Main Page
Main News Page (1)
Jhalkari Bai
Freedom Fighter
Guest Book
Mein Bhi Kaain Kehno Chha
Goaar Darpan
Goaar Forum
What Do You Think?
Special Mail
Gor History
Dr. Tanaji Rathod
Pradeep Ramavath-1
Goaar Goshti
Religious Persons
Political Persons
Social Reformers
Goaar Chetna
Goaar Ratan
Ian Hancock (Part-1)
Ian Hancock (Part-2)
VaniaDeGila Kochaniwski
Gypsy Caravan: A Movie
Yvonne Slee
Article of Bajram Haliti
Caravan of Indo Romany Writers (Part-1)
Caravan of Indo Romany Writers (Part-2)
International Roma Day
Bajram Haliti
Goaar History
Goaar Writers
Attn: Researchers
About Us

COPYRIGHT T.A. Acton (ed.)

Scholarship and the Gypsy Struggle: Commitment in Romani Studies Hatfield: University of Hertfordshire Press (2000), pp. 1-13.


Ian Hancock

If we hypothesise that Rajputic, the military lingua franca of the army camps a thousand years ago, became reshaped differently as its speakers separated and moved into different areas, we can account for its Islamisation in Muslim areas, where it became the Persian- (and increasingly Arabic-) influenced Urdu (written in Arabic script), and its Hinduisation in Hindu areas, where it became Sanskritised (and written in Devanagri script). But some Rajputic speakers “remained in jungles and Chambal Khore, a valley in Rajasthan, [and] came to be known as Lamans or Banjaras . . . the group which fled to the forests and remained there are the present-day hill-tribes called the Banjaras” (Naik, 1978:5). Under such circumstances, their original Rajputic did not develop into either Hindi or Urdu, but emerged into a distinct but related new language—Ghorbati (also called Ghorboli, Lamani or Banjari). The speech of those Ghor (Banjaras) who subsequently left Rajasthan for other parts of India to the South underwent further change in the direction of the surrounding languages, as Grierson noted, paralleling the re-shaping of Rajputic into Romani in the Greek-language environment.

The movement out of India by those Rajputs who were the ancestors of the Romanies is accounted for in Banjara history; less easily explained is why they did not return to India. Possibly they continued to maintain themselves as a mercenary military force, engaging in conflicts with the Muslims along the Silk Road and the Caspian littoral; the linguistic evidence points to this as their route to the West. Soulis (1961:163) suggests that Seljuq raids in the area at the end of the eleventh century were a factor [in Hancock, 2004, I bring the “Seljuq Factor” into the discussion as providing an account of how the pre-Romanies moved from India to Anatolia]. Marushiakova and Popov (1997:63) attribute their eventual movement into Europe to the twelfth and thirteenth century Ottoman invasions. Significantly, and as further support of their original profession as (besides soldiers) shiviranuchara or camp followers they document those early Romanies as “servants in the auxiliary detachments or as craftsmen servicing the army.” Their separation from India caused something of a linguistic and social trauma. Even within India, the Rajputs were able to maintain a distinctiveness from the surrounding peoples but lacking even that normalising factor, the ancestors of the Romanies were left completely isolated with their language and mixed ethnic identity, becoming increasingly remote from their homeland. As with the Rajputs before them, the sense of being a composite population on one level, but a distinct and “special” population on another, came in time to characterise a distinctively Romani world view which permitted the Romanies to continue to accept new people into their number, providing they took on Romani identity, and a Romani linguistic trait which allowed the language to incorporate and assimilate new grammatical and lexical material. This capacity to absorb and modify, socially as well as linguistically, represents a continuum from the time of the creation of the Rajputs, and remains an integrally defining aspect of Romani identity.


In Hancock (1995), I demonstrated that, contrary to the established belief, Romani and Domari have distinct and unrelated histories, separated by over five hundred years. It was possible to show this by examining the Iranic lexical content of both, and concluding that the percentage of shared items was far too low if both languages had been part of the same migration through Persia. If the third member of the conventional trilogy comprising this theory is included—Lomavren, spoken in Armenia—then there is no single Iranic item shared by all three branches. Domari is clearly of an old-Indic type, and differs significantly from Romani in its phonology, grammar and lexicon.

While it can be demonstrated that Domari, spoken in several dialects throughout the Middle East, but particularly in Syria, is not related to Romani (other than that they are both Indian languages—which could equally well be said about Romani and Sinhalese), accounting for its origin presents a different problem.

The movement of the ancestors of the Romanies out of India has traditionally been explained by looking to Firdausi’s account of the gift of several thousand musicians in the fifth century from the Indian king Shankal to the Sassanid shah of Persia. That such an episode occurred is quite likely, since it has been documented in a number of places (Hancock, 1999). But it does not hold up linguistically, since the musicians were “Sindhian,” from the north-western Prakritic area, and Domari shows much closer lexical similarities with languages of the Central group (Nseir, 1998). According to The Encyclopedia of Islam (Minorsky: V:818), there is a group which identifies itself as Kurdish east of Bohtan in Persia, “which bears the suggestive name Sindi or Sindiyan (the Sindhis).”

