designation of a number of Iranian dialects spoken in the center of Persia, roughly between Hamadān, Isfahan, Yazd, and Tehran, that is, the area of ancient Media Major, which constitute the core of the western Iranian dialects.


CENTRAL DIALECTS, designation of a number of Iranian dialects spoken in the center of Persia, roughly between Hamadān, Isfahan, Yazd, and Teh­ran, that is, the area of ancient Media Major, which constitute the core of the western Iranian dialects. They are most closely related to 1. the band of Ṭāleši and Tati dialects that stretches from the areas of Vafs and Āštīān (Tafreš) in the northwestern center of Persia to northern Azerbaijan (area of ancient Media Minor) and Ṭawāleš, and 2. the now largely extinct dialects of the Tehran-Ray area (ancient Raga), and the dialects of Semnān and surrounding villages to the east of Tehran (area of ancient Qūmes). They are more dis­tantly related to the Gōrānī dialects, spoken in the area north and northeast of Kermānšāh in the Māhīdašt area (Māhīdašt, lit. “the Median plain,” with Māh < Māδa-) and east of Mosul in Iraq. More distantly, yet, they are related to Kurdish and Baluchi (see baluchistan iii). Closely related to the Central dialects are the dialects spoken in Ḵūr and some other villages on the southern border of the central desert (Dašt-e Kavīr) and in Sīvand north of Persepolis in Fārs, both of which, however, also contain features that group them with Kurdish to the west and Baluchi to the east.

The Central dialects thus constitute the southern­most group of the so-called Northwest Iranian dialects, with the exception of Baluchi. To their south and west so-called Southwest Iranian or Perside dialects are spoken, including the dialects spoken in Luristan, Fārs, Lārestān, and Bašākerd.

As elsewhere in Persia, the dominating influence of Persian has resulted in the ousting of the local dialects not only in the major cities, but also in many towns and villages, where only the older generation speaks, or remembers, some of the earlier local dialect (Majīdī). In the northern and western parts of the Central dialects area the expanding Turkic dialects have played a similar role, beginning in the early Islamic centuries, in replac­ing local Iranian dialects (see kalaji). There is consider­able resistance to the intrusion of Persian and Turkic, however, and the local dialects have been retained in many locations. Information on the dialects of many cities where Persian is now spoken comes from religious minorities. In particular, the Jewish communities in such cities as Hamadān, Kāšān, Golpāyegān, Nehāvand, Ḵomeyn, Isfahan, Yazd, Kermān, etc. (see Yar­shater, 1974) and the Zoroastrians in Yazd and Kermān (See behdīnān dialect) have retained the Central dialect spoken in their cities, at least until recently. There are also a few texts, mostly poems, written in urban dialects that have been preserved in manuscripts.

Classification of the Central dialects.

Research on the Central dialects began in the 1880s and continued into the 1930s. Little was then done until the interest in these dialects was renewed in the mid­1970s. Since then much new information has been gathered, enabling scholars to understand the strati­fication of the Central dialects and the complex mor­phology especially of the northern Central dialects (Yarshater, 1974, 1985; LeCoq, 1989, Krahnke, 1976).

The internal grouping of the Central dialects has presented considerable problems because of numerous intersecting isoglosses and the lack of clear isoglottic bundling, a pattern which is probably the result of repeated internal migrations, probably dating back to before Islam, as no major population moves are re­corded for central Persia at least since the 7th century of our era.

Major isoglottic patterns. The systematic study by Karl Krahnke has put our understanding of the dy­namics of the synchrony and diachrony of the Central dialects on a firm basis. Having identified 36 distinctive phonological, morphological, and lexical isoglosses, he has shown that two major intersecting patterns can be recognized: one in which a set of linguistic features are spreading along a northwest-southeast axis, another in which they spread along a north-south axis starting in the Kāšān area. In addition to these two primary patterns, there are a number of secondary ones, in which isoglosses radiate from various centers.

The northwest-southeast pattern correlates with the topography of the area and, consequently, old and modern communication channels. The north-south pattern emanating from the northern Kāšān area, with its center around Qohrūd, reflects a conservative area with boundaries defined by topography and lines of communication. In the center of this area are the highest mountains in this part of Persia, and no main roads penetrate it, leaving the area relatively isolated. The isoglosses constitute wave-like patterns, rather than bundles, with the exception of the southeast of the area, where a northeast-southwest diagonal bundle of isoglosses, including the 1st singular ending -i and the absence of the past marker be-, separates the Nāʾīni group from the remainder of the dialects. Krahnke hypothesizes that the wave-like pattern may reflect an old genetic boundary between a northwestern dialect group and a southwestern one—now blurred but preserving its orientation—with centers somewhere along a line drawn from Ardestān to Isfahan.

The following grouping of the Central dialects is largely based on Krahnke’s observations (pp. 259-65), but also on the notes by Yarshater (1985) and LeCoq (1989). It coincides essentially with Krahnke’s northwest-southwest patterns. As Krahnke notes, given the isoglottic diversity, the members of each of these “groups” share only a few traits exclusively with each other, while there are intersections with other groups. Accordingly, the groupings reflect areal tendencies rather than discrete groups. The following schematic classification of the Central dialects may be proposed:

1. a northwestern group, including Āštīāni, Āmoraʾi, Kahaki;

2. a western group consisting of Maḥallāti, Vānīšāni, Ḵᵛānsāri;

3. a large northern-central group in the Kāšān area consisting of Ārāni-Bīdgoli, Delījāni, Našalji, Abūzaydābādi, Qohrūdi, Kāmūʾi, Jowšaqāni, Meymaʾi; further also Abyānaʾi, Farīzandi, Yārandi, Sohi, etc., and Bādi, Naṭanzi, Kašaʾi, Tāri, Ṭarqi, etc.

4. a southern group in the Isfahan area, which may be subdivided into Gazi, Sedehi, to the west and center, Ardestāni, Zefraʾi, Nohūji to the northeast, and Sajzi, Kūhpāyaʾi, Jarqūyaʾi, Rūdašti, Kafrūdi, to the east and south;

5. an eastern group consisting of Tūdeški, Keyjani, Abčūyaʾi, Nāʾīni (and the closely related Zoroastrian and Jewish dialects of Yazd and Kermān), Anāraki, Yazdi.

It should be noted that speakers of many dialects in the Kāšān area identify their own dialects as rāyejī/rājī, a term which may simply mean “local” or “current” dialect or may imply a distant reminiscence of the city of Raga/Ray, thus relating the dialects of this northern central area to those of the now extinct dialects around Tehran. In either case, the term shows that the speakers are aware that their dialects are distinct from others. Another local term used to distinguish dialects is buro-beše, literally “come!—go!” According to Krahnke (pp. 55-56, 263), this term is used by speakers in the village of Abčūya to designate the dialects just north of them in the Naṭanz area, where buro and beše are the typical imperatives of “come” and “go,” respectively (see also Yarshater, 1985, p. 745). In turn the speakers of Abčūyaʾi designate their own dialect as osme-siga, literally “now—this way,” two words typical of their own dialect. This distinction made by native speakers reflects the recognition of the major division that separates the southeast central Nāʾīni-Abčūya-Anāraki dialects from the others and which may be the remnant of an old dialect boundary in central Media Major.

