Historisk-filosofiske Skrifter udgivet af Det Kongelige Danske Videnskabernes Selskab Bind 4, nr. 3. Hist. Filos. Skr. Dan. Vid. Selsk. 4, no. 3 (1966) THE DIALECT OF AWROMAN (HAWRAMAN-I LUHON) Grammatical sketch, texts, and vocabulary BY d. n. Mackenzie København 1966 Kommissionær: Ejnar Munksgaard Det Kongelige Danske Videnskabernes Selskab udgiver følgende publikationsrækker: The Royal Danish Academy of Sciences and Letters issues the following series ofpublications : Bibliographical Abbreviation Oversigt over Selskabets Virksomhed (8°) Overs. Dan. Vid. Selsk. (Annual in Danish) Historisk-filosoflske Meddelelser (8°) Hist. Filos. Medd. Dan. Vid. Selsk. Historisk-filosofiske Skrifter (4°) Hist. Filos. Skr. Dan. Vid. Selsk. (History, Philology, Philosophy, Archeology, Art History) Matematisk-fysiske Meddelelser (8°) Mat. Fys. Medd. Dan. Vid. Selsk. Matematisk-fysiske Skrifter (4°) Mat. Fys. Skr. Dan. Vid. Selsk. (Mathematics, Physics, Chemistry, Astronomy, Geology) Biologiske Meddelelser (8°) Biol. Medd. Dan. Vid. Selsk. Biologiske Skrifter (4°) Biol. Skr. Dan. Vid. Selsk. (Botany, Zoology, General Biology) Selskabets sekretariat og postadresse: Dantes Plads 5, København V. The address of the secretariate of the Academy is: Det Kongelige Danske 'Videnskabernes Selskab, Dantes Plads 5, Kobenhavn V, Denmark. Selskabets kommissionær: Ejnar Munksgaard’s Forlag, Nørregade 6, København K. The publications are sold by the agent of the Academy: Ejnar Munksgaard, Publishers, 6 Norregade, Kobenhavn K, Denmark. Historisk-filosofiske Skrifter udgivet af Det Kongelige Danske Videnskabernes Selskab Bind 4, nr. 3. Hist. Filos. Skr. Dan. Vid. Selsk.4, no. 3 (1966) THE DIALECT OF AWROMAN (HAWRAMAN-I LUHON) Grammatical sketch, texts, and vocabulary BY d. n. Mackenzie København 1966 Kommissionær: Ejnar Munksgaard PRINTED IN DENMARK BIANCO LUNOS BOGTRYKKERI A/S Introduction Near the turn of this century three Europeans in succession visited the mountainous district of Awroman (Hawrnmân), in the west of the Persian province of Kur­ distan (Ardalân), and made notes on the dialects spoken there. The first to publish his not very considerable notes was J. de Morgan,1 in 1904. By that time both Åge Meyer Benedictsen and Oskar Mann had returned with the fruits of their labours, but neither immediately published his material. Although both these scholars actually visited Awroman, the greater part of their linguistic notes was made beforehand in the town of Sanandaj (Sinn). It is possible that Mann became aware of a certain discrepancy between his notes and the dialect of Awroman proper, due to the shortcomings of his informant. At least, whatever the reason, he did not mention the dialect specifically in the plan for the appropriate part of his great Kurdisch-Persische Forschungen. Nevertheless his notes were published, after his death, by Hadank.2 It will, therefore, be necessary to return to the question of their accuracy. The most important collection of notes and texts was undoubtedly that made by Benedictsen. Unfortunately it was not until 1921, when in his own words ‘les textes ne m’étaient plus bien familiers, et que bien des passages m’étaient devenues incompréhensibles’, that his material was published, ‘grâce aux lumières et au zèle inlassable de M. Arthur Christensen’.3 In the light of new material it is now possible to give a clearer picture of the morphological processes of the language. These are of a surprising complexity for a modern West Iranian dialect. They are far more intricate, for example, than those of any Kurdish dialect, though Hawrami forms an island in a Kurdish sea. Indeed, in many ways Hawràmï has the aspect of a somewhat archaic Middle Iranian dialect. Not the least of its claims for attention is the light it throws on the history of neigh­ bouring Kurdish dialects.4 1 In Mission scientifique en Perse, tome V, Etudes linguistiques, Paris, 1904. 