Acta Orientalia Academiae Scientiarum Hung. Volume 67 (4), 453 – 465 (2014) DOI: 10.1556/AOrient.67.2014.4.4 THE BALOCHI DIALECT OF THE KOROSH HABIB BORJIAN* Center for Iranian Studies, Columbia University 450 Riverside Drive, New York City, USA e-mail: hb146@columbia.edu Koroshi is the Balochi dialect spoken by the Korosh (Koroš), a group associated with the Qashqa’i tribes of Fārs in southwestern Iran. Entirely isolated from the main body of the Baloch habitat, Koroshi distinguishes itself in grammar and lexicon among Balochi varieties. The phonology of Koroshi demonstrates a solid Balochi pedigree but not without major mutations. Likewise, the nominal case-number system of Koroshi shows significant deviation from most other Balochi dialects. In verb morphosyntax a salient peculiarity is the coexistence of two parallel systems of the imperfective, which appear to be stabilising in an evolutionary process of the Koroshi aspect system. Borrowing from the neighbouring languages is salient in the lexical domain, where Persian, the Fārs dialects, and Qashqa’i Turkish each play a part as the source language. Given all these peculiarities the degree of mutual intelligibility between Koroshi and other Balochi dialects is yet to be established. Key words: Iranian languages, Balochi, Fārs dialects, Qashqa’i Turkish, dialectology, comparativehistorical phonology, morphology, syntax, loanwords, endangered languages. In memoriam Hamid Mahamedi (1935–2004) The Korosh used to work as camel keepers for the Qashqa’i tribal confederation along the tribe’s seasonal migrations in the province of Fārs. The Korosh are now predominantly settled in the outskirts of various villages and towns of Fārs and adjoining provinces, mostly among other Qashqa’i settlers. Pockets of the Korosh have been * The author would like to express his thanks to the anonymous reviewer of this article for his/her comments. 0001-6446 / $ 20.00 © 2014 Akadémiai Kiadó, Budapest 454 HABIB BORJIAN identified around Kāzerun in Bālādeh, Korošābād, Dādin, Jadval Torki, Šur-e ʿAbdolxāni, and, southeasterly along the Zagros valleys, as far as Lārestān and beyond (Salāmi 2004a, p. 23). Their population was initially estimated as forty to fifty families (Mahamedi 1979), then as two-hundred households (Salāmi, op. cit.), but the emerging data suggest even more. Even so, as the Korosh no longer live in their traditional tribal context and because of the predominance of Persian education and media, Koroshi should be regarded as an endangered language. The Korosh are not documented in the ethnographies on the Qashqa’i known to this author. There is, however, a 19th-century mention of a homonymous tribe (Kuruš, Koriš, etc.) among the Baloch on the Indus border (Bellew 1891, p. 143). No evidence exists as to when the Korosh might have joined the Qashqa’i tribe; those interviewed by Mahamedi (1979) in the 1970s remembered at least three generations. A long presence among the Turkish speaking Qashqa’i is not suggested by internal linguistic evidence (see section 23, below); on the other hand, the fact that Koroshi has diverged from the Balochi of Baluchistan in significant traits and that the Korosh are Shi’ite attest to their relatively long isolation from the main body of the Baloch. Other self-designation of the Korosh is Dārḡa, a term interpreted by a local tradition as echoing their bygone profession as dāruḡas, or police officers under the Safavid rule (1501–1722) (Salāmi 2004b).1 The Koroshi dialect was discovered by Hamid Mahamedi (1979) (henceforth HM) during his field visits in Fārs in the 1970s. His Korosh informants used their native vernacular at home while their “cultural” language was Qashqa’i Turkish, in which they told stories and sang songs. More recently ʿAbd-al-Nabi Salāmi (2004a) (henceforth AS) has published a dictionary of hundreds of words, appended by some verbal paradigms