Talk:Man'yōgana
This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | ||||||||||||||||||||||||||||
|
place name examples
Antonius contributed this sentence:
- Man'yōgana continues to exist in some regional names of present-day Japan. In these cases, the pronunciation of these kanjis is obcure.[sic]
I am not sure what you had in mind when you wrote this. Could you please give examples in your next edit? Are these extremely rare cases? --69.212.98.38 09:59, 5 Aug 2004 (UTC)
- Places like Nara (奈良, what?-good), for example. See also many place names in Hokkaido, borrowed from Ainu, but spelled with Kanji. A good many place names have kanji that seem to be completely senseless, and are probably the result of such kanji-spellings. Also, there are "respelled" names like Hitachi Province, historically Hi-tachi (Sun-rise), but spelled Hita-chi (常陸), literally "Always land".
- this is really interesting- does anyone have more examples? brain 04:02, 11 April 2007 (UTC)
Addition on 26 Jun 06
I've added a section from the Japanese article on manyogana. I welcome any corrections or expansions to it. I was considering adding the chunk "江戸時代の和学者、春登上人は「万葉用字格」(文化15年)の中で、万葉仮名を五十音順に整理し<正音・略音・正訓・義訓・略訓・約訓・借訓・戯書>に分類した。" but:
- I'm not sure how to romanize 春登上人. Haruto Shounin?
- Note how he came up with 8 categories for classifying manyogana, but there are really only 7 in the article. I'm not exactly sure how 春登上人's categories correspond to those in the article, which is the odd one out, or how best to translate those category names.
adamrice 15:49, 26 June 2006 (UTC)
- 春登 is to be read as Shuntō, not Shūtō. There is no need to transliterate 上人, which is not a name but rather means monk or priest. He was a phonetician living between 1769 and 1836. Bendono 08:19, 5 October 2006 (UTC)
- 春登上人 is not a standard Japanese name, it is a given scholastic name, the proper reading for this is : Shūtōshōnin. Sudachi 12:11, 27 June 2006 (UTC)
Re: Ateiji
A phenomenon similar to man'yōgana, called ateji (当て字), still occurs, where words (including loanwords) are spelled out using kanji for their phonetic value: for example, 倶楽部 (kurabu, club).
While I am not knowledgeable about the use of the kanji 倶楽部 to denote a "club" in Japanese, I would like to point out that 倶楽部 originates from the Chinese, and in fact does mean a "club". Therefore, I wonder whether this is indeed a good an example of Ateiji.
- Interesting. Does the Japanese 倶楽部 in fact come from the Chinese, or is it a case of convergence? I have no idea what the etymology of this word is in Japanese. We could certainly change it to something else (although this example has propagated to other articles and other language versions of this article already, so that horse has left the barn). Also, please sign your comments by typing 4 tildes at the end like this (without the spaces): ~ ~ ~ ~. adamrice 13:13, 17 August 2006 (UTC)
- Wow, that is interesting... the meaning of the kanji used there could certainly quite logically be taken to mean "club," but they also just so happen to have the readings of ku-ra-bu. 面白い。LordAmeth 08:29, 1 October 2006 (UTC)
- 倶楽部 is not a Chinese word, or at least it's not in any online Chinese dictionary. Considering a common alternative spelling is クラブ I would think that it's more likely an English borrowing, with kanji applied after the fact. Compare 阿蘭陀 / 和蘭陀 which was borrowed form Portuguese. If 倶楽部 is occasionally used in China, it could be borrowed from Japanese. Shinobu 02:46, 5 September 2007 (UTC)
- Wow, that is interesting... the meaning of the kanji used there could certainly quite logically be taken to mean "club," but they also just so happen to have the readings of ku-ra-bu. 面白い。LordAmeth 08:29, 1 October 2006 (UTC)
orientation of charts
Notice the katakana chart runs in one orientation, and the hiragana chart runs in the other... I think it would be clearer if there were more parallelism between the two charts. Maybe nitpicky, but there it is. brain 03:59, 11 April 2007 (UTC)
Table of 80