Closely related to the Central dialects are:

6. Sīvandi, spoken in a linguistic enclave in Fārs (north of Persepolis), and

7. the dialects of the oases of Ḵūr and Farroḵī and some other villages in the central desert (Dašt-e Kavīr).

Sīvandi. Sīvandi shares most of its characteristics with the Central dialects, but being surrounded by Persian and Fārs dialects, it also contains words exhibiting typically Perside features and therefore prob­ably borrowed from Persian or neighboring dialects (e.g., dās “sickle” and asiow “watermill,” Pers. āsīā[b], with Perside *θr > s). Sīvandi has not only *rz > l (see above), but also *rd > l in vel-e “flower,” as commonly in the Fārs dialects (Pers. gol).

Two phonetical developments in Sīvandi set it apart from the rest of the Central dialects (for examples see below): intervocalic *č > š and initial (also secondary) *hw-/*xw- and *hu-/*xu- > f- (also in Ḵūri). Morphologically, Sīvandi has features that link it to the Central dialects, as well as features from the surround­ing dialects, such as the deictic particle -ū (but feminine -e) instead of -é. Of the personal endings Sīvandi shares the 1st singular -i (e.g., m-as-i “I sleep”) with the eastern, Nāʾīni, group of the Central dialects, but the 2nd plural -ige/-ike (e.g., m-as-ike “you sleep”) with Qohrūdi, Kešaʾi, Vānīšāni and Ḵᵛānsāri of the central and western group, with which it also shares the auxiliary gen- to form the passive. Sīvandi also pos­sesses intransitive/inchoative formations in -y-, how­ever (e.g., intrans. eškiy- versus trans. ešken- “to break”). As in the Lori and Nāʾīni dialects there is no prefix be- in perfective forms in Sīvandi, and the imperfective prefix *ma- is the same as in Persian and the Fārs dialects. The Sīvandi present and past perfect of transitive verbs, in which the agent affix precedes the auxiliaries -en “is” and - “was,” are of the Lori type (e.g., gort-em en, gort-em bī “I have/had taken”). The perfect marker -en equals the copula -en (e.g., xaš-en “it is good”), as in the Fārs and Lori dialects, as well as in Ḵūri.

Sīvandi has preserved a number of words found neither in the Central dialects nor in the neighboring Fārs dialects, such as šet “milk” (< OIran. *xšwifta-; Pers. šīr). It has a number of words in common with Kurdish, for instance, pird “bridge” (< *pṛtu-), com­pare Kurdish pīrd (but Pers. pol), and uir “fire,” compare Kurdish agir. As in Kurdish long ā is not rounded before nasals (e.g., 1st, 2nd, 3rd person plural endings -emā, -etā, -ešā < -ān).

It appears, thus, that today’s Sīvandi is the product of multiple inputs, which have considerably restructured the language of the original immigrants of Sīvand.

Ḵūri. Ḵūri (and its neighbor Farvī) exhibits the following phonetical changes not found in the other Central dialects: *hw- > f-, secondary hw > f in fīn “blood.” An interesting morphophonological alternation is that of h and f in be-hos- (subjunctive) “sleep” (< *hufs-) versus faft- “slept” (< *xwaft) and ho “sleep, dream” (exact origin uncertain). Initial *- becomes g- as in Baškardi and Baluchi; inter­vocalic š > `; initial *x- > k- (e.g., kerūs “rooster,” cf. Pers. xorūs), as also in Kurdish and Baluchi; as well as four Perside developments: *θr > s, initial *dw- > d-, initial *y > j, and retention of *ft (after low vowel). A typical Farvī development is, apparently, that of ō > wa, e.g., dwahd “daughter” and šwahš “six,” Ḵūri dōd, šōš (P. O. Skjærvø, personal communication). The far-deictic pronoun ev “that, he/she” (< *awa-) and the near-deictic em “this” (< *ima-) group Ḵūri with Old Persian and Kurdish, which has av- and am- (Solay­mānīya, Warmāwa; other dialects aw-, with typical Kurdish change of intervocalic m > w/v). The imperfec­tive markers de-/ti- are found in most Kurdish dialects. The 3rd singular copula is -en as in Sīvandi and the Lori and Fārs dialects.

Typical Ḵūri lexical items include the following: different present and past stems of “to say”: d-uš-ān/got-a-m “I say/said” (also in Kurdish); heyg (cf. Sīvandi xui, Fārs dialects xāg, Baluchi hāyk, Kurdish hāk “egg”); ayer “fire” (cf. Sīvandi uir and Kurdish agir); nāg “nose”; nenang “hen,” Farvī niyang.

Table 29 contains a selective list of isoglosses intend­ed to indicate the diversity of the area (T = Āštīāni; Tafreši; M = Maḥallāti to Ḵᵛānsāri; K = Kāšāni dia­lects; I = Isfahani dialects; N = Nāʾīni and the Beh­dīnān dialect of Yazd and Kermān; K = Ḵūri; S = Sīvandi; for more detail, see below). They are the prefix *fra- in the verb “to sell” (Mid. Iran. frōš-frōxt-); the perfective prefix; the imperfective prefix or affix; the ending of the 1st singular present, etc.; the pronoun “we”; the near demonstrative “this”; the verb “to become”; the adverb “now”; and the adjective “big.”

Dialectology of the Central dialects.

1. Phonology.

Vowels. Most of the dialects have not yet been accurately recorded, and the material for a phonemic analysis of the vocalic systems is insufficient and permits only tentative conclusions about distribution patterns and historical developments in the Central dialects as a whole, such as the retention of the long mid vowels ē and ō, raising of low vowels (e.g., e > i), and fronting of back vowels (e.g., a > ä, e).

Consonants. There is considerable homogeneity in the development of the consonants (see Table 30). Among the most significant and oldest northwestern features are the following:

IE. *k’ > s, e.g., Abčūyaʾi massa “big” (cf. Pers. meh “bigger”), Ḵūri āsk “gazelle” (cf. Pers. āhū); IE. *k’w > sp, Qohrūdi, Sīvandi espa “dog” (cf. Pers. sag); *> z, e.g., zōn-/zūn- “to know” (common; cf. Pers. dānestan); Kāšāni, etc., (h)eze, Sīvandi zī-rā “yester­day” (cf. Pers. dī-rūz); OIran. *θr > (h)r, e.g., Naṭanzi pur “son, boy” (cf. Pers. pesar); initial *dw > b, e.g., Naṭanzi bar “door” (cf. dar = Pers.); initial *y > y, e.g., Ārāni yo “barley” (cf. Pers. jow). Ḵūri does not share the developments of *θr > (h)r, initial *dw > b, initial *y > y (cf. Ḵūri pos “son,” dar, jow).

Some later changes, as well, mainly involving clus­ters, are shared by many or most Central dialects: *rz > l, e.g., Vānīšāni al “to let, put” (Pers., Fārs dialects hel-), Sīvandi mol “neck” (also Fārs dialects), Ḵūri gol “flower” (with g- < w-; cf. Pers. gol); xt > (h)t, e.g., Farīzandi dota, Sīvandi det “girl, daughter,” Ḵūri rēda “dropped” (< *rēxt). Other changes are spread widely, *xr mostly > (h)r, e.g., Ārāni hrin “to buy,” Sīvandi sir “red” (Pers. sorḵ), Ḵūri -hrīn- “to buy”; *xm mostly to (h)m, e.g., Ārāni hmar- “to break.”

