Status of Japanese requests

Mark Crispin mrc+ietf at Panda.COM
Sat Sep 26 00:38:40 CEST 2009


On Fri, 25 Sep 2009, John Cowan wrote:
>> Also, is the name to be written in Hepburn or kunrei-shiki?  If the 
>> latter, it would be "kunrei-siki"...
> "Kunreisiki" according to ISO 3602, which does not employ hyphenation.

For what it's worth, Wikipedia says that you're right.

It's "kunrei-shiki" in English, "kunreisiki" in kunrei-shiki, and 
訓令式ローマ字 ("instruction type Roman characters) in Japanese script.

FWIW, kunrei ("instruction") is as in "order", "command" as opposed to 
pedagogical.

>> I also note that Hepburn, Nippon-shiki, and kunrei-shiki all require the 
>> use of macrons for long o.  Hepburn and Nippon-shiki use U+0304; 
>> kunrei-shiki uses U+0302.  AFAIK, the use of "ou" is so-called "word 
>> processor romanization"; "oo" is Jorden.
>> Supposedly, long a, u, and e also use macron forms, but I don't recall 
>> ever seeing them (what does ISO 3602 say?).
> Circumflexes only: â ê î ô û.

In case it wasn't clear, I was using "macron" as a generic term for both 
macron and circumflex.  For the purposes of romanizing Japanese, the 
difference is cosmetic and they're the same thing.  Sorry for being fuzzy 
like that in front of a bunch of linguists... :)

But are you sure that all long vowels are written with circumflexes?

Jorden says that â û ê ô are used, but I don't recall ever seeing the 
circumflex form for anything other than ô.

Jorden says that î is not used.  I am certain that I have never seen ôkî 
or ookî (for 大きい / おおきい, "large") as opposed to ookii.

Both ee and ei forms occur for long e, and oo and ou forms occur for long 
o.  Some (but not all) native speakers make a difference in how these are 
pronounced.

I am pretty sure that I have never seen kirê (for きれい, "pretty") as 
opposed to kirei (everybody except Jorden) or kiree (Jorden).  Jorden says 
that ê is used for "ee" but not for "ei".

I was going to say that the circumflex is only use for ou and not for oo, 
but then I remembered that I do see Ôsaka (大阪/おおさか, "Osaka") 
whereas only Jorden uses "Oosaka".

>> I almost always see long e written as "ei", but of course Jorden uses
>> "ee" to be different.
> You obviously understand this better than I do, but FWIU the kind of
> differences you are talking about can't cause errors in reading, unlike
> the higher-level ones between kunrei, Nippon, and Hepburn.  You (or your
> conversion software) just understands that ee = ei, and that's that.
> There's nothing else it could possibly mean.

Actually, no.  ee and ei are different.  Jorden coerces most occurances of 
ei to ee, independently of how it's written in Japanese script. 
Everything in Jorden is about the spoken language (in Tokyo dialect).

Conversion software may not necessarily understand that "ei" was intended 
when "ee" is written.  Same for "ou" vs "oo", and the mess with "di"/"du" 
(see below).

>> The differences between Nippon-shiki and kunrei-shiki are subtle.  Other 
>> than the macron difference (and I don't think that anyone who romanizes 
>> Japanese with macrons pays attention to which one they use), AFAICT 
>> the only difference is that Nippon-shiki preserves (as "di" and "du") the 
>> kana ぢ and づ that are otherwise pronounced identically (as "zi"/"ji" and
>> "zu") to じ and ず. 
> According to the non-normative note in 3602, Nippon uses ha and never wa,
> he and never e, wo and never o, and di, du, dya, dyu, and dyo.

OK, that makes sense.  So Nippon-shiki follows kana, and kunrei-shiki is a 
simplification based upon pronunciation (in Tokyo dialect).  Hepburn use 
consonant values that an English speaker would expect (e.g., "fujitsu" 
instead of "huzitu").

dya, dyu, and dyo are simply the twisted forms of di with ya/yu/yo:

Hepburn:	da,  ji,  zu,  de,  do,  ja,  ju,  jo
Nihon:		da,  di,  du,  de,  do, dya, dyu, dyo
kunrei:		da,  zi,  zu,  de,  do, zya, zyu, zyo

I've also seen Hepburn text using:
 		da, jhi, tzu,  de,  do, jha, jhu, jho
by analog with
 		ta, chi, tsu,  te,  to, cha, chu, cho

What a mess!

-- Mark --

http://panda.com/mrc
Democracy is two wolves and a sheep deciding what to eat for lunch.
Liberty is a well-armed sheep contesting the vote.


More information about the Ietf-languages mailing list