Origin of Roma

At the end of the eighteenth century, linguistic comparisons of Romani with Indic Indo-European languages proved the Indian origin of the Roma. The following occurrence, which perhaps everyone who is interested in Roma history knows about, supposedly pointed researchers in the direction of India as home of origin. The first person to describe it was the enlightened Slovak intellectual Samuel Augustini ab Hortis in his work "Zigeuner in Ungaren (…)" (1775).

"On November 6, 1764, I was visited by the erudite printer Štefan Pap Nemethi, who told me in an interview what he had learned from the Calvinist preacher Štefan Váli who, at that time, was working in Almáš in the district of Komárno. When Váli was at the university in Leyden, he became friendly with three Malabar youths(…) When Váli noticed that their language bore more than a slight resemblance to the language of our Gypsies, he took advantage of the opportunity and wrote down more than a thousand of the Malabar words they used along with the words' meanings(…) When Váli returned to his country, he wanted to ascertain the meanings of those Malabar words; the Gypsies effortlessly translated them." (ab Hortis 1994. p. 54)

What language were the Malabar students speaking? The land of Malabar lies in what today is the southwest coastal Indian state of Kerala. There they speak Malayalam, a Dravidian language which has nothing in common with Indo-European languages (Hindi, Bengali, Punjabi, Marathi, Gujarati - Romani! – and others). If the students had been speaking Malabari – or Malayalam, Váli would not have been able to perceive a resemblance to Romani. The students undoubtedly belonged to the highest social class, which used – mainly in written expression – the language of the intellectuals, the old cultural literary language of India, Sanskrit. Most of the roots of Sanskrit words appear in modern Indic (Indo-European) languages – including Romani.

There is another possible explanation. The southwest coast of India was a cape where, from time immemorial, there were relations between Indians and the western world. Christianity set down roots there at the time of St. Thomas the apostle, who allegedly died in southern India. It is possible that Indian Christians from all over the extensive region of India came to study right there, the center of Indian Christianity. Their native languages might have been not Romani, but some other Indo-European languages similar to Romani. (Elsewhere in the land, Christianity did not take hold.) From that centre, some left for Europe to finish their Christian education.

Whatever the case may be, Váli's remarkable discovery came "into circulation" and many linguists, including the German Johannes Rüdiger and the Englishman William Marsden, took notice of it. They and those who followed them in the next century irrefutably ascertained that Romani is a modern Indic language. At the beginning of the twentieth century, Ralph Turner ascertained the closest relationship between Romani and Hindi, Rajasthani dialects or Punjabi. Meanwhile, his classification was generally accepted.

It has been judged that approximately 70% of the basic vocabulary of Romani is related to other Indic (Indo-European) languages. Below are examples of two sentences and one proverb. Further examples follow in illustrations.

  • mire bala kale hin (R.) / mere bal kale hain (H.) [My hair is black]
  • ajsi bari rakli! (R.) / ajsi bari larki! (H.) [Such a big girl!]
  • Me raňi, tu raňi, ko pherela paňi? (R.) / Me(n) rani, tu rani, kon bharega pani? (H.) [I (am) a lady, you (are) a lady – who will fetch the water?]

The original "Indic" words in Romani, in comparison with borrowed words, have the greatest morphosyntactic potency, which means that it is possible to create a number of other words with specific suffixes.

The following are examples of derivatives of the word bari - large, big. They are derived from the base (bar-):

Romani English
bari-pen (m) size, distinction, pride
bar-ikan-o pretentious, proud
bar-ikaň-ipen (m) pretentiousness
bar-ikaň-ar-el pes to primp and preen, to overdress
bar-ar-el to enlarge
bararel avri to raise children
bararel pes to brag
bar-uv-el (barol) to grow
bar-av-el to cultivate
bar-eder officer (from a semantic extension of the comparative "bigger")

Apart from the lexicon (words), Romani also has a number of grammatical and derivative formants – endings and suffixes - in common with Indic languages.

Example: the singular adjectival ending

  • m. – o (Romani, Marvari): baro (R. Marv.) large, big; kalo (R., Marv.) black
  • f. – i (R., Hindi, Marvari, etc.) bari (R. Ind.) large, big; kali (R., Ind.) black

The genitive ending (m. –ka [Hindi] / ko [Marvari, R.]; f. –ki [Hindi, Marvari, R.]) is the same as in Vlach Romani except that, in Indic languages, the ending is connected to the root word while in Romani it is connected to the extended stem:


  • bijav-es-ko dives (Vlach, R.) / bjao-ka divas (H.) [wedding day]
  • bijav-es-ki rat (Vlach, R.) / bjao-ki rat (H) [wedding night]

The dative ending –ke, the locative endings –te and –e (archaic) and a few verbal endings are the same in Romani and Bengali.


