i. Lori Dialects

These are spoken by both settled and migratory folk over a large area of western Iran, including parts of Hamadan Province (at least from Nehāvand southward) through Lorestān to Khuzestan, Čahār Maḥāl and Baḵtiāri, Kohgiluya and Boir Aḥmadi, and Fārs. They belong, together with Persian, to the southern branch of Western Iranian (for features that distinguish southwestern Iranian languages from related languages, see Sundermann 1989).

The Lori dialects divide naturally into two quite distinct groups corresponding to the distinction between the so-called Greater Lors (Lor-e bozorg) and Lesser Lors (Lor-e kuček). In the following, Greater Lori dialects will be referred to as Southern Lori (SLori), Lesser Lori dialects as Northern Lori (NLori). 

Reported SLori dialects are Baḵtiāri (henceforth Baḵ; Zhukovskiǐ, 1923; Mann, 1910; Lorimer, 1922, 1930, 1954, 1955, 1963) and Mamasani of the Boir Aḥmadi group (Mam, B-A; Mann, 1904, 1910), other Boir-Aḥmadi (Loeffler, 1977; Lamʿa, 1970), Kohgiluya (Mann, 1910). The dialects spoken in settled communities of Khuzestan such as Rāmhormoz and Masjed Solaymān are a form of SLori (author’s field notes). More distantly related are Dezfuli and Šuštari (D-Š; MacKinnon, 1974).

Reported NLori dialects are Giōni (henceforth Gi; MacKinnon, 1977), Ḵorramābādi (Ḵor; Izadpanāh, 1964; MacKinnon, 2002), Čagani (henceforth Ča; Blau, 1993), and Bālā Gerivāʾi (BGr; Amanollahi and Thackston, 1987; see also Mann, 1904 and 1910, for early material on “Feili,” a traditional term used to denote speakers of NLori, and from Nehāvand).

The geographical dividing line between NLori and SLori appears to be the river Āb-e Dez (q.v.). The Lori dialects and their relationships with one another are not well investigated. Studies of individual dialects are few in number and vary in quality. The last serious field work on Baḵtiāri, for example, the most extensively reported of the Lori dialects, dates from the early 20th century and, though good for its era, suffers from conceptual and other handicaps. Scholars at the time had no notion of the phoneme, for example, nor did they attempt thoroughgoing analytic investigations of the dialects they were studying, preferring instead to furnish extensive texts and translations into European languages and occasionally Persian. Oskar Mann and David Lorimer also produced useful but short notes on phonology, morphology, and syntax. (Kerimova, 1982, is an attempt to synthesize what is reported of Baḵtiāri and other Lori dialects including NLori; MacKinnon 2002 compares features of reported NLori dialects.) Nevertheless, despite the differences in quality and reporting styles of the descriptions, it is possible to come to some firm conclusions about the language spoken by the Lori group and arrive at a kind of West Persian typology.

All Lori dialects closely resemble standard Persian and probably developed from a stage of Persian similar to that represented in Early New Persian texts written in Arabic script. The sole typical Lori feature not known in early New Persian or derivable from it is the inchoative marker (see below), though even this is found in Judeo-Persian texts. There is furthermore no need to assume a common “proto-Lor” stage for the dialects grouped under the term Lori. NLori and SLori, though they share many features, probably developed separately though along parallel lines. Despite their similarities to standard Persian, the Lori dialects share features that set them apart as a group from the standard language. It is in their phonology that the Lori dialects diverge most noticeably from Persian.


Vowels. The vocalic systems of all Lori dialects may be derived from one resembling that of Early New Persian, consisting of five tense vowels, /ī/, /ē/, /ā/, /ō/, /ū/, and three lax, /i/, /a/, /u/. The vocalic inventories of SLori dialects are similar in their articulation to that of standard Persian.

NLori dialects, however, are quite different. They exhibit (1) front rounded vowels [ö], [ü], (2) central articulation of lax midlevel /e/, (3) strong diphthongization, including the development of a diphthong with a front round base vowel [ö] followed by a front round rising glide [j], and (4) a phone, tense front [Ī], that is acoustically close to but not identical with /ī/. The articulatory position of NLori [Ī] probably results from the crowding of phonological space resulting from the development of front rounded phones. Dezfuli, on the border of the NLori area, shows a strong tendency to front /ū/ and /o/ and shares other features with NLori (as noted below).

