xiii. Iranian Languages
Central Asia was the ancient homeland of the Iranians and therefore also of the Iranian languages. There is no clear and indisputable evidence of the languages spoken on the plains and in the mountains of Central Asia before the 1st millennium b.c., when the entire area was probably already inhabited by Iranians. Nevertheless, some scholars have supposed that in certain Iranian languages there are recognizable traces of a substratum of pre-Iranian Central Asian languages that are typologically similar either to those of the Burushaski (q.v.) or Himalayan type or to Dravidian (e.g., Edel’man, 1980, 1983). This view would imply that, before the appearance of the Aryans (i.e., the common ancestors of the Indo-Aryans and Iranians) or Iranians in Central Asia, non-Indo-European languages were spoken there. On the other hand, it should be noted that there is a complete absence of any placenames that can be demonstrated to be of non-Iranian origin, even in the remotest mountain valleys like Wāḵān, which is immediately adjacent to Burushaski-speaking regions. Indeed, most names with clear etymologies seem to have originated in Old Iranian times, including the native form of Wāḵān itself: Wux, which is undoubtedly from Old Iranian *Wahwī- “good, beneficent,” an ancient river name (cf. Av. Vaŋᵛhī Dāitiiā, the name of a river in Airiianəm Vaējah, the mythical homeland of the Iranian tribes; differently Morgenstierne, 1938, pp. 433, 465, < *Waxšu).
Historical development. The so-called Eastern Iranian languages were and still are spoken over a wide area, from the Caucasus, where modern Ossetic is spoken, to Chinese Turkestan, where in medieval times descendants of the “Saka” languages were spoken in Khotan and other places and the modern Pamir language Sarikoli is still spoken (for overviews see, e.g., Edel’man, 1986, pp. 3-6; Oranskiĭ, in Osnovy I, pp. 17-26; R. Schmitt in Compendium, pp. 25-31; N. Sims-Williams, ibid, pp. 165-72; P. O. Skjærvø, ibid., p. 370), and may have been even more widespread before the Achaemenid period, embracing southern Russia and all those areas of Central Asia where mainly Turkic languages are now spoken (see, e.g., Schmitt, op. cit., pp. 92-93). To the west and the north of this area Iranians had contact with the Slavs and Finno-Ugrians, and a considerable number of loanwords from Old or Middle Iranian have been identified in both the Eastern Slavonic (mostly Russian) and Finno-Ugrian languages (see, e.g., R. Bielmeier, in Schmitt, ed., pp. 236-45); so far there is no evidence of reciprocal borrowing in Iranian. On the other hand, contacts between Northeast Iranian and Indo-Aryan have been continuous throughout history, with borrowing in both directions. In the modern languages it is often impossible to distinguish between ancient loanwords and more recent ones (Morgenstierne, 1974; 1975). A typical instance of Indo-Aryan influence is the common occurrence of retroflex (“cerebral”) consonants in most Iranian languages spoken in the region of closest contact (see, e.g., R. E. Emmerick in Schmitt, ed., p. 209).
At the time of the Achaemenids all Iranians spoke “approximately the same language, with but slight variations” (Eratosthenes apud Strabo, 15.2.8), that is, the dialects were probably still mutually comprehensible. It is commonly believed that the language, or languages, of the Avesta belonged to eastern Iran and that the oldest parts of the work were composed somewhere in Central Asia or Afghanistan/Sīstān (see avesta; avestan geography). In contrast to the Rigveda, which was largely compiled on the Indian subcontinent and contains words that are not Indo-Aryan, non-Iranian words are absent from the Avesta. This fact has led some scholars to suggest further that Iranians moved through Central Asia on the heels of the Indo-Aryans, as a second wave of Aryan migration (Burrow, p. 140; Morgenstierne, 1975, p. 433).
The following (Middle) Iranian languages are attested from a.d. the 1st millennium in Central Asia: Scythian and Sarmatian (only in proper names), Sogdian, Bactrian (q.v.), Choresmian (q.v.), Parthian, Bukharan (Henning, 1958, pp. 84-86), and perhaps Ferghanan (the latter two known only from coins; see, e.g., Sims-Williams, in Schmitt, ed., pp. 165-72).