and dozens of elicited sentences. Salāmi maintains that the Koroshi of Kāzerun, especially that of Bālādeh, has been far less affected by Persian syntax than the Koroshi spoken in Čāršurak (in Lārestān), while the Korosh residing further south in Sedāḡ have shifted to the local Lārestāni dialect.2 Beside these two sources, Carina Jahani and Maryam Nourzaei (2011) (henceforth J-N) have published a Koroshi text, interlinearised and translated, but without specifying its provenance. Examples in the present study are from Salāmi unless cited with other author’s initials. The following diachrony and sketch grammar rest on the above-mentioned sources. The data show significant variation, to the extend that one tempts to consider different dialects of Koroshi, but a closer examination reveals that much of the variation is justifiably superficial, and variation occurs even within the speech of the same speaker. This peculiarity could have been intensified by areal contact with the neighbouring tongues as well as an active process of restructuring within Koroshi itself (see e.g. section 16, below). The description here is made with keeping an eye on other Balochi dialects,3 especially those spoken in Iran (Ir. Bal.; summarised by Spooner 1967 and described by Jahani – Korn 2009). 1 The view expressed in some online blogs that the word Koroš has derived from Old Persian Kuruš (Cyrus the Great) is no more than folk etymology. 2 Personal communication with ʿAbd-al-Nabi Salāmi in March 2013. 3 On the Balochi dialects, see Elfenbein (1989); Jahani – Korn (2009). Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH 455 Phonology 1. The Koroshi sound system differs from most other Balochi dialects in that it lacks retroflexes. Koroshi seems assimilating to its current environment by adopting from the neighbouring Fārs dialects the fricative [δ], a postvocalic allophone of /d/. /w/ in Mahamedi’s documentation corresponds to /v/ in other sources; both may represent bilabial spread [β]. [ž] is heard infrequently; it seems an allophone of /j/ in jan/žan “woman”. /g/ may soften to /y/ intervocalically, as in magašagant/mayašayant “they say”. A detailed treatment of the synchronic phonology of Koroshi awaits the publication of more texts. 2. The dialect has lost historical vowel-length distinction, partly by diphthongising long vowels, as in *čēr/čīr > či(e)r “under”. This must be a late development, for it also occurs in borrowed words, e.g., bied “willow” (cf. Bal. ged). 3. Within comparative-historical phonology4 several layers can be identified in the development of consonants. Isoglosses most specific to Balochi, i.e., preservation of Old Iranian (OIr.) postvocalic stops and affricates and plosivisation of the fricatives (Korn 2005, pp. 77– 82), are in general applicable to Koroshi (Table 1), but with these innovations: Bal. p > f (from both OIr. *-p- and *f), Bal. k > x (partially), and Bal. t > d (from OIr. *θ).5 The Old Iranian cluster *xt yields (with metathesis) Southern Balochi tk,6 which is also a characteristic of Koroshi. The cluster is further developed to kk in Koroshi past stems: pakk- “cook” (Bal. patk- < *paxta-), dokk- “sew”, rekk- “pour”, and, with further affrication, sočč- “burn”. This rule may also explain ākk-/ātk“come” (if from *āxta < *āgata-, proposed by Elfenbein 1989). 4. OIr. *w- > g- (Bal. g(w)-) (Korn 2005, pp. 98–101) affects the inherited vocabulary of Koroshi in bulk; examples are gād “wind”, gānk “voice”, gāfanda “weaver”, gaš- “say”, gen- “see”. The exceptions appear to be loanwords; for instance, bis “twenty” (for the expected *gis) is also found in other Balochi dialects (Korn 2006), an indication that the word was borrowed from Persian into Balochi when the Korosh were not yet parted from their ancestral homeland. Other examples of loanwords from Persian are bey “moat” and bied “willow” (< OIr. *waiti-), both failing not only 4 For a throrough treatment of Baluchi historical phonology, see Korn (2005). The first two are probably independent of Eastern Balochi which has changed all postvocalic stops to fricatives (Elfenbein 1989). 