Romany Language
The Romany Alphabet

    Romany language has not been written until recent times and remained for centuries a spoken tongue only. The first book published in Romany was the Bible, of which there are several versions in both Latin and Cyrillic characters, and although not having a unified writing system, the texts may be taken as reference for a standard alphabet in both scripts.

    In spite of the general tendency to romanization of non-Latin alphabets, some attempts have been made to invent unsuitable alphabets containing glyphs that cannot be found in any kind of keyboard, making the normal writing very difficult if not impossible. Actually, the most appropriate script for Romany is Cyrillic, however, also Latin alphabet is completely sufficient for a correct spelling of this language, since the characters having diacritics (namely caron/hácek or acute accent) may be replaced by digraphs or trigraphs when such graphic marks are unavailable.

    Owing to the fact that Roma have a longstanding presence in every European country and mostly in Slavic nations, a large number of Roma population is more familiar with Cyrillic rather than Latin alphabet, and both may become official – as it was the case of Serbo-Croatian, now separated into Serbian and Croatian for political rather than strictly linguistic reasons. Therefore, here we present the Romany alphabet developed in such a way as to cover all the possible phonetic variations existing today, in the simplest and most accurate way.

 

The Romany Alphabet in Latin and Cyrillic Characters

Latin Script

 

Cyrillic Script

A

a

A

a

B

b

Б

б

C

c

Ц

ц

Ć

ć

Ћ, Чь

ћ, чь

Č

č

Ч

ч

D

d

Д

д

E

e

E, Э

e, э

 

ê

 

є

F

f

Ф

ф

G

g

Г

г

H

h

Һ

һ

I

i

И

и

K

k

K

к

Кh

kh

Къ

къ

L

l

Л

л

M

m

M

м

N

n

H

н

O

o

O

o

P

p

П

п

Ph

ph

Пъ

пъ

R

r

P

p

Ř

ř

Ръ

S

s

C

c

Š

š

Ш

ш

T

t

T

т

Ť

ť

Тъ

тъ

U

u

У

у

V

v

B

в

X

x

X

x

Y

y

Й

й

Я

я

Ë

ë

Ю

ю

Ы

ы

Z

z

З

з

Ź

ź

Ж

ж

Ž

ž

Џ , Ӂ

џ , ӂ

Phonetic Description Of Each Letter  
   
A, a  
   
Vowel, equivalent to:
Russian, German, Spanish ‹ a ›,
Hungarian ‹ á ›.
Examples: akaná / aкaнa (now); rat / paт (blood).
 
   
   
B, b / Б, б  
   
Consonant, voiced bilabial plosive.
Equivalent to:
Russian ‹ б ›,
German, English ‹ b ›. In some dialects it sounds closer to ‹ v ›, almost like Spanish ‹ b ›.
Examples: balbal / балбал (wind); abiáv / aбияв (wedding).
 
   
 
C, c / Ц, ц  
   
Latin alternative spelling: Ts, ts.
Consonant, voiceless alveolar affricate.
Phonetically equivalent to:
Russian ‹ ц ›,
German ‹ c ›, ‹ z ›,
Hungarian ‹ c ›, ‹ cz ›,
Italian ‹ zz ›.
Examples: lanco / лaнцo (chain); cirde / цирдэ (pull).
 
   
   
Ć, ć / Ћ, ћ  
   
Cyrillic alternative spelling: Чь, чь (Russian keyboard). Latin alternative spelling: Cj, cj; also Tj, tj.
Consonant, voiceless alveolo-palatal affricate.
Phonetically equivalent to:
Serbian ‹ ћ ›,
Croatian ‹ ć ›,
Approximately similar to Hungarian ‹ ty ›.
Examples: ćin / ћин (buy); bući / бући (work).
This phoneme is rare and almost lost in several dialects; having become closer either to ‹ č / ч › or else to ‹ k / к ›,
so that the examples above may be pronounced respectively:
čin / чин or else kin / кин (buy); buči / бучи or buki / буки (work).
 
   
   
Č, č / Ч, ч  
   
Latin alternative spelling: Ch, ch.
Consonant, voiceless post-alveolar affricate.
Phonetically equivalent to:
Russian ‹ ч ›,
German ‹ tsch ›,
English, Spanish ‹ ch ›,
Hungarian ‹ cs ›.
Examples: čačes / чaчэc (truly); niči / ничи (no).
 
