GYPSY i. Gypsies of Persia




Gypsies are generally referred to by the term kowli in Persian, seemingly a distortion of kāboli, i.e., coming from Kabol, the capital of Afghanistan. It is not at all certain, however, that all the groups referred to as kowli are authentic gypsies; nor that only the groups referred to as kowli should be considered as gypsies. The fact is that almost everywhere in Persia there are groups with characteristics similar to those of the Gypsies, but they are called by different names, sometimes designating their geographic or ethnic origin, sometimes their social status, and sometimes their profession: abdal (Ḵuzestān), āhangar (several places), čegini or čingāna (Azerbaijan, the Caspian provinces), fiuj (Ḵuzestān), ḡarbālband (Tehran), gowdari (Baluchestan), jat (Khorasan, Afghanistan), jugi (Central Asia), Karači or Karāči (Azerbaijan), kāvol (Lorestān), kowli (Tehran, central province, Lorestān, Baḵtiāri, Fārs), ḡorbati (Lorestān, Baḵtiāri), lom (Tajikestan), luli (several places), luri (Baluchestan, center, not to be confused with the inhabitants of Lorestān), luṭi (Kermān, Lorestān, not to be confused with the 19th-century urban thugs, see Migeod, 1959), mazang, mul- tāni (Central Asia), moṭreb (Lorestān), qerešmāl (Khorasan), širāzi (Baḵtiāri), sudāni (Persian Gulf), suzmāni (Kermānšāh), tušmāl (Baḵtiāri), žott (Baluchestan, Ḵu-zestān) (Sykes, 1902, pp. 437-38; Minorsky, 1931; Ama-nolahi and Norbeck, pp. 3-4; Ḏokāʾ, passim).

The identity of these groups being uncertain, there are no statistics about them; at best they are estimated to be “two or three thousand people” (Amanolahi and Norbeck, p. 2). Their origins are just as obscure. According to a legend reported in Šāh-nāma (Moscow, VII, pp. 451-52) and repeated by several modern authors (e.g., Bausani; Goeje), the Sasanian king Bahrām V Gōr (q.v.) learned towards the end of his reign (421-39) that the poor could not afford to enjoy music, and he asked the king of India to send him “ten thousand luris, men and women, lute playing experts” (l. 2558). When the luris arrived, Bahrām gave each one an ox and an ass and an ass-load of wheat so that they could live on agriculture and play music gratuitously for the poor. But the luris ate the ox and the wheat and came back a year later with their cheeks hollowed with hunger. The king was angered with their having wasted what he had given them, ordered them to pack up their bags on their asses and go wandering around the world.

This interesting legend only partially reflects the real life of the kowlis and assimilated groups. These groups, however, possess several common characteristics. They are usually small groups of two or three nuclear families on the move, living in tents or temporary dwellings, and following seasonal migrations of pastoral nomadic tribes to which they are more or less attached (Barth, 1961, pp. 91-92; Digard, 1981, pp. 21-22), or moving from village to village and to town suburbs. That is why they have been described as “peripatetic” or “peripatecians” (Rao, 1985). There are, however, districts in some cities (e.g., Birjand, Nišāpur, Sabzavār) where the kowlis live permanently (Ivanow, 1920, pp. 282-83; idem, 1926, p. 157).

The economy of the kowlis is based on supplying the nomads and/or the settlements they frequent with services or manufactured articles against money or goods, hence their designation by some authors as “service nomads” (e.g., Barth, 1960; Kieffer; Olesen, passim). Their professional specialties are blacksmithery (āhangari), peddling, making small articles of everyday use such as sieves made of wood and gut, weaving and carding combs made of wood and metal, spindles and tops made by turning wood, wicker baskets and straw mats (kowli), etc. They also work as musicians (luri, luṭi, tušmāl, etc.) and, in a more marginal sense, fanfare comedians as well as performing in animal shows (Filmer, passim; Westphal-Hellbusch, 1964, passim; Drukker).

These groups are organized in economically autonomous domestic production units that are related to each other within each professional guild; each profession corresponds to a group and the groups are in general endogamous. Within the group the transmission of social status and technical specialization is patrilineal.

Although the kowlis are Muslims, Shiʿite or Sunnite as the case may be, and partly adopt the language (see below) of the village or tribal communities for which they work, they are much despised by these communities. The generic word ḡorbati (lit: “stranger”), by which they are designated in many regions, is very pejorative. The segregation of these artisans and musicians is also manifested by their prohibition to marry outside their group. In some areas, for instance in the Baḵtiāri region, they are prohibited to practice the same activity of production as the tribe (e.g., animal husbandry, weaving) or wear the tribal dress (Digard, 1981, p. 211).