6 Cf. Western Balochi ht, Eastern Balochi xt (Korn 2005, pp. 111 – 112, 184 – 185). 5 Acta Orient. Hung. 67, 2014 456 HABIB BORJIAN Table 1. Diachrony of Stops, Affricates, and Fricatives OIr. Bal. Kor. Examples Counterexamples *-p- p f āf “water”, šaf “night”, taf “fever”, tāfak hovog “co-wife” “pan”, tafar “axe”; and probably daf “mouth” (Bal. hopak/g) (Bal. dap) and lāf “stomach” (Ir. Bal. lāp), whose etymologies are unknown *-t- t t pot “weft”, dantān “tooth”, morta “dead” *-k- k k x giryak “weep”, šodik “hungry”; Bal. -k > x in magas “fly”, etc. heyx “egg” (Bal. hāyk), gox “cow” (Ir. Bal. are probably guk), ganux “simpleton” (Marw Bal. ganok) loanwords *-b- ? ovort- “bring” (= Ir. Bal. aort-) *-d- d d pād “foot”, mud “hair”, ōdān “there”, kad “when”, fād “salt” (Bal. wād < *hwāda-), šod“wash” *-g- g g probably rogen “ghee” (Bal. rogin) *-č- č č roč “day”, či(e)r “under”, sučin “needle”, gečin “sieve”, pač- “cook”, doč- “sew”, reč“pour”, soč- “burn” *-j- j j derāj “long” (Bal. drāj) *f p f geft- “seize” (Bal. gipt-) *θ t d gud “excrement” (Bal. gūt), modag “migration” (Bal. mētag “settlement”) *x k k kar “donkey”, kan-, kannak “smile”, karān- ː karānt- “scratch” va(ː)d- “self”, berād (Bal. brāt) “brother”, kad “when” nāxon “nail” (Bal. nākun), mih “nail” (Bal. mek) the *w- > g test but also that of the retention of the Old Iranian *-t-, which is another Balochi-specific diachronic feature.7 See more examples in section 22. 7 Should gō (in gō-ham “together”) be related to Pers. bā “with” (< Middle Pers. abāg) – a not very likely proposition – it would attest to a relatively late chronology of *w- > g sound change. See also Korn (2005, p. 181). Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH 457 5. OIr. *hw- normally yields w-/v- (another Balochi trait), as in va(ː)d- “self”, var- ː vart-8 “eat”, vān- ː vānt- “read”. There seems a further change of w- to g- in gāhār “sister” (as in some other Balochi dialects)9 and to f- in faːš (Pers. xwaš) “fine, happy” and probably in fād “salt” (Bal. wād). “Sleep” yields mixed results: fō (noun), vasp- ː vaft- (intransitive verb), fāfen- ː fāfent- (causative verb).10 6. Common West Iranian isoglosses. Proto-Iranian sibilants from proto-Indo-European *k’, ǵ(h) persist in kassān “small”, mazzan “big” (< *kas, *mas), zi “yesterday”, zān- ː zānt- “know”, while some keywords are Perside: āhu “gazelle”, darvā “sea”, dumāδ “bridegroom”. Another Northwest trait is preservation of OIr. *j in jan “woman”, jan- ː jad- “hit”. Other key changes are Southwest Iranian, as in general Balochi: OIr. *θr > s in sa “three”, āhos “pregnant” (< *āfos < *ā-puθra-), ās “fire”; *dw- > d in dar “door”, diga “other”; *y- > j in jō “barley”, joδā “separate”, ji “yoke”. Noun Phrase 7. The Balochi case–number system has been decoupled in Koroshi to a threesome case system independent of number. Formal plurals are still infrequent in Koroshi, but the dialect seems to have been developing, as an areal feature, a regularised plural marked with the suffix -obār for animate and inanimate nouns alike, as in sibobār “apples”, mišobār “ewes”, ādamobār “people”. There are also mardingal and janingal that appear to be signifying groups of men and women respectively. 