There is considerable inconsistency, or “turbulence,” in other developments, many of which involve changes that have not affected all lexical items. In most instances there are distinct, areally defined, patterns, however. The following is a selective list of such changes:

Old palatals: As mentioned, *g’ > z throughout, e.g., zūn- “to know,” but > h in bohu/boʾi “arm” in a small area from Qohrūd to Gaz and Zefra, in Vānīšāni, Nāʾīnī, Anāraki, Ḵūri, Sīvandi, versus bazi in neighbor­ing Ḵᵛānsāri.

Intervocalic Old Iranian *č > j or ` throughout, e.g., vāj-/vāž- “to say,” with the exception of Sīvandi, where it becomes š (e.g., paš- “cook, bake” < *pač-); Old Iran. *j mostly > j or `, e.g., Naṭanzi jan “woman,” Abūzayd­ābādi žen(g), Sīvandi, Ḵūri žen, `inū, Farvī jen (cf. Nāʾīni, Anāraki enjū); but > y in yan “woman, wife” in Ārāni, Qohrūdi and Sohi, and Persian zan in Maḥallāti, Vānīšāni, and Ḵᵛānsāri.

Initial *w > v in vā(d) “wind” in most locations; in vī(d) “willow” in the north, but Persian bīd in the Isfahan and Nāʾīn areas; varg “wolf” in the northern and southeastern Kāšān area, but Persian gorg else­where. Only Ḵūri and Farvīgī have g- (e.g., gā(d) “wind”; -gof- “weave,” cf. Pers. bāf-), a development it shares with some Baškardi dialects and Baluchi (where *w- > g- or gw-).

The development of intervocalic and final d < OIran. *t is fairly regular: It remains as d in many words, forming an isoglottic line from Ḵᵛānsār, Jowšaqān, Abyāna, Bād, but changes to a glide, mostly > y, to the south of this area. It is lost in the past participle/infinitive in all dialects, e.g., Farīzandi parsāa “asked,” with the exception of Qohrūdi persoda. Note also Nāʾīni, Anāraki biyār “brother.”

Intervocalic š is voiced in Ḵūri, e.g., gūž “ear.”

Initial *hw- becomes f- in Ḵūri (e.g., de-fer-ān, be-fard-am “I eat, I ate,” cf. Pers. mī-ḵᵛoram, ḵᵛordam; far “sun,” fār “sister,” cf. Pers. ḵᵛor-šīd, ḵᵛāhar) and Sīvandi (e.g., fet “slept” < Mid. Iran. *xwaft, Pers. ḵoft; fey- “self, own” < *xwad; fešk “dry”< *hušk; fird “small” < *xwurd; farm “sleep” < Mid. Ir. *xwamn/xwarm). This development must have taken place fairly late in the history of Sīvandi, as it also affects secondary initial hw-/wh- (e.g., fīt “sifted” < *whīt < *wīht < *wēxt; fīn “blood” < OIran. *wahuni-; feše “hungry” < *wišna < *wṛsna-). Intervocalic *hw becomes -w/-u or is lost (e.g., muari “I am eating” < *ma-xwar-, contrasting with past tense fard < *xwart; and present mas- “sleep” < *ma-hwas- < *hwafs-, versus past tense fet). In Ḵūri secondary hw became f in fīn “blood”; differently from Sīvandi, f is retained after prefixes (see the examples above).

Clusters. š is lost in the groups šm and šn in čašm “eye” > čam “eye” in the middle area from Maḥallāt to Zefra, and in a few other words in individual dialects, e.g., Kešaʾi pōina “heel” and Abyānaʾi enoyn- “to hear” and tena “thirsty” (usually pāšna, ešnav-, and tešna or similar forms). A similar development is found in Baškardi, Kumzari, and Baluchi (see EIr. III/6, p. 635, III/8, p. 848).

Old Iranian št remains throughout the Kāšān area but becomes st and ss elsewhere, e.g., Naṭanzi bavašt ­“ran” but Gazi bevesse-.

Old Iranian ft remains or becomes (h)t, depending on the words and area. It becomes t in the Kāšān dialects in most words, but generally ft in the Isfahan dialects, e.g., Qohrūdi derkat- “fell,” Sīvandi fet “slept” (cf. Pers. ḵoft), Ḵūri got- “said” (cf. Pers. goft). But Gazi derkaft-, Ḵūri -geraft- “took” (< *grift).

Initial *fr- mostly becomes (h)r-, e.g., Ārāni rūd “sold,” Sīvandi rut, Anāraki -hrāt, Ḵūri -raveš- (cf. Pers, forūš-, forūḵt), but Perside fr is found in this word in the Isfahan area up to Zefra, e.g., Zefraʾi frōš-; in *frada- “tomorrow” fr- becomes h- in the northern Kāšān area, e.g., Qohrūdi hiyō, but Persian fardā (beside other words) is found elsewhere. Exceptionally hr has become xr in Kešaʾi xrasn- “to send” (< *frēst-ēn-, cf. Pers. ferest-ād) and Qohrūdi a-xrūš- “to sell.”

2. Noun morphology and syntax.

Gender. Recent research (Yarshater, 1985) has shown that the retention of the distinction of feminine gender is much more widespread than assumed so far, especially in the north, that is, the Tafreš (as in the neighboring Tati dialects) and the Kāšān area, for instance, Āštīāni and Āmoraʾi vīnīy-a “nose,” Āštīāni masc. ān “that one,” fem. āna. As Yarshater has shown, factors of syntax and semantics, as well as focus of discourse, produce considerable differences in the marking of gender in these dialects (as in others where gender is retained). Such factors include the degree of agentivity (most common with human agents, less with other animate agents, and least with inanimate and generic agents), definiteness (most common with def­inite words, less with indefinite ones), and number (most common with singular words, less with plural ones), as well as word class (such as, in decreasing order: noun, adjective, indefinite article, demonstrative adjec­tive, demonstrative pronoun, past intransitive verbs, past transitive verbs, and the copula; see also cases). For example, in Jowšaqāni the feminine gender is distinguished in proper names, nouns, demonstrative pronouns, the copula, the past of intransitive and transitive verbs, and, uniquely among the Kāšān dialects, in proper names and verbal forms based on the present stem. Examples: Hasan-e vagertā versus Zeynab-e vagertā-a “Ḥasan/Zaynāb returned”; nen quč-­a quč-a xubi-a “this ram is a good ram” versus nen böz-e böz-e xubi-asta “this goat is a good goat,” aga nö bavöz-­e versus aga nön-a bavöz-ea “if he/she runs.” In Sīvandi gender is marked in definite eżāfa-constructions, e.g., kor-i me “my son,” žen-a me “my wife,” and before suffix, e.g., žen-a-š “his wife”; the deictic suffix is stressed -ū for masculines, e.g., quč-ū “the ram,” but unstressed -e for feminines, e.g., usúr-e “the horse.”

Oblique case. As in the case of gender distinctions, the Tafreš and northern Kāšān dialects tend to retain the distinction between direct and oblique cases. For example, in Āštīāni the following case endings are distinguished:

  Singular Plural
  Masculine Feminine  
direct (no ending) -a -gal
oblique -i -o -gal-ān

In this and other dialects the oblique case functions as possessor, specific direct and indirect object, as well as agent in the past tenses of transitive verbs, for example: Rostam-e sang-eš biyānd “Rostam (agent) threw a stone.”