Consonants. The consonantal inventories of the Lori dialects are close to that of Standard Persian. The following differences are noteworthy: Most Lori dialects maintain the distinction between /q/ and /γ/. Thus, Ḵor, BGr qapō “scales,” γurc “wild sheep,” D-Š qabul “accept,” γarib “foreign.” In Baḵtiāri initial /γ/ usually merges with /q/; otherwise the distinction is maintained: rōγan “oil” čaqū “knife.” Giōni merges the two sounds into the voiceless velar /q/: qazā “food,” qār “cave,” qarü “strange.”

The NLori dialects have developed a strident lingual, /ł/, produced by bringing both the tip and sides of the tongue into contact with the alveolar ridge and allowing breath to escape around the tip of the tongue. This results in a hissing sound with a palatalized quality: pił “money,” łöj “lip.” Baḵ, Mam have [w] for Standard Pers. [v] in intervocalic position. Baḵtiāri postvocalic, spirantized /d/ is apical, not dental. Alone among Lori dialects, Baḵtiāri often changes intervocalic /m/ to /w/: dōwā (Lorimer, 1922), dūwā (Mann, 1904) “bridegroom,” aveid-om “I came” (Lorimer 1922). Dezfuli and Šuštari have acquired pharyngeals from Arabic, which occur in Arabic loanwords (Šuš ʿajīb “strange,” baʿd “after,” ḥōš “enclosure, house”) and have also been extended to some Persian items (Šuš ʿas “bone”).

Common Lori phonological developments. An ensemble of historical phonological rules shared in one form or another among all Lori dialects (and to a degree with other non-Lori dialects such as the Fārs dialects (see FĀRS viii) to the southeast and Laki to the northwest) have restructured the Lori phonemic inventory and given these dialects their distinctive coloring. The main processes are raising, merger, fronting, and laxing of vowels, and the spirantization or complete loss of post-vocalic voiced stops.

Raising of midlevel vowels. In all Lori dialects the tense mid-level back vowel /ō/ is raised and merged with /ū/: Baḵ dūst, D-Š dūs, Ḵor dūs, Gi dūs “friend.” Lorimer (1922) reports tense [ō] in Baḵ durō “lie” and items such as pōud-an “cook,” dōhd-an “to sew; milk.” These last, however, likely exhibit lax /o/ lengthened by dropping of the following [h] (see below, Laxing before -ḵt). The corresponding D-Š items are poḵt-a and doḵt-a, Ḵor, Gi poḵt-e, doḵt-e.

In NLori the corresponding tense front vowel /ē/ remained distinct: Ḵor, Gi, BGr mĪš, “ewe.” In SLori the situation is more complex. Dez preserved /ē/ (Dez mēš), Šūš did not (Šuš mīš). Lorimer (1922) and Mann (1904) report Baḵtiāri possessing both [ī] and [ē] as reflexes of Early New Persian.

Raising before nasals. All Lori dialects raise tense /ā/ before /n/, /m/. In NLori /ā/ becomes tense midlevel /ō/: Ḵor, Gi, Ča, BGr hōna-mō “our house.” The BGr reflex of /ā/ is reported to be close to but not identical with [ū]. SLori /ā/ merges with /ū/: D-Š ḵūna, B-A hūne “house,” Mam zūnī “knee,” dūmā “bridegroom.” Lorimer reports variations between [ō] and [ū] in Baḵ: dōnist-an ~ dūnist-an “to know,” hōna ~ hūna “house,” šūm ~ šōm “evening meal.”

Fronting of /ō/, /ū/. In all Lori except Gi, original /ō/, /ū/ were fronted and merged with /ī/: Ḵor, Ča pił, D-Š, Baḵ, B-A pil “money.” In Gi, original /ō/, /ū/ were fronted but retained rounding as in Gi püł, a process which added a phone /ü/ to the Giōni vocalic inventory. Fronting likely occurred before the raising in SLori of /ā/ to /ū/ before nasals as D-Š ḵĩ, Baḵ hīn “blood,” but D-Š ḵūna, Baḵ hūna “house” would suggest. Fronting appears to have occurred before dental stops and historical -y, but not before labials, strident dentals, or velars. This suggests the rule was an assimilation process in which original /ō/, /ū/ were fronted in anticipation of the following dentals.