8. Cases. The nouns and pronouns are expressed in three cases, direct, oblique, genitive; alternatively, subject, object, possessive are equally relevant designations for our purpose. The last two are marked with -ā and -ī/-ey respectively; so the declension of “house” goes: log, lṓgā, lṓgey. The personal pronouns are declined in Table 2. The genitive precedes the head noun: log-ey dar “house door”.11 It also occurs with postpositions (see section 11, below). The oblique is a used as direct or indirect object. Examples: (direct object) in kōhobār-ā az idān bozoret “pick up these stones from here”; (indirect object, with or without preposition) a madress-ā ākey log-ā “he came home from school”. The genitive and oblique cases co-occur in čir-e deraxt-ey nār-ā nešte “he is sitting under the pomegranate tree”. Note that the oblique marker -ā is suppressed by clitics: miš mā hāḡol‿en “the ewe is in the cote”; J-N vad‿et bege! “hold yourself!” 18 A colon symbol separates the present and past stems. I.e. a dissimilation of *hwahār > *wahār parallel to the change that happened in Greek (Morgenstierne 1927, p. 97; Korn 2005, p. 123). 10 Comparable with the Biābānak group of dialects in central Iran (see Borjian 2013). 11 In the phrase deraxt-ey nār “pomegranate tree”, the possessive morpheme acts in lieu of the Persian eżāfa marker. 19 Acta Orient. Hung. 67, 2014 458 HABIB BORJIAN Table 2. Personal Pronouns Subject Object Possessive man manā mani 2 ta ta(rā) 3 ā āši, āy, ayi āši mā mārā may 2 šomā šomārā šomāy 3 āšān āšānā āšāni Sg. 1 Pl. 1 9. Adjectives. Attributive adjectives precede the head noun and normally receive the suffix -eyn. Examples: sōz-eyn deraxtobār “green trees”, sih-eyn asp “black horse”, mazzanoyn bač “big boy”, gott-eyn kō “large stone”. The adjectival suffix is superseded by clitics: consider gott “large” in Jalāl-i bāḡ xeyli gott‿en1 “Jalal’s garden is1 very large”. 10. Pronouns and Deixis. There are two sets of personal pronouns: independent and enclitic (Tables 2 and 3). The independent pronouns are declined according to the three-case normal system stated above, in section 8. Enclitics are used (1) as possessives (see below); (2) with the reflexive base va(ː)d-, e.g., vaːd-i doros‿iko “himself built [it]”; and (3) as agents in the passive construction of transitive verbs in the past tenses (see section 18, below). Table 3. Verbal Endings, Copula, and Pronominal Suffixes Verbal endings Copula Enclitic Pronouns -ān -an, ān, -am -om 2 -ey, -ay -ey, -ay, -ē -et 3 -i(t) (pres.); -ø, a(t) (past) -an(t), -en, -ø -i, -ey, -ay Pl. 1 -en -en -en 2 -et -at -u/-o 3 -an(t) -ant -eš Sg. 1 Koroshi has three strategies in using possessive pronouns: (1) With freestanding pronouns, e.g., i bozobār maːni/ašāni‿eyn “these goats are (lit. is) mine/theirs”. (2) With enclitic pronouns, e.g., ketāb-ay “his book”. (3) Combining both the Balochi and Persian strategies: Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH 459 mani1 lōg-om2 ba mazzani-e aši3 lōg-āy4 nayn “my1,2 house is not as big as his3,4 house” hāmmo aši1 raftār-ā-h-ey2 nārāzi‿ant (-h- is epenthetic) “everybody is unhappy with his1,2 behavior”. Demonstratives are i “this”, ā “that”, išān “these”, and āšān “those, they”. Demonstrative adverbs are īdān “here”, ōdān/ūdān, ā́ ŋa “there”; hano “such, thus”. Interrogatives are kad “when”, kaye “who”, ko “where”, koyen “where is”, čont “how much”. 11. Adpositions. Contrary to the general trend in Balochi, Koroshi is essentially prepositional, probably under areal influence, and employs both Persian and Balochi prepositions, such as a(z) “from” (for Bal. ča, ša, ač, aš) (Jahani – Korn 2009), gu “with” (cf. Bal. gōn), mā “in, into” (Marw Bal. mā), či(e)r “under”. A peculiar adposition is kaːnek(ā) “beside, near” (cf. Farvi kenāgā), which functions as both preposition and postposition: čekkobār kaːnek-e hōz-ā nešteyadant “the children were sitting by the pool” čerāḡ difār-ey kaːneka‿yn “the light is near the wall” may deh-ay kaːnekā ruxāna maravag “a river passes by our village” Note the genitive case of the governed noun (hōz-ā) after this preposition, compared with the oblique case (section 8) with other prepositions (difār-ey, de-ay). Verb Phrase 12. Stems and Verbal Nouns. Verbs are constructed on the binary stems traditionally called past and present, which in the case of Koroshi