In Āštīāni, as well as in Āmoraʾi and Kahaki, the case distinction is also retained in certain pronouns (Table 31).

The same distinction has been retained in Abyānaʾi. Abūzaydābādi has reflexes of the singular oblique -i in eżāfa-constructions of nouns with final stressed -á, for instance, bar key-é < bar keyá-i/e “the door of the house.”

Plural. The plural marker is generally derived from the Middle Iranian plural oblique marker *-ān or from the Middle Iranian plural (originally abstract?) ending *-īhā. The Lori-type marker -gal is found in Āštīāni and Āmoraʾi among the Tafreš dialects (but Kahaki -iya), -gel-o in Zefraʾi among the Kāšān dialects, and -gar in Sīvandi. A reflex of the former plural direct case marker -e/-i is found in Abūzaydābādi, where nouns in final stressed -a have -e (but not elsewhere), for instance, pāk lūwé < lūwá-i/e “all intestines.” The unmarked form (< Middle Iranian plural direct case or generic singular) is also often used for the plural, for instance, böz xodō šāx-yō jang a-kerän “the goats fight with their horns.”

Definiteness. Most dialects appear to have a vocalic marker similar to colloquial Persian topicalizing -é, for instance, Judeo-Isfahani kuwen-é “the kohen.” Ḵūri has -ū, e.g., mardū, `inū, and Sīvandi has masculine -ū, feminine -e. All dialects have an indefinite marker -i/e.

Eżāfa construction. The word order is the same as in Persian and other dialects in southern Persia, that is, noun + adjective and possessed + possessor. Although the Persian-type connective -i, -e (eżāfa) occurs in several of the recorded dialects, too little is known about its distribution to determine whether it is a genuine part of their morphology. In Abūzaydābādi the possessor is in the oblique case (“genitive”), e.g., bar kēy-é (see above; there is a similar construction in Abyānaʾi), which was probably the original pattern in all the Central dialects. Both Abūzaydābādi and Abyānaʾi, possibly also several other dialects, have a compound-noun construction as well, in which -a- ­serves as connector, e.g., Abyānaʾi dot-a-xāla “daugh­ter of maternal aunt.”

Adpositions. The Central dialects have prepositions and postpositions, as well as ambipositions. Stilo (1985) has shown that the Central dialects, together with Northern and Central Kurdish, Gōrānī, and the southernmost Tati dialects, lie in a buffer zone between the languages in the north, which consistently use post­positions, and those in the south, which consistently use prepositions. In particular, in many dialects the position of de (< dar) determines its function: as a preposition it marks direction, as a postposition it marks location or ablative; for instance, Meymaʾi de-kul-eš-eš na “he put it on his shoulder” versus man šomā-de betarsām “I was afraid of you” (cf. Pers. man az šomā mītarsīdam). In some Central dialects be, which generally marks the indirect object, has assumed the same functions as de in the other dialects (see also on affixes below).

Direct object. The Persian-type marker -(r)ā occurs in many dialects, but with varying frequency, and too little is known about its use, especially in past transitive sentences, to identify its range of functions. It appears that at least in some dialects it is not used if the noun phrase is already defined by a personal suffix.

Independent pronouns. In the Central dialects per­sonal and demonstrative pronouns have only one case form (with the exception of the 1st and 2nd singular in Āštīāni, Āmoraʾi, and Kahaki, as well as in Abyānaʾi, see above). When unmarked by adpositions the pro­nouns function as subject and as indirect object with verbs semantically implying animate/human recipients, for instance, Zefraʾi īčkī ū-š n-īˊe-tō “nobody (īčkī . . . -š n-) would give (īˊe-tō) him (ū) (anything)” (-š marks 3rd singular past agent).

Personal affixes (enclitic pronouns). All central dia­lects have personal affixes whose function it is to mark the possessor, the object of prepositions, the indirect object, the direct object in the present, and the agent in the past tenses of (mostly) intransitive verbs; examples: Judeo-Isfahani dim-aš “his face,” īn harfā-rā ve-šūn vāt “(he) said these words to them,” Zefraʾi em­-et-gō “I must” (em- “for me,” et-gō “it is necessary”), Nāʾīni kār-i xūb-ī-š-om vé-ārt “I (-om, agent)” did (lit. brought about) something good for him (-š, indirect object).”

There are several major isoglottic splits in the pro­nouns. The personal affix of the 3rd person singular is either -i/-e (from OIran. *-hai, him, etc.; cf. Av. hōi/) or -š (from OIran. šai, šim, etc.; cf. Av. šē, OPers. šaiy). The form -i/-e is found in the northern part of the Kāšān area, including Ārāni, Abūzaydābādi, Abyānaʾi, but also in Sedehi among the Isfahan dialects, and in Ḵūri. The 3rd plural forms are generally formed from the singular forms by adding the oblique plural marker *-ān, for instance, Abūzaydābādi singular -e, plural -- ōn. In Nāʾīni, Anāraki syncopated forms were recorded by Ivanow: -m’n-, -t’n-, -š’n-, which have now been reduced to -m-, -t-, -š-, coinciding with the singular forms. The two isoglosses intersect in the Kāšān area, where both Ārāni and Abyānaʾi use -i/-e for possessor, perfective past agent, and present direct object, but š-e for the indirect object and the agent in the imperfective past, for instance, Ārāni š-e-dā “he gave him,” Abyānaʾi š-e-gī “he wants, he must” (lit. “for him it is necessary”). Pronouns derived from *hai, etc., are also typi­cal of Kurdish, Baškardi, and Baluchi; pronouns from *šai, etc., are typical of Perside dialects, as well as Caspian dialects, Semnāni, and so on.

Another major isoglottic split is found in the demonstratives. In general the demonstratives can be derived from Middle Iranian forms such as Middle Persian ēn and ān; only in a wedge-like isogloss in the northern Kāšān dialects, the boundary of which stretches from Jowšaqān south to Soh, Naṭanz, and north to Bād, excluding Meyma and Keša, do we find forms with initial n- used as demonstratives, as in the southeastern-most Southern Tati dialects (Eštehārdi) and Sangesari in the Semnān region. A near-deictic form (“this”) in d- (< OIran. *aita-) is found in Nāʾīni, Anāraki, etc., and a similar plural demonstrative inn is found in Kešaʾi—just south of the n-isogloss—and Vārāni in the northwest of the central area. Forms derived from OIran. *ima- are found only in Zoroastrian Yazdi (m-) and in Ḵūri (em “this”). The far-deictic pronoun in Ḵūri is ev < OIran. *awa-.

Several Central dialects, among them Zefraʾi and Nāʾīni, appear to distinguish three sets of demonstrative pronouns (“hic, iste, ille”), but this question has not yet been fully examined. Table 32 shows forms from Vānīšāni, Zefraʾi, Abyānaʾi (including the oblique forms and the prefixes of the imperfect, see below), Nāʾīni, Ḵūri, and Sīvandi. The distinction between the 3rd person pronouns and the two sets of demonstratives is semantically not always clear.