   
 
D, d / Д, д  
   
Consonant, voiced alveolar plosive.
Phonetically equivalent to:
Russian ‹ д ›,
German, English, Spanish, Hungarian ‹ d ›.
Examples: dur / дур (away); dey / дэй (mother).
In some dialects, when preceding ‹ e › or ‹ i › it is pronounced as a voiced alveolo-palatal affricate, like Serbian ‹ ђ ›, Macedonian ‹ ѓ ›, Croatian ‹ đ ›.
However, it is not useful to add this phoneme as a separate character to the alphabet, as it is only a dialectal variation of ‹ d ›.
 
   
   
E, e / Э, э  
   
Vowel, equivalent to:
Russian ‹ э ›,
German, Spanish ‹ e ›,
Hungarian ‹ é ›.
Examples: kher / къeр (house); cirde / цирдэ (pull).
 
   
ê / є  
   
Vowel, never placed as initial, so that the upper case (Ê / Є) is written only when all characters are capitalised.
Depending on the dialect, it may be equivalent to ‹ e › or ‹ o ›.
Examples: bêrš / бєрш (year); šêro / шєpo (head).
These examples may be pronounced respectively berš / бeрш or borš / бoрш (year); šero / шepo or šoro / шopo (head).
The addition of this character aims at covering both phonetic variations.
 
   
   
F, f / Ф, ф  
   
Consonant, voiceless labiodental fricative.
Phonetically equivalent to:
Russian ‹ ф ›,
German, English, Spanish, Hungarian ‹ f ›.
Examples: foro / форо (town, city); eftá / eфтa (seven).
 
   
 
G, g / Г, г  
   
Consonant, voiced velar plosive.
Phonetically equivalent to:
Russian ‹ г ›,
German, Hungarian ‹ g ›,
English ‹ g › as in “give”,
Italian ‹ g › as in “gatto” or ‹ gh › as in “ghiro”.
Examples: gad / гaд (shirt); guglimós / гуглимос (sweetness).
 
   
   
H, h / Һ, һ  
   
Consonant, voiceless glottal fricative.
Phonetically equivalent to:
German, English ‹ h ›.
Examples: halyar / һальяp (understand, feel); hodina / һодина (rest).
This phoneme is not very common.
 
   
   
I, i / И, и  
   
Vowel, equivalent to:
Russian ‹ и ›,
German, Spanish ‹ i ›,
Hungarian ‹ í ›.
Examples: čiriklíирикли (bird); iven / ивен (winter).
 
   
   
K, k / К, к  
   
Consonant, voiceless velar plosive.
Phonetically equivalent to:
Russian ‹ к ›,
German, English, Hungarian ‹ k ›.
Examples: kokalo / кокало (bone); šukar / шукар (beautiful).
 
   
   
Kh, kh / Кь, кь  
   
This digraph represents a voiceless velar plosive consonant.
It is a strongly aspirated ‹ k ›, ‹ к ›.
Examples: kham / кьам (sun); kher / кьер (house).
 
 
L, l / Л, л  
   
Consonant, alveolar lateral approximant.
Phonetically equivalent to:
Russian ‹ л ›,
German, English, Spanish, Hungarian ‹ l ›.
Examples: lil /лил (letter, written paper); balbal / балбал (wind).
In some dialects it is pronounced in a velarized way (“dark l”), almost like Polish ‹ ł ›.
   
   
M, m / M, м  
   
Consonant, bilabial nasal.
Phonetically equivalent to:
Russian ‹ м ›,
German, English, Spanish, Hungarian ‹ m ›.
Examples: mišto /мишто (well); amal / амал (friend).
 
   
   
N, n / Н, н  
   
Consonant, alveolar nasal.
Phonetically equivalent to:
Russian ‹ н ›,
German, English, Spanish, Hungarian ‹ n ›.
Examples: nak / нaк (nose); angar / aнгaр (coal).
 
   
   
O, o  
   
Vowel, equivalent to:
Russian, German, Spanish ‹ o ›,
Hungarian ‹ ó ›.
Examples: mol / мол (wine); oxto / oхтo (eight).
 
