User talk:Eirikr/2020

Not much going on here at present. Feel free to drop a line. Eiríkr Útlendi | Tala við mig 16:41, 31 March 2006 (UTC)

Archives (old, deprecated)

2006-2007

2011-2012

Archives
2006 – 2007 – 2011 – 2012 – 2013 – 2014 – 2015 – 2016 – 2017 – 2018 – 2019 – 2020

---

Special:Contributions/2409:8900:1D51:C835:45F0:E00:C7EC:C01

User:M. I. Wright mentioned on the Wiktionary Discord server that these need a cleanup. It seems like you're good with historical stuff, so I leave you this message. —Suzukaze-c◇◇ 01:10, 3 January 2020 (UTC)

@Suzukaze-c: Thank you! Vetted and reworked as needed.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 01:18, 5 January 2020 (UTC)

天文

Hi there. I was wondering if you could help me update the Japanese entry here with the reading and definition at Tenbun? ---> Tooironic (talk) 13:20, 13 January 2020 (UTC)

• @Tooironic, how's that?  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:04, 13 January 2020 (UTC)

  Excellent. Thanks for your hard work! ---> Tooironic (talk) 07:48, 14 January 2020 (UTC)

山井

Sorry to bother you again, but I have another question. Can 山井 also be read as Yamanoi? For example in the name of the Japanese scholar Yamanoi Tei 山井 鼎 (1670–1728)? If so, would you mind adding this reading when you get time? Thanks! ---> Tooironic (talk) 08:02, 15 January 2020 (UTC)

@Tooironic: No worries. Expanded the entry. Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 16:42, 15 January 2020 (UTC)

Thank you so much. ---> Tooironic (talk) 01:56, 16 January 2020 (UTC)

Trying to resolve Proto-Japonic verb lemmas

(Pinging @Kwékwlos for input too) I think we should move Proto-Japonic verb pages to their roots, instead of guessing their shūshikei. It not only illustrates the verb class better (e.g. we move *kaku to kak-, exposing it as a godan verb, and *intu to inte-, exposing it as ichidan) it saves us much wrangling on whether to apply syncope to the Proto-Japonic forms themselves (a massive issue when we run into ichidan verbs, again), and I haven't found a source that explicitly reconstructs specific forms (unlike say Indo-European languages), instead reconstructing roots and suffixes separately in a format "root-suffix-suffix". mellohi! (僕の乖離) 15:24, 15 January 2020 (UTC)

@mellohi!, no strong thoughts here. I haven't read enough of the academic literature to know if there's any emerging consensus re: reconstructed shūshikei for Proto-Japonic. I'm fine moving these to the root forms.

However, when doing so, we should make sure we're moving to the right location. For instance, modern Japanese 出る (deru, “to come out”, intransitive) is from historical Old Japanese 出づ (idu), clearly showing that this is 下二段 with a Proto root form of int-. As an additional point, we would need the root int- in order to explain the existence of modern Japanese 出す (dasu, “to take out”, transitive).

There are various other cases where the modern language has an 一段 verb, but the lexicon as a whole has evidence of cognates that imply the existence of a 四段 root, even if that root no longer exists. I am reasonably certain that the 二段 and 一段 verbs are all defective derivatives of 四段 roots.

That said, if we have a 二段 or 一段 OJP verb without any such cognates, and where the Ryūkyūan branch also only evidences reflexes of that same 二段 or 一段 verb paradigm, then naturally we can only reconstruct a vowel ending for the Proto root form.

HTH! ‑‑ Eiríkr Útlendi │^{Tala við mig} 16:53, 15 January 2020 (UTC)

Proto-Japonic final *o

We all know this yields OJp. o₁, but in light of the comparison of OJp. ko₁ and PR *kuwa for the word "child" (an irregular retention thereof), should they all be reconstructed as *-uwa (or *-ura)? In any case this would fit the Kara word twol if the given form for *to (“door”) was from an earlier *tuwa or *tura. I think this would also apply for the accusative *wo, which may come from an earlier *uwa. Cheers. Kwékwlos (talk) 22:34, 12 February 2020 (UTC)

• @Kwékwlos, re: kuwa, I suspect the PR may be an innovation. Notably, the reconstruction of OJP ⟨o₁⟩ as /wo/ is not entirely settled, as we can see in the table at w:Old_Japanese#Vowels. For instance, Vovin reconstructs OJP ⟨o₁⟩ as /o/, contrasting with ⟨o₂⟩ as /ə/. See his 2011 paper Sources of Old Japanese i2.

We know that JA /o/ usually corresponds with RYU /u/. We also see various RYU terms that end in 子 that show this reflex ku instead of [k]kwa, as listed here at JLect (just search for 子), such as 船子 funaku, 団子 dāgu, 梃子 tiku, etc.

We also know that JA /wa/ usually corresponds with a longer vowel in RYU /a/, as in our discussion regarding Okinawan まーすん.