3. Verbal morphology and syntax.

Prefixes. All the Central dialects have prefixes that largely retain their original directional meaning, for instance, *var- “on, out,” *dar- “in,” and *- “to, from” (< OIran. *frā- “forth, away”). The most elaborate system appears to be that of Abyānaʾi (Lecoq, 1975, pp. 57-58): ar- “upward” (e.g., ar-geratan “to pick up”), ber- “downward” (e.g., ber-katan “to fall down”), bar- “out” (e.g., bar-ammayan “go out, leave”), var- “forward” (e.g., var-ammayan “to ad­vance, come forth”), dar- “in,” also intensive action (e.g., dar-petan “to wrap (in)”), - “back” and repeti­tive and “enlarging” action (e.g., vā-xardan “to drink”), and - with no clearly definable meaning (e.g., hā-­geretan “to take”).

Stem gradation (umlaut). A reflex of the Old Iranian stem gradation is found in the variation between intransitive and transitive stems of *čar-/čār-(ēn)- “to graze” (intrans./trans.) in many of the dialects, e.g., Vānīšāni čer- versus čōr-n-.

Personal endings. There are three major variants of personal endings. The ending of the 1st singular is -i/-e in the Nāʾīni, dialects and Sīvandi, as opposed to -ān > -ōn/-ūn in the remainder of the dialects, except Ḵūri, which has -ām(-ã, -ān). The ending of the 3rd singular is -e/-a (from Mid. Iran. *-ad/-ēd < OIran. -at-i/-ayat-i) in most dialects, but Āštīāni among the Tafreš, and the southern dialects, including Vānīšāni, Ḵᵛānsāri, and the Isfahani dialects, have a rounded back vowel -ū/-ō of uncertain origin (possible analogy from forms like *b(av)am/* “I/he will be” < *bawam/bawad, *š(av)am/*šō “I go/he goes” < *šawam/šawad; the form may also have been influenced by the 3rd person singular demonstrative ū, which typologically is not unusual). Most varied are the endings of the 2nd plural: Āštīāni, Kahaki, -īn, Āmoraʾi -ite; the dental is also found in Sedehi -ide, Meymaʾi ida, Ḵᵛānsāri -adi; -g- (of unknown origin): Vānīšāni, Qohrūdi, Kešaʾi, Sīvandi -ige/-ike; or -iya (with regular change of intervocalic d > y).

The verb “to be.” As in most Iranian dialects the enclitic copula agrees with the intransitive present endings. However, there is considerable diversity in the 3rd person. Most notable are the forms found in marginal areas, for example, in Sīvandi: (h)änd-ām, -ī, -0, -īme, -īke, -0 or -īne (note that the endings -ām and -0 are those of the past tenses, as in Pers. hast-am, hast-ī, hast). Ḵūri has 3rd sing. enclitic copula -en (e.g., . . . johun-i-e[n] “it is a nice . . .”) versus existential ha “(there) is, exists.”

The existential/locational verb “to be” also shows great variety, including differentiation as to animacy. For example, Vānīšāni has -ū “is,” identical with the verbal ending, but also bissū “there is” (probably related to, or influenced by, *st- “to stand”) and animate ōyū. In Nāʾīni the marker of existence or location is (e.g., kiä vä ādamīzād? dī vä ādamīzād “Where is the human? This is the human.” Farīzandi has a fairly elaborate system, distinguishing between interrogative and declarative, as well as gender and number, for example, ko šo/štä/štändä “where is he/she/they?” (with šo, etc., probably related to, or in­fluenced by, š- “to go”) versus an-de darä/darea/darända “he/she/they are (in) there” (with dar < *dar “in”).

The basic patterns of the verbal endings are illus­trated in Table 33.

Inflectional passive/intransitive. Most Kāšāni dialects have retained passive/intransitive forms, usually formed by adding -y- to the present stem (from Mid. Iran. -ī(h)- < OIran. -ya-; also found in some Lori-type dialects), but to considerably various degrees; Ḵᵛānsāri and Abyānaʾi have forms in -k/g-. Examples: Meymaʾi be-š-pič-i-y-e “it will (be) cook(ed),” Vārāni be-ranj-i­-y-a “he suffered,” Judeo-Kāšāni vā-darz-i-y-a “it was sewn,” Judeo-Hamadāni na-hmar-i-y-a “it is not (being) broken,” Abūzaydābādi a-kar-i-y-o “it is (being) done” and m-a-kar-i-y-o “it was (being) done,” Sīvandi eškiy- ­“to break.” Only one form, ba-hmar-i-y-a, has been recorded for Kamūʾi in the center of the Kāšāni dialect area and Ārāni to the north of Kāšān. Ḵᵛānsāri (e.g., et-kuš-k-ān “I am being killed”) and Abyānaʾi (e.g., e-kar-g-ān and ba-kar-g-ōy-ān “I am being made/I was made”).

In the other dialects the passive is formed by means of periphrastic constructions similar to Persian košta šodan “to be killed.” Different auxiliaries are employed, however: gert- (in the east: Nohūji, Keyjani, Nāʾīni, Abčūyaʾi, Anāraki), gel- (Farīzandi, Bādi), gan­- (Maḥallāti, Vānīšāni, Ḵᵛānsāri, Qohrūdi) and gen­- (Sīvandi), and the rest b- “be, become.”

Present and past imperfective. The formation of the imperfective present and past shows two distinct iso­glottic patterns, by the prefix *at- (of uncertain origin, perhaps from OIran. *aiwa-da “at the same time, all the time”(?); also found in the northwestern Central and southernmost Tati dialects, such as Vafsi) or the enclitic particle -e. The prefix *at-, which appears as ed- in Tafreš (e.g., Āštīāni ed-gīro “he takes,” Āmoraʾi ed-ām [< *ed-vām] “I say”; but Kahaki to-, e.g., to-vāje “he says”), as et- in Ḵᵛānsāri (e.g., et-kuš-ān “I kill”), as a-, prevocalic (a)t-, in the Kāšān dialects (e. g., Meymaʾi yā t-ār-on “I remember,” lit. “I bring to place/here”) and sometimes contracted with following vowels (e.g., Vārāni y-owr-a “he brings,” but t-ey “he comes”); in the southeastern Kāšāni and the Nāʾīni dialects it is preserved only after verbal prefixes and verbal nominals, e.g., Zefraʾi šūn “I am going,” but -ī- ­in dir-īe-bend-ōn “I am hitting,” gird ī-ker-ōn “I gather”; Anāraki mu e-wīn-ī “I see.” In Ḵūri the imperfect marker is de- (e.g., de-kar-ām, de-kard-a-m “I am/was doing”), but ti before vowels (e.g., tiām, ti-avō-h-ām “I was coming”). Sīvandi has ma-, as in the Fārs dialects.

The enclitic particle -e (of uncertain origin; perhaps from *ēw as in Class. Pers. -ī and Pers. (ha) < Mid. Pers. ham-ēw, or from Mid. Iran. 3rd sing. optative of “to be”: *) is found in the Isfahani dialects (e.g., Gazi šōn-e “I am going”). In the Ḵᵛānsāri and Vānīšāni area it is used together with the marker et- as an optional -e/-i (cf. Class. Pers. hamē raft-ī “he used to go”). Note also the Āmoraʾi plural endings -īm-i, -īt-e, -en-e, and Āštīāni 3rd plur. -on-e. The Zoroastrian dialect of Yazd and Kermān has the prefix et- in the positive, but the enclitic -e with negative (e.g., et-ō-t “comes,” but n-ō-t-e “he does not come”; See behdīnān dialect).