   
 
P, p / П, п  
   
Consonant, voiceless bilabial plosive.
Phonetically equivalent to:
Russian ‹ п ›,
German, English, Spanish, Hungarian ‹ p ›.
Examples: papin / пaпин (goose); piperi / пипepи (pepper).
 
   
   
Ph, ph / Пъ, пъ  
   
This digraph represents a voiceless bilabial plosive consonant.
It is a strongly egressive ‹ p ›, ‹ п ›.
Examples: phen / пьeн (say); phral / пьрaл (brother).
 
   
   
R, r / P, p  
   
Consonant, rhotic alveolar trill or alveolar tap.
Phonetically equivalent to:
Russian ‹ p ›,
English, Spanish, Hungarian ‹ r ›.
Examples: ruv / pув (wolf); baró / бaрo (big).
 
   
   
Ř, ř / Pъ, pъ  
   
Latin alternative spelling: Rh, rh.
The guttural rhotic consonant, namely, the “guttural r”.
Phonetically equivalent to:
German, French ‹ r ›.
Examples: řom / ръoм (man, husband); manřo / мaнръo (bread).
In most dialects this phoneme is not present and is pronounced in the same way as ‹ r ›.
 
   
 
S, s / C, c  
   
Consonant, sibilant voiceless alveolar fricative.
Phonetically equivalent to:
Russian ‹ c ›,
English, Spanish ‹ s ›,
German ‹ s ›, ‹ ß ›, Italian ‹ ss ›,
Hungarian ‹ sz ›.
Examples: sap / сaп (snake); asaimós / acaймос (laughter).
 
   
   
Š, š / Ш, ш  
   
Latin alternative spelling: Sh, sh.
Consonant, voiceless palato-alveolar fricative.
Phonetically equivalent to:
Russian ‹ ш ›,
English ‹ sh ›,
German ‹ sch ›,
Hungarian ‹ s ›.
Examples: šon / шон (month); mašo / мaшo (fish).
 
   
   
T, t / T, т  
   
Consonant, voiceless dental-alveolar plosive.
Phonetically equivalent to:
Russian ‹ т ›,
German, Spanish, Italian, Hungarian ‹ t ›.
Examples: tató / тaтo (hot); vorta / вopтa (straight, correct).
 
   
   
Ť, ť / Tъ, тъ  
   
Latin alternative spelling: Th, th.
It is a strongly egressive ‹ t ›, ‹ т ›.
Approximately similar to Hungarian ‹ ty ›.
Examples: ťem / тъeм (nation, people); ťav / тъaв (thread).
In some phonetic variants it is pronounced in the same way as ‹ c ›.
 
   
 
U, u / У, у  
   
Vowel, equivalent to:
Russian ‹ у ›,
German, Spanish, Italian ‹ u ›,
Hungarian ‹ ú ›.
Examples: purum / пурум (onion); suno / суно (dream).
 
   
   
V, v / B, в  
   
Consonant, voiced labiodental fricative.
Phonetically equivalent to:
Russian ‹ в ›,
German ‹ w ›,
English, Spanish, Hungarian ‹ v ›.
Examples: vurdon / вурдон (wagon, car); iv / ив (snow).
 
   
   
X, x  
   
Consonant, voiceless uvular fricative.
Phonetically equivalent to:
Russian ‹ x ›,
German ‹ ch ›,
Rioplatense/Castilian Spanish ‹ j ›.
Examples: xaben / xaвeн (food); xoxaimós / xoxaймос (lie).
 
   
   
Y, y / Й, й - Я, я - Ë, ë - Ю, ю - Ы, ы  
   
In Latin alphabet it is a semivowel, equivalent to:
German, Hungarian ‹ j ›,
English ‹ y ›,
Spanish final ‹ y ›.
Examples: řayo (heaven); kšey (daughter).
In Cyrillic alphabet they correspond to diphthong structures as follows:
The character ‹ й › represents the equivalent sound in Russian, and is written after a tonic vowel (aй, eй, oй, yй).
Examples: рьайо (heaven); кшeй (daughter); рьoй (spoon); сaпуй (soap).
The characters ‹ я ›, ‹ ë › and ‹ ю › represent the same diphthongs as in Russian, respectively equivalent to ‹ ya ›, ‹ yo › and ‹ yu › in Latin script.
The character ‹ ы › is a semivowel, equivalent to the same glyph as in Russian and represents some dialectal variants in which the Latin character ‹ y › is pronounced as German ‹ ü ›.
 