Scythian and related languages were spoken over an enormous territory, from the Black Sea to China (see, e.g., V. I. Abaev, in Osnovy I, pp. 272-364; Bielmeier, op. cit.). The names of two Central Asian Saka tribes are mentioned in ancient sources, including the Achaemenian inscriptions: Saka tigraxauda (Sakas with pointed caps) and Saka haumavarga (Sakas who eat hauma?). No texts in these ancient languages have been discovered from Central Asia proper, but the related Khotanese and Tumshuqese languages spoken in Chinese Turkestan (q.v.) are known from texts from about the 7th to the 10th centuries. Among modern Iranian languages those spoken in the Pamirs belong to the same group as the ancient languages of the Saka, and some are perhaps directly descended from the dialects of antiquity (see below). For example, Wāḵī shares some important phonological features with Khotanese (see, e.g., Skjærvø, op. cit., pp. 375).
Sogdian was spoken in Sogdiana proper, in and around the Zeravshan (Zarafšān) valley, but was also the lingua franca of merchants (many of them Sogdians) along the Silk Road (see, e.g., I. M. Steblin-Kamenskiĭ, in Osnovy II, pp. 347-50; Sims-Williams, in Compendium, pp. 173-75). Most Sogdian texts have been excavated from the sites of colonies established in Chinese Turkestan and date from a.d. the 4th century. Some shorter inscriptions, as well as a substantial collection of documents from the 8th century discovered at Mount Mug on the upper course of the Zeravshan river and coins with Sogdian legends from the 2nd-8th centuries, have been excavated in Sogdiana itself. The language must have survived the Arab conquest and even has a modern descendant, Yaḡnōbī, also called “modern Sogdian” or New Sogdian (see, e.g., Skjærvø, ibid., with refs.).
Bactrian (q.v.) was the language of Bactria in the Kushan period (ca. a.d. 46-225); it is known from inscriptions from Bactria itself, as well as from coins and some short and damaged inscriptions excavated in southern Uzbekistan and Tadjikistan (see also Sims-Williams, in Compendium, pp. 230-35; Steblin-Kamenskiĭ, in Osnovy II, pp. 314-46).
Choresmian (q.v.) was the language of the Iranians who settled in ancient Choresmia (Ḵᵛārazm) on the lower course of the Amu Darya. An inscription in Aramaic script, consisting of the single word asp(a)bārak “rider,” was found at Koĭ-Krylgan-kala and may date from the 4th-2nd centuries b.c. It is the oldest written monument known from Central Asia. With the exception of coins, most of the relevant archeological finds from Choresmia remain unpublished. The major sources for the language remain glosses in Arabic in manuscripts of Moqaddamat al-adab by Maḥmūd Zamaḵšarī (d. 538/1143), Ketāb Qonyat al-monya by Moḵtār Zāhedī Ḡazmīnī (d. 658/1260), and others (see also H. Humbach, in Compendium, pp. 193-203). Several Choresmian loanwords have been identified in such modern Turkic languages as Uzbek and Karakalpak (Livshits, p. 140).
Parthian was spoken in ancient Parthia, to the southeast of the Caspian Sea (southern Turkmenistan and Khorasan). It is known from ostraca of the 1st century b.c. found in the ruins of the Arsacid capital Nisa (near Ashkhabad), which are probably written in early Parthian, possibly with an admixture of Aramaic; from bilingual Sasanian royal inscriptions of a.d. the 3rd century, written in both Parthian and Middle Persian; and Manichean texts from Chinese Turkestan (probably of a.d. the 3rd/4th-8th centuries; see, e.g., V. S. Rastorgueva and E. K. Molchanova, in Osnovy II, pp. 147-232; W. Sundermann, in Compendium, pp. 114-37). The language of ancient Margiana (medieval Marv) was also closely related to Parthian (Livshits, pp. 144-45).
Persian was originally the language of the province of Fārs; through contacts with northwestern vernaculars, Middle Persian acquired many new features and was transformed into a sort of koine, with both southwestern and northwestern components (see, e.g., Lentz). Its expansion into Central Asia began perhaps in Sasanian times, but it was greatly intensified after the Arab invasion, when many settlers and refugees migrated from the west. Middle Persian was also one of the Iranian languages used by Manichean and Christian communities in Central Asia and Chinese Turkestan (see, e.g., Rastorgueva and E. K. Molchanova, in Osnovy II, pp. 6-146; Sundermann, in Compendium, pp. 138-64). Manichean Middle Persian texts are known from the 3rd to the 9th centuries; the earliest Christian Manichean text is a fragmentary Psalter that must have been written between the 6th and 7th centuries but may contain earlier material (see Skjærvø, in Studia Iranica 12/2, 1983, p. 178). Few details of the diffusion of Persian are known, but it formed the basis both for a new literary language, modern Persian (Farsi, Dari), and for the modern Perso-Tajik dialects. Many loanwords from Eastern Iranian languages have entered both literary Persian and modern Tajik, especially local dialects.
