Revised request: Japanese transliteration variants

Peter Constable petercon at
Wed Sep 2 03:33:40 CEST 2009

In terms of inherent mnemonics, the subtag "strict" has a generic meaning. (There could be lots of other scenarios involving sub-variants that could be called "strict".) There have been debates here in the past over generic tags. If you look in the list of registration forms

you'll see that we have managed to avoid generic subtags. (IIRC, "western" and "eastern" were proposed for Armenian variants, and alternatives were chosen instead.) 

So, what about something like "nihonshi" or "nihon"? (Prefix would still be "ja-Latn-kunrei"


-----Original Message-----
From: ietf-languages-bounces at [mailto:ietf-languages-bounces at] On Behalf Of Frank Bennett
Sent: Tuesday, September 01, 2009 2:38 PM
To: Phillips, Addison
Cc: ietf-languages at
Subject: Re: Revised request: Japanese transliteration variants

On Tue, Sep 1, 2009 at 11:46 PM, Phillips, Addison<addison at> wrote:
> Hello Frank,
> Please note that "nihonshiki" is too long to be a valid subtag (the limit is eight characters). You should propose a shorter alternative.

Thank you, and my apologies for the back-and-forth over this.

As "Nihon-shiki" transliteration is formally registered with the ISO as a variant of "Kunrei", it should be treated as a sub-variant of that transliteration method.  I would like to revise the proposal as follows.  To avoid confusion, please refer to this as "Frank's second revised proposal":

     Type: variant
     Subtag: hepburn
     Description: Revised Hepburn romanization.
     Prefix: ja-Latn

     Type: variant
     Subtag: kunrei
     Description: Kunrei-shiki romanization, as defined in ISO-3602.
     Prefix: ja-Latn

     Type: variant
     Subtag: strict
     Description: Nihon-shiki romanization, as defined in ISO-3602 Strict.
     Prefix: ja-Latn-kunrei

Frank Bennett

> Regards,
> Addison
> Addison Phillips
> Globalization Architect -- Lab126
> Internationalization is not a feature.
> It is an architecture.
>> -----Original Message-----
>> From: ietf-languages-bounces at [mailto:ietf-languages- 
>> bounces at] On Behalf Of Frank Bennett
>> Sent: Tuesday, September 01, 2009 6:25 AM
>> To: ietf-languages at
>> Subject: Revised request: Japanese transliteration variants
>> Thanks to everyone for their feedback on the earlier request.
>> Hepburn
>> is what I know, but the comments reminded me of the importance of 
>> precision and authority, and I have revised the request to provide 
>> more background, and added tags for two systems that some persons 
>> locally are likely to demand receive equal billing with Hepburn.
>> Request for variant registration
>>   1. Name of requester: Frank Bennett
>>   2. E-mail address of requester: bennett at
>>   3. Records Requested:
>>      Type: variant
>>      Subtag: hepburn
>>      Description: Revised Hepburn romanization.
>>      Prefix: ja-Latn
>>      Type: variant
>>      Subtag: kunrei
>>      Description: Kunrei-shiki romanization.
>>      Prefix: ja-Latn
>>      Type: variant
>>      Subtag: nihonshiki
>>      Description: Nihon-shiki romanization, as defined in ISO-3602 
>> Strict.
>>      Prefix: ja-Latn
>>   4. Intended meaning of the subtag:
>> Indicates the target content is Japanese text, romanized according to 
>> the rules set forth in the document cited in the relevant 
>> Description.
>>   5. Reference to published description of the language (book or
>> article):
>>       English (primary)
>>           Revised Hepburn: ALA-LC Romanization Tables (available for 
>> download)
>>           Kunrei-shiki: ISO-3602 (available for purchase)
>>           Nihon-shiki: ISO-3602 Strict (available for purchase)
>>       English (secondary)
>>       Japanese (primary)
>>           Revised Hepburn: Japan Transport Ministry Bulletin no.
>> 490
>> of 26 July 1947.
>>           Kunrei-shiki: ローマ字のづづり方 [Method of Romanization], Cabinet 
>> Notice no. 1 of 9 December 1954.
>>       Japanese (secondary)
>>   6. Any other relevant information:
>> The immediate need for this is in the context of bibliography 
>> management, where alternate representations of a title, name or other 
>> field must be offered for sorting or display purposes.  Japanese has 
>> a very orderly phonetic structure and native logographic 
>> representation, but there are several different methods of 
>> romanization, none of which have succeeded in dislodging the others.
>> Broadly speaking, existing romanization methods fall into two camps.
>> Revised Hepburn and its variants well exploit the latin script, to 
>> produce a representation from which it is easy to approximate 
>> Japanese pronunciation with a minimum of exposure to the language.
>> Kunrei-shiki and Nihon-shiki adhere closely the logographic 
>> representation used in native Japanese writing.  As a result, these 
>> methods require some training in order to learn how to map the roman 
>> character combinations used onto the spoken language.
>> Apart from these relatively small practical differences, their 
>> differing origins (or, more precisely, the differing image of their
>> origins) fuels a patriotic division of loyalties between the two
>> camps: for Hepburn is named for a 19th century Christian missionary 
>> to Japan, and Kunrei-shiki is named for a government order issued 
>> during the period of direct Imperial rule, in 1937.  The friction 
>> between the two camps will presumably persist indefinitely.  
>> Therefore, it is probably prudent to add representatives of both 
>> camps to the standard at this point, to avoid future controversy.
>> =======
>> hepburn
>> =======
>> Revised Hepburn is the most common Japanese romanization method in 
>> use.  While there are several variants of Hepburn, the ALA-LC 
>> Romanization Tables provide clear and comprehensive guidelines likely 
>> to be acceptable to any consumer of text that expects Hepburn 
>> romanization.
>> (The Traditional Hepburn system dates from the late 19th century, and 
>> has been superceded in modern publishing by Revised Hepburn.  I am 
>> not aware of any publisher that requires it, and have therefore left 
>> it out of this submission.)
>> (There is an "extended Hepburn" system, which avoids the use of 
>> macrons.  This is not in common use, and is documented, as nearly as 
>> I can tell, only in journal papers published by a single academic.  I 
>> have therefore left it out of this submission.)
>> ======
>> kunrei
>> ======
>> Kunrei-shiki is the officially recognized method for romanizing 
>> Japanese in the Cabinet Office and many government ministries, 
>> including the Ministry of Education.  Most adults in Japan were 
>> taught Kunrei-shiki romanization in elementary school, before being 
>> exposed to Hepburn in later schooling or universitiy.  Kunrei-shiki 
>> is definitely a minority method in real-world use, but as the 
>> successful campaign to have Hepburn dropped from the ISO standards, 
>> and Kunrei-shiki set up in its place, the proponents of Kunrei-shiki 
>> are eager partisans, and sensitive to the attitude of standards 
>> bodies.
>> I'll be happy to address any questions or concerns that the group may 
>> have.
>> Thank you for your time,
>> Frank Bennett
>> _______________________________________________
>> Ietf-languages mailing list
>> Ietf-languages at
Ietf-languages mailing list
Ietf-languages at

More information about the Ietf-languages mailing list