Talk:Yotsugana

Untitled
So, how are these pronounced, exactly? How do all four differ when they are distinct? InnocentOdion (talk) 16:57, 11 August 2008 (UTC)

This article would be greatly improved if an expert incorporated the answers to the followings questions.

Using Hepburn romanization:

1) Can we assume that when they are distinct (≠) or dominant (=) じ is always "ji" and ず is always "zu"?

2) What are the romanizations of ぢ and づ?

3) In cases 1 and 3, which dominate? Kazenorantaro (talk) 19:17, 13 April 2011 (UTC)

An educated guess can be made concerning the Hepburn romanization of the four phonemic cases of modern yotsugana, based upon the English Wikipedia articles "Yotsugana" and "Romanization of Japanese", and the Japanese Wikipedia article 「四つ仮名 」.

1: ji＝ji＝ji＝ji 2: ji＝ji≠zu＝zu 3: ji＝ji≠zu≠(zzu or dzu) 4: ji≠gi*≠zu≠(zzu or dzu)

"ga", "gui", "gu", "gue", "go".
 * "gi" is early 17th century romaji for ぢ, not modern romaji for ぎ. The corresponding early romaji for the voiced "ka gyou" : are

Kazenorantaro (talk) 08:59, 21 April 2011 (UTC)

Nihon-shiki?
Why is this using the Nihon-shiki romanization when everything else on Wikipedia (and most works) used the Hepburn romanization? — Preceding unsigned comment added by FamAD123 (talk • contribs) 00:03, 20 February 2017 (UTC)


 * Becaues the Hepburn system conflates them: じ=ぢ=ji, ず=づ=zu, so a convention is needed to make the distinction in roman letters, and Nihon-shiki is a reasonable choice (possibly it should be called Kunrei-shiki)... Imaginatorium (talk) 03:41, 20 February 2017 (UTC)

Voiceless equivalents
In Tohoku and Okinawa, where all four of the sequences in question are merged, are the voiceless pairs merged, or partly merged, as well? Do any dialects merge す and つ, or し and ち, or all four? —Mahāgaja · talk 17:44, 21 January 2019 (UTC)

Phonemic merge of vowels?
In the northeastern dialects where all four are homophones, are /i/ and /u/ merged altogether? ThighFish (talk) 16:59, 4 May 2021 (UTC)
 * Not merged, but neutralized. They still contrast when not following an alveolar obstruent. Nardog (talk) 14:54, 7 May 2021 (UTC)
 * seems to imply that the phonemes /i/ and /u/ are generally neutralised, not only after alveolar obstruents, but isn't clear on this. Should be phrased unambiguously. --Florian Blaschke (talk) 12:46, 20 January 2024 (UTC)
 * By definition neutralization is a conditional, incomplete merger, so there shouldn't be such a thing as "generally neutralised". The description in that article is unsourced anyway, but the Japanese counterpart, which is better referenced, also confirms it's neutralization and not a complete merger. Nardog (talk) 01:19, 21 January 2024 (UTC)