The modern period. The modern linguistic map of Central Asia is characterized by a very complex pattern of interaction between two genetically different language groups, Turkic and Iranian. On the whole, Turkic languages have been supplanting Iranian ones, and vast territories have become bilingual. Russian has recently become very influential as well and is forcing out indigenous languages, especially in the official sphere.
The following Eastern Iranian languages are currently spoken in Central Asia (cf., e.g., Morgenstierne, 1958; Skjærvø, op. cit., p. 371; Lazard, ibid., pp. 289-90, refs. p. 293).
1. Yaḡnōbī, a descendant of Sogdian (see above), is spoken on the upper course of the Zeravshan river. The Yaḡnōbīs were forced to migrate to the steppe in the winter of 1970, but are now gradually returning home (see, e.g., A. L. Khromov, in Osnovy III/2, pp. 644-701, on the migration p. 644; Bielmeier, in Compendium, pp. 480-88, on the migration, p. 480, n. 2 with refs.).
2. The Pamir languages spoken in the mountain valleys of the western Pamirs (see, e.g., J. Payne, in Schmitt, ed., pp. 417-44). They include several representatives of the so-called Šuḡnī-Rošanī subgroup (D. I. Edel’man, in Osnovy III/2, pp. 236-347): Šuḡnī, with the dialects of Ḡund, Šahdara, and Bajū; Rošanī, with the dialect of Ḵuf; Bartangī (q.v,); and Rošorvī (formerly called Orošorī). Yazḡulāmī (ibid, pp. 348-407) and Sariqōlī in the eastern Pamirs, the latter spoken by relatively few individuals, probably immigrants from the Sarikol valley in Chinese Turkestan. Both are closely related to the Šuḡnī-Rošanī subgroup. Some local Tajik and Turkic dialects have only recently supplanted Eastern Iranian vernaculars of the Pamir type, as is clear from place names and other words in local dialects in the valleys of Darwāz and Karategin (Rosenfeld). A few words of a western Pamir language were recorded about a century ago among elderly people in the valley of Wanj, where the Tajik dialect spoken today contains many words and grammatical peculiarities found in the neighboring Pamir languages. Many speakers of the latter are bilingual and also know the local form of Perso-Tajik.
3. In the Wāḵān valley Wāḵī is spoken. It shares some isoglosses with Khotanese, but is also more archaic than this Middle-Iranian dialect, and shows strong influence from Indo-Aryan from ancient times (see above; Skjærvø, op. cit., p. 375; Morgenstierne, 1974, 1975).
4. The two closely related dialects Iškašmī and Sanglīčī are spoken to the west of the Wāḵān valley and south of the Yazḡulāmī-Šuḡnī-speaking area.
None of these languages is used in writing, though scholarly collections of texts in transcription and dictionaries have been published. The principal phonetic characteristics of indigenous Iranian languages are those common to the entire eastern branch of the Iranian language family: the development of Old Iranian b, d, and g into fricatives β, ’, and y; the voicing of Old Iranian ft and xt as βd and yd; the disappearance of h (a tendency also in modern Tajik dialects, probably influenced by the Eastern Iranian languages); and so on (see, e.g., Sims-Williams, in Compendium, pp. 167-69). The morphology is characterized by more complete retention of older inflections than in Western Iranian languages, conservation of gender distinction, absence of eżāfa constructions, and placement of attributes before nouns (see, e.g., Skjærvø, op. cit., pp. 371-74).
5. Baluchi (q.v.) is spoken in the vicinity of Marv (in the district of Yolatan) among immigrants from Sīstān who settled in Turkmenistan at the beginning of the 13th/20th century.
6. The Tajik form of New Persian is represented by many local dialects, or subdialects, which are commonly categorized as northern (spoken around Bukhara and Samarqand and in Ferghana), central, and southern, or southwestern and southeastern, spoken in Tajikistan. The local dialects have been especially strongly influenced by Turkic. In the verbal system the extensive use of modal verbs is reminiscent of similar constructions in Turkic languages (cf. Efimov et al., in Osnovy III/1, pp. 193-94; see also xiv, below).
7. Small groups of recent immigrants along the Persian and Afghan borders include speakers of different Persian dialects from Sīstān, Herat province, and Badaḵšān. Kurds living near Firuza, to the south of Ashkhabad, came to Turkmenistan from Khorasan, where they had been settled in the 11th/17th century (Livshits, p. 158).