Perfective and subjunctive. The general marker of the perfective tenses, including the perfective subjunctive, is ba-/be- (e.g., Kafrāni be-rit “dropped”), except in stative verbs (see below). The present and past perfect, as well as the perfect subjunctive, are generally based on a perfective participle ending in -a, followed by the present, past, and subjunctive of “to be” (as in Persian rafta ast, rafta būd, rafta bāšad). Forms without be- are used throughout the Central dialects area to express resultative aspect, for instance, Ḵūri be-dia-g-e “he has seen” (-e = agent) versus em eškassa-g-en “this is broken” (-en = 3rd sing. of the existential/locational “to be”; these two examples also exemplify the common use in Ḵūri of -g- as intervocalic connector).

With transitive verbs the agent is usually marked by the personal affix, inserted after ba-/be-, for example, be-š-vāt(-a) “he (has) said” (-š- = agent). The major exception is Sīvandi, in which *be- has been lost and the agent affix is inserted after the main verb, before the auxiliary, for instance, gort-em-en “I have taken”

(-em = agent, -en = 3rd sing. copula), following the pattern of the neighboring Fārs dialects. In several of the Nāʾīni dialects be- has been reduced to a front vowel (e.g., Nāʾīni mī-m-i-ka and Anāraki mu-m-i-ka “did” < *-bi-ka).

Stative and resultative verbs. Most dialects contain a small set of “stative” verbs, including “to have” and “to be able, can,” which lack both the imperfective and perfective markers and from the subjunctive by means of a periphrastic construction (e.g., Kešaʾi dōrt būn “may, should have” and zōm būn “may, should be able to”). Abūzaydābādi has a set of defective resultative verbs (e.g., present āvo, past avade “he is/was seated,” but čīn/češt “to sit [down]”), for which corresponding verbs may possibly be found in other dialects.

Future. The future tense is marked by *kām- (origi­nally “to wish” or “it is my, etc., wish”) in an area about Kāšān including Ārāni and Bīdgoli, Abūzaydābādi, Qohrūdi, Abyānaʾi, Farīzandi and Yārandi, and Jowšaqāni (e.g., Abyānaʾi kam-om šo “I will go”). This marker also is used in conditional clauses. Abyā­naʾi in addition has a separate near future, marked by the combination of the subjunctive/perfective prefix with the imperfective prefix (e.g., b-e-kar-ān “I will do (soon),” but e-kar-ān “I am doing” and ba-kar-ān “[that] I should do”) and a future imperative marked by -- (e.g., ba-kar-iš-ä “you will do!” as opposed to the regular ba-ka “do!”).

Narrative tenses. Abūzaydābādi distinguishes what may be called historical present forms, that is, a narrative imperfect and past in some intransitive verbs. These forms are based on the present subjunctive and imperfect, respectively, and are marked by the lack of vocalic harmony: narrative imperfect á-šo “he was going (then),” but present o-šó “he is going,” narrative bá-šo “he went (then),” but subjunctive bo-šó “that he go, he may go.” Jowšaqāni sabah b-at-am “tomorrow I will come” and b-at-eme-am, probably “I would come,” may be examples of the same constructions, and other instances may yet be found in other dialects.

Ergative. Like most West Iranian non-Persian-type dialects, all the Central dialects have retained a form of the so-called “split” ergative, traditionally called “passive construction” (by some also termed “agential construction”), in which the agent, or logical subject, in the past of transitive verbs is marked by the oblique case of nouns and/or pronouns in dialects where the oblique is still distinguished and/or by pronominal affixes (e.g., Kešaʾi čemō-d-em kö be-dī . . . “when I saw your eyes . . .,” lit. “eyes-your-I when saw”; this example also demonstrates the “raising” of a personal affix, here the agent, from the dependent relative clause into the main clause: čemō-d be-m-dī > čemō-d-em kö be-dī).

The patient, or “logical,” object is in the direct case in dialects which have preserved case distinction, and the verb agrees with the patient, rather than the agent, for instance, Abyānaʾi ba-š-köšt-ān “he killed me” (< “I was killed by him”). In Farīzandi and Yārandi agreement is retained in both feminine gender and number, for instance, dot-am be-šī-ä “I took the girl” (lit. “girl-I took-her”), while in Abūzaydābādi agree­ment is confined to the singular (e.g., ī xarguš-am be-­köšt-e “I killed a rabbit,” lit. “one rabbit-I killed-her”).

4. Lexicon.

There is considerable homogeneity in the lexicon of the Central dialects. They all have typically north­western verb stems like *vāj-/vāt- “to say” (Pers. -/goft), and *k(af)-/ka(f)t “to fall” (Pers. oft-/oftād). Other lexical items, however, show isoglottic distri­bution. For example, “to hit” is *der-band- in the northwest, but *xus- in the southeast, with a turbulent boundary area along the line from Gaz to Naṭanz. Roughly the same boundary area is found for the distinction between northwestern *čīn-/šīn- and south­eastern *ni(n)g- “to sit.” Of nouns, *isba “dog” is found in the northwestern area, including the Ḵᵛānsāri dialects, Qohrūdi, and Abyānaʾi but *kuya elsewhere (except Yazdi svaka). The distribution of “very” shows a fairly similar pattern, *xeylī in the north and north­west, *mālī in the southeastern-most Kāšāni dialects, the eastern Isfahani dialects, and the Nāʾīni dialects (but Yazdi xeyli). The word for “now” is *hat in the northwest, largely following the same patterns, and *zo­- in the southwest, whereas the Nāʾīni dialects have *osm­- (but Yazdi mene). The words for “big, great” are *mas in the Ḵᵛānsāri dialects in the west and the Nāʾīni dialects in the east, but *gord in the northern Kāšāni dialects down to Meymaʾi and Kešaʾi, and *bela in the southeastern Kāšāni dialects and in the Isfahani dialects. In roughly the same area as *bela the word for “finger” is *engul-, versus *angušt- elsewhere (but also engul- in Naṭanzi to the north of this area). Other words show even more variety, such as “to send”: Qohrūdi kīn, Kešaʾi xrasn-, Zefraʾi endōr, Yazdi nv-; “nose”: Vānīšāni, Qohrūdi, Sedehi, etc., domāḡ, domōḡ, Nāʾīni lüṇč, Anāraki loṇč, Ḵūri nāg, Sīvandi pet; “mouth”: Vānīšāni dān, Qohrūdi dohūn, Nāʾīni dehen, Anāraki āī, Kešaʾi and Zefraʾi eyn, Sedehi āīn, Sīvandi kap (also Fārs dialects, impolite usage).

There is also some semantic differentiation, notably Farīzandi-Yārandi animate šīn-/šī- versus inanimate bar-/bard- “to take away,” animate ānī-/ūnī versus inanimate ār-/ārd “to bring.” In Zoroastrian Yazdi-Kermāni there are some distinctions between “ahuric” and “daevic” verbs, for instance, ahuric hem-t-ōšt ­versus daevic hem-par- “to stand up,” reflecting the lexical split already found in Avestan.



General: F. C. Andreas, Iranische Dialektaufzeichnungen aus dem Nachlass von F. C. Andreas, ed. K. Barr, W. B. Henning, and A. Christensen, Abh. der Gesellschaft der Wissenschaften zu Göttingen, phil.-hist. Kl., 3 Folge, Nr. 11, 1939.

H. W. Bailey, “Persia. II. Language and Dialects,” in EI1 III, pp. 1050-58.

A. Christensen, Contributions. D. I. Edel’man, Osnovnye voprosy lingvisticheskoĭ geografii. Na materiale indoiranskikh yazykov, Mos­cow, 1968.

