1. Marcel Courthiade is a non-Rom professor of the Romani language at the Sorbonne, INALCO. He developed a theory of strata of the Romani dialects. In his opinion (which has become an international dogma), dialects of the Arli type take first place, while those of the Kalderash type take last place. He based his analysis upon only 9 Romani words. He compared the 9 Romani words without relating them to their original source (Indian languages, of course). It is in such a manner that the best Romanian dialect was decided, by comparing the dialects among themselves, without reference to Latin, for instance. Even though the linguist Morris Swadesh, creator of Glottochronology (a), developed a list (b) of 100 words (originally 215) which represented the base vocabulary of all the languages of the world, M. Courthiade compared 9 Romani words between themselves, with their variants. In any case, the lexicon of the Romani language contains around 8000 words, about as many words as the Hebrew Bible.

List of criticisms of M. Courthiade:

1) „Phirdiom” – „I walked”

He did not choose the best verb. All serious linguists choose the verb “to do”. If this verb were chosen, one would see that in India it is pronounced KĂR (“do!”, imperative), exactly as the Kalderash Roma pronounce it (stratum III). The Arli Roma say KER, a less frequent pronunciation in India. Thus, M. Courthiade’s theory is damaged. The pronunciation -diom versus -dem is not conclusive, because in India, the conjugation of the verb “to do” is VO KĂRDIYA – “he did”, just as the Kalderash Roma pronounce it. In India, the pronunciation “ kărdea” also exists, but I will be discussing this in a future article. In Hindi, the conjugation KĂRDIYA is used for the first, second and third persons singular. Thus, Indians say: “me kărde(o)m” – “I did”, “ tu kărdean” – “you did”, “vo/voi kărdia” – “he/she did”. In a following article, I will show that the Roma inherited these forms of pronunciation and conjugation from India.

2) -imos, the cornerstone, and bone – kokalo

The suffix -imo(s) has been the stumbling block for all linguists and tsiganologists. They assumed that the Roma acquired it from Greece following a long stay in the Byzantine Empire. A noun is constructed from an adjective with this suffix: shukar – beautiful, shukarimos – beauty, alternatively shukaripen. Or, mishto – well, mishtimos/mishtipen – kindness. M. Courthiade and all the tsiganologists and professors assumed that the Kalderash Roma acquired the suffix in Greece. Therefore, they placed the Arli-type dialects in first place because the -ipen suffix is Indian. Even though Mr. Gheorghe Sarau’s book (1) gives no origin for the -ipen suffix; a book which examines in detail the theory of strata developed by Marcel Courthiade.

But, because only those who seek may find, I have discovered that, in fact, the suffix -imos has a Sanskrito-Prakrit origin. Had the tsiganologists practised Yoga, they would have found that certain mystical Yogic powers (2) use the suffix -ima, for example “lagh+ima” – “easiness”, (lokh+imo) in Romani.

“Kokalo – bone” also thought to have been borrowed from Greek into Romani, appears as “kagkala” in Sanskrit – “skeleton”: = SE & script = HK & link = yes & beginning = 0

A changing into O

In India, the final A changes to O. For example, in Hindi we find the word “kala – black” (masc.); in Bengali this word takes the form “kalo – black”, and this form (kalo) is used by the Roma. This explains the alteration of the Sanskrit -ima form into the Romani -imo form. But Romani has retained the Sanskrit -ima form in some cases, such as “patsima – opinion”.

What M. Courthiade did not know was that the Kalderash Roma of Caras-Severin use the suffixes -imos and -iben in parallel. The Kalderash Roma say “hamos – food” and also “haben – food”. In this case, M. Courthiade’s theory is invalid. And there are tens of thousands of Kalderash Roma who use the suffix -ben. But M. Courthiade did not know this, and consequently (due to lack of interest or research), declared that the Arli-type dialects were the most archaic. Moreover, the Lovara Roma also use the suffixes -imo/-ipo in parallel.

The reference words with which M. Courthiade classified the Arli-type dialects as the most archaic examples are incorrect. If one chooses the Hindi word “pani” as the reference, why not choose the Hindi word with the pronunciation “kărdiya – he did”, because almost everywhere in India it is pronounced “kar”, not “ker”, as the Ursari Roma pronounce it, although in some parts of India it is pronounced “ker”. In all linguistic works, the verb “to do” is generally examined the most often; it is the most common example of conjugation in all grammars of Sanskrit, Hindi, Bengali and in comparative Indo-European studies. We see that the Kalderash Roma use the suffix -ben, although the Ursari Roma use the suffix -imos: it is not a matter of giving preference to one group of Roma based only upon a certain suffix, nor on an -imos suffix borrowed from Greece as M. Courthiade’s table suggested. It would have been better had the French researcher chosen the word “haben – food”, but then it would have been clear to see that the Kalderash Roma use it, and consequently, that the theory had cracks! The suffix -imo is not Greek but Sanskrit, as we learn from the work of Mr. Madhusudan Mishra. Mr. Mishra did not comment on Romani in his document, but we have found evidence there that proved that the theory that the Romani -ibe suffix was more archaic, to the disfavour of the -imo suffix, was incorrect.