• Long /a/: Japanese 皮 (kawa, “hide, skin”) ↔ Okinawan かー (kā, “hide, skin”), classical Japanese 合わす (awasu, “to bring things together”) ↔ Okinawan あーすん (āsun, “to bring things together”)
• Long /o/: Japanese 弱る (yowaru, “to become weak”) ↔ Okinawan よーゆん (yōyun, “to become weak”), Japanese 添わる (sowaru, “to be added onto, to go with”) ↔ Okinawan そーゆん (sōyun, “to go with”)
• Long /u/: Japanese 壊す (kowasu, “to break”) ↔ Okinawan くーすん (kūsun, “to break”), Japanese 壊れる (kowareru, “to become broken”) ↔ Okinawan くーりゆん (kūriyun, “to become broken”)

Thus, we'd expect Proto kuwa to become Okinawan kū, not kwa.

Further, we know that JA /ra/ is reflected in some Okinawan words as /wa/, as in Japanese 枕 (makura, “pillow”) ↔ Okinawan まっくゎ (makkwa, “pillow”).

I propose instead that the various Ryukyuan terms reflecting Proto-Ryukyuan *kuwa are derived not directly from Proto-Japonic *kuwa -- which I don't think existed for the meaning of "child; small thing" -- but instead from Proto-Japonic *ko ("child") + Proto-Japonic -ra (plural and generalizing suffix), with the suffix construed more as a generalizer in the Ryukyuan branch. This parallels the etymologically-plural modern Japanese 子供 (kodomo), also used as a singular.

I've discovered I'm not alone in the above argument: see also this essay on JLect similarly discussing the derivation of the Ryukyuan branch.

• Re: Kara twol, the ⟨wo⟩ notation in Koreanic romanizations does not necessarily indicate /wo/, but may just mean /o/. See also the 돌 (dol) entry mentioned at w:Gaya_language, where the Middle Korean romanization ⟨twolh⟩ is spelled in Hangul as 돓, using the regular /o/ vowel character. I suspect this may be due to one of various possibilities:

• The Middle Korean romanization at our 돌 (dol) entry is mistaken and should not include that "w".
• The romanization is in accordance with English-language academia, and reflects a practice whereby "wo" is used to indicate /o/.
• Modern Korean /o/ reflects older /wo/.

Whatever the case for the romanization or the underlying vowel value, any argument for cognacy between OJP ⟨to₁⟩ and MK 돓 (tolh? twolh?) would have to explain what happens to the final -lh in the Middle Korean -- was this a Korean innovation? What does it mean? If it's an integral part of the word, what happened to it in Japanese? In addition, Middle Korean final -h often evolved via lenition from earlier -k, if my memory is correct, which further complicates efforts to link these two. Add to that the fact that we really don't know the phonetic realization of the Kara word at all, due to the way that Chinese was used both phonetically and semantically for the various Koreanic pre-Hangul languages.

Anyway, it's been fun working on this. Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 20:26, 13 February 2020 (UTC)

For kuwa: dango does not actually derive from a final *kwo, but rather an alteration of danki, so it has no relevance to the word. The kanji is used for its pronunciation, not its actual derivation. The term funaku may be correct in assuming PR *ko, but we cannot exclude a borrowing from Japanese. The form tiku may or may not be derived from an assumed PR *ko, or it could be simply a Japanese borrowing for such a lexically complex term.

When we reconstruct PJ, sometimes irregular correspondences show up. An example is EOJ nwozi, which would imply Japanese noji, but the actual form is niji, contrasting Okinawan nūji, which agrees with EOJ. Particularly notable is the case of *mi (“three”) with modern Ryukyuan reflexes of mi rather than m or n.

If my theory is correct, we can explain the o₁ in so₁ra as an earlier *suwara, as a form *sora would yield *sura.

About the alternate derivation with PJ *-ra, the PR form would be *kora, that would yield Okinawan *kura. But the form kkwa appears to have been from *kura, but if PR *ku existed it would have been derived from PJ *ku. So the irregular conservatism seems to be more likely for me.

About the word twol, perhaps it would make sense to derive a phonemic /dor/ from an earlier /*dora/. The Kara language you described may actually represent Peninsular Japonic, which of course a relationship to Korean is tenuous at best, and so the Koreanic romanizations are what I believe to be imperfect, since they may be based of sinograms that were transcribed centuries earlier. If the *l is to be expected, then it would make sense for me to derive *to from *tura via an intermediate *tuwa. Of course, for the word *tonari, EOJ has to₁nari₁, which directly implies a preform *tuwanari. Kwékwlos (talk) 21:07, 13 February 2020 (UTC)

@Kwékwlos --

• Re: dango, apologies for my confusion, I'd forgotten that that was a more recent term. I'm not sure what you mean by "lever" being lexically complex. As a compound, it's "hand" + "diminutive thing". That said, we have other examples of JA 子 ko not corresponding to Ryukyuan kwa, such as Japanese 蟻子 (ariko) ↔ Okinawan あいこー (aikō), Japanese 女子 (onnago, wominago) ↔ Okinawan 女子 (winagu) ↔ Yoron 女子 (funagu).
• Re: the reflexes of mi "three", I'm not sure how that's relevant? I also disagree with the statement at Proto-Japonic *mi that Proto mi "three" is cognate with Proto mu "six". "Related" I could agree with, but "cognate" means "these were originally the same word", which I doubt.
• Re: the Kara term, if the Kara matched the Old Korean phonetics from which Middle Korean tolh arose, the term might have been something more like tolk. See also this discussion re: nara (search for "old Korean locative" to find the relevant bit discussing final -lh). There's no reason I can see to reconstruct any such tura.
• Re: Eastern Old Japanese noji, I believe that's a hapax legomenon, appearing solely in Man'yōshū book 14, poem 3414, and thus to be taken with a grain of salt. The original spelling also uses the glyph 努, which you'll see over at w:Man'yōgana was confusingly used to represent both nu and no₁.
• Re: "irregular conservatism", I can't tell what you mean by that.
• Re: this theory that OJP ⟨o₁⟩ derives from Proto-Japonic uwa, as best I can tell, the only rationale for reconstructing Proto-Japonic w here is the presumed /wo/ phonetic value for Old Japanese o₁. Old Japanese o₁ is not universally reconstructed as /wo/. Is there some other justification for this w that you keep introducing in the Proto terms? Also, the additional syllable seems ... problematic, at best.