  • dža! (R., H., Beng.) [go!]
  • amen geľam (R.) / amra gelam (Beng.) [we went]

It is understandable that, for a thousand and more years, since the time the Roma left their original Indian homeland, the phonetic and grammatical structure of Romani has changed. Most of the changes are subject to rules and, therefore, in words that are pronounced in another way, we can find an etymological connection. For example, the voiced aspirates (bh, dh, gh, and others), which other Indic languages have kept, have disappeared from Romani and generally have been changed to unvoiced aspirates:


  • gham (H.) [heat of the sun] / kham (R.) [sun]
  • dudh (H.) / thud (R.) [milk]

Retroflex (articulated with the tip of the tongue curled upward and back against or near the juncture of the soft and hard palates) sounds characteristic of Indic languages have, for the most part, changed to "r" or their retroflexion has been lost:


  • doi (H.) [wooden spoon]
  • roj (R.) [spoon]

We could devote many more pages to the linguistic correspondence between Romani and other modern Indic languages; they are so important and numerous that there can be no doubt about the Indian origin of Roma.

The autonym Rom is itself Indic. Etymologically it is connected with the caste designation Dom (Ddom).

Apart from the language, many other important signs point to the Indian origin of the Roma: their anthropological type (including the prevalent frequency of blood groups), the traditional professions of Roma and Doma; important socio-cultural institutions and unwritten laws and customs, including remnants of specific religious practices. One example is the institution of the group court called kris of the Lovara and Kalderaš Roma which functions like the panchayat, the traditional tribal court in Indian villages. Laws about "caste distance" – endogamy and the prohibition of commensality (eating together) – until recently and, perhaps, still kept up today in traditional Roma society, are typical of traditional Indian society (compare žužo Roma, Rom, jati). According to the command of endogamy, members of one jati – one Roma clan or one Roma sub-ethnic group – may not marry a girl from a clan where a different "caste" profession prevails or which follows different rules (or different degrees of keeping to the rules) of ritual cleanliness. The prohibition against commensality states that a member of a jati (caste) who consider themselves as "higher" or ritually "cleaner" will not eat in the home of a member of a caste who are considered "less clean", "lower".

It is thus indisputable that Roma come from India. When, though, they appeared in Byzantium (probably in the eleventh century, perhaps earlier), the empire extending over the vast Eurasian region, either they themselves forgot their Indian origin or the Byzantines did not find out about them – or there was an understandable lack of communication between the members of such different cultures. If we can judge from the appellations by which the Roma in Byzantium were labeled, the Byzantines identified them with members of a sect of magicians – Athingani – (from which later appellations such as "Cikán", "Tsigane", "Zigeuner", "Zingari", etc.) or with Egyptians (hence "Gypsy", "Gitan", "Jifti", etc.) Both exonymous terms for Roma were already used in Byzantium: first, Athingani, later, both Athingani and Giptoi/Jifti.

If Roma forgot their Indian origin during their migration to Europe, the question "Why?" arises. We often come across this question.

There are several answers. In the first place, we might ask ourselves where our great-great-grandfathers came from. If we are not especially interested in the genealogy of our family, we will probably not know. If the genealogy of our own family does interest us, we will have to search through birth, marriage and death registers, which means written documents. If the Czechs know the beginnings of Czech history, the Germans the beginnings of German history, the English English history, it is because they learned their history in school or their parents told them what they themselves remembered from school. The knowledge of a limited circle of specialists versed in letters was disseminated by various institutional channels (first of all, in school) to the broad public. If history is not saved in written documents – "litera scripta manet" (what is written remains) – or in elaborate oral accounts passed on by special "bards", then generally "family memory" does not reach back farther than four or five generations.