Laxing before -ḵt. Lori dialects lax the tense midlevel vowels before the sequence /-ḵt/: Ḵor, Gi, D-Š poḵt “cooked,” reḵt “poured.” Amānolāhi and Thackston report suḵtan “to burn” with lax [u], but rĪḵt-an “to pour” with tense [Ī].

Postvocalic spirantization. Voiced stops occurring after vowels, particularly in words of Iranian origin, were often spirantized or dropped altogether. Ḵor, Gi willow, Gi “smoke,” D-Š mār “mother,” berār “brother.”

In NLori the sequences /ab/, /āb/ retained their front-back opposition after spirantization, becoming Ḵor, Gi [öj], [ow]: łöj “lip,” šöj “night,” but aftow “sun,” towessō “summer,” ketow “book.”

In SLori the two sequences merged with varying results in the different dialects. In Dezfuli and Šuštari monophthongs resulted, front rounded in Dez: šö “night,” “lip,” aftö “sun’; back rounded in Šuš: šō, , aftō. Baḵtiāri merged the two sequences into the diphthong [au]: šau, lau, aftau.

Dental stop assimilation. In Ḵor, Gi, Ca, BGr, D-Š, Baḵ sequences /-st-/ and /-zd-/ become /-s:-/, /-z:-/. Ḵor, Gi towesso, Dez töwessũ:, Šuš towessũ: “summer.”

Prothesis. Corresponding to the Middle Persian initial sequences /sC-/, /šC-/ are Lori /VsC-/, /VšC-/ as opposed to standard Pers. /sVC-/, /šVC-/. Thus, Ḵor, Gi espĪ “white,” eškesse “to break,” ešgaft “cave,” Dez esbē “white,” ešgam “stomach,” ešgaft “cave,” Šuš esbī, ešgam, ešgaft. The phenomenon is well known in early New Persian texts (Lazard 1963, pp. 175-76).

Lowering of /ḵ/. In NLori and sporadically Baḵ initial /ḵ/ becomes /h/. Ḵor, Gi hard-e “to eat,” hāss-e “want,” hoftī-e “sleep,” han-e “read,” Baḵ hin “blood,” handist-an “laugh” hoya “egg,” hīš (in qaum o hīš), hušk “dry,” but ḵāst-an “want,” ḵārd-an “eat,” ḵausīd-an “lie down,” ḵerīd-an “buy.”

Deletion of /f/. NLori dialects delete /f/ in the sequence /-ft/ in three words, corresponding to Standard Persian gereft “took,” goft “said,” and raft “went,” with attendant lengthening of the preceding vowel. Ḵor, Gi gere:t, go:t, ra:t. In Baḵ, Mam the sequence in the same three words goes to /-(h)t/, /-(h)d/, sometimes with changes in vowel articulation: Baḵ girēd-an “to take,” go(h)d-an “to say,” rahd-an “to go,” Mam gero:t-an “they got,” rāht-īm “we went.” The process did not occur in Dezfuli and Šuštari.

Specifically Northern Lori phonological developments. All Lori dialects participated in the above processes, though variations in how those processes were carried through distinguish the dialects from one another and the NLori group from SLor. The following historical processes, in which SLori did not participate, further set NLori off from SLor.

Tensing before /h/. The sequence /ih/ becomes NLori /Ī/. Ḵor, Gi “village,” tĪrō “Tehran,” gerĪ “knot.” In Dezfuli and Šuštari exactly the opposite occurs: tense vowels occurring before /h/ are laxed but lengthened: D-Š pe: “fat,” ra: “road,” ko: “mountain.”

Laxing of /ā/ before /-nd/. Ḵor, Gi, BGr man “remained,” ran “drove,” han “read” (Pers. mānd, rānd, ḵānd). Dezfuli has undergone a similar process, though after raising of /ā/: Dez mond, rond, ḵond.

Deletion of /d/, /g/ after /n/. Voiced stops /d/ and /g/ are deleted after /n/: Ḵor, Gi čan “how much,” denō “tooth,” gan “rotten,” man-e “to remain,” töfaŋ “gun,” jaŋ “war.”

Fronting of lax /o/. NLori dialects front lax midlevel /o/ in the vicinity of the dental consonants /t/, /d/, /n/, /s/, /z/. Ḵor, Gi “two” nök “beak,” “morning,” šöste “wash,” šötör “camel” zö:r “noon,” Ča “you (sg.),” düm “tail.” The environment in which this rule operated suggests an assimilation process similar to that in which /ō/, /ū/ were fronted.