may best be characterised as imperfective and perfective stems. (The pair is shown here separated by the colon symbol.) As in other Iranian languages, many of the Koroshi past and present stems, such as bann- ː bass- “bind”, afford no obvious synchronic relationship. Regular past stems, on the other hand, are formed by suffixing to the present stem -ed (or -t before nasals and approximants), a productive formant in all causative past stems. The causative present stem is formed by adding -(i)en- to the present stem of an intransitive verb. Example for the verb “glue”: intransitive transitive časpčaspien- ː ː časpedčaspent- The infinitive and the past participle are formed by suffixing -ag to present and past stems respectively. These two forms in -ag- are called “long stems” by Mahamedi, since they form the base for imperfective mood and perfect periphrastic, both formed with copula as auxiliary (Tables 3 and 4). -ag- tends to soften to ay and shrink to a/ā in the final position and to g before nasals. (The final a and ā remain indistinctive in many inflections.) In some transitive verbs the past stem shortens when not suffixed. Example for “do”: present stem kan-, infinitive kanag; past stem kut-, past participle kuta(g), with shorter forms ku, ko, ke. The infinitive and the past Acta Orient. Hung. 67, 2014 460 HABIB BORJIAN Table 4. Verb Structure Imperative = be- + stem pres. + ending (sg. -ø, pl. -et) Pres. subjunctive = be- + stem pres. + ending Pres. imperfective = a‿ + stem pres. + ending Past imperfective Preterit Pres. perfect Pluperfect Perfect subj. intr. ma- + infinitive = ma- + infinitive + copula pres. + copula past trans. = ma- + infinitive + -ad- + encl. pron. intr. = stem past + ending trans. = stem past + encl. pron. (fronted optionally) intr. = past part. + copula pres. trans. = past part. + encl. pron. (fronted optionally) intr. = past part. + copula past trans. = past part. + -ad- + encl. pron. (fronted optionally) intr. past part. + copula pres. + bi = participle are also used as verbal nouns: pakkā “cooked”, dokkā “sewn”, J-N man bowṓ-ey sā́ z-ey janág-ā yād-om mohấ “I remember instrument playing of Father”. 13. Mood and Aspect Markers. The subjunctive and imperative are marked with the modal prefix b(V)-, unless one of the lexical preverbs dar-, ber-, er- is present. While the perfective aspect remains unmarked, the imperfective12 is expressed in two parallel structures, with the prefix ma- or proclitic a‿ (Table 4). The proclitic a shows remarkable stability in Koroshi, compared with its uncertain semantic stance in most other Balochi dialects (Elfenbein 1989). Unstressed, it stands between the verb and the preceding word. It stays on the verb only at clauseinitial position or after vowel-final words; else it is absorbed by the word that precedes the verb. Thus, for instance, the object (which ends in the vowel -ā)13 never absorbs the imperfective marker a. Examples of the position of the proclitic a (data from J-N): On the preverb or the nominal part of compound verbs: dar‿a kafī (for Pers. dar-mi-oftad) “it gets engaged in”, ber‿a dā “it throws down”, er‿a kanant “they set [it] onto”, zorr‿a gī “it returns”. On the subject: howr‿a jant “the rain strikes”. On the verb after a vowel: ā́ ŋa a‿čarant “they graze there”; a modag-ā 12 13 For the imperfective, see section 16. See section 8, above. Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH 461 ā‿mānīt “it is left behind from migration”. At clause-initial position: a‿šīt… “he says…”. 