   
 
Z, z / З, з  
   
Consonant, voiced alveolar fricative.
Phonetically equivalent to:
Russian ‹ з ›.
Examples: zor / зoр (strength); zeleno / зeлeнo (green).
 
   
   
Ź, ź / Ж, ж  
   
Latin alternative spelling: Zh, zh.
Consonant, voiced alveolo-palatal fricative.
There are two main pronunciation patterns for this character, according to the dialect:
1) As a palatalized ‹ r ›, phonetically similar to:
Czech ‹ ř ›,
Polish ‹ rz ›,
Sicilian ‹ r ›;
2) In the same way as the Rioplatense Spanish consonant ‹ y ›, ‹ ll ›,
close to Hungarian ‹ zs ›.
Examples: źuvlí / жувли (woman, female); vuźó / вужo (clean).
 
   
   
Ž, ž / Џ, џ - Ӂ, ӂ  
   
Latin alternative spelling: Dzh, dzh; Cyrillic alternative spelling: Дж, дж.
Consonant, voiced postalveolar affricate.
Phonetically equivalent to:
Russian digraph ‹ дж › as in “джазз”,
German ‹ dsch ›,
English ‹ j › as in “jazz” or ‹ g › as in “gentle”,
Italian ‹ g › as in “già”.
Examples: žin / џин (read, count); ažuker / aџукэр (wait).
In some dialectal variants it is pronounced as a voiced palatal plosive, like Hungarian ‹ gy ›.
 
   
 

 

Other Alphabets

The Latin alphabet used by Yaron Matras -Romlex- is as well quite accurate phonetically, covering a wide variety of dialectal utterances. However, we prefer to replace the character ‹j› by ‹y›, as the first one may induce to misinterpretation within Roma who are familiar with graphemes of English, French, Spanish and other tongues. Indeed, even though ‹j› is a semivowel equivalent to a short ‹i› in most European languages, it has also a fully consonantal value in the ones mentioned above, being equivalent to ‹ž› in English, ‹ź› in French or Catalan, ‹x› in Castilian Spanish. The grapheme ‹y› does not present this difficulty.

The so-called "standard" alphabet, promoted by the IRU, besides the lack of accuracy and the presence of nonsensical characters that cannot be found in any keyboard, contains also other improper graphemes such as ‹rr›, that is meaningless in Romany, or ‹q›, which has an unnecessary double value besides the fact that such character is used in conventional phonetics to represent the Arabic "qof" and similar sounds, which in Romany is exceptionally present in some dialectal pronunciations of ‹ć› and is not related to the alleged equivalences intended by this alphabet. There are other flaws that we will not consider now, as this alphabet is to be discarded as a practicable Romany script.

Other unsuitable script which has been suggested for Romany language is Devanāgarī, just because this abugida writing system is in modern times related to Sanskrit, which in turn has contributed to the formation of Romany language along with other tongues spoken in the Indus Valley in early times, such as Greek. Like many other theories regarding Roma, this one is based on erroneous patterns and anachronistic historical incongruity. Just to mention a few:
·It is absolutely certain that Roma have never used Devanāgarī, as this writing system did not exist until more than one century after Roma had left the Indus Valley and by the time it appeared Roma were already sojourning in Asia Minor, on their way to Europe.
·Devanāgarī evolved from Brāhmī, which most scholars agree that developed from Imperial Aramaic, the script used by the Persian Empire and widely known in the Indus Valley during the Achaemenian rule. Subsequently, Greek alphabet has been the most used in that region.
·Devanāgarī is connected to Sanskrit only in recent times, since the British colonial rule over India. Indeed, before that time Sanskrit had not any standard alphabet or writing system whatsoever.
Therefore, Imperial Aramaic and Greek alphabets were the most widely used during the development of Romany language, not Devanāgarī.
As we have stated above, there is a general tendency to romanization of non-Latin alphabets, and in some languages with a longstanding literary and cultural tradition expressed in other writing systems, these had been replaced by European alphabets – as for example the Turkic tongues, in which the Arabic-based script has been superseded by Latin or Cyrillic. The idea that Roma might adopt Devanāgarī is unrealistic.