Like the hiragana and katakana often shown in the "Table of 50", it may be a bit less confusing if the Man'yōgana are shown in the "Table of 80" (arising from the old 8-vowels a, i, ï, u, e, ë, o and ö), as in the early Man'yōgana the vowel harmony was still retained, while in their later usages, the vowel harmony was lost and the 8 vowels were consolidated into the current 5 vowel system. If I find a reference, I will add it, but if someone ones to start one, that would be good, too. CJLippert (talk) 22:35, 15 December 2007 (UTC)
- Found some. CJLippert (talk) 02:23, 16 December 2007 (UTC)
/g b d z/
When did distinct characters for voiced syllables disappear? It seemes peculiar that, if man'yōgana had distinct voiced characters that modern kana should have lost them, and created a diacritic for that purpose. Nik42 (talk) 02:51, 18 December 2007 (UTC)
- They did not disappear. However, initially there was no orthographic way to indicate it with kana. That developed a little later. Bendono (talk) 02:57, 18 December 2007 (UTC)
- But, what I mean is, this article shows distinct man'yogana characters used for, e.g., /ga/ as opposed to /ka/. What happened to those characters? Did they become used interchangeably? Did the voice-specific ones fall out of use? Nik42 (talk) 03:04, 18 December 2007 (UTC)
- During the transition from man'yōgana to hentaigana, a particular set of characters were used more frequently than others, and without a concrete way to indicate voicing, these characters were used for both voiceless and voiced sounds. Thus, /ka/ could be [ka] or [ga]. Dakuten marks can be found during the Heian period in limited specialized texts, but it is not until the late Kamakura period until it becomes fairly widespread and common. Bendono (talk) 03:27, 18 December 2007 (UTC)
i2 and o2
Were these distinct vowels? Or dipthongs? 69.182.52.86 (talk) 01:40, 4 May 2008 (UTC)
- Short answer: we don't know. Initial research suggested various vowel types, but research since the 70s has been generally suggesting that the kō/otsu distinctions represented a kind of glide (/j, w/). There is no consensus within the academic community and it is strongly debated today. Bendono (talk) 09:10, 7 May 2008 (UTC)
Hiragana origin picture
Isn't this hentaigana?Moocowsrule (talk) 05:37, 3 October 2008 (UTC)moocowsrule
Derivation Of Katakana Characters Pictures
Hi,
From the original:
http://upload.wikimedia.org/wikipedia/commons/1/11/Katakana_origine.png
to
http://upload.wikimedia.org/wikipedia/commons/0/0c/Katakana_origine.svg
I noticed a few mistakes:
The last (bottom) stroke in the top radical in the kana for "sa" in the svg is highlighted red. That is incorrect
"te" also has a problem: the length of the descending left stroke between the horizontal strokes are highlighted red. This is also incorrect. Can you amend the svg, upload, replace, delete the links to the files in the corresponding articles (wikipedia and anywhere else), and copypasta? Thanks.174.3.103.39 (talk) 05:14, 9 June 2009 (UTC)
What's the difference between "complete" and "partial"?
In the section "Types", in the subsections of "Shakuon kana" and "Shakkun kana", there are these tables giving some examples of how characters are mapped to certain sounds. But I have to ask, why do these tables have "partial" and "complete" characters? VulpesVulpes42 (talk) 18:19, 25 August 2016 (UTC)
Use of modernized kanji?
@Thibaut120094: You recently reverted various edits by an IPv6 anon who had changed certain kanji from the modernized shinjitai to the more-historically-accurate and pre-reform kyūjitai. We know that various glyphs such as 満 and 真 were simplified in relatively recent times from the older forms of 滿 and 眞. For an article discussing the origins of man'yōgana and describing usage within its historical context, it would seem to make more sense to use the older, historically attested forms, no? ‑‑ Eiríkr Útlendi │Tala við mig 17:12, 15 June 2022 (UTC)
- Hello Eirikr,
- Sorry, I reverted because it didn't correspond to the Japanese Wikipedia's table and some sources I found: [1][2][3].