Synchronic Phonology. Deletion of final /n/. All Lori dialects delete final /n/. Ḵor, Gi “bread,” nōn-yā “loaves of bread,” venō “they,” venōn-Ī ke ḵeri-em “the ones I bought.” SLori dialects show strong nasalization of the preceding vowel (D-Š nũ:, “bread,” towessũ: “summer,” Baḵ duhũ: “mouth”), NLori dialects do not.

Also noteworthy are processes of umlauting, though these take various forms in the different dialects. Thus: Gi emā-m “I came,” emū-ī “you (sg.) came,” Dez dār-om “I have” dör-ī “you (sg.) have,” Šuš piā “man,” piō “a man” (piā + ). The NLori dialects have developed an opposition between ordinary /l/ and a lateral fricative /ł/ (see above) along with a parallel opposition between a flapped and rolled /r/. As in Kurdish and other NW dialects, the strident, longer sounds occur initially, the softer ones do not. The strident sounds also occur in contact with high, front, tense vowels (Ḵor mē-īł-em “I allow,” bořī “he cut”).


Lori morphological categories are close to those of Standard Persian, though the devices differ somewhat.

Nominal Morphology. Noun. The plural is marked in Lori according to the same rules as Standard Persian. The plural morphemes vary, however. All NLori dialects use the stressed suffixes -ō(n) for animates, -yā for inanimates (the latter is known from Early New Persian; see Lazard 1963, p. 196); Čagani in addition to these uses -(a)l; D-Š use -ū(n) for animates, -(h)ā for inanimates; Baḵ, Mam gal.

All Lori dialects possess (1) the eżāfa, unstressed lax -e (e.g., BGr berār-e gap “big brother”); (2) the relative marker (Ḵor, Gi, Ča, BGr unstressed , Dez ē, Šuš, Baḵ , e.g. Gi ow-Ī ke hard-em “the water that I drank”); (3) the indefinite marker (phonologically quite like the relative marker though of different origin, e.g., BGr darvĪš-Ī uma “a dervish arrived”); and (4) the definite object marker (in all dialects unstressed -ne/-na, e.g., Mam dūmā-na āwurd-īm “we brought the bridegroom”).

None of the Lori dialects differs from Standard Persian in the use of these morphemes, though in their phonology the dialects differ. Thus, in NLori the eżāfa is realized only after consonants (in Ḵor also after -Ī). In Dezfuli and Šuštari it occurs in all contexts and in Šuštari raises a preceding low vowel: Šuš piā “man,” but piō gap “big man.” In most Lori dialects the definite object marker is realized as -ne, -na only after vowels; otherwise it is -e, -a (in Ḵorramābādi it is realized as -ene in all positions except after enclitics or nouns ending in -n: tenāf-ene boři-em “I cut the rope,” but ser-eš-e boři-em “I cut its head”).

Demonstrative Adjective. Corresponding to Pers. īn, ān are common Lori ī, ū. (See Demonstrative Pronouns below for plural forms.)

Demarcative Suffix. NLori -(e)ka, D-Š -aka, Baḵ -ekū. The Demarcative Suffix marks a singular or plural noun usually to contrast it with some other noun or nouns. It is often suffixed repeatedly in the discourse to the same noun. It corresponds to Tehrani Persian stressed final and exhibits the same syntax: the eżāfa between a noun and its following qualifier must be deleted when the construction is marked with -aka as must the eżāfa in comparable phrases in Tehrani. Thus, Šuš kwak-e gap “big boy” (Pers. pesar-e bozorg), but kwak gap-aka “the big boy” (Pers. pesar bozorg-é).

Deictive Suffix. In Ḵorramābādi and Dezfuli-Šuštari stressed -a may be attached to a noun, often in conjunction with the demonstrative adjectives. The suffix appears to emphasize the distinctness of the noun or the deictive function of the construction without necessarily contrasting the noun with some other noun: Šuš ū čūa “that piece of wood.” As with the Demarcative Suffix, eżāfa must be deleted in constructions so marked: Ḵor mōn-an vā čū gap-a “they hit with the big stick,” Šuš ū kīf gap-ana bī-ār dā-m “bring that big sack and give it to me.”