When the Roma left India, they did not write chronicles of their history nor did they have "bards" who, through oral accounts, would keep people from forgetting noteworthy stories. For that matter, not keeping chronicles was typical. On the other hand, it is interesting that, even today, members of some Dom jatis (castes of Roma relatives), e.g. the Dom-Mirasis, specialise as "bards". To musical accompaniment they sing songs of the history of their jajmans (patrons). These generally belong to a varna (class) of Brahmans (originally priests and intellectuals) or Kshatriya (originally rulers and fighters) or Muslim farmers. Dom-Mirasis earn their living as professional bards. More precisely, they are employed by "customers/clients" whose family history the "bards" must know by heart and sing about. The history of a particular clan is then passed on only in entertaining stories which are forgotten over the next few generations.

A further reason that Roma do not say that "India" is the land they come from is that they very probably do not know the term "India", just as even Indians did not know the name or, until recently, use it - especially Indians without an education. The term "India" is exonymous, i.e., not the inhabitants but members of other nations used it for the Indian subcontinent – first the Persians, then the Greeks, the Romans, the Arabs, the Byzantines, the English and, gradually, the whole world adopted it. The name for the whole country - or rather the geopolitical area to which various western states came in contact in the course of time – was taken from the river "Indus" and then passed on and disseminated. Greek and Roman travellers, historians and geographers had taken the name from the Sanskrit word sindhu, meaning "ocean", "great river".

The ancestors of today's Roma could hardly know the name "India" and could scarcely identify with some Indian state as imagined by Europeans. Until recently, and, perhaps, still today, people in a traditional Indian village identified, first of all, with their jatis, then with the place/area they came from, possibly with their ethnic origin (Punjabi, Bengali, Marathi, etc.) – and, earlier, with the rajah's or maharaja's mini-state in which they lived.

It is surprising that in two documents which we know about now, the "Indian origin" of Roma is presented. One is in a chronicle of the Italian city of Forli from 1422: "Aliqui dicebant, quod erant de India" – ("Some said that they are from India") (Muratori, Script. Rerum italica XIX, 8. – Horváthová 1964).

The other document was pointed out to us by the chief rabbi of the Czech Republic, Mr. Karol Efraim Sidon. In Hebrew writing, the "Chronicle of the World" speaks of the events in 1602 when King Philip banished the Spanish kushim – blacks. (Today's Roma in Spain are called Kale – Blacks!)

"King Philip of Spain banished all kushim (blacks) who lived in his kingdom so they would travel to their land, India, the place where they were born, their ancient land. There were more than 200,000 of them. And he did it against their will. And most of them died of hunger along the way and many of them were killed by robbers." (Romano džaniben 1-2/2000, pp. 6-7).

For the reasons described above, it seems unlikely that a few Roma would have stored their original "Indian" homeland in their memories. India could be a symbol for Europeans of a distant, unknown land inhabited by dark-skinned people with an incomprehensible language and a strange "barbarian" culture. And so thoughts of India as the original homeland of the Roma were just as accidental as speculations about the origin of Roma being Nubia, Ethiopia, Egypt or even "the lost continent of Atlantis".

When and why the Roma left India has not yet been convincingly clarified. The American linguist T. Terrence Kaufmann asserts that the Roma left India in the fourth century B.C., that is, at the time the army of Alexander the Great invaded northern India. Where, then, did the Roma live for fifteen hundred years before they appeared in Byzantium?

Almost every historical publication about the Roma cites a legend recorded by the Persian poet Firdausi in his extensive work, "Shah Namah" ("The Book of Kings") (1011). According to the legend, an Indian ruler was supposed to give king Bahram V (420-438) 10,000 Luri (musicians) who were to entertain his working people with music. Shah Bahram gave the musicians corn seeds for sowing and beasts of burden so they would settle and begin to farm the land. The musicians apparently ate the seeds and the following year returned for more. Naturally, the shah ordered them to roam through the world on their donkeys since they were unwilling to farm. Other historians recorded a similar legend, e.g. the Persian historian Hamza Isfahani in a work (961) and an Arabic historian at Talibi (1020). In all three works the number of musicians differs as well as the name of the Indian ruler. For the bestowal of the musicians, apart from the name Luri, they used the name Zott, which is the Arabic way of pronouncing the Indic word Jat. Today's Jats in India are farmers; they are thought to be of Scythian origin. During the reign of Bahram V. they belonged to the Sindhi population; some cultivated water buffalo; some fished. Since the Jats were the first inhabitants of "India", Arab raiders met them and probably gave their ethnonym to all Indians. At the present, the Russian Roma linguist, Lev Cherenkov, is inclined to accept the hypothesis of "musicians given to Bahram".