In reference to a similar context among the Swat in Pakistan, Fredrik Barth maintains that these “despised professions in the service of land-owners” can be likened to castes. This notion is, however, rejected by the scholars (including the present writer) who believe that castes could not exist outside a social structure in which all members are categorized in a system of castes (see Bouglé; Dumézil; Hocart; Bailey, 1957; idem, 1961; Balandier, pp. 99-107). The gypsies of Persia seem rather to fit into a “system of patronage and clientele,” as Dumont put it (p. 265), complementing a class society (Digard, 1973).

The rough picture drawn of the groups of artisans living in symbiosis with nomadic communities or villages should, however, be modified in each individual case. A number of these groups should in fact be treated as particular cases. For instance, the musicians (luri, luṭi, tušmāl, etc.) benefit from a special status, which distinguishes them from both the kowlis and ḡorbatis present in the same area, as well as from the musicians of other regions. Among the Baḵtiāris, for example, the tušmāls (Digard, 1974) all belong to the same faction of ʿamala, traditionally at the service of the Khan and an integral part of the tribe. They are at the bottom of the social hierarchy and must marry among themselves, but they have the right to wear Baḵtiāri dress. The name tušmāl, by which they are called is an outdated title among the Baḵtiāris, but it is still in use elsewhere, notably in Lorestān, where it is an equivalent to kad-ḵodā (chief). The situation of the tušmāls is singularly in contrast with that of the musicians of Mamassani, Boir Ahmadi and Lor tribes culturally and geographically closest to the Baḵ-tiāris, where they have the unenviable status of “foreigner” (ḡorbati) and are pejoratively called luṭi or moṭreb (Minorsky, 1931; Bahman Beygi; Nadjamadbadi, passim).

The ethnic identity, real or supposed, of the professionally specialized groups that live in contact with the nomadic tribes of the southwest of Persia is, therefore, crucial to their status; however, it remains a very ambiguous question. Even the vocabulary is ambiguous. In fact the terms used, not only in Persia (Amanolahi and Norbeck; Minorsky, 1931; Sykes, 1902) but in all the Middle East (Burton; Anastase, 1902; Littmann; Massignon, passim; Lewis and Quelquejay; Kenrick, 1975, 1994; Berland) to designate these groups refer to very distant notions, ranging from socio-economic status (e.g., abdal “slaves,” ḡorbati “foreigner”) and profession (āhangar “blacksmith,” luṭi, “musician”) to regional or supposedly ethnic origins (čingāna “from Čangar or Žingar, in India,” kowli, “from Kabol,” žott or jat “Indian,” sudāni, “black”).

Such confusion is not exceptional. Similar examples date back to most ancient times, e.g., the Khalybes of Herodotus, which is not clear whether it refers to some people or it was a professional designation (blacksmith, Planhol). The same ambiguity exists in our times among the Baḵtiāris, where the ethnonym kowli refers to a specified socio-professional group, that of very mobile pedlars who visit the nomads to offer them various light objects of their manufacture (e.g., weaving combs, sieves, wooden spoons, spindles, etc.) and leave them immediately thereafter, contrary to other ḡorbatis who pre-ferably remain in contact with the same portion of the tribe and move with it. Some of these ḡorbatis are referred to by the Baḵtiāris by the name of a profession, āhangar (blacksmith), but they call themselves širāzi (from Shiraz).

Here, the linguistic criterion itself is of little help. However, one thing is certain: For all the groups in question, Persian or the languages of the tribes with which they associate has, until recent progress in schooling, been a second language, a working language (Amanolahi and Norbeck). Beyond that, the linguistic studies of these groups (Amanolahi and Norbeck; Digard, 1978; Ivanow, 1914; Macalister; Sykes, 1902; Wirth) are too piecemeal and too partial to permit definite conclusions as to the nature of their language. Except in two cases (Ivanow, 1922; Turner, 1978), their language cannot be considered as an argot so common among many Persian guilds. In all other cases (languages called Darviši, Luṭiuna, Āhangari, Širāzi, etc. in various places) one is struck by the existence of common terms among several groups, such as sanuta or sanufta for dog; tirang, ox; nahur, eye; mana, bread; dontaz, sister; bri, brother; dāqis, mother; bāqis, father; kala, son; gowari, chief; and so forth (Amanolahi and Norbeck). Despite their flimsiness, these few similarities seem to indicate the existence and permanence in Persia of at least a partially common language and culture among the groups of kowlis and assimilated groups, despite the strong loss of culture brought about by fragmentation and geographical dispersion.