Personal Pronouns. Lori pronouns in the singular are close to identical among the various groups, essentially NLori me, SLori mo; NLori , . SLori to; common Lori ū. Third person singular and plural are the demonstrative adjectives.

The plural items vary. Bālā Gerivāʾi and SLori show a tendency to regularize the paradigm by various analogical means. NLor: Gi īmā, šemā, ūnō; Ḵor Īmā, šemā, ūnō; Ča emā, šemā, ūnō; BGr īmū, šemū, ūnō. SLor: Dez omũ, šomú, ūšũ; Šuš amā, šamā, ūšũ; Baḵ īmā, īsā, yunun; Mam īmā, īšā, īnūn.

Suffix Object/Possessive. NLori -em, (SLori -om), -et -eš (Baḵ -es), -omō(n), -etō(n), -ešō(n) (SLori -emū(n), -etū(n), -ešū(n); Baḵ esū(n)).

Possessive. Most Lori dialects are reported to build possessive pronouns corresponding to Persian māl-e man, māl-e to “mine, yours, etc.” on a base hi(n)- plus possessing suffixes. Thus, Ḵor hĪn-e me “mine,” hĪn-e tö “yours” and the like.

Reflexive. In all dialects based on ḵ- plus the personal suffixes.

Demonstrative. Lori dialects have built a set of Demonstrative Pronouns on the demonstrative adjectives. Gi, Ḵor ya, va, “this,” “that,” and plurals yenō(n), venō(n), are probably demonstrative adjectives plus the Deictive marker -a and an intrusive [n] in the plural. BGr has (near) hi, henō(n), (far) hu, honō(n), Dez (near) ī, īà, (sg.), hanũ (pl.), (far) ū, ūã, (sg.), hūnũ (pl.); Šuš (near) ī, īã, īhe(n) (sg.) (ī)henũ (pl.), (far) ū, ūā, (ū)hũ (sg.), (ū)hūnũ (pl.). Baḵ and Mam show ī, ū, yūnū, , hūnū. Mam also has plurals in-gal, un-gal, un-gal-un.

Verb Morphology. Verb. Lori verbal categories are close to those of standard Persian. Verbal constructions are based on three stems, present, past, and perfect. The final syllable of the past and perfect stems is stressed. Presents are stressed on the personal ending except when negated or in the subjunctive mood, in which case stress shifts to the prefix. (D-Š subjunctive prefix is zero.) As with standard Persian, Lori dialects mark true direct objects in the past tense and show no sign of ergativity.

Present. All Lori dialects build a present tense on the present stem plus unstressed modal prefix and stressed personal suffixes. The prefix is NLori mĪ-, D-Š be-, Baḵ, Mam, BA ī-, e-. Personal suffixes are: Gi -em, -ī -a, -īm, -īt, -an; Ḵor -em, -ī -a, -Īm(on), -Īt(on), -en; Ča -em, -ī -a, -ymo/-mo, -et/eto, -an; BGr -em, -ī -a, -Īm, -Īt, -an. SLor: Dez -om, , -a, -ēm, , -en; Šuš -om, , -a, -īm, -īt, en; Baḵ -om, , -i, -īm, -īn, -an/-in; Mam -om, , -a, -īm/-īmū, -ītū/-tū/-īn, -an/-in.

Ongoing action may be emphasized by using various devices. In NLori may occur with the verb: Ḵor me hā mo-hor-em “I’m eating (right now),” BGr hā yā “he’s coming.” Like colloquial Persian, D-Š use forms of “have”: mo dār-om be-ḵūn-om “I’m reading (right now).” Ḵorramābādi may use forms of “have, keep” as well.

Imperative. Except in Dezfuli and Šuštari imperatives are formed with the prefix stressed ba- and the unmarked present stem in the singular, with the 2nd person plural suffix in the plural. Gi barĪz, barĪzīt “pour!” In Dezfuli and Šuštari the imperative marker is zero: D-Š “read!” NLori form the imperative of the verb “go” anomalously without prefix: Ḵor řōj “go” singular, řē-to, plural, Gi řōj, řü-it, Ča řē-to (plural), BGr raw (singular). For negation see below.

To be. The present copula, like that of Persian, is identical to the personal suffixes of the present tense verb. Lori also shows forms that parallel Persian hastam, hastī etc., built on stems in h-: Ḵor, BGr hā- , Gi hĪ-, Dez hē-, Šuš hī-, Baḵ hēd.