2 Mundarten der Gûrân, besonders das Kandûlüî, Auramânî und Bâdschalânî, bearbeilet von Karl Hadank, Berlin, 1930, pp. 367-395. 3 Les dialectes d’Awromdn et de Pâwa, textes recueillis par Âge Meyer Benedictsen, revus et publiés avec des notes et une esquisse de grammaire par Arthur Christensen, Copenhagen, 1921 (Hist. Filol. Mecld. Dan. Vid. Selsk., VI, 2) [cited B-C]; also ‘Some New Awromânï Material prepared from the collections of Åge Meyer Benedictsen’, by A. Christensen, BSOAS, 8, 467-476 [B-C, II]. 4 See my ‘Origins of Kurdish’, TPS, 1961, 68-86. In that article the transcription of Hawràmï ex­ amples was brought into line with that used in my description of Kurdish (see below). In the present work 1* 4 Nr. 3 In 1957 I had the good fortune to meet in London a young man from Awroman, Tahsin, son of Muhammad Amin, originally of the district of Luhdn (see Map). Now when Benedictsen visited ‘Naw-e-Sütà’ (recte Naws'uda) in 1901 he was the guest of the ‘sultàn-e-Lohün’ [sic], who, however, regarded him with ‘une méfiance crois­ sante’ and in the end ‘une inimitié ouverte’ which forced him to retire from the scene sooner than he would have wished. It was to our common amusement, therefore, that Tahsin and I decided this ‘sultan’ must have been his great-grandfather, namely Hama Sadd San (= Sultan), father of Jâfar San, father of Muhammad Amin San.5 My friend proved ready and willing to make amends for his ancestor and the following notes are based entirely on information kindly supplied by him in the course of many months of periodic questioning. Thanks are also due to the School of Oriental and African Studies, of the University of London, which made his co-operation possible. Hawrâmï is a Gôrànï dialect and, as will appear, probably the most archaic and best preserved of the group. The dialect described here is that of Hawràmàn-ï Luhôn, and more specifically of its chief village Nawsüda. I his was the home of Jatar San and his family until 1932, when he went into exile in Iraq. There he died, in Halabja, in about 1943. Tahsin was born in a Hawrâmï village in Iraq. On the family’s return to Persia his father went to live in Pâwa, which he therefore considered his home. Nevertheless he was well aware of the differences between his mother tongue and the dialect of Pâwa. To avoid any chance of confusion, however, I preferred to restrict my enquiries to his own dialect. It must be said further that his second language is the Kurdish of Suleimaniye. While it is always possible that this may have in­ fluenced his Hawrâmï, his remarkable consistency under cross-examination leads me to think not. I have felt obliged to reverse the process and transcribe the occasional quotations from Kurdish according to the scheme demanded by the description of Hawrâmï. 5 See the family tree given by C. J. Edmonds, Kurds, Turks and Arabs, London, 1957, p. 155, to which work the reader is also referred for a full description of Awroman (see Index, s. v. Hewrâmân). Nr. 3 5 IRAQ IRAN Key to the map. (Villages identified from the 1/4 inch: 1 mile map by Tahsin.) a Daray mar. d Berwdz. g Balxa. b Gulp. e Häna garmta. h Bidra. c ? Bäxa kon. f Xarpäni. Luhon. Nawsuda. la Zdiwar. Id Nods a. lg Safakdn. lb Sosakän. le Bemrwe. lh Dar a iv Hajijid le Sosme. If Narivi. li Wurd. Hajij- Nr. 3 6 Jwanro. ja Hirwe.* jd Komadara.* jg Di§a. jb Daribar* je Nijdr* jh G(i)rdla. jc Ddridn, jf Sawalxe.* ji Nuridw.* Pawa. Pdwa. pa Xdnagd. pd Nosma. Pf Sarkrdn. pb Durisdn.* Pe Darmur. Pg Bmdara. pc Darabayan.