Minutes of the meeting of the UNGEGN Working Group on Romanization Systems
New York, 1 May 2014
- Preparations for the 11th UNCSGN
- WGRS and liaison with ICANN and ISO TC 46/WG 3
- Updates to the WGRS website
- Proposal to create a sub-group to address transcription systems for regions in Africa
- Any other business
1. Preparations for the 11th UNCSGN
- For which languages/scripts for which a romanization system was previously approved but not implemented could a (new) romanization system be endorsed before the 11th UNCSGN (2017)?
- For the syllabic South Asian scripts (Indian group of languages), no romanization systems have been implemented at this point of time. No contacts have been established with Indian authorities as yet, and contacts with Pakistan have been sporadic and inconsequential. It can therefore not be expected that systems for these scripts can be offered for recommendation in 2017. The system presently most viable for transliteration of Hindi and Urdu is the Hunterian system, which might perhaps be acknowledged as an official standard in due time.
The Working Group will resume its attempts to contact experts in the countries concerned.
- In Ethiopia, the UN approved romanization system for Amharic seems not to be implemented anymore. A simplified system is currently used instead.
- For Khmer, instead of the UN recommended system a new provisional system is now in general use. The Cambodian Mapping Agency will try to endorse this before 2017.
- The UN approved system for the transliteration of Macedonian Cyrillic in the Former Yugoslav Republic of Macedonia, dating from the time this country was part of Yugoslavia, is no longer in general use. BGN/PCGN recently approved a number of minor changes to the system formerly employed, in accordance with the current use. The details will be shared with the WG. The convenor of the WG maintains contacts with experts from the FYROM, but sees no prospects for endorsement before 2017. In linguistic circles, a quite different system appears to be in use.
- As for the romanization of Arabic, the system approved in the conference held in Beirut in 1972 was done so with the provision that it would be valid “within those Arabic-speaking countries where this system is officially acknowledged”. In reality, even the latter countries presently do not seem to apply it anymore.
- Saudi Arabia submitted geographical names for the UNGEGN Database in a modified system of 2007, which was adopted by most countries that formerly used the 1972 Beirut system. The 2007 system in fact reverts to a system preceding the implementation of the Beirut system: it was devised by Arabic linguists in 1971. It differs from the Beirut system in the representation of five consonant phonemes, for which underscores are applied; the most radical difference is the application of dh instead of z-cedilla. The use of underscores may cause encoding problems when names are to be entered into digital data files.
- Mr. Brahim Attoui clarifies that the confusion of the two competing systems of romanization arises from the fact that the Beirut system was adopted in the absence of linguistic experts from Arabic-speaking countries except Lebanon, although it was accepted by the Arab League. Moreover, some countries in which Arabic has official status (specifically Chad and Eritrea) are not included in the Arab League.
- The use of the new system, which is generally accepted by the delegates of the Arabic Division, should be recommended by UNGEGN. Contrary to the Beirut system, however, it still lacks support from the Arab League. The Arabic Division should think about how to involve the Arab League. The only country as yet applying the new system on maps is Jordan.
- Next to the Beirut 1972 system and the new romanization forwarded in 2007, some countries still apply traditional French transliterations. International oil extracting companies generally apply either English or French transliterations of Arabic names.
- In September there will be a meeting in Algeria, which may offer an opportunity to forward a solution. To be continued.
- Perspective systems for languages/scripts on the agenda:
- Armenian: a simplified system is endorsed by the government, but this is not yet recommended by UNGEGN because the omission of the apostrophes meant to discriminate between consonant phonemes renders this system irreversible. A draft has been made in Armenia to make it reversible; perhaps it can be made ready before the next meeting.
- Dzongkha: previous contacts with an expert in Bhutan seemed promising, but they have been lost. There is a romanization system in use.
- Kirghiz: the national government accepted a romanization system, which was also adopted by BGN/PCGN. The system is reversible.
- Maldivian: the national romanization system has been modified. It is largely reversible. To non-English speakers, the representation of vowel sounds is somewhat uncommon. The WG thus far failed to establish contact.
- Georgian: the proposed system is not yet universally accepted within Georgia. The use of apostrophes to discriminate between consonants meets opposition because some names would contain multiple apostrophes. There will be no initial apostrophes, as they can only follow consonants. It is hoped but not quite certain that the problems can be solved and the system endorsed by 2017.
- Competing systems:
- Japanese: the Kunrei-siki system of romanization (1954) is officially endorsed by the Japanese government and supported by linguistic experts, but it has never become extensively used. The prevailing system is as yet the Modified Hepburn System. The latter is also applied in sea charts. This WG meeting is not attended by experts from Japan.
- Korean: the National System of Romanization of the Republic of Korea has been universally accepted within the Republic of Korea, and was also accepted by BGN/PCGN. As yet no agreement was reached between the Republic of Korea and the Democratic People’s Republic of Korea: after 2007 there have been no bilateral meetings about this subject anymore.
- Languages/scripts lacking a national system of romanization: no new information about these has been passed to the WG.
- Inclusion of romanization systems for regional and minority languages:
- If a country wishes to submit romanization systems for regional languages, we can include them in the UNGEGN list for international adoption.
- In Canada, romanization systems for minority languages using a non-Roman script (i.c. the Inuktitut syllabary) are relatively new. Systems that for instance 5 years after their adoption are not/no longer under debate could be taken into consideration. It is up to Canada to put this forward. Romanization of Inuktitut is also being discussed by BGN.
- If a minority language in one country is the official language or a minority language of another country, the two countries should agree upon the proposed romanization system before it can be recommended by UNGEGN.
- In Poland, a system of romanization is in use for the Cyrillic-written Ruthenian dialect Lemko. The application of this system cannot be extended to the Ruthenian languages/dialects spoken in for instance Slovakia and Vojvodina, as these are said to linguistically differ.
- Another script that could qualify for internationally standardized romanization is Tifinagh, used to write the Berber languages in several North African countries. Berber minority languages have official status in Morocco and Libya and national status in Algeria. In Morocco they are written in Tifinagh. Tifinagh romanization is currently under discussion at PCGN.
- The subject has an overlap with the work of the WG on Geographical Names as Cultural Heritage.
2. Liaisons maintained by the WG
- ICANN is concerned with the subject of romanization because it needs to set rules about the characters permitted in Internet domain names. In some cases the expertise of WG members might be asked for.
- ISO Technical Committee 46/WG 3 (Conversion of written languages): the convenor of the WG joined this committee in 2013. Where possible the efforts of both organizations should be coordinated, although they have different scopes and fields of application. ISO devises transliteration systems for scripts, not for languages. ISO systems are therefore not donor-generated. ISO standards are strictly reversible. ISO also maintains sets of simplified transliterations.
- Unicode: contacts are maintained by the WG on Data Files and Gazetteers (Ms Caroline Burgess)
3. Updates to the WG website
- Updates to the status report will be made in two stages: a) alphabetic systems; b) syllabic systems. There could be problems in updating the Arabic section as more information is needed about the present use of romanization systems in different Arab countries.
4. A proposal to establish a Subgroup for transcription of (unwritten) indigenous languages in West and Central Africa made by Mme. Élisabeth Calvarin will be taken in consideration. As the languages concerned do not make use of a non-Roman script, maybe the Task Team for Africa should be involved.
5. Other business: input from experts in the WG is asked for the possible development of a mechanism to measure the level of implementation of systems of romanization.
Notes taken by Mr. Tjeerd Tichelaar