Past. The past tense is built on a past stem plus personal suffixes, the same as the present tense suffixes, though unstressed, except for third singular, which is unmarked.

Perfect Tenses. Present perfects are based on the past tense + 3rd person singular of “be”: Ḵor hard-em a, hard-ī a “I have eaten, you have eaten,” Dez goft-om a, Šuš ḵward-om a, Mam gōt-om a “I have said,” Baḵ kird-im a “I have done.”

SLori past perfects are formed analogously with “it was”: Šuš ḵward-om bīd “I had eaten,” Baḵ ašnīd-om bī “I had heard.” BGr affixes , possibly a truncated form of , to the past tense to form past perfect: rat-am-ī “I had gone.” Giōni and Ḵorramābādi form past perfect essentially as in standard Persian: Ḵor hard-ī-m, Gi hardüīm “I had eaten” (harda + bi- + -m), though in Ḵorramābādi may be suffixed to all persons except 3rd singular: hard-im-i, hard-it-i; plural: hard-ī-Īmō(n-ī), hardīĪtō(nī), hard-ī-n-ī.

A separate perfect form to indicate current state is reported in Ḵorramābādi, Bālā Gerivāʾi, Dezfuli, and Šuštari, built on the past participle plus forms of “be.” Typically the verbs used in this construction mean “stand,” “sit,” “sleep” and the like: Ḵor nešessā´-m “I am sitting” (lit. “I have sat down”), Dez versī́´d-am, Šuš versā́´d-am “I’m standing,” both with stress on the final syllable as in Ḵor. Ḵor uses forms of hām “I am” etc. D-Š uses forms of the basic copula, -m, “I am,” –i  “you are,” etc., D-Š the copula.

Subjunctive. A present subjunctive is formed by prefixing stressed ba- to the present stem (Ḵor mo-hon-em “I read,” bo-hon-em “(that) I read”) except in D-Š, where the un-prefixed stem indicates subjunctive mood: be-ḵun-om “I read” ḵun-om “(that) I read.” A perfect subjunctive is built on the perfect stem and the subjunctive of “be”: Ḵor gere:t-ū-em (gere:ta + bū-em) “(that) I have taken,” Šuš geroft-a būw-om.

A suffix (BGr -ay) is attached to verbs expressing unrealizable wishes: Šuš kāš ke morγ-ī bīd-om-ē, andar havāb perhess-om-ē “I wish I were a bird, [that] I could fly into the air.”

Ḵorramābādi forms a past unreal based on the past perfect prefixed with ba- and suffixed with ; an object may occur before the suffix: Ḵor kāš ba-han-ī-m-eš-ē (ba + hana + + -m + -eš + ) “I wish I had read it.” In BGr the prefix ba- is reported to be optional with most verbs.

Optative. An optative Mood (see Lazard 1963, pp 338-39), formally distinct from the subjunctive, is reported in SLori dialects, but not NLor. It is formed on the present stem + ā: Šuš van-h-ā-m “may I be placed,” Dez ʿaziē ger-ā-t “may mourning seize you!” Baḵ ne ven-ā-ī tīr “may you not fire a shot.”

Preverbs. Most Lori dialects structure preverbal verbs as in standard Persian: Ča jager-š ö pöf-eš dar mi-y-ār-a “she takes out its liver and lungs.” Dezfuli and Šuštari, however, prefix both the negative and modal particles, in that order, before the preverb: Šuš na-m-der-vord-en-eš “they wouldn’t take it out.”

Negation. All Lori dialects negate indicative verbs with the stressed prefix na-, which may not co-occur with the subjunctive/imperative marker ba-. Dezfuli and Šuštari, which use be- as indicative modal prefix, negate by prefixing na- to the indicative and nasalizing the modal particle: be-ḵwar-om “I eat,” na-m-ḵwar-om “I don’t eat.” Negation of the imperative is Ḵor, Ča, BGr stressed na-, Gi, D-Š, Baḵ stressed ma-: Ḵor na-za “don’t strike,” Ča geriva na-ko “don’t cry”; Gi ma-röj “don’t go,” D-Š ma-osũ “don’t take,” Baḵ pušt ma-dih-in sō “don’t turn your back on them.”