W. Geiger, “Kleinere Dialekte und Dialektgruppen. III. Central Dialekte,” in Grundriss I/2, pp. 381-406.

K. Krahnke, Linguistic Relationships in Central Iran, Ph.D. dissertation, University of Michigan, 1976.

P. Lecoq, “Les dialectes du centre de l’Iran,” in Schmitt, ed., pp. 313-26.

D. N. MacKenzie, “The Origins of Kurdish,” TPS, 1961, pp. 68-86.

M.-R. Majīdī, Gūyešhā-ye pīrāmūn-e Kāšān wa Maḥallāt, Entešārāt-e Farhang-e Zabān-e Īrān 12, Tehran, 1354 Š./1975.

O. Mann, Kurdisch-persische Forschungen, pt. 3, vol. 1, 1926.

G. Morgenstierne, “Neu-iranische Sprachen,” in HO I/1V, Iranistik I, 1958, pp. 155-78.

Idem, “Feminine Nouns in -a in Western Iranian Dialects,” in W. B. Henning and E. Yarshater, eds., A Locust’s Leg. Studies in Honour of S. H. Taqizadeh, London, 1962, pp. 203-08.

V. S. Rastorgueva, ed., Opyt istoriko-tipologicheskogo issledovaniya iranskikh yazykov I-­II, Moscow, 1975.

G. Redard, “Notes de dialectologie iranienne, II: Camelina,” in Redard, ed., Indo-Iranica. Mélanges présentés à Georg Morgenstierne, Wiesbaden, 1964, pp. 155-62.

Idem, “Other Iranian Languages,” in Current Trends in Linguistics 6, 1970, pp. 97-135.

R. Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989.

D. Stilo, “Ambi­positions as an Areal Response. The Case-Study of the Iranian Zone,” in Seventh South Asian Languages Analysis Roundtable, Ann Arbor, 1985, Ann Arbor, 1987, pp. 308-36.

Tedesco, “Dialektologie.” G. L. Windfuhr, “Isoglosses. A Sketch on Persians and Parthians, Kurds and Medes,” in Monumentum H. S. Nyberg II, Acta Iranica 5, Tehran and Liège, 1975, pp. 457-72.

Idem, “New West Iranian,” in Schmitt, ed., pp. 251-62.

E. Yarshater, “The Jewish Com­munities of Persia and their Dialects,” in Ph. Gignoux and A. Tafazzoli, eds., Mélanges Jean de Menasce, Louvain, 1974, pp. 455-66.

Idem, “Distinc­tion of Grammatical Gender in the Dialects of Kashan Province and the Adjoining Areas,” in Papers in Honour of Professor Mary Boyce II, Acta Iranica 25, 1985, pp. 727-45.

Zhukovskiĭ, Materialy, 1888-1922.

Idem, “Prevaritel’nye zametki o nekotorykh "persidskikh nerechiyakh",” Zapiski vos­tochnago otdeleniya . . . 1, 1887, pp. 23-29.

Studies on individual dialects: Abčūyaʾi: Krahnke.—Abūzaydābādi: Krahnke. P. Lecoq, “Le dialecte d’Abu Zeyd Abad,” in Monumentum H. S. Nyberg II, Acta Iranica 5, Tehran and Liège, 1975, pp. 15-38.

E. Yarshater, in EIr. I/4, pp. 401-02.

—Abyānaʾi: ʿA. Bolūkbāšī, “Barrasīhā-ye lahja-ye Abyāna, M.A. thesis, Tehran, 1349 Š./1970.

P. Lecoq, “Le dialecte d’Abyāna,” Studia Iranica 3, 1975, pp. 21-63.

E. Yarshater in EIr. I/4, pp. 404-05.

—Āmoraʾi: M. Moḡdam, Gūyešhā-­ye Vafs o Āštīān o Tafreš, Īrān-kūda 11, 1318 Yazdegerdi/1949.

—Anāraki: V. Ivanov, “Two Dia­lects Spoken in the Central Persian Desert,” JASB, N.S. 21, 1926, pp. 405-31.

Idem, “Notes on the Dialect of Khur and Mihrijan,” Acta Orientalia 8, 1927, pp. 45-61.

H. Kanus-Credé, “Notizen zum Dialekt von Anārak,” Iranistische Mitteilungen 5, 1971, pp. 10-22.

G. L. Windfuhr, “Anāraki” in EIr. II/1, pp. 2-3.

—Ārāni and Bīdgoli: N. Ḏokāʾī Bay­żāʾī, “Zabān-e rāyejī-e Ārānī,” in Majmūʿa-ye ḵeṭā-bahā-ye taḥqīqātī-e īrānī II, 1351 Š./1972, pp. 342-45.

E. Yarshater, “Bīdgol and bīdgoli,” in EIr. IV/3, pp. 247-49.

Idem, “The Dialects of Ārān and Bīdgol,” in Études irano-aryennes offertes à Gilbert Lazard, Paris, 1989.

—Ardestāni: W. Bailey, “Mod­ern Western Iranian: Infinitives in Gazi and Soi,” TPS, 1935, pp. 73-74. Yarshater, 1985.

—Arisman: Krahnke.

—Āštīāni: Ṣ. Kīā, Gūyeš-e Āštīān I: Vāža-­nāma, Tehran, 1335 Š./1956. Moḡdam.

Bādi: Krahnke.—Bāḡoli: A. Ṭabāʾīzāda, Gūyeš-e Bāḡol, M.A. thesis, University of Tehran, 1348 Š./1969.

—Barzūki: Majīdī.—Bīdgoli, see Ārāni.—Bījagāni: Majīdī.—Borūjerdi, Jewish: Yarshater, 1974.

Delījāni: Majīdī. Yarshater, 1985.

Farīzandi: Christensen, Contributions.—Farvī (Farvīgī), see Ḵūri.

Gazi: W. Bailey, “Modern Western Iranian: Infini­tives in Gazi and Soi,” TPS, 1935, pp. 73-74.

W. Eilers and U. Schapka, Westiranische Mundarten aus der Sammlung W. Eilers II: Die Mundart von Gäz, 2 vols., Wiesbaden, 1979.

Zhukovskiĭ, Materialy II.—Golpāyegāni, Jewish: Yarshater, 1974.

Hamadāni, old: C. Huart, “Le dialect persan de Sivend,” JA, 9th ser., 1, 1893, pp. 241-65.

—­Hamadāni, Jewish: Abrahamian (see Isfahani).

H. Sahīm, Gūyeš-e yahūdīān-e Hamadān, M.A. thesis, University of Tehran, 1353 Š./1974.

Isfahani, old: M. Adīb Ṭūsī, “Se ḡazal-e Eṣfahānī az Awḥadī Marāḡī,” NDAT 15/4, 1342 Š./1963, pp. 387-400.

A. Tafażżolī, “Eṭṭelāʿāt-ī dar bāra-ye lahja-ye pīšīn-e Eṣfahān,” Ī. Afšār, ed., Nāma-ye Mīnovī, Tehran, 1350 Š./1971, pp. 86-104.

—Isfahani, Jewish: R. Abrahamian, Dialectologie iranienne. Dialectes des Israelites de Hamadan et d’Ispahan et dialects de Baba Tahir, Paris, 1930.

