift before V-initial object pronouns; 1. no shift, 'masak-a ‘he grabbed him’: Kirenawa, 2. shift: ma'sak-u: Mardin. (68) Form of 2MSG object pronoun; 1. -ak, beet-ak ‘your house’ (MB), 2. -ik, beet-ik (Siirt). (71) Doubled verbs 3MSG; 1. CaCCa, tamma ‘he finished’: Kirenawa, 2. CaCC, tamm: CB. (72) weak Wnal verb, vowel suffix; 1. -Vt, -o/u, nis-at ‘she forgot: Kirenawa, (p.280) 2. (i)y-Vt, nisy-ət: Mardin. (73) Initial vowel of imperfect of aCaC verbs; 1. prefix + aa, yaaxud ‘he takes’: Kirenawa, 2. prefix + oo, yooxud or yooxuz; Uzbekistan

Dialects used in survey (number represents the code):

Western Sudanic Arabic: 1. Kirenawa; 2. Mada; 3. Aajiri; 4. Wulaad Eeli; 5. Amm Timan; 6. Umm Hajar; 7. Abbeche; 8. Atia I; 9 Atia II.

Mesopotamia (dialect subgroup as in Table 5.4 given in brackets): 10. Christian Baghdadi (Baghdad qultu); 11. Jewish Baghdadi (Baghdad qultu); 12. Muslim Baghdadi (gilit); 13. Mardin (Anatolia); 14. Daragözü (Anatolia); 15. Siirt (Anatolia); 16. Kha weetna (non-Baghdadi qultu); 18. Hiit (non-Baghdadi qultu); 19. southern Mesopotamia (gilit).

30. Uzbekistan; 31. Shukriyya.