Inchoative. An inchoative marker (cf. Mid. Pers. -īhist) in various forms is in use in all Lori dialects to indicate change of state in the subject. In NLori it is -y- attached to the present stem, and -ess attached to either present stems of verbs whose infinitives end in -ī-e, -ā-e (e.g., čarḵĪ-e “to turn”) or past stems of other verbs to indicate past tense. Ḵor: mI-čarḵ-y-a “it turns,” čarḵ-es “it turned,” mĪ-sūz-y-a “it is burning,” soxt-es “it burned,” soxt-ess a “it has gotten burned.” Cf. BGr larisa/lar-y “to turn.”

In Dezfuli and Šuštari -ehess is attached to the present stem to form inchoatives: D-Š be-čarḵ-ah-a “it turns,” čarḵ-eh-es “it turned.” The inchoative often marks intransitives such as verbs of motion: D-Š be-ras-oh-om “I (shall) arrive,” ras-ess-om “I arrived.” Cf. also Baḵ tarak-ist-an “to split,” čīr-ist-an “to drip.”

Causative. All Lori dialects exhibit a morpheme (Gi -enā, Ḵor -ōn/-an, D-Š -enī, Baḵ -n), which attaches to the present stem of intransitives to form transitive’s: Gi mĪ-sūz-en-em “I burn,” sūz-en-ā-m “I burned’; Ḵor mĪ-res-ōn-a “he sends,” resan “he sent”; D-Š be-časb-en-om “I attach,” časb-en-īd-om “I attached.”

Be, become.” It is striking that Lori dialects, like Middle Persian but unlike standard Persian, do not use verbs of motion such as šodan, gaštan, āmadan and the like to indicate change of state or process. Rather, all Lori dialects use forms of “be,” bī- or bū-, marked with various verbal morphemes, generally the continuous prefix, to indicate “become”: Ḵor garm m-u-a (mĪ + bū + -a) “it’s becoming warm,” Šuš garm e-bu-a id. Dezfuli and Šuštari form the past of “become” with the inchoative marker: Dez bi-ess-ã, Šuš bu-ess-ã. Baḵ prefixes vā-: vā-bī “it became.” It is also striking that unlike Standard Persian no Lori dialect uses the stem bāš- to indicate subjunctive or imperative of “be.” (Middle Persian uses bāš- as an imperative, though not a subjunctive stem). All use stems bī- or bū- and subjunctive markers: Ḵor bu-em, Šuš bu-w-om “(that) I be.”


Much of the everyday vocabulary of the Lori dialects sets them off from the standard language. Note such items as: D-Š ešgaft “cave,” gap “big,” mīn “in,” korr “boy,” nehā/nohō “in front of,” hanī “another,” piā “man,” Ḵor remesse “to collapse,” BGr rumissa “be destroyed,” D-Š rom-n-īd-ã, Baḵ rum-ni-n “to destroy,” “for,” Ḵor, Gi, BGr šana, Dez šondã, Šuš šund-ã, Baḵ šund-an “throw,” Ḵor, Gi, Ča van-e, D-Š vand-ã “throw.” Note various forms of Arabic ḥauš “enclosure” to indicate house: Ḵorjš, Dez ḥöjš. Strictly NLori are bard “stone,” de “from,” hār “down,” łā “direction,” vārd “with,” īse “now,” ūse “then,” īče “here,” ūče “there,” nĪāye “put.” Unknown to NLori is D-Š kuak “boy,” D-Š tar-ess-ã, Baḵ ter-est-an “to be able” (all NLori use tōnesse).

 Sample Text (Ḵorramābādi Dialect)

Īmā hoft-ī-Īm nezīk-yā sāat-e čār ö čār ö nīm. vano ke pā taš b-ī-n ö biār b-ī-n hama-ne biār kerd-en ö ve sörat hama xo-šōn-e āmāda kerd-en sī ḥarekat ve taraf-e kola.

amā qerār nĪā-n ke yak dö nafar bē-Īs-an de ešgaft ö baqia rö-an vīr-e kola. čūnke asās-yā ziād-Ī dāšt-en ke ne-mĪ-tōness-en vā xo-šō bö-jr-an-ešō, dö nafar man-en de ešgaft ke ham bē-Īs-an tĪ asā-s-yā va ham enāštā-ne hāzer kon-an ke möjqē ke īn-ō ver-mĪ-gerd-an bo-hor-an.