One is also struck by the presence, in these tongues, of words close to Hindi, Romani and Manouchian (spoken by Manouch gypsies of Europe and America), such as čekel, earth/Romani čik, mud; gohrā, horse/Hindi qorā, horse/Roman huro, colt; loh, iron/Hindi loha, same mean-ing, Lohar, blacksmith; mārez, man/Romani more, man (interjection), murš, male; ponnawi, rain/Hindi and Manouchian pānī, water; potor, son/Hindi putra, same meaning (literary), and so on (Digard, 1978). These similarities clearly seem to indicate that the groups in question, their language(s) and their culture(s) could be considered as part of the vast nebula known as “gypsy.” Such are the conjectures, much more than definite conclusions, based on the fragmentary data currently available.



I. Afšār Sistāni, Kowlihā: pažuheš-i dar zamina-ye zendagi-e kowlihā-ye Irān wa jahān, Tehran, 1377 Š./1998.

S. Amanolahi and E. Norbeck, “The Luti, an Outcast Group of Iran,” Rice University Studies 61, 1975, pp. 1-12.

P. Anastase, “Les Nawar ou Tziganes d’Orient,” Al-Mashrik 5, 1902, pp. 865-76, 932-40, 966-75, 1031-37, 1077-86.

Idem, “Luristan, or Land of the Luri,” Journal of the Gypsy Lore Society (JGLS) 7, 1913-14, pp. 298-319.

H. Arnold, “Some Ob-servations on Turkish and Persian Gypsies,” JGLS 46, 1967, pp. 105-22.

M. Bahman-beygi, ʿOrf o ʿādat dar ʿašāyer-e Fārs, Shiraz, 1324 Š./1945; tr. V. Monteil as “Moeurs et coutumes des tribus du Fars,” in V. Monteil, ed., Les tribus du Fārs et la sédentarisation des nomades, Paris and The Hague, 1966, pp. 97-152.

F. C. Bailey, Caste and the Economic Frontier, London, 1957.

Idem, “Tribe and Caste in India,” in L. Dumont and D. Pocock, eds., Contributions to Indian Sociology 5, Paris and The Hague, 1961, pp. 7-19.

G. Balandier, Anthropologie politique, Paris, 1967.

F. Barth, “The System of Social Stratification in Swat, North Pakistan,” in E. R. Leach, ed., Aspects of Caste in South India, Ceylon and North-West Pakistan, Cambridge Papers in Social Anthropology 2, Cambridge, 1960.

Idem, Nomads of South Persia: The Basseri Tribe of the Khamseh Confederacy, Oslo, 1961.

A. Bausani, The Persians, from the Earliest Days to the Twentieth Century, London, 1972.

J. C. Berland, No Five Fingers Are Alike: Cognitive Amplifiers in Social Context, Cambridge, Mass., 1982 (Gypsies of Pakistan).

G. F. Black, “The Gypsies of Armenia,” JGLS 8, 1913-14, pp. 327-30.

C. Bouglé, Essai sur le régime des castes, Paris, 1908.

R. F. Burton, The Jew, the Gypsy and el-Islam, n.p., n.d.

J.-P. Digard, “Histoire et anthropologie des sociétés nomades: le cas d’une tribu d’Iran,” Annales: Economies, Sociétés, Civilisations 28, 1973, pp. 1423-35.

Idem, Baxtyâri, nomades de la montagne, Paris, 1974.

Idem, “‘Tsiganes’ et pasteurs nomades dans le sud-ouest de l’Iran,” in J.-P. Liégeois, ed., Tsiganes et nomades: tendances actuelles de la recherche, Paris, 1978, pp. 43-53.

Idem, Techniques des nomades Baxtyâri d’Iran, London etc., 1981.

R. and B. Drukker, “Turkish Gypsy Bear-leaders,” JGLS 47, 1968.

G. Dumézil, “La préhistoire indo-iranienne des castes,” JA 216, 1930, pp. 109-30.

L. Dumont, Homo hierarchicus: essai sur le système des castes, Paris, 1968.

H. Field, Contributions to the Anthropology of Iran, Chicago, 1939.

H. Filmer, The Pageant of Persia, Indianapolis and London, 1937.

M. J. de Goeje, Mémoire sur les migrations des Tsiganes à travers l’Asie, Leiden, 1903.

M. J. de Goeje and J. Sampson, “The Gypsies of Persia,” JGLS N.S. 1, 1907-08, pp. 181-83.

F. H. Groome, “Persian and Syrian Gypsies,” JGLS 13, 1891.

A. M. Hocart, Les castes, Paris, 1938.

W. Iva-now, “On the Language of the Gypsies of Qāināt (in eastern Persia),” J(R)ASB 10, 1914, pp. 439-55.

Idem, “Further Notes on the Gypsies in Persia,” ibid., 16, 1920, pp. 281-91.