Instead, the presence of the Dom (Domari speakers) as well as of other ‘Gypsy’ populations in the Middle East such as the Karači, whose linguistic affiliation has still to be determined (Kenrick, 1976 c,d) might well be accounted for using the same theoretical framework as that proposed here for Romani, i.e. composite military troops moving westwards out of India. Like Romanies, the Dom also refer to non-Dom as kajja “civilians.” To justify this, it must be shown that there was an invasion of India in this area at this time, and that Indian troops left India to engage them.

This proves to be the coming of the Turkic-speaking Huns who, starting from Bactria, eventually occupied other parts of Persia and India, but who were defeated and driven back by the Persians in the sixth century (Thapar, 1966:142), and who were gradually absorbed into the local population in India by the end of the millennium. While their cultural impact was not extensive, the linguistic consequences of their invasion were far-reaching:

Prakrit is of linguistic interest as illustrative of the linguistic evolution from Prakrit to Apabhramsha (literally ‘falling down’). Apabhramsha, a corrupt form of Prakrit, dialect, is believed to have originated in the north-west and travelled from that region with the migrations of people who scattered and settled in central and western India after the Hun invasions (Thapar, 1966:257).

There are accounts of the invasion of the Huns, and references to their being “pushed back” in a “brave defence” by King Skanda Gupta who, “for the duration of his twelve-year reign . . . was obliged to ward off their predatory assaults” (Wolpert, 1977:94):

From the middle of the fifth century a new barbarian invader, the Hun, made his ominous appearance, as his kindred were doing in Europe. For a generation or so the Guptas succeeded in holding off this menace from the northwest. But toward the close of the century it reappeared in the persons of Toramana and Mihirakula, the latter holding Kashmir, western India and part of the Gangetic basin. In fifty years the Huns were pushed back to Kashmir and parts of the northwest, and they never again became a threat, losing their identity among the Rajput clans of later fame.” (Wright, 1969:301; see also Thapar, 1966:140).

The White Huns, or Ephthalites, or as the Indians called them, Hunas, were barbarians in the sense most painful to settled societies... While Attila’s Huns streamed across Eastern Europe to attack Italy in the mid-fifth century, a southward wave overwhelmed the Sassanid Empire in Persia and the Kushan remnants of the Indian north-west. Skanda Gupta (455-470) is credited with a brave defence of his western frontiers in face of this irruption, but by about 500 a Hun chieftain was recorded as far south as Ujjain . . . The Gupta dynasty is taken to extend, though interrupted in the fifth century by Hun invasions, from 320 to about 450 . . . [the Buddhist university city of] Taxila, beyond the Gupta boundaries, [was] in the fifth century devastated by the Huns (Watson, 1988:60,62,66).

There are also some references to the possibility that the Indians who moved out to confront the Huns at this time were from Kabul rather than Sindh, and that the two events have been confused over time. Harriot (1830:524-5) reported that a Persian historian, Fateh Ali Khan, told him that Firdausi’s musicians were from Kabul, not Sindh, and Gobineau’s Kauli informants in Persia in the 1850s assured him that their ancestors were from Kabul, not Sindh, and that their name was originally Kabuli (1857:690). At the time of the Hun invasions, Kabul was not just the city it is today, but an entire Hindu kingdom (also called Kapisha) which extended from the Kabul River Valley to the Hindu Kush. Siraiki, Multani and some other languages transitional between the central and northwestern groups are now spoken in this area, though we need to reconstruct the linguistic situation as it was there fifteen centuries ago.


The mixed nature of Romani, and the social and linguistic clues evident in an examination of (particularly) its lexicon, make a strong case for its having taken its initial form as a military koïné which left India with its speakers, subsequently developing outside of its homeland. Work begun on an examination of its historical phonology and grammar also supports the fact of its multi-source origins. The identity of its first speakers, and the circumstances of its emergence, established a sociolinguistic ‘character’ which continues to typify the Romani people and language.

The same type of military historical scenario may also explain the presence of Indian languages and peoples such as the Domari throughout the Middle East.


Abbas, Khwaja A., 1969. “A link language for the common man,” Language and Society in India, 1969:29-36.

Alderson, A.D., and Fahir Iz, eds., 1959. The Concise Oxford Turkish Dictionary, The Clarendon Press, Oxford.

Bakker, Peter, and Maarten Mous, eds., 1994. Mixed languages: 15 case studies in language Intertwining. IFOTT, Amsterdam

Baskakov, N.A., 1969. Vedenie v izuchenie Tjukskixjazykov, Vysshaya Skola, Moscow.

Bright, William, ed., 1966. Sociolinguistics, Mouton, The Hague

Chatterji, Suniti K., 1931. “Calcutta Hindustani: A study of a jargon dialect,” Indian Linguistics, 1(2-4):77-234.

Clough, James Cresswell, 1876. On the Existence of Mixed Languages, Longmans, Green and Co. London.

Comrie, Bernard, ed., 1990. The World’s Major Languages, The University Press, Oxford.

Fraser, A., 1995. (2nd ed.). The Gypsies, Blackwell, Oxford.