Separately, and relatedly, I am growing concerned about your edits regarding Proto-Japonic forms, as they seem to be based on your personal thoughts (such as "if my theory is correct" above), without regard to broader academic research. While Wiktionary is not as strict as Wikipedia when it comes to original research, we do try to be clear about what is citable and what is conjecture or synthesis. I would be happier with some of your edits if you could clarify what is "probable" and what is actually citable to this or that academic author, ideally more than one.

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 08:05, 17 February 2020 (UTC)

I've been paying more attention to mathematics, where they are actually defined: such as a cube being a square prism. To be fair, I am a very curious amateur linguist. You may want to pay attention to "The Language of the Old-okinawan Omoro Soshi: Reference Grammar, With Textual Selections" which will help alot in our studies regarding Japonic. Besides, Japonic linguistics is done mostly by Japanese speakers who have little experience in the English language, unlike those of the Indo-Europeanists who could atleast speak English, French, German, or Russian. Kwékwlos (talk) 10:05, 17 February 2020 (UTC)

Sources to be trusted

For decades Japanese was thought to be an Altaic language, but recently it is no longer thought of as an actual family, but rather a Sprachbund. Of course there may be relationships, but they must be borrowings. With the dissolution of Altaic as a reliable proposal some have turned to Korean or Koreanic. While words like seom and shima are obviously related, it strikes to me that some of these native Korean words may actually represent Japonic borrowings, or that the Japonic words are borrowed from Old Korean.

From what I know, Robbeets obviously wants a relationship with "Transeurasian", which is just a rebranded Altaic, and even proposes AN etymologies for some Japonic words.

Vovin, on the other hand, once supported the Altaic theory but later became a critic. An example of a redeemable decision, I must say. He now focuses on finding loan relationships among the former "Altaic" languages.

Among the amateur scholars, some even from Japan, people like Mayuki Kikasane (twitter @MajukyiSanapey) fancy a reconstruction of so-called "Yasiman", or even a "Kanade" PJ reconstruction. Of course they partner with people like (twitter @cicada3301_kig) who propose their own interpretation, clearly something that isn't to be trusted upon, as they love to study languages that are extinct or created.

With Robbeets and any amateur scholar out of the picture, what can you say about reliable sources on Japonic (with the rise of English-speaking Japanese and Okinawans and English words found predominantly in Japan used in their emerging English dialect, such as HP and busjacking)? Kwékwlos (talk) 21:35, 13 February 2020 (UTC)

@Kwékwlos, by way of authors in English, my reading has been less extensive than I'd like, especially in recent years as things IRL have taken up more of my time. That said, I'm mostly familiar with Vovin, Martin, Robbeets, Starostin, Unger, Frellesvig (though only indirectly from what others have told me, I have yet to get my hands on his History of the Japanese Language). Subjectively, Vovin seems the most solid, perhaps joined by Unger and Martin. From what I've seen directly, and what I've read from others, Robbeets has been prone to some serious flaws in her analysis of Japanese and Japonic. I'd take anything from her with a large grain of salt, and seek to find agreement from other authors. Starostin doesn't know Japanese etymologies at all, and as best I can tell, his exploration of Japonic has been based purely on word lists and accidental resemblances -- occasionally he finds interesting things, but he's basically useless as a primary source, as everything needs further corroboration. Frellesvig, meanwhile, sounds to be savvier with regard to Japanese and Japonic, and more likely to be on-point and not just accidentally correct.

From my own perspective, Vovin's turnabout from Altaic proponent to opponent seems to be due to the sloppiness of past writings attempting to establish a Japonic—anything-else relationship, where further research into the claims made ultimately reveals a whole lot of not much. While I don't know that I yet fully accept Vovin's current statement that Korean and Japanese are wholly unrelated (claiming that Japonic and Koreanic grammatical and morphological similarities are due solely to areal proximity raises the question of why Korean is not a whole lot more like Chinese), I do take his core point that any attempt at showing a relationship must do more than simply point out resemblances, and must derive specific, repeatable, and predictive rules for sound correspondences, etc.

Another good English-language source I should mention is Shibatani's The Languages of Japan, part of the Cambridge language series. He provides a good overview of Japanese, Japonic, and Ainu. (For that matter, it's been several years, I should re-read that.) I particularly appreciated his analysis of accent patterns and dialectal shift, describing the "prestige" lects of the capital regions in the Kansai area and the less-prestigious provinces and explaining some of the mechanics of regional differences in a way that made more sense to me.