Idem, “An old Gypsy-Darwish Jargon,” ibid., 18, 1922, pp. 375-83.

Idem, “Notes on the Ethnology of Khurasan,” JRGS 67, 1926, pp. 143-58.

D. Kenrick, “Romanies in the Middle East,” Roma (Chandigarh) 1/4, 1976, pp. 5-9; 2/1, 1977, pp. 30-36; 2/2, 1977, pp. 23-39.

C. M. Kieffer, “A propos de la circoncision à Caboul et dans le Logar: note de dialectologie et d’ethnographie afghanes,” in Festchrift für Wilhelm Eilers, Wiesbaden, 1967, pp. 191-201.

J. Koch-anowski, Gypsy Studies, 2 vols., New Delhi, 1963.

H. C. Köprülüzade Mehmet Fuad, “Abdal,” in Türk halk edebiyati ensiklopedisi, Istanbul, 1935.

G. L. Lewis and C. Quelquejay, “Čingāne,” in EI2, pp. 41-42.

E. Littmann, Zigeuner arabisch, Bonn, 1920.

R. A. S. Macalister, Language of the Nawar or Zutt, the Nomad Smiths of Palestine, London, 1914.

Massé, Croyances. L. Massignon, Annuaire du monde musulman, Paris, 1954.

H. G. Migeod, “Die Lûtis, ein Ferment des städtischen Lebens in Persien,” Journal of the Economic and Social History of the Orient 2, 1959, pp. 82-91.

V. Minorsky, “Les Tsiganes Lûlî et les Lurs persans,” JA 218, 1931, pp. 281-305.

S. Nadjmabadi, “Die Shiravand in West-Lorestan: Mit besonderer Berücksichtigung des Verwandtschaftsystems,” Ph.D. diss., Heidelberg, 1975.

A. Olesen, Afghan Craftsmen: The Culture of Three Itinerant Communities, Copenhagen, 1994.

A. G. Paspates, Études sur les Tchingianés ou Bohémiens de l’Empire Ottoman, Constantinople, 1870.

K. P. Patkanoff, “Some Words on the Dialects of the Transcaucasian Gypsies,” JGLS N.S. 2, 1908-09.

X. de Planhol, “Geographica pontica,” JA 251, 1963, pp. 293-309.

ʿA. Qāʾem-maqāmi, “Kowlihā ki hastand wa kojā hastand?” Armagān 29, 1339 Š./1960, pp. 262-70.

A. Rao, Les Gorbat d’Afghanistan: aspects économiques d’un groupe itinérant “Jat,Paris and Tehran, 1982.

Idem, “Des nomades méconnus: pour une typologie des communautés péripatétiques,” L’Homme 25, 1985, pp. 97-119.

J. Sampson, “On the Origin and the Early Migrations of the Gypsies,” JGLS 19, 1923.

A. T. Sinclair, “The Oriental Gypsies,” JGLS N.S. 1, 1907-08, pp. 197-211.

P. Sykes, “Anthropological Notes on Southern Persia,” Journal of the Royal Anthropological Institute 32, 1902, pp. 339-49.

Idem, “The Gypsies of Persia: A Second Vocabulary,” ibid., 36, 1906, pp. 302-11.

Idem, “Notes on Musical Instruments in Khorassan, with Special Reference to the Gypsies,” Man 9, 1909, pp. 94-103.

Idem, “Persian Jats,” JGLS N.S. 4, 1910.

D. Tripler, “The Banjara: Notes on Some Cousins of the Gypsies,” JGLS, 3rd Series 44, 1965.

R. L. Turner, The Position of Romani in Indo-Aryan, Lodon, 1927.

Idem, “Notes on ‘A Note on the Secret Language of the Traditional Musicians of Iran,’ by S. Amanullahi” JGLS,4th Series 1, 1978, pp. 283-86.

S. Westphal-Hellbusch, “Randgruppen in Nahen und Mittleren Osten,” Baessler Archiv 53, 1980, pp. 1-59.

S. Westphal-Hellbusch and H. Westphal, The Jat of Pakistan, Berlin, 1964.

Idem, Zur Geschichte und Kultur der Jat, Berlin, 1968.

G. L. Windfuhr, “European Gypsies in Iran: A First Re-port,” Anthropological Linguistics 13, 1970, pp. 271-92.

A. Wirth, “A Persian Gypsy Vocabulary,” JGLS, 3rd Series, 6, 1927, pp. 88-95.

Y. Ḏokāʾ, Kowli wa zendagi-e u, Tehran, 1337 Š./1958.

(Jean-Pierre Digard)

Originally Published: December 15, 2002

Last Updated: February 24, 2012

This article is available in print.
Vol. XI, Fasc. 4, pp. 412-415