- As for the kyūjitai/shinjitai characters, the characters changed by the IPv6 (齒, 滿, 眞, 蘿 and 鷄) do not appear in the Japanese sources I found, if you have a reliable source where we can find these characters, feel free to add it, otherwise it would be WP:OR. Thibaut (talk) 11:46, 16 June 2022 (UTC)
- Cheers @Thibaut120094, thank you for the reply.
- Digging through my local copy of the Man'yōshū (MYS) HTML files from http://jti.lib.virginia.edu/japanese/manyoshu/, I get the following hit counts (note that there are a couple instances where one poem might have multiple hits):
- 齒: 28 hits in 8 of the MYS books
- Book 1, poems 1, 10 -- used for phoneme ba in the modern reading gloss (in poem 1) -- might be a modernization, where it was actually pronounced as pa in the Old Japanese; c.f. theorized derivation of conditional -ba. Used for yo in poem 10, apparently from the alternative reading yowai related to youth and weakness.
- Book 2, poem 210 -- used for phoneme ba in the modern reading gloss -- might be a modernization, where it was actually pronounced as pa in the Old Japanese; c.f. theorized derivation of conditional -ba
- Book 3, poem 284 -- used for phoneme ba in the modern reading gloss -- might be a modernization, where it was actually pronounced as pa in the Old Japanese; c.f. theorized derivation of conditional -ba
- Book 4, poem 674 -- used for phoneme pa
- Book 7, poems 1134, 1135, 1137, 1240, 1382, 1400, 1405 -- used for phoneme pa (with one instance parsed as conditional -ba in the modern reading gloss)
- Book 11, poems 2624, 2696, 2773, 2782 -- used for phoneme pa, conditional -ba, and in case for yo
- Book 12, poems 2947, 2986 -- used for phoneme pa (with one instance parsed as conditional -ba in the modern reading gloss)
- Book 16, poems 3860 through 3869 in the [左注] notes (which appear to be written in kanbun)
- 滿: 0 hits. However, shinjitai version 満 appears 104 times, making it likely that the kyūjitai version 滿 appears in other editions.
- 眞: 0 hits. However, shinjitai version 真 appears 417 times, making it likely that the kyūjitai version 眞 appears in other editions.
- 蘿: 14 hits in 5 of the MYS books:
- Book 2, poem 113 title, 228 -- used as an ideograph for koke "moss", not as man'yōgana for the sound ra
- Book 6, poem 962 -- used as an ideograph for koke "moss", not as man'yōgana for the sound ra
- Book 7, poems 1120, 1214, 1334 -- used as an ideograph for koke "moss", not as man'yōgana for the sound ra
- Book 11, poems 2630, 2750 -- used as an ideograph for koke "moss", not as man'yōgana for the sound ra
- Book 13, poems 3227, 3228 -- used as an ideograph for koke "moss", not as man'yōgana for the sound ra
- 鷄: 12 hits in 6 of the MYS books:
- Book 2, poem 105 -- used in compound 鷄鳴 (literally "birdcall") as a kind of jukujikun for 暁 (akatoki, "dawn")
- Book 3, poems 312, 382, 460, 479 -- used for phoneme ke1, and as an ideogram for "bird / chicken" (tori) in poem 382
- Book 4, poem 553 -- used for phoneme ke1
- Book 5, poems 813, 814 in the title notes (which appear to be written in kanbun)
- Book 6, poems 963, 1015 -- used for phoneme ke1
- Book 7, poems 1081, 1413 -- used for phoneme ke1
- 齒: 28 hits in 8 of the MYS books
- The two characters 滿 and 眞 might be used in the Nihon Shoki, Kojiki, or other ancient texts, but I do not have similar searchable access to copies of these. If you're aware of any such copies, I'm all ears. :) ‑‑ Eiríkr Útlendi │Tala við mig 18:52, 16 June 2022 (UTC)
Table showing evolution from kanji->grass style->hiragana is off
Not to be mean to anyone, but the characters depicted in red in this table are all very poorly draw depictions. The table should be replaced with authentic depictions of the 草書 characters. I don’t know how to do that, otherwise I’d do it myself. — Jim_Lockhart (talk) 15:01, 15 June 2024 (UTC)