HTH as a good starting point.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:50, 17 February 2020 (UTC)

唐名

Hey there. When you get time, would you mind creating a Japanese entry for this term? My Japanese dictionary provides the following definition: からな 【唐名】① 中国での名称。② 日本の官職を唐制の呼び名に当てたもの。太政大臣を相国（しようこく），中納言を黄門と称するなど。とうめい。③ あだ名。別名。「横車とはな…おのれがやうな女の―よ」〈浄瑠璃・十二段長生島台〉. Many thanks! ---> Tooironic (talk) 04:03, 17 February 2020 (UTC)

• @Tooironic: Ya, will do! I see three readings for this: からな, とうみょう, and とうめい. I'll have a go some time in the next few days -- tomorrow even, if I can make the time.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 05:07, 17 February 2020 (UTC)

  Look forward to seeing the fruits of your labour. Cheers! ---> Tooironic (talk) 05:30, 17 February 2020 (UTC)

@Tooironic, looks like @Poketalker created the entry before I got to it. I just added some minor details a moment ago. Hope that scratches your itch.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:14, 18 February 2020 (UTC)

Thanks everyone! ---> Tooironic (talk) 01:34, 18 February 2020 (UTC)

@Tooironic: by the way, you're exactly quoting the Daijirin in which there is the equivalent online version available if you never checked. ～ POKéTalker（═◉═） 06:54, 18 February 2020 (UTC)

I see. Thank you! ---> Tooironic (talk) 22:03, 18 February 2020 (UTC)

OJP /yi/ and /wu/

Vovin says that /yi/ and /wu/ are impossible sequences, and would preclude a PJ *wu (“to sit”) reconstruction. In fact it doesn't exist as a syllable in the Man'yōgana syllabary. Do you have any opinion on this? Kwékwlos (talk) 21:42, 2 March 2020 (UTC)

Replied at Reconstruction_talk:Proto-Japonic/wu#Lemma_form.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 01:16, 3 March 2020 (UTC)

友達

Thanks, I understand the system now. Hkbusfan (talk)

Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:49, 6 March 2020 (UTC)

Japonic terms for resting

A surface inspection seems to uncover a root yoko/yuko for resting. It is amply attested, in Okinawan yukuyun, Kunigami yuhurun and yuhumin, Amami yuhumuri, Miyako yukū, Yonaguni dugun, and Japanese ikou. This root, when used verbally, doesn't seem to appear bare and is generally combined with a derivational suffix. In Japanese it appears with -pu, in Okinawan and Kunigami it appears with *-aru, in Amami and Kunigami it appears with the denominative verbalizer *-amu. Does this establish the existence of a primary root meaning "to rest" or are there secondary natures in the way? mellohi! (僕の乖離) 17:32, 7 March 2020 (UTC)

@mellohi! -- Very interesting. I haven't encountered this term before.

Looking this up in Daijirin, I see a note 〔漢文訓読系の語〕, suggesting that this may have been a higher-register or more formal term. I searched through the Man'yōshū, which tends more towards poetry rather than academia, and found no instances of the word, which might reinforce the DJR comment's suggestion. Meanwhile, in the KDJ, we see that the term appears in the Nihon Shoki of 720, so it's quite old. We also see senses and spellings that point towards the root ik- as in 息 (iki, “breath, breathing”).

There appears to be a strong correlation between ik- "breath" and ik- "living", as in 生きる (ikiru, “to live, to be alive”, older form iku), 生き (iki, “living, alive”). I suspect this also overlaps with 行く (iku, yuku, “to go”), from the basic idea that living things move. The Ryukyuan cognates you've found would seem to reinforce the idea of a y- initial, not just for 行く (iku, yuku) but also for 息 ([y]iki).

There are various terms in Japanese that may have a lingering initial y- at the root level that disappears in certain forms, since the yi phoneme doesn't really happen in Japanese. See for instance the likely cognacy of 稲 (ine) and 米 (yone), the alternation in 行く (iku, yuku), etc. I think I recall seeing a paper or two about that recently... here's one, The Syllable yi in Old Japanese, Thomas Dougherty, 2011. Vovin's 2011 paper On one more source of Old Japanese i₂ doesn't discuss yi directly, but he does have some examples showing how a yu phoneme might collapse in certain contexts into i. I think I've seen this discussed elsewhere as well, but I can't find it at the moment. (Plus, my Chrome installation blew a tire the other day when my laptop battery crapped out, and I lost tons of history, including two windows with multiple tabs that had various papers open in them... so much for persistent session management, I'm back to Firefox now.)

Anyway, I hope that provides some leads for further exploration. Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:36, 7 March 2020 (UTC)

"I searched through the Man'yōshū, which tends more towards poetry rather than academia, and found no instances of the word, [...]"

See this, only one logographic attestation in the Manyoshu. Chuterix (talk) 02:56, 31 May 2023 (UTC)

@Chuterix, very interesting -- it appears that that line is subject to interpretation, perhaps unsurprisingly, given the logographic spelling.