The English Rom Donald Kenrick also believes that Roma could be the descendants of the Zotts, who later settled with their water buffalo in a marshy area to the south of Basra. Between 820 and 834, they even created a kind of independent small "state". They rose up against the Arabs, but their rebellion was suppressed by the Arab caliph Mutasim. Many rebels were executed and the rest were deported to Ainzarba, a city on the Byzantine border. In 855, Ainzarba was conquered by the Byzantines and the Zotts were deported to the interior of Byzantium.

These scholars and other specialists base their hypothesis on linguistic theory. In Romani, there are a great number of words taken from Persian, which means that the Roma must have been in contact with the Persian-speaking population over an extended time. On the other hand, Romani contains few Arabic words. The above-mentioned scholars have concluded that Roma were already living in Persia in the fifth and sixth centuries – under the reign of Bahram V and his descendants of the Sassanidae dynasty, that is, before Persia was conquered by the Arabs in the seventh century.

There is a convincing argument against this theory: on one hand, the Arabs did conquer Persia, but Arabic never penetrated into normal society, nor did it replace Persian in its function as an official and literary language. Arabic was used by a few intellectuals (e.g. Hamza Isfahani, who wrote in Arabic) and primarily religious officials. Roma did not have to come in contact with any Arabic on the Persian lands at that time or afterwards when the Arabs conquered the land.

Roma did not have to come into meaningful contact with Arabic even if some clans of metalworkers, musicians and traders of beasts of burden went into the service – or were forced to go into the service – of Arab armies. Actually, in an "Arab" army usually only the leaders were ethnic Arabs while the rank-and-file mercenaries were recruited from various ethnic groups, mainly from the Iranian population.

Roma professor Ian Hancock le Redžosko of Texas University in Austin (USA) believes that Roma – originally Rajput fighters whose army was composed of a great variety of castes – left India because of the Muslim invasions. Troupes led by General al Qasim began the invasions with the conquest of Sindh in 712. The wars peaked with twenty-one border raids led by Mahmud of Ghazni (beginning of 11th century). Hancock dates the departure of the Roma from India back to exactly the time of Mahmud. Hancock bases his theory on linguistics: according to him, modern Indic languages lost their neuter gender, and neuter words were absorbed by the masculine gender. Hancock (2001) presents a table in which he compares the genders of Romani words with related Hindi words. His theory is interesting, but if Athingani (very probably Roma) were already in Byzantium in the eleventh century, they would scarcely have left India in the same century.

It is not out of the question that the exodus of the Roma from India was caused by several factors at once (Hübschmannová 1972). One of them could have been famine. Even today, famine periodically strikes the desert land of Rajasthan and drives out the starving population to look for food in other parts of India. Communities of relatives (clans or "breakaways" from various jati) migrated together. In Rajasthan, perhaps also in neighboring Punjab, ancestors of Roma clans apparently dwelled quite long before they left India. (The resemblance of Romani to the main dialect of Rajasthan, Marvari, has not yet been researched.)

Why would members of musical, metalworking, trading and other closely related castes head back to the west? Probably for some suitable chance of employment. Perhaps they had the possibility of plying their traditional trades in the army or in commercial caravans. Both the army and commercial caravans needed to be accompanied by blacksmiths, musicians, traders of beasts of burden, healers, entertainers who would chase away the destructive boredom of the "team" during the inevitable idle time on the road. And it is exactly these professions that make up the traditional livelihoods of Roma clans and sub-ethnic groups (jatis). Byzantine sources and, later, various documents from all over Europe speak of these trades in connection with Athingani (Roma). Still today, Roma earn their living as musicians, blacksmiths, horse dealers, circus performers and puppeteers wherever they can. Therefore, it is quite certain that they brought their professions from India. What is left is to ascertain why and how.

Let's assume that one part of the Roma (ancestors of today's Roma) left India either to accompany commercial caravans or as recruits in Muslim armies. (Just a remark: If we reproach them for insufficient "patriotism", that would be an anachronism. "Patriotism" – identification with one's homeland – is a much later development) However, the majority of "close relatives" of the Roma remained in India.