Gobineau, Count Joseph de, 1857. “Persische Studien I: Die Wander- stamme persiens,” Zeitschrift der Deutschen Morgenlandischen Gesellschafft, 11:689-699.

Gumperz, John J., and Robert Wilson, 1971. “Convergence and creolization: A case from the Indo-Aryan/Dravidian border in India,” in Hymes, 1971:150-167.

Hancock, Ian, 1980. The Language of the Texas Afro-Seminoles, Publication of the Seminole Negro Scout Historical Association, Brackettville.

Hancock, lan F., 1995. “On the migration and affiliation of the Rom, Dom and Lorn Gypsies,” in Matras, 1995:29-59.

Hancock, Ian F., 1999. The Origins and Westward Migration of the Romani People. Occasional Paper of the International Romani Archives, No. 5,675,989.

Hancock, Ian 2004. “On Romani origins and identity: questions for discussion,” in Marsh & Strand (2004).

Harriot, John, 1830. “Observations on the oriental origin of the Romanichal, or tribe miscalled Gipsey or Bohemian,” Transactions of the Royal Asiatic Society, 2:518-88.

Hymes, Dell, ed., 1971. Pidginization and Creolization of Languages, Cambridge University Press, Cambridge.

Isaac, Jeanette, 1984. “The Banjara: Nomads of time,” Swagat Magazine, July, pp. 52-9.

Jahr, Ernst Hakon, ed., 1992. LanguageContact: Theoretical and Empirical Studies, Mouton DeGruyter, Amsterdam.

Kachru, Yamuna, 1990. “Hindi-Urdu,” in Comrie, 1990:470-89.

Kelley, Gerald, 1966. “The status of Hindi as a lingua franca,” in Bright, 1966:299-308.

Ketalaer,Josua, ca. 1698 [1743]. [Grammatical description of Hindustani] private, Leiden.

Khubchandani, Lachman M., 1963. The Acculturation of Indian Sindhi to Hindi, a Study of Language in Contact, Unpublished doctoral dissertation. The University of Pennsylvania.

Khullar, K.K., 1995. “Urdu, the language of national unity,” India Perspectives, March, pp. 21-25.

Maenchen-Heffen, J. Otto, 1973. The World of the Hun, The University of California Press, Berkeley.

Marsh, Adrian and Elin Strand, eds., 2004. Contextual, Constructed and Contested: Gypsies and the Problem of Identities. In prep.

Marushiakova, E., and Vesselin Popov, 1997. “The Romanies in the Balkans during the Ottoman Empire,” Roma, 47:63-72.

Matras, Yaron, ed., 1995. Romani in Contact: The History, Sociology and Structure of a Language. Amsterdam: John Benjamins

Mesthrie, Rajend, 1993. “Koïnéization in the Bhojpuri-Hindi diaspora, with special reference to South Africa,” in Siegel, 1993:25-44.

Minorski, V., 1954.“Luri” in The Encyclopedia of Islam. Leiden University Press, Leiden.

Minturn, Leigh, and John T. Hitchcock, 1966 The Rajputs of Khalapur, India, Wiley, New York.

Naik, Ranjit, 1978. Banjara (Indian Roma) from Barothan, Report presented at the Second World Romani Congress, Geneva, April 8- 11.

Nida, Eugene A. and Harold W. Fehderau, 1970 “Indigenous pidgins and koïnés,” International Journal of American Linguistics, 36:146- 155.

Nseir, Amjad ibn, 1998. The subgrouping of Domari within the Aryan languages: A Morphosyntactic analysis. Research Paper, Department of Linguistics, The University of Texas at Austin.

Petersen, Walter , 1912. “Vedic, Sanskrit and Prakrit,” Journal of the American Oriental Society, 32:414-28.

Rathod, Jasvantsinh N., 1988. Report on the Survey of Banjara Castes, Tribal Research and Training Centre, Ahmedabad.

Rathore, B. Shyamala Devi, 1997. “Gypsy Music” Paper presented at the Conference on Romani Studies, University of Greenwich, London.

Reinecke, John, 1937. Marginal Languages: A Sociological Survey of the Creole Languages and Trade Jargons, Unpublished doctoral dissertation, Yale University.

Reinecke, John, Stanley Tsuzaki, David DeCamp, lan Hancock and Richard Wood, eds., 1975. A Bibliography of Pidgin and Creole Languages. The University of Hawaii Press, Honolulu.

Siegel, Jeff, 1985. “Koïnés and koïnéization,” Language in Society, i4(3):357-78.

Siegel, Jeff, ed., 1993. Koïnés and Koïnéization (= International Journal of the Socioloogy of Language, Special Issue, N0.99) Mouton de Gruyter, Berlin.

Soulis, George C., 1961. “Gypsies in the Byzantine Empire and the Balkans in the late Middle Ages,” Dumbarton Oaks Papers, 15:142-65.

Wright, Desmond, 1969. History of the World: Prehistory to the Renaissance, Two vols., Hamlyn. London.

Zograf, G.A., 1963. “Kalkutsij xindustani knotsa,” Kratkii Soobshchenija Instituta Narodov Azhii, 61:142-53.