The MYS version available via the University of Virginia (direct link: MYS 11.2645) doesn't give the reading ikopu for the 息 glyph. Here's their text (with / added to show stanza breaks, since our templates munge the whitespace):

• 宮材引(みやぎひく) / 泉之追馬喚犬二(いづみのそまに) / 立民乃(たつたみの) / 息(やむ)時無(ときもなく) / 戀渡可聞(こひわたるかも) [Man'yōgana]
  宮材引(みやぎひ)く / 泉(いづみ)の杣(そま)に / 立(た)つ民(たみ)の / やむ時(とき)もなく / 恋(こ)ひわたるかも [Modern spelling]

  miyagi piku / idumi no sama ni / tatu tami no / yamu toki mo naku / kopiwataru ka mo

  (please add an English translation of this example)

@Mellohi!, if the association with the 息 (“breath”) kanji with the ikopu reading is valid, then this would suggest a connection to root yuk- / ik-, a very productive root, with various derived meanings:

• to go: 行く (yuku, iku)
• to live: 生く (iku)
• breath: 息 (iki)

The "breath" sense is pretty clearly a conjugated from of verb iku ("to live", by extension "to breathe"). From that, ikopu would be iku → irrealis conjugation stem ika- → ablaut shift to iko- → + iterative suffix -pu = "to be breathing", presumably also "to take a breather". HTH! ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:04, 31 May 2023 (UTC)

ぬ

Currently the usage notes start with "In Old Japanese and continuing in classical and later Japanese, the ending -nu and the ending -ta", but this is anachronistic. -ta suffix certainly has not existed yet in Old Japanese and Middle Japanese. -nu suffix should be compared here not with -ta suffix, but with -tu (つ) suffix (which I would rather prefer to romanize as -tu, not -tsu which does not match reconstructed Old/Early Middle Japanese pronunciation). I am aware that t part of -ta is distantly cognate with t of -tu (-ta < -taru < -tari < -te + ari, where -te is 連用形 of -tu), but -ta suffix appeared probably in Early Modern Japanese. The other part of the usage notes ("mainly for verbs that indicated transitive or intentional actions") also matches classical -tu suffix and not modern -ta suffix.

Would you prefer to just change -ta to -tu, without {{lang|ja|...}}? — This unsigned comment was added by Arfrever (talk • contribs).

@Arfrever: Apologies, I thought you were only replacing the romanization with the kana. Best practice for Japanese is to specify the reading in romaji unless the text is specifically talking about the graphemes, so other templates like {{ja-r}} or {{m}} should be used instead of {{lang}}. Also, I didn't catch that the description had -ta instead of -tsu. I've just fixed the text; please have a look.

Separately, please remember to sign your posts with four tildes, like this: ~~~~

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:49, 7 March 2020 (UTC)

p- reintroduction in onomatopoeic words

Hey! Thanks for explaining the thing, it's really what I could not even dare think about. Could you say if there is any literature on the topic I can get familiar with? —Fayanzar (talk) 20:37, 20 March 2020 (UTC)

• @Fayanzar: Glad that was helpful for you. I don't have my usual English-language references to hand, but if you can read Japanese, the JA WP article at w:ja:半濁音 describes the outlines. That explains the /p-/ initial as arising during the Edo period, but I'm pretty sure I've read elsewhere that it probably started showing up -- in some words at least -- during the preceding Muromachi period. It's also possible that I got my wires crossed, and I may be mixing this up with the word-medial /-p-/, which the JA WP article does mention as starting during the Muromachi.

If you're interested in the history of the Japanese language, I've read Shibatani's The Languages of Japan, which was a good solid academic overview, and that also touches upon Ainu. I've also heard good things about Bjarke Frellesvig's A History of the Japanese Language, but I haven't read that one myself. Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 21:12, 20 March 2020 (UTC)

薔薇

But it's biological organism which is usually written with katakana?Shen233 (talk) 01:38, 29 March 2020 (UTC)

@Shen233: Biology terms are only usually written in katakana in biology contexts. In other contexts, kanji or hiragana generally prevail. See also ja:w:ネコ科, for instance, where we find katakana ネコ used in the context of taxonomy and biology, but 猫 or ねこ in pretty much everything else. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:07, 29 March 2020 (UTC)

True, but for 薔薇 in particular, at least in my experience, バラ is seems more common than ばら. what do you think? Shen233 (talk) 06:22, 29 March 2020 (UTC)

@Shen233: Checking stats at Google Books, I get 198K for the kanji spelling, 230K for hiragana, and 184K for katakana.

However, hit counts are still a separate issue. For the yomi of native or nativized words (such as Middle-Chinese derived on'yomi terms, not strictly native), Wiktionary uses hiragana for kana renderings and for ruby text. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:38, 29 March 2020 (UTC)

Thanks for the info. Would you mind explaining what ruby text is? Shen233 (talk) 06:42, 29 March 2020 (UTC)

See ruby character. —Μετάknowledge^{discuss/deeds} 06:48, 29 March 2020 (UTC)

続く

This term is usually rendered in the hiragana つづく; the form w/the kanji is seldom seen. That's why I put it there. WorldwideBallcaps (talk) 06:41, 29 March 2020 (UTC)

@WorldwideBallcaps: ... except the kanji spelling is not rare. Checking a couple conjugation forms, the kanji spelling is actually more common on Google Books. What context are you describing? If manga, that's a special case and is generally not relevant for much of what we do here. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:57, 29 March 2020 (UTC)

It's also used in non-anime media, such as movies and live action shows, so it should merit some mention, I feel. WorldwideBallcaps (talk) 06:58, 29 March 2020 (UTC)

I have seen both and it seems interchangeable for me, though in media toward younger people つづく is more common, but that shouldn’t be an issue? Shen233 (talk) 07:01, 29 March 2020 (UTC)

@Shen233: Indeed, media targeting younger audiences will use fewer kanji as a general trend. That is not lexically important, however, and has no bearing on Wiktionary.