Although Mr. Mishra (3) was not concerned with Romani, because the latter is an Indian language, it is governed by the same grammatical principles. Moreover, the Romani -imo suffix is considered by all linguists studying the Romani language to be borrowed from Greek, thus suggesting that the groups who use it are not speaking correctly. The discovery that the -imo suffix is Indian is ours, based on the work of Mr. Mishra. With the help of Mr. Mishra’s work, I have discovered that the Romani suffixes -imo, -ibe, and -po are in fact present in Sanskrit, in forms practically identical to those in Romani. The Indian suffix -ana (in “khana” – food) is in fact analogous to the Indian suffixes -ima, -pan and -po. In fact, the Romani word habe/habo/haipo/hamos – “food”, is suffixed differently to the Indian word “khana”, but all the suffixes are Indian, including -imos, not only the -iben and -ipen variants. The Lovari Roma, for example, indiscriminately use haben, hamos, hamo, hapo! I would be curious to know how M. Courthiade would classify the Lovari Roma. Moreover, the work of Mr. Gheorghe Sarau, in which are cited M. Courthiade’s examples, does not state the geographical origins of the -ibe/-ipo suffixes. The irony is that the -imo/-ima suffix is, if not the most ancient, for it is found in Apabhramsha Prakrit, at least as ancient as suffixes in the -iben group. Regarding the alteration of the final -a to -o (from the Prakrit suffix -ima), it reminds us of the alteration of the alteration of the Hindi adjective “kala – black” to “kalo” in Bengali, Nepali, Rajasthani and Romani. The Roma sometimes pronounce the suffix -imo as -ima (rishima – “conflict”, in the Crimean Roma dialect), just as in Sanskrit, and -imoh or -imas (feminine noun “anklimasta” – “exit” in Gurbet dialect; “patsima” – “opinion” in Kalderash).

In Mr. Mishra’s book about Apabhramsha Prakrit, we find for example “vank+ima”, deformation. In Sanskrit, there is “vankra – crooked”, but in the language of Nepal we find “bango – crooked”, deformed, just as in Romani. For how many centuries must the Roma have remained in India for the Sanskrit word “vankra” to have evolved into “bango” in Romani, just as in modern Indian languages?

3) Elision of the -n. Pani, a masculine noun.

  1. Courthiade decided that the dialects in which the word for “water” takes the form “pani” were more archaic than those which use the form “pai”. But in Sanskrit, this word takes the form पाय “paya” – “water” (4). What he forgot to say was that both in Hindi and Romani, the noun “pani” has the feminine ending (ni), but is masculine. In Romani, the masculine gender is indicated by adding the definite article “o” and in Hindi, it is known by long tradition (5). This phenomenon is considered as an exception in Hindi. In Romani, it is also exceptional. The conclusion is that, despite the hundreds of years that have elapsed since the exodus from India, the form of the word “pani” – “water” has been retained as masculine, despite its feminine ending, in both languages.

The Romani language confirms it itself

It is not necessary to look to the Indian languages to demonstrate the elision of the “n” in “pani/pai”. The Kalderash Roma name their language, “Romani shib”, as “Roma’i shib”. The Kalderash Roma also say “harku’i” – “copper”, in place of “harkuni”. Thus, the elision of the “n” is common. We may suppose that the Roma prefer easier pronunciations, abandoning the forms “khoni – tallow”, “kuni – elbow”, in favour of “khoi” and “kui”.

The comparative degree

  1. Courthiade chose the forms po/mai. “Po” is of Slavic origin and “mai” is Romanian. Examples: “mai mishto / po mishto” – “better”.

The only correct dialects are those which use -eder to express the comparative degree: “mishteder” – “better”. This form is derived from the Sanskrit comparative form “-tara”. M. Courthiade took no account of this issue.

The definite articles

  1. Courthiade did not manage to remember that all Romani dialects use the Greek definite articles “o” and “i”. Therefore, the dialects which he declared to be Stratum I are comparable to those of Stratum II or III. In consequence, all the dialects are equal, because they use the same Greek definite articles, just like the -imos suffix. But, as we have just shown, the -imos suffix is also Indian.

The article (6) on the linguistic archaeology of the -imo suffix, in which it was demonstrated that its origin is Indian and not Greek (Indo-European), was published in the Yearbook of the General Society of Anthropology of Prahova, no. 2/2017 under the title “The Roma people, the people of the language. Aspects taken into account in attempting to establish a literary language”, written by the current author.

1-Sarău, Gheorghe, 1997, Romii, India și limba romani, București: Kriterion.


3-Mishra, Madhusudan, 1992, A grammar of Apabhramsha, Delhi: Vidyanidhi Prakashan Delhi.


5- : „But exceptions exist. for e.g., the following end in -ī (-ee) but are masculine: पानी (paanī) = water पक्षी (pakshī) = bird.”

6- Marian Nuțu Cîrpaci, „Poporul rom, popor al limbii. Aspecte de luat în seamă în încercările de a stabili o limbă literară” în Anuarul societății prahovene de antropologie generală, nr. 2/2017, Centrul jedețean de cultură Prahova, Editura Mythos.


Glottochronology is a technique to calculate the temporal separation between two dialects.



Translated by Natalie Winter (Ivend), native Romanichal speaker.


Lasă un răspuns

Completează mai jos detaliile tale sau dă clic pe un icon pentru a te autentifica:


Comentezi folosind contul tău Dezautentificare /  Schimbă )

Fotografie Google

Comentezi folosind contul tău Google. Dezautentificare /  Schimbă )

Poză Twitter

Comentezi folosind contul tău Twitter. Dezautentificare /  Schimbă )

Fotografie Facebook

Comentezi folosind contul tău Facebook. Dezautentificare /  Schimbă )

Conectare la %s