Kinship with Roma is being sought, e.g., by political representatives of the Banjaras, ambulant tradesmen who, among others, furnished the army with salt. Other "kin" are Gade Lohars, ambulant blacksmiths who work in the same way, i.e., bešindos – in a sitting position - as Roma blacksmiths. Sapéré/sanpválé may also have common Roma ancestors. Sapéré (sanp – snake, comp. with Romani sap) exhibit dancing cobras, but, mainly, they can heal snake bites and work with poison. Several Byzantine sources mention this Athingani profession. But since the type of snakes Sapéré worked with do not exist in Europe, the "snake-keeping" trade disappeared. (It has been kept up in Egypt.) Undoubtedly various castes of musicians are related to the Roma. The Dom-Mirasis provided the music for the most important religious holidays such as the birthdays of "Krishna" and "Rama"; Badis are musicians and also, in some castes in India, circus performers. (Cannot the old Romani last nameBadi originally have been a caste name?); Biharis are also circus performers (analogy between caste names and the common Roma last name Bihari?)

Members of all of these castes are considered descendants of the Adivasis, the original population of India, who lived in the time of the "Mohenjo-Daro" civilisation, one of the oldest in the world (comp. Rom), and so it can be said that they were co-founders of that civilisation. Various ethnic groups were involved in its creation. We will probably never know their original names, but those names have nothing in common with the names of contemporary castes. Or have they?

The Indian origin of the Roma, on one hand, was irrefutably established nearly two and a half centuries ago. On the other hand, we still do not know precisely when the Roma left India, why they left, or to which level of society they belonged. It is decidedly not simple to verify, but through the cooperation of linguists (experts in Indic languages in their various stages of development and, of course, experts in Romani), historians (and those include specialists in Indian history as well as Persian, Byzantine and Arabic history), ethnologists and others, it might be possible to arrive at a conclusion which comes as close as possible to the truth.

Although specialists have known for a long time where the Roma come from, this fact has not yet penetrated into the general consciousness. Even many Roma have not had the opportunity of learning (e.g., in school) about their land of origin. In popular-science literature half truths are still passed on about how Roma belonged to the lowest, "untouchable" castes on Indian soil. That statement does not take into account the length of Indian history, in which today's "untouchables" formerly belonged to the co-creators of an advanced civilisation. Only after they were gradually subjugated by "Aryan" Indo-Europeans, did they, as defeated people, find themselves on the edge of society. Exactly because the pre-history of the Roma has not been properly evaluated, some Roma deny that India is their land of origin and look for their origin, e.g. in Palestine, in Egypt, etc. On the other hand, there are more Roma who find in their Indian origin one of their sources of national pride. One of these, for example, is the Russian Roma poet (born in Latvia) Leksa Manush. He wrote a paraphrase of one of the two oldest Indian epic poems, "Ramajanam". Naturally even Roma historians such as the above-mentioned Ian Hancock, the late Bartoloměj Daniel or Jana Horváthová and others are aware of the Indian origin of Roma and try to contribute to filling in the picture of "Indian" Roma history.

In India there is a very strong interest in Roma. In the capital of the state of Punjab, Chandigarh, a Roma centre has been functioning for the past 70 years. It was originally founded by politician and diplomat W.R. Rishi. It publishes the magazine "Roma" and Roma literature. It published, for example, Manush's modern version of "Ramajanam".

On Indian soil there have been international Roma cultural festivals. One of them took place in Chandigarh in 1993. It was addressed by the then prime minister, Indira Gandhi.

A good understanding of Indian culture and history can contribute to the national pride of the Roma.


ab Hortis, Samuel Augustini (1775-1776 / 1994) Cigáni v Uhorsku 1775 / Zigeuner in Ungarn 1775. Bratislava.
Fraser, Angus (1998) Cikáni. Praha.
Hancock, Ian (2001) K upřesnění doby odchodu mluvčích protoromštiny z Indie. In: Romano džaniben 3/4, pp. 6-12.
Horváthová, Emilia (1964) Cikáni na Slovensku. Bratislava.
Hübschmannová, Milena (1972) What can Sociology suggest about the Origin of Roms. In: Archiv orientální 40/1, pp. 51-64.
Kenrick, Donald (1994) Gypsies: from India to the Mediterranean. Paris.
Majumdar, RC (ed.) (1960) The classical accounts of India. Calcutta.
Matras, Yaron (1999) Rüdigerův příspěvek k objasnění původu romštiny. In: Romano džaniben 3/4, pp. 80-102.
(2000) Rozhovor s Karolem Efraímem Sidonem, vrchním rabínem v České republice. In: Romano džaniben 1/2, pp. 3-11.
Image Printable version
Image Choice of Indian Etyma