you can login to your account or register with kal

Kurdish Alphabets and Transliteration

Dr. S. Rezaei Durroei,

Where systems of writing become identified (as often happens)
with unreasoning and unreasonable political, national or religious
passionsthere is little that the linguist can or should do.
(Berry 1968)

The signing of Kurdish accord in Washington in September has created a new hope for Kurds and it has risen the important debate of unifying Kurdish scripts. A new discussion has begun in Britain in the monthly Hetaw, and on VOA Kurdish service about the Alphabet and the need for the phonemes /gh (غ)/, /h (ح)/ and /' (ع)/ which are being outlawed by some extremist nationalists (because, they claim, they are borrowed from Arabic).

In one extreme they are people who argue for the need for a new alphabet (see Izady's proposal). Such proposals are more like spelling revolution than spelling reform. On another extreme there are those conservatives who believe that one should not change the achievements of the past and there is no need for reform or new alphabets. New alphabets are required principally for 3 purposes:

  1. to provide for the first time a means of writing languages as yet unwritten or virtually so.
  2. to provide auxiliary alphabets for languages which have already a standard script (e.g. Roman systems for Hebrew, Japanese, etc. ).
  3. to provide alternatives to standardized but for some reason inadequate writing systems (e.g. Spelling Reform).

In the case of Kurdish, there are already 3 alphabets in use: Arabic/Islamic, Latin/Roman and Cyrillic. The Cyrillic alphabet for Kurdish is the most specified and linguistically motivated alphabet for Kurdish which covers most of the necessary symbols for Kurdish, and uses a very limited set of letters with diacritics.

On the other hand the Arabic/Islamic script for Kurdish is the most under-specified script with many ambiguities in pronunciation, but with a rich history of Kurdish calligraphy. It is the only alphabet which has made its way to the teaching, education in the universities and official administration at the present time.

The Kurdish Latin script comes in between the two and its advantage is that it is closer to the spelling of English than the two others. Its main disadvantage is that it lacks some symbols. Even in Kurmanji literature like (Rizgar 1993) the need for such extra letters have been promoted. The claim of those (such as Izady in his proposal) that those aspirated sounds are not used in Kurmanji is not justified and Baran Rizgar has clearly shown by evidence from Kurdish Kurmanji corpus that those aspirated letters are being used and are needed.

If all the Kurds have used the Cyrillic alphabet, there was no need for such spelling reforms, because it is the only Kurdish script which has all those aspirated sounds. But it is wrong to assume, as it sometimes is done, that the choice of an orthography can be determined solely on grounds that are linguistically or pedagogically desirable. Where systems of writing become identified (as often happens) with unreasoning and unreasonable political, national or religious passions there is little that the linguist can or should do (Berry 1968).

In spelling reform, the majority of letters needed for a new alphabet can and should be drawn from the traditional alphabet in general use throughout the area (e.g. Roman, Cyrillic, etc.), and precise values should be assigned to each letter as far as possible in accordance with customary usage. The graphematic deficiencies of the source-alphabet can be supplied if necessary by such additions and modifications as

  1. extra letters from other scripts
  2. diagraphs and trigraphs (e.g. Welsh "ll", German "sch").
  3. diacritically modified letters (with supraliteral tilde, tashdid and subliteral macron, cedilla and postliteral circl, apostrophe).
  4. ad hoc supplements (restored, borrowed, adapted, invented).

Besides social and political factors, there are many scientific factors involved. Measures of typographical economy have been suggested from time to time and have included proposals for the abolition of capital letters. But capitals are signs usually non-phonemic of importance (Berry 1968).

The use of diacritically modified and conglomerate letters has been criticized on many grounds but especially on those of typographical, pedagogical, and psychological inadequacy. But the attitude in each case is clearly a matter of training and habit. the objections to diacritical marks may be summerized briefly: it is said

  1. that a multiplicity of such marks lessens the legibility of type by blurring the bold outline, especially the top outline of a word, and by demanding an individual circumnavigation of letters by the eye impedes word recognition and slows reading (Berry 1968),
  2. that letters with diacritical marks must separately cast and that such letters are less durable. One should also mention the problems in scanning such letters by automatic computer tools such as Optical Character Recognition (OCR).