@WorldwideBallcaps: Again, the kanji spelling is not rare. Please stop adding usage notes that erroneously state that it is. This is trivially confirmable in Google searches. ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:16, 29 March 2020 (UTC)

下午

Would you mind fixing this Japanese entry when you get the chance? ---> Tooironic (talk) 05:18, 1 April 2020 (UTC)

@Tooironic: Done.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:47, 1 April 2020 (UTC)

Many thanks! ---> Tooironic (talk) 22:24, 1 April 2020 (UTC)

重箱読み

quick question on 重箱読み, I know it's 2-kanji compound where first is on and second is kun,but does it apply to a yojijukugo where first two is on and second two is kun like in 滅多矢鱈? Shen233 (talk) 06:53, 1 April 2020 (UTC)

@Shen233: Ya, for 重箱読み or 湯桶読み, we only consider two-kanji compounds, same structure as the source words 重箱 or 湯桶. Anything bigger is no longer 重箱読み or 湯桶読み.

(As a minor point, please don't forget to add -- or keep -- whitespace lines in the wikicode. Wiktionary convention is to keep some whitespace in the code to make it easier for editors to read.)

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:50, 1 April 2020 (UTC)

thanks, and as for the adjective/adjectival noun business, I think it's just the use of terminology. Adjectival noun is same as na-adjectives, and I seen many entries uses "adjectival noun" for na-adjectives like 綺麗 or 安静. Also https://en.wikipedia.org/wiki/Adjectival_noun_(Japanese)　Shen233 (talk) 18:43, 1 April 2020 (UTC)

@Shen233: Cheers, ya, those are stale entries. In past discussions, the consensus emerged to use more standard POS vocabulary, considering that our readership consists of English readers who may not be familiar with alternative labeling schemes like "adjectival noun" or "stative verb" (one idea I've run into for what to call -i adjectives, since they can be used predicatively), but who are probably familiar with the basic English terms for parts of speech like "adjective". I've been slowly cleaning up "adjectival noun" entries to use "adjective" instead, but I haven't been actively looking for them. Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:51, 1 April 2020 (UTC)

こたえる reading for 堪える

I tried adding another reading for 堪える, would you mind clean that up if you have time, thanks! Shen233 (talk) 23:43, 3 April 2020 (UTC)

賦能 - any possibility from Japanese?

Hi there. I was wondering if there is any possibility the Chinese term 賦能 might have come from Japanese? The term does not exist natively in Chinese, and for now I can only assume it was calqued from English empower. ---> Tooironic (talk) 06:21, 7 April 2020 (UTC)

@Tooironic: FWIW, I've never encountered this. Doesn't seem to be used at all in JA. See lack of any entry at https://kotobank.jp/word/賦能 or at https://www.weblio.jp/content/賦能 and the lack of any likely meaning in our single-character entry at Japanese 賦. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:27, 7 April 2020 (UTC)

Thank for checking that. I guess 賦 is not a common character in Japanese at any rate. In my Japanese-English dictionary, the only entries that come up are for 賦課, 賦与 and 賦課, which are red links for now. ---> Tooironic (talk) 11:13, 7 April 2020 (UTC)

About 百合 and あしび

Hello. I'm glad to see you for the first time. My name is Japoniarra. Let me ask you straightforwardly, on what basis were these two items "百合" and "あしび"'s rollback? Please let me know. If there is no basis, the item will be reverted. --ハポニアラ (talk) 05:19, 24 April 2020 (UTC)

Hello @ハポニアラ, let me address each in turn.

• At Japanese 百合, you added two surname readings, Moai and Momai.

• I cannot confirm these surname readings. The closest I can find is Momae, as listed in ENAMDICT here.
• Separately, these readings have separate etymologies and pronunciations from the yuri reading section that you'd added these to.

Given these two problems, I reverted your edit.

If these readings can be confirmed, then the entry structure must be changed to include at least one separate etymology section, ideally one separate section for each separate reading.

• At Japanese あしび, you removed the kanji spelling 馬酔木. This spelling is confirmable as a surname with the reading Ashibi at ENAMDICT here.

Since this spelling for the surname is confirmable, I reverted your removal.

Please let me know if you have any further questions or concerns. Responsibilities IRL are keeping me plenty busy, so please note that replies may not be immediate. ‑‑ Eiríkr Útlendi │^{Tala við mig} 16:37, 27 April 2020 (UTC)

To @Eirikr

Thank you for taking the time out of your busy schedule to respond.

I'll explain each of these step-by-step.

• Regarding 百合, I have added two more readings. It's "Moai" and "Momai".

• You said that the reading of these surnames could not be confirmed by ENAMDICT, but I have my reasons. If you search for "百合" on this page of the website "Dictionary of Actual Surnames (実在苗字（名字）辞典)", you will find "yuri (ゆり)", "moai (もあい)" and "momai (もまい)" in hiragana.
• As for the other point you made about having to separate the etymological sections when the readings are different, I didn't study it. I'll be careful from now on.