ḥarekat kerd-Im. nezīk-e ye sāat-I ra rat-Im tā rasess-Im ve kola. alvatta xēł-I sard bī. barf am omō-ī. ye tīkI am bard-yā kö sör bī-īn-ī. va hamī dałīł Imā yavāš yavāš ö vā saxtī mI-rat-Im ve ra.

“We slept until close to four or four-thirty. Those who were around the fire and who were awake awakened everyone and we quickly made ourselves ready to set out for the ambush.

“But they arranged that one or two persons remain in the cave and the rest set out for the ambush. Since they had a lot of equipment which they could not take with them, two persons remained in the cave both to stay with the equipment and to prepare breakfast so that when they return they might eat.

“We set out. We walked for about an hour until we arrived at the ambush. Of course, it was extremely cold. Also snow had fallen. And some of the stones of the mountain had become slippery. It was for this reason that we went very slowly and with difficulty.”

See also BAḴTIĀRI TRIBE ii. Baḵtiāri dialect; BOIR AḤMADI ii. The Dialect; DEZFUL ii. Dezfuli and Šuštān dialects; GIŌNI; KHUZESTAN, dialects.



Sekandar Amānolāhi and W. M. Thackston, Tales from Luristān (Matalyā Lurissú): Tales, Fables and Folk Poetry from the Lur of Bālā-Garīvā, Cambridge, 1987.

Joyce Blau, “Le kurde lori, Etudes irano-aryennes offertes à Gilbert Lazard, Cahier de Studia Iranica 7, 1989, pp. 37-58.

Idem, “Le čagani: Lori ou Kurde-,” Studia Iranica 22, 1993, pp. 93-119.

Bahrām Dāvari, Żarb-al-maalhā-ye batiāri, Tehran, 1964.

Hamid Izadpanāh, Farhang-e lori, Tehran, 1964. Idem, Farhang-e laki, Tehran, 1988.

A. A. Kerimova, “Lurskie i Bakhtiyarskie Dialekty,” in Osnovy III, 1982, pp. 287-315.

Gilbert Lazard, La langue des plus anciens monuments de la langue persane, Paris, 1963.

Idem, “Le dialecte Laki d'Aleshtar (Kurde meridional),” Studia Iranica 21, 1992, pp. 215-45.

Manučehr Lamʿa, Farhang-e ʿāmiāna-ye ʿašāyer-e Buyer-Amadi o Kogiluya, Tehran, 1970.

Pierre Lecoq, “Les dialectes du sud-ouest de l’Iran,” in Rüdiger Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989, pp. 341-49.

D. L. R. Lorimer, The Phonology of the Bakhtiari, Badakhshani and Madaglashti Dialects of Modern Persian, London, 1922.

Idem, “A Bakhtiari Prose Text,” JRAS, 1930, pp. 347-64.

Idem, “Popular Verse of the Bakhtiāri of S.W. Persia,” BSOAS 16, 1954, pp. 542-55; 17, 1955, pp. 92-110; 26, 1963, pp. 55-68.

Idem, “A Bakhtiari Persian Text,” in George Redard, ed., Mélanges Georg Morgenstierne, Wiesbaden, 1964, pp. 129-33.

Colin MacKinnon, The Phonology and Morphology of Dezfuli-Shushtari: A Study in West Persian Dialectology, unpublished Dissertation, Ann Arbor, 1974.

Idem, “The Dialect of Gio,” Studia Iranica 6, 1977, pp. 211-47.

Idem, “The Dialect of Xorramābād and Comparative Notes on other Lor Dialects,” Studia Iranica 31, 2002, pp. 103-38.

Oskar Mann, “Kurze Skizze der Lurdialekte,” Sitzungsberichte der Königlichen Preußischen Akademie der Wissenschaften, Berlin, 1904, pp. 1173-93.

Idem, Die Mundarten der Lur-Stämme im südwestlichen Persien, Berlin, 1910.

Werner Sundermann, “Westmitteliranische Sprachen,” in Rüdiger Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989, pp. 106-13.

V. A. Zhukovskiǐ, Materialy dlya izucheniya persidskikh narechiǐ III. Narechie Bakhtiyarov Chaharleng i Kheftleng (Materials for the study of Persian dialects III. Dialect of the Čahār-lang and Haft-lang Baḵtiāri), Petrograd, 1922.

(Colin MacKinnon)

Originally Published: January 7, 2011

Last Updated: January 7, 2011