The need for alphabetical economy is widely recognized. The specializations of form for initial, medial, and final, and compound letters in the Arabic alphabets are traditional examples of alphabetical extravagance. Alphabetical economy may be achieved notably by such devices as (1) letter groups, (2) redundant letters, (3) contextual conventions. Objection to the use of letter groups is usually on the grounds that they are wasteful of effort and can cause ambiguity. But some find them less than ideal, but preferable to the use of diacritics or special letters.

Izady in his proposal argues for the adequacy of his alphabet and he is sensitive to the use of anything present in Turkish alphabet and standards in a Kurdish script! Language and scripts are medium for communication, and such passionate nationalist arguments ignore the role of language as a social phenomenon and the extra needs for training, the history of the reforms and the present Kurdish scripts.

His proposal also ignores the linguistic justification for the presence of aspirated letters in Kurdish on the ground that some of them are only used in South Kurmanji dialects. If it is the case then why such letters exist in the Cyrillic alphabet which is used by Kurmanji speakers? As we mentioned Baran Rizgar in his dictionary also gives the statistics of the use of such letters in Kurdish and the need for representing them in Kurdish dictionary. Michael Chyet (Chyet 1997) also discusses the importance of aspirated letters for Kurdish language in his review of Kurdish Kurmanji dictionaries.

In passing it should be noted that it is no advantage for a language to be of an Indo-European language family and it is not a disadvantage for a language to belong to the Turco-Mongolian, East Asiatic tongues. The spelling of Western European languages have also many shortcomings and the necessity for such reform for those languages have been raised in the literature. There is no point for us to discard an already available system for a new one for reasons such as affinity between Kurdish and Indo-European languages. Kurdish belongs to the Iranian branch of Indo-Iranian branch and both Persian and Hindi use scripts which are not even close to the Roman or Latin scripts. But Izady rightly mentions that the Arabic script for Kurdish causes many problems for representing Kurdish.

Another criticism to Izady's is that he argues against the use of diacritics and he himself uses diacritics for vowels in his proposal. Last but not least, the proposal does not give any information about the precedence of the letters and the alphabetical order.

One moderate approach to extending Kurdish script is to consider the Latin Script of Kurdish and add a limited set of symbols for representing Kurdish letters. As part of a Machine Translation project for Kurdish - See (Rezaei-Durroei & Fatah 1998) and (Rezaei-Durroei 1998) - we have considered some extra letters for the Kurdish Latin script. A transliteration scheme that covers all letters and details of Kurdish script has also been developed. This transliteration is also useful for Kurdish communication on the Internet and it may also be considered as a basis for a Kurdish script with no use of capital letters or diacritics.

As we mentioned earlier, measures of typographical economy have been suggested in the literature from time to time and have included proposals for the abolition of capital letters. But we agree with Berry (1968) that capitals are signs usually non-phonemic of importance, especially for marking Proper nouns. But they can be captured by another mechanism. see October Issue of Kurdica Newsletter for further details on the transliteration scheme.


  1. Berry, J. (1968) The making of Alphabet. In Fishman J. A. (1968) Readings in the Sociology of Language. The Hague, Paris: Mouton. pp. 737-753. Appeared in the proceedings of the 5th Int. Congress of Linguistics (Oslo, Univ. Press, 1958), pp. 752-764.
  2. Chyet, Michael L. (1997) Kurmanji Kurdish Lexicography: a Survey and Discussion.
  3. Izady, Mehrdad R., A Pan-Kurdish Alphabet .
  4. Rezaei Durroei, Siamak; Fatah M., Rebwar (1998), Role of Kurdish Language and Planning for It, In the proceedings of the 2nd conference on the Regional Dimensions of Kurdish Identity, Princeton University, Princeton, New Jersey, March 20-21, 1998.
  5. Rezaei Durroei, Siamak (1998), Kurdish Media, Language and Technology, Centenery of Kurdish Media, London, April 1998.
  6. Rizgar, Baran (1993) Kurdish-English, English-Kurdish Dictionary, 1993.


Your rating: None Average: 3 (2 votes)