Therefore, I would like to have two new sections in "百合".

The same can be said about あしび.

• In other words, you said on the ENAMDICT page that the surname "Ashibi" is written with the Chinese character "馬酔木", but when I look at the "Dictionary of Actual Surnames", I see "馬酔木" is read as "Asebi(あせび)", "Masuki(ますき)" and "Maseki(ませき)" (as an aside, I did not find the reading "Maseki" when I searched on Google, but I found the reading "Masebo" on this page of the "The Japanese surnames etymology dictionary (日本姓氏語源辞典)", so the manager of the "Dictionary of Actual Surnames" probably misread the phone book).

By the way, have you ever heard of the term "ghost surname (幽霊名字)" proposed by Japanese surname researcher Hiroshi Morioka (森岡浩)? This term refers to a surname that is listed in a surname dictionary but whose existence cannot be confirmed.

What I am saying is that the ENAMDICT you are basing your argument on may contain many non-existent readings of real surnames, so to speak "ghost readings".

Or, conversely, it is possible that there are many readings of actually existing surnames that ENAMDICT does not list.

And therein lies the reason why I base my "Dictionary of Actual Surnames (First Names)" on it. This website, as the name implies, is a site that tries to list only real surnames and their readings. And the rationale for that, too, is shown here, by the order in the phone book and the names on the Internet. Therefore, I thought that the rationale on this site would be easy to revisit and highly reliable

However, if you still claim that there is no surname that reads "百合" as "moai" or "momai", or that there is a surname that reads "馬酔木" as "ashibi", then I think you should do an exact match search on a search site for the kanji and readings (hiragana, katakana, romaji) of the surname. If you do that, I think you'll find most of the readings of surnames that actually exist.

However, since the proof of the thing (surname) which does not exist is difficult and becomes "the devil’s proof", I think it is better to do the proof of the thing (surname) which exists basically.

Of course, if you ask me for more details, I will immediately prove the existence of the surnames "Moai" and "Momai" with materials other than the "Dictionary of Real Surnames".

I think the important thing is to make sure that no matter what site we use as a source, the last name is real or not.

I apologize for the lengthy post.--ハポニアラ (talk) 08:12, 30 April 2020 (UTC)

Almost any surname you claim to exist is questionable. If you're going to claim to exist, prove it with actual directories on the web, not a site with a collection of surnames on it. -- ハポニアラ (talk) 13:48, 12 May 2020 (UTC)

@ハポニアラ: As you mentioned above, this is trying to prove a negative. Various sources online list these readings and spellings that have been at issue. Even if they are ghost readings / spellings, we may still have entries for them, ideally indicating that these are ghost readings / spellings and listing any corroborating references. Just as you state that we have no guarantee that a reading or spelling listed in ENAMDICT or myoji-yurai is authoritatively correct, we similarly have no guarantee that the lack of a reading or spelling in your reference of choice is authoritative proof that such reading or spelling does not exist.

If you would like to bring these up for further discussion with a broader audience, I would suggest creating relevant threads at WT:RFV. If you would like to discuss references and criteria for inclusion, I would suggest creating relevant threads at WT:BP. ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:53, 13 May 2020 (UTC)

"proving a negative": like sending a non-existent term to WT:RFV? —Suzukaze-c◇◇ 23:58, 13 May 2020 (UTC)

If it's about finding verification, or confirming lack thereof, then yes, request for verification would be the correct forum. My points above are that there does appear to be verification for the readings and spellings at issue. ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:11, 14 May 2020 (UTC)

with DEFAULTSORT at the top, other `sort` args aren't needed

Would that it were true, but see, for example Category:ja:Film, where コンテ is now sorted under コ rather than こ.

As far as I have worked out, DEFAULTSORT does seem to propagate to {{bor}} / {{inh}} /{{der}}, but not to {{lb}} or to {{top}} / {{C}}. I'm not sure about other templates. Cnilep (talk) 04:49, 1 June 2020 (UTC)

@Cnilep: Oh, dear. Thank you for cluing me in. That inconsistency is rather distressing. I suspect that the {{lb}}, {{top}}, {{C}} templates / modules need updating, but I haven't the foggiest how. ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:56, 1 June 2020 (UTC)

難しい (muzukashii, “difficult”) and probable Ryukyuan relatives

@Kwékwlos There seems to be multiple Ryukyuan terms related to this adjective, somehow, despite being a -ka + -shik- adjective which you'd usually think has no business in Ryukyuan. Interestingly, Jarosz's relatively recent paper "The reflexes of Proto-Ryukyuan close vowels in a post-nasal position" reconstructs Proto-Ryukyuan *motika- of identical meaning as the proto-form of several similar Ryukyuan adjectives, like Okinawan むちかさん (muchikasan). I assumed these were early-stage borrowings (note that *u after *t often merges with the reflex of *i), but who knows? mellohi! (僕の乖離) 13:12, 9 June 2020 (UTC)