* Taxt. Sari Hawrdmdn. ta Kamdld. tc Dizli. te Tifli. tb Darawki. td Bdrdmdwa. tf S(i)len. Razaw. Razdw. ra Dagaga. rd Dalamarz. rf Asparez. rb Mdzibm. re Zom. rg Kalji. rc Biridar. * Kurdish-speaking population. Phonology § 1. The phonemic system of Hawrami is remarkably similar to those of the sur­ rounding Kurdish dialects of Suleimaniye and Sina. There are 26 consonants (in­ cluding two semi-vowels), which may be represented by the same symbols as have already been adopted for the Kurdish dialects.1 The vowel phonemes, however, 10 in number, present a problem of representation. For various reasons, particularly the close relationship between ‘long f, a and the corresponding semi-vowels, the diacritics traditionally used are inadequate for our purposes. The symbols used be­ low, therefore, may appear strange at first sight. Dental, Palato- Pharyn- Labial Velar Uvular Glottal Alveolar Alveolar gal P / k g Stop and Affricate.................... 6 b d / 0 1 s2 è h h Fricative....................................... X z t ( Nasal.............................................. m n Lateral......................................... l l Vibrant ......................................... r r Semi-vowel................................... w y Consonants. § 2. Of the consonants it is the dental and alveolar series which requires most de­ finition. (a) d, normally a voiced dental stop, is realized in post- and inter-vocalic position, and between r and a following vowel, as a continuant [a], as in the Kurd, of Sul. and Sina.3 This sound, represented in the notes of Andreas4 and Benedictsen by d, 1 See my Kurdish Dialed Studies, London, 1961, I, § 1. 2 As in Sul. Kurdish (v. KDS, I, § 10b), the Arabic emphatic consonant ^ s is heard in the two words Sas ‘60’ and sad ‘100’ only. 3 KDS, I, §§ 8 b, 37. 4 Iranische Dialektaufzeichnungen, aus dem Nachlass von F. C. Andreas, zusammen mil Kaj Barr . . . bearbeitet und herausgegeben von Arthur Christensen, Berlin, 1939, p. 215. Nr. 3 8 without further definition, is a half-close central continuant caused by the tip of the tongue approaching the upper teeth without making contact. Thus it is not a fricative and will be written here d, i. e. as a continuant allophone of d. An exception is the d of the 2nd person plural verbal ending -de. Although Benedictsen gives -da as a variant of his -da, Tahsin never realized this as other than a dental stop in any context. hard ‘he did’, karda ‘done’, karde ‘you do’ In one context d is entirely absorbed, viz. by a following i, e. g. -idi- > -i- in blidia [1 bi:e] ‘look!’ (b) The same continuant [a] is heard as the normal realization in final position of the 2nd singular personal pronoun suffix, in contrast with the non-final form -(/)/. The continuant is accordingly written f. hdy-t-a ‘art thou awake?’ dizman-it ‘thy enemy’, ada-[ ‘thy mother’ (c) The phoneme n comprises, beside the normal voiced dental nasal, a velar allophone [q]. This is never heard without a following g (k), however fleeting. angusa [arj'gusa, ag^usa] ‘finger’ dang [daqg, daqg/k] ‘fame’ In a few numerals it is realized as a nasalization of the preceding vowel a, thus ydnza [jar'za] ‘eleven’. (d) / and l, r and f, correspond exactly to the same phonemes of Sul. Kurd.5 / is a front and / a back, velarized, dental lateral, r is an alveolar flap, while r is rolled. Neither l nor r ever occur in initial position in a word. mila ‘mountain pass’ : mila ‘mouse’ maru ‘I bring’ : maru ‘I break’ (e) h and c, occurring mainly in loanwords from Arabic, are quite distinct from li and the glottal stop [?] respectively. horia ‘houri’, hdzir ‘ready’ t-vmr ‘age’, (¿)amra ‘order’ Vowels. § 3. Seven of the vowel phonemes are ‘long’, i e e a o o u, and three ‘short’, i a u. (a) Of the short vowels i, a slightly open, central [±], is particularly unstable. In unstressed position between consonants it is frequently not realized. 6 KDS, I, §§ 13-16.