14. Person Marking. Depending on the tense, mood, and transitivity, person is designated by three means: verbal endings proper, the copulative verb, and enclitic pronouns (Table 3). The endings and copula are in general agreement, though each shows considerable variation. Singular and plural are used interchangeably in the second and third persons (Elfenbein 1989 observed similar in other Balochi dialects). The third singular shows considerable variation; some stems ending in /r/ or /n/ take a short -t (for example, bu-(w)ar-t “that he eat”, a‿na-twān-t “they cannot”), while verbs of high frequency may omit the ending when the distinction it affords is not explicitly required: berra “that he go”, māhā ≈ mā-hāg–en “he comes, is coming”, vafta ≈ vaftagen “he has slept”. Enclitic pronouns are used in the transitive past tenses (see section 18, below). 15. Conjugations. Verb forms are summarised in Tables 4 and 5. For transitive past, see section 18. 16. The Imperfective. Koroshi is in a historical process of stabilising its verbal aspect distinction. The imperfective is expressed by two constructions (Table 4): (1) with the stem and proclitic a,14 inherited from Balochi, and (2) with the infinitive and prefix ma-, which is an amalgam of Balochi and Persian structures (for Balochi forms, see Jahani–Korn 2009, section 4.5.3). (There is no progressive with the verb “have”, as is the case in Persian.) A rather loose semantic association for each form can be deduced: the first form usually expresses the present indicative, future and habitual action while the second is used in progressives. Examples: Present/future: aga alʾān berren hatman a āši a‿ras-en “if we go now we will definitely catch him”. Habitual: bačč-et kad a madressā a‿key logā? “when does/will your son come home from school?” Progressive present: beya duši navaftayay ke hālā čort ma-janay–ay? “have you not slept last night that you are napping now?” Progressive past: vaxti ke sel āk(k), āšān če kār-i ma-kaney–adant? “what were they doing when the flood came?” There are however many examples in which the form with ma- is used to express the present-future and the habitual, e.g., četaːr āši pa-jā namārayay? “how come you don’t recognise him?” sabā sohbi marava (a contraction of ma-ravag–en) “he will go tomorrow morning” kākā-m madrasā namarava “my brother doesn’t go to school” 14 For its syntactic position, see section 13. Acta Orient. Hung. 67, 2014 462 HABIB BORJIAN Table 5. Summary of Verb Forms (2nd and 3rd sg.) “sleep” (intrans.) 2nd sg. “eat” (trans.) 3rd sg. 2nd sg. 3rd sg. Infinitive vasb-ag Imperative bu-asb-ø – bu-a(r)-ø – bu-asb-ay bu-asb-i bu-ar-ay bu-ar-t Pres. indic. a‿wasb-ey a‿wasb-i *a‿war-ey ? Pres. prog. ma-vasvay–ay ma-vasba(g–en) ma-varay–ay Imperfect ma-vasvay–aδay ma-vasvay–aδ ma-verey-ad‿ay ma-verey-ad‿ø Preterit xaft-ay vārt‿et vārt‿i vārtay‿et vārtay‿i vārtay-ad‿et vārtayad‿i vārtay‿et bi vārtay‿i bi Subj. pres. 15 Pres. perfect vafta(y)–ay Pluperfect vaftay–aδay Perfect subj. vaftay–ay bi var-ag xaft-ø vafta[g–en] vaftay–a vafta bi δ ma-vara(g–en) It appears that the muddle in the present imperfective of Koroshi is influenced by the construction of the present-future tense with mi- in modern Persian. The past imperfective is marked in all available data only with ma-, indicating a merge between the simple and progressive past, e.g., HM ma-rrawag–adan “I would go, I used to go, I was going”, ma-gašag–adan “I would say, I was saying”. See also section 18. 17. The Perfect Periphrastic. The intransitive present perfect and pluperfect are constructed by the infinitive and past participle plus the present and past copula respectively, e.g., HM ātkag–am, AS ākkay–ān “I have come”; HM ātkag–adan, AS ākkey–adān “I had come”. The perfect subjunctive is the present perfect plus bi, the third singular of “be/become”, e.g., ākkayān bi (for Pers. āmada bāš-am) “I may have come”. 18. Transitivity. As in many other Iranian languages, Koroshi transitive past employs pronominal enclitics as subject or agent markers. These may either stay on the verb in lieu of personal suffixes, or they may front the sentence onto a preceding word, which can be the preverb, the nominal component of compound verb, or the object. These two parallel constructions can be exemplified in berenj geft‿om ≈ berenj‿om geft “I bought rice”. Mahamedi (1979, p. 287) found no instance of the agent on the 15 Data from Mahamedi (1979). Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH 463 verb, but Salāmi’s documentation offers plenty of examples with both constructions. As regards the imperfect, there is no example with a fronted agent in the data at hand. As demonstrated by the formulas in Table 4, the structure of the transitive preterit and present perfect differs from those of intransitive in that the personal suffix (ending) is replaced by the enclitic pronoun (agent), with fronting as an option. In the imperfect and pluperfect, however, not only the past copula is replaced by the enclitic pronoun (agent) but also the morpheme -ad- (variant: -at-) is inserted between the verbal noun and the agent, or if the agent is fronted in the pluperfect -ad becomes word-final. This morpheme equals the past copula third person singular ad. (Note: there are many instances of intransitive endings being attached to transitive verbs, e.g., az pirāri aši neyδay-ān “I haven’t seen him since two years ago”.) Examples for (1) the agent staying on the verb, (2) the agent is fronted to a preceding word: Preterit: (1) man vaːd-om did‿om “I saw [it] myself”; i qašanguveyn pirāhāmā aː ko geft‿et? “where did you buy this pretty shirt?”; sibobārā aː deraxt-ā čed‿eš “they picked the apples from the tree”. (2) kāyad manā gānk‿i jad? “who called me?”; kāmā log-ā vːad-i doros‿i‿ko? “which house did he build himself?” Imperfect: xiār-ā ma-borey-ad‿om “I was cutting the melon”; ma-genay-ad‿ø “he would see” (note the third singular is unmarked instead of taking the expected pronominal enclitic ‿i). Pres. perfect: (1) čikkada pul gu vad-et overtey‿et? “how much money have you brought with you?” (2) tarā a log-ā dar‿eš gekā “they have dismissed you from the house”. Pluperfect: (1) gāštag-od‿om “I had said”; xorāk-ey nesf-ā vārtey-ad‿om “I had eaten one half of the meal”. (2) HM dorəs‿ət kodag-at “you had made”; āf‿i wārtait “you had drank water”. 19. Negation. The negation morpheme ná- is prefixed to the stem or to imperfective markers, e.g., J-N a‿na-twān-t bār-ā bebā “it cannot carry the load”; na-ma-ken-ēn “we are not doing”. The prohibitive is marked with má-, as in makan “do not!” 20. Be, Become, Have. The substantive verb or copula is normally suffixed to nouns. The past substantive is formed by infixing -ad- to the present copula. The copula may be prefixed with b(V)-, in which case it conveys either “be” or “become”. The list in Table 3 for the copula (together with clitics and verb endings) serves verb conjugation; in general usage the substantive verb suffixes are most commonly found as: Present: sg. -ān, -ay, -en, pl. -en, -et, -ant; Past: sg. -adān, -aday, -ad/-at, pl. -aden, -adet, -adant. Examples: i lug‿en “this is a/the house”; i asp esbieδ‿en “this horse is white”; i bāmerd kā‿yen? “who is this man?”; Ali ko‿yen “where is Ali?”; panā bod (Pers. penhān šod) “it hid”. Existence and ownership are conveyed by attaching the enclitic copula (or enclitic pronouns) to the base ass-/ast-, e.g., sabā-šāmi mā logā assām “I will be home tomorrow night”; ā bāḡ mahsul-i gahter-i assen “that orchard has better Acta Orient. Hung. 67, 2014 464 HABIB BORJIAN produce”; siheyn bolut ōri ni “a black cloud has no rain”; HM čon bačč-et assen? yeki assen‿em “how many sons do you have? I have one”. Lexis 21. Inherited Balochi words form the lexical stock of Koroshi; some examples are: bot “louse” (Bal. boṭ, bot, boḍ) (Rossi 1979, section E13), dāti “maternal aunt” (Bal. tātī “paternal aunt”; borrowed from Indic), dim “face” (Ir. Bal. dēm), dogār “earth” (Marw Bal. ḍigar), fād “salt” (Marw Bal. wād), gahr “cold” (Ir. Bal. gwahr), gamz “wasp” (Sarāvāni gwavz, Ir. Bal. gudar), hor or ōr “rain” (Ir. Bal. haur), jahlād “low” (Ir. Bal. jahl(ā)), jogen “mortar” (Bal. jogun; also in Lori, Lārestāni), leyp “game, dance” (Ir. Bal. leib), marōči “today” (= Marw Bal.), nagan “bread” (= Bal.), šaːnek “kid” (Ir. Bal. šenik, Marw Bal. šinik), aškon- ː aškont- “listen”, pa-jā ārag “to recognise”. 