All speakers of modern Iranian languages in Central Asia are Muslims (see ii, above), and all the languages contain large numbers of Arabic loan words, mainly acquired through New Persian.
See also on individual languages.
Bibliography : H. W. Bailey, “Languages of the Saka,” in HO I, IV/1, 1958, pp. 131-54. T. Burrow, “The Proto-Indoaryans,” JRAS, 1973, pp. 123-40. I. M. D’yakonov, “Vostochnyĭ Iran do Kira” (Eastern Iran before Cyrus), in Istoriya iranskogo gosudarstva i kultury (History of the Iranian empire and culture), Moscow, 1971, pp. 122-54. D. I. Edel’man, “K substratnomu naslediyu tsentralnoaziatskogo yazykovogo soyuza” (On the substratum heritage of the Central Asian languages), in Voprosy yazykoznaniya 5, 1980, pp. 21-32. Idem, “Nekotorye osobennosti glagola v vostochnoiranskom areale (k rekonstruckii tipologii substrata)” (Some special features of verbs in the Eastern Iranian area [On the reconstruction of substratum typology]), in Geneticheskie, arealnye i tipologicheskie svyazi yazykov Azii (Genetic, regional, and typological relations of Asian languages), Moscow, 1983, pp. 131-38. Idem, “History of the Consonant Systems of North-Pamir Languages,” IIJ 22, 1980, pp. 287-310. Idem, Sravnitel’naya grammatika vostochnoiranskikh yazykov. Fonologiya (Comparative grammar of the Eastern Iranian languages. Phonology), Moscow, 1986. A. L. Grjunberg and I. I. Steblin-Kamenskiĭ, “Etnolingvisticheskaya kharakteristika Vostochnogo Gindukusha” (Ethnolinguistic characteristics of the eastern Hindu Kush), in Problemy kartografirovaniya v yazykoznanii i ètnografii (Problems of cartography in linguistics and ethnography), Leningrad, 1974, pp. 276-83. J. Harmatta, “Studies in the Language of the Iranian Tribes in South Russia,” Acta Orientalia Hungarica 1, 1950-52, pp. 261-314. Henning, “Mitteliranisch.”
V. A. Livshits, “Iranskiye yazyki narodov Sredneĭ Azii” (Iranian languages of the peoples of Central Asia), in Narody Sredneĭ Azii i Kazakhstana I, Moscow, 1962, pp. 131-58. G. Morgenstierne, Indo-Iranian Frontier Languages II: Iranian Pamir Languages (Yidgha-Munji, Sanglechi-ishkashmi and Wakhi), Oslo, 1938. Idem, “Neu-iranische Sprachen,” in HO I, IV/1, 1958, pp. 155-78. Idem, “Early Iranic Influence upon Indo-Aryan,” in Commemoration Cyrus I, Acta Iranica 1, Tehran and Liège, 1974, pp. 271-79. Idem, “Ancient Contacts between N. E. Iranian and Indo-Aryan?” in Mélanges linguistiques offerts à Émile Benveniste, Paris, 1975, pp. 431-34. I. M. Oranskiĭ, Die neuiranischen Sprachen der Sovjetunion, 2 vols., Janua Linguarum, Series Critica 12, The Hague and Paris, 1975. Idem, Iranskie yazyki v istoricheskom osveshchenii (Iranian languages in historical interpretation), Moscow, 1979. Idem, Vvedeniye v iranskayu filologiyu (Introduction to Iranian philology), 2nd ed., Moscow, 1988. Osnovy iranskogo yazykoznaniya, 4 vols., Moscow, 1979-87. V. S. Rastorgueva, “Different Types of Bilingualism among Iranian Peoples of the USSR,” in VII International Congress of Anthropological and Ethnological Sciences, Moscow, 1964. A. Z. Rozenfeld, “K voprosu o pamirsko-tadzhikskikh yazykovykh otnosheniyakh” (Problems of Pamiri-Tajik linguistic relations), in Trudy instituta yazykoznaniya VI, 1956, pp. 273-80. R. Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989. I. M. Steblin-Kamenskii, “"Knees" and "Elbows" of the Pamir Substratum,” Peredneaziatskiĭ sbornik 3, Moscow, 1979, pp. 212-14. For bibliographies on individual languages see articles on these languages. See also afghanistan v.
(Ivan M. Steblin-Kamenskij)
This article is available in print.
Vol. V, Fasc. 2-3, pp. 223-226