D. N. MacKenzie, “Jewish Persian from Isfahan,” JRAS, 1968, pp. 68-­75. Yarshater, 1974.

Jowšaqānī: A. K. S. Lambton, Three Persian Dia­lects, London, 1938. Yarshater, 1985.

Kafrūdi (Kafrāni): Zhukovskiĭ, Materialy II.—Ḵᵛānsāri: Y. Baḵšī, Tarānahā-ye Ḵᵛānsār, Tehran, 1334 Š./1955.

W. Eilers and U. Schapka, Westiranische Mundarten aus der Sammlung W. Eilers I: Die Mundart von Chunsar, Wiesbaden, 1976.

Mann, 1926. M.-Ḥ. Tasbīḥī, Gūyeš-e Ḵᵛānsār, Rawalpindi, 1395 = 1354 Š./1975.

Zhukovskiĭ, 1888.—Kāšāni, old: M. Moḥaqqeq, “Ašʿār-ī be lahjahā-ye maḥallī,” FIZ 7, 1338 Š./1959, pp. 247-52.—Kāšāni, Jewish: Yar­shater, 1974.

Zhukovskiĭ, Materialy II.—Kermāni, Jewish: G. Lazard, “Le dialects des Juifs de Kerman,” in Monumentum Georg Morgenstierne II, Acta Iranica 22, Leiden, 1982, pp. 333-40.

Yarshater, 1974.—Kermāni, Zoroastrian: V. Ivanov, “The Gabri Dialect Spoken by the Zoroastrians of Persia,” Rivista degli Studi Orientali 16, 1934, pp. 31-97; 17, 1938, pp. 1-39; 18, 1939, pp. 1-59.

See also behdǰnān dialect.­—Kešaʾi: Zhukovskiĭ, Materialy I.—Keyjāni: Krahnke.—Kūhpāyaʾi: Krahnke.—Ḵūri, Farvī, Mehragāni: Š. ʿA. Faqīhī Ḵūrī, “Ṣarf-e zabān-e ḵūrī,” Yaḡmā 20/4, 1346 Š./1967, pp. 204-12.

B. Farahvašī, Vāža-nāma-ye ḵūrī, Tehran, 2535 = 1355 Š./1976.

R. Frye, “Report on a Trip to Iran in the Summer of 1948,” Oriens 2, 1949, pp. 204-11.

Idem, “Notes on Farvi, a Dialect of Biyabanak,” Oriens 2, 1949, pp. 212-15.

Idem, “Notes on a Trip to the Biyabanak, Seistan and Baluchistan,” Indo-Iranica 6/2, 1052, pp. 1-6 (“Safar-e Bīābānak wa Sīstān wa Balūčestān,” Dāneš 2/10-11, 1331 Š./1952, pp. 528-33).

Idem, “Taṭbīq-e lahjahā-ye ḵūrī wa balūčī,” Mehr 8/3-4, 1331 Š./1952, pp. 142-44, 217-21.

M. Ḡolāmreżāʾī, “Nokta-­ī čand dar bāra-ye lahja-ye ḵūrī,” Majmūʿa-ye ḵeṭābahā-ye taḥqīqātī-e īrānī, 1351 Š./1972, pp. 369-­75.

M. Honari, “Importance du palmier-dattier dans la vie des habitants de Xor,” Objets mondes 11, 1971, pp. 49-58.

V. Ivanov, “Two Dialects Spoken in the Central Persian Desert,” JASB, N.S. 21, 1926, pp. 405-31.

Idem, “Notes on the Dialect of Khur and Mihrijan,” Acta Orientalia 8, 1927, pp. 45-61.

Ṣ. Kīā, “Yāddāšt-ī dar bāra-ye gūyeš-e Farvīgī,” MDAT 2, 1334 Š./1954, pp. 34-41.

G. Redard, “Le palmier à Khur,” in W. B. Henning and E. Yarshater, eds., A Locust’s Leg. Studies in Honour of S. H. Taqizadeh, London, 1962, pp. 213-19.

Maḥallāti: Majidi. Mann.—Mehragāni, see Ḵūri.—Meymaʾi: Lambton (see Jowšaqāni). Majīdī.

Nāʾīni: Mann, 1926. A. Querry, “Le dialecte persan de Nayin,” MSL 9, 1896, pp. 1-23.—Narāqi: Majīdī.—Našalji: Majīdī. Yarshater, 1985.—­Naṭanzi: Christensen, 1930. Mann, 1926.—Nohūji: Majīdī.

Qalhāri: Majīdī.—Qohrūdi: E. Browne, A Year Amongst the Persians, Cambridge, 1893. A. Houtum-Schindler, “Beiträge zum kurdischen Wortschatze,” ZDMG 38, 1884, pp. 43-116. Mann, 1926. Yarshater, 1985. Zhukovskiĭ, Materialy I. Idem, 1922.

Sagzī (Sajzī): Krahnke.—Sedehi: B. Farahvašī, “Taḥlīl-e sīstem-e feʿl dar lahja-ye sedehī,” MDAT 10/3, 1342 Š./1963, pp. 311-24. Mann. Zhukovskiĭ, Materialy II.—Sīvandi: Andreas. W. Eilers, Westiran­ische Mundarten aus der Sammlung W. Eilers III: Die Mundart von Sivänd, Wiesbaden, 1988. C. Huart, “Le dialect persan de Sivend,” JA, 9th ser., 1, 1893, pp. 241-65. P. Lecoq, Le dialecte de Sivand, Wiesbaden, 1979.

Mann. E. Molchanova, “Sivandi: Main Trends of Development,” paper presented at the 32nd International Congress for Asian and North African Studies (Hamburg, 1986), Moscow, 1986.

G. Morgen­stierne, “Stray Notes on Persian Dialects,” NTS 19, 1960, pp. 73-140.

Zhukovskiĭ, Materialy II.—Sohi: W. Bailey, “Modern Western Iranian: Infinitives in Gazi and Soi,” TPS, 1935, pp. 73-74.

Mann.—Sohi: Andreas. A. Houtum-Schindler, “Beiträge zum kurdischen Wortschatze,” ZDMG 38, 1884, pp. 43-116.

Ṭarqi: Krahnke.

Vānīšāni: Zhukovskiĭ, Materialy I.—Vārāni: Majīdī. R. Nīlīpūr and M.-T. Ṭayyeb, “Tawṣīf-e sāḵtmānī-e dastgāh-e feʿl-e lahja-ye Vārān,” Majalla-ye zabānšenāsī 2/1, 1364 Š./1985, pp. 51-92; 2/2, 1364 Š./1985, pp. 81-92.

Yārandi: Christensen, Contributions.—Yazdi, Jewish: A. Romaskevich, “Lar i ego dialekty,” Iranskie yazyki 1, 1945, p. 45. Yarshater, 1974.—­Yazdi, Zoroastrian (see Kermāni, Zoroastrian). See also behdīnān dialect.

Zefraʾi: Zhukovskiĭ, Materialy I.—Zori: Majīdī.

Figure 13. The extent of the area where Central dialects are spoken

Figure 14. Central dialects

Table 29. Select isoglosses

Table 30. Synopsis of major sound changes

Table 31. Case distinctions in personal pronouns

Table 32. Pronouns

Table 33. Verbal endings

(Gernot L. Windfuhr)

Originally Published: December 15, 1991

Last Updated: October 10, 2011

This article is available in print.
Vol. V, Fasc. 3, pp. 242-252