22. Loanwords are predominantly from Persian, as is the case with other Balochi dialects. Koroshi carries some additional Perside words that are identifiably from surrounding dialects native to Fārs: merzeng “eyelash” (≠ Ir. Bal. močāč), ālāyen “visible”, ārma “longing of pregnant women”, bey “moth”, čey “thing”, debr “rough surface”, parčal “dirty”, pā-sed “ladder”, šekāl “wild goat”, xenj “claw”, zuzuk “hedgehog”, zaːla-m mira “I fear” (tars is not used). Areal lexicon dominate family terms: doyi “mother” (cf. Bal. māt, Marw Bal. mās), bovā(y) “father” (Marw Bal. pis), daδa, gāhār “sister” (Ir. Bal. gohār, Marw Bal. gwar, dādā, Sarāvāni warg), kākā, berād “brother” (Bal. brāt, Marw Bal. brās, lālā, lālayk), jānek “daughter” (Marw Bal. dutag). 23. The inventory of Turkish/Qashqa’i loanwords is surprisingly small; examples are ālma “apple” (Azeri Turk. alma), bolut “cloud” (Turk. bulut), buḡānāx “whirlwind” (Azeri Turk. boğanaq “storm”), čoqqor “pit” (Turk. çukur), došān “rabbit” (Azeri Turk. dovşan), gālen “bride” (Azeri Turk. gəlin), qaδaḡan kang “to entrust”.16 24. Among idiosyncrasies one finds šehid “thirsty” (lit. “martyr”); inspired by the Shi’i Moḥarram rituals, the word is also found in Farvi, a dialect of Xur-Biābānak district in central Iran. Abbreviations Bal. = Balochi (Baluchi) Ir. = Iranian Ir. Bal. = Balochi dialects spoken in Iranian Baluchistan 16 words. The author would like to thank Kioumars Ghereghlou who helped identify the Turkic Acta Orient. Hung. 67, 2014 THE BALOCHI DIALECT OF THE KOROSH Kor. OIr. part. Pers. pl. pres. sg. Turk. 465 = Koroshi (Koroši) = Old Iranian = participle = Persian = plural = present = singular = Turkish Symbols: hyphens (-), en-dashes (–), and concave underscores (‿) signify clitics in a decreasing order of boundedness. A colon symbol separates the present and past stems. References Bellew, H. W. (1891): An Inquiry into the Ethnography of Afghanistan. London. Borjian, H. (2013): Farvi Dialect. In: Encyclopaedia Iranica Online; at http://www.iranicaonline. org/ articles/farvi-dialect. Elfenbein, J. (1963): A Vocabulary of Marw Baluchi. Naples, Istituto Universitario Orientale di Napoli. Elfenbein, J. (1989): Balōčī. In: Schmitt, R. (ed.): Compendium linguarum iranicarum. Wiesbaden, Reichert, pp. 350 – 362. Jahani, C. – Korn, A. (2009): Balochi. In: Windfuhr, G. (ed.): The Iranian Languages. London, Routledge, pp. 634 – 688. Jahani, C. – Nourzaei, M. (2011): A Folktale in Koroshi Dedicated to Joy Edelman. In: Leksika, ėtimologija, jazykovye kontakty. Moscow, pp. 62 – 70; at http://www.lingfil.uu.se/digital Assets/133/133434_die-sbornik-jahani-nourzaei-2.pdf. Korn, A. (2005): Towards a Historical Grammar of Balochi: Studies in Balochi Historical Phonology and Vocabulary. Wiesbaden, Reichert. Korn, A. (2006): Counting Sheep and Camels in Balochi. In: Bogoljubov, M. N. et al. (eds): Indoiranskoe jazykoznanie i tipologija jazykovyx situacij. Sbornik stat’ej k 75-letiju professora A. L. Grjunberga (1930 – 1995). St. Petersburg, Nauka, pp. 201 – 212. Mahamedi, H. (1979): On the Verbal System in Three Iranian Dialects of Fars. Studia Iranica Vol. 8, pp. 277 – 297. Morgenstierne, Georg (1927): An Etymological Vocabulary of Pashto. Oslo, Dybwad. Rossi, A. V. (1979): Iranian Lexical Elements in Brāhūī. Naples, Istututo Universitario Orientale. Salāmi, ʿA. Pur-e Davān (2004a): Ganjina-ye guyeššenāsi-e Fārs. Vol. 3. Tehran, Farhangestān. Salāmi, ʿA. (2004b): Barrasi-e ejmāli-e guyeš-e koroši. Guyeššenāsi Vol. 1, No. 3, pp. 39 – 56. Spooner, B. (1967): Notes on the Baluchī Spoken in Persian Baluchistan. Iran Vol. 5, pp. 51 – 71. Acta Orient. Hung. 67, 2014