First of all, if we expect native terms in Ryukyuan (take Shuri Okinawan for example), you have attestations of *midare > Okinawan njariyun, which presupposes a reduction of *mi- > 'n-. But then you have terms like *miti > Okinawan michi instead of expected *nchi, which may suggest earlier *meti if we follow the fact that *ke > ki and *ki > chi, or else a borrowing from Japanese. The same could be said for PR *mu, which has the same reflex in cases such as *mukade > Okinawan nkaji. Hence he reconstructs *motika-, but how would you expain the non-palatalization of *k here after an *i? Therefore I conclude this word probably entered Ryukyuan as a Japanese loanword after the sound changes *k > ch (after *i) and *e, *o > i, u, since we would have expected *muchichasan if it was indeed an inherited word, which is unlikely since this word has no known Old Japanese cognate. Kwékwlos (talk) 14:09, 9 June 2020 (UTC)

• @mellohi!, the ～しく vs. ～く distinction in the adverbial for Old Japanese adjectives may have been the innovation. I see, for instance, the Shuri Okinawan adjective あさまさん corresponding with mainland 浅ましい (asamashii) (both presumably from a proto verb asamu reflected in JA 浅む (asamu)), just as Shuri あささん corresponds with mainland 浅い (asai). Similarly, Shuri あたらさん corresponds to (now-obsolete) mainland 惜し or 可惜し (atarashi), Shuri うとぅなさん and mainland おとなしい (otonashii), Shuri うれーまさん and mainland 羨ましい (urayamashii), Shuri をぅかさん and mainland おかしい (okashii), etc. etc.

There appears to be enough lexical overlap for Shuri -asan adjectives and mainland -ashii adjectives that I suspect these are cognates and not borrowings. We also seem to see similar morphological development where at least some of these appear to derive from verbs in the -a stem form, plus an adjective-forming suffix.

‑‑ Eiríkr Útlendi │^{Tala við mig} 22:04, 9 June 2020 (UTC)

• @Kwékwlos, mu in mainland Japanese also collapses to n in certain contexts -- albeit not in word-initial position. The most common of these was the suppositional suffix, evolving into either -u or -n in Classical Japanese. Meanwhile, medial -mi- in verbs collapses to -n- regularly to produce the modern conjunctive ～んで forms, from older ～みて, for verbs like 絡む (karamu), 込む (komu), 窄む (tsubomu), etc.

Looking at Shuri, contrasting initial mi- in midaru → njariyun and initial mi- in miti → michi, I wonder if pitch accent might be a factor. Perhaps also the derivation of miti as honorific prefix mi- + base noun ti (“path, way”).

Re: palatalization of Ryūkyūan *k, I'm only familiar with that when the consonant is followed by a front vowel, not when preceded by it. Proto-Ryūkyūan -ika- would still yield modern -ika-, while Proto-Ryūkyūan -aki- would yield modern -achi-. See also Shuri ちかさん corresponding to mainland 近い (chikai). Or have I missed something?

‑‑ Eiríkr Útlendi │^{Tala við mig} 22:04, 9 June 2020 (UTC)

For Shuri, you have the doublet chichasan, which is inherited. The form with -k- is irregular or a likely Japanese borrowing. There appears to be no evidence that pitch accent did influence the sounds. Kwékwlos (talk) 22:23, 9 June 2020 (UTC)

There are other Shuri words that have -ika-. C.f. てぃーじかさん, likely corresponding to mainland 手近い (tejikai), むちかさん at issue in this thread (albeit from likely root form mutuku), あんじかび with alternative form んちゃび, ちむしからーさん of uncertain derivation, etc. It does not appear that -ika- uniformly resulted in affricated forms. ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:19, 9 June 2020 (UTC)

I also note that there appear to be doublets for the words related to midare:

• Mainland 乱れ (midare, noun) → みだり ↔ んじゃり
• Mainland 乱れる (midareru, verb) → みだりゆん ↔ んじゃりゆん

Not a doublet, but another term exhibiting initial mida- rather than nja-:

• Mainland 乱れ世 (midare yo) → みだりゆー, apparently as used on 久米島

I suspect that the doublets represent the same words at different stages of sound change. ‑‑ Eiríkr Útlendi │^{Tala við mig} 03:55, 10 June 2020 (UTC)

Re: honorific prefix mi-, a quick search reveals that this remains in Shuri みくち (honorific 御 + 口　“mouth”) and みくし (honorific 御 + 腰 “hips”). This may explain why the initial mi- persists in Shuri michi, and has not collapsed into n-. ‑‑ Eiríkr Útlendi │^{Tala við mig} 04:28, 10 June 2020 (UTC)

@Kwékwlos: I have a high belief that Jarosz's *motika- should be reconstructed as *motuka- instead, which not only resolves the palatalization problem but also provides a better vowel correspondence to 難しい (muzukashii) itself. mellohi! (僕の乖離) 04:49, 10 June 2020 (UTC)

Revisiting this thread after a while. @Kwékwlos, you stated above,

Therefore I conclude this word probably entered Ryukyuan as a Japanese loanword after the sound changes *k > ch (after *i) and *e, *o > i, u, since we would have expected *muchichasan if it was indeed an inherited word, which is unlikely since this word has no known Old Japanese cognate.

Although Japanese 難しい (muzukashii) itself is only cited to around 1000, it does have an Old Japanese derivation. The adjective is listed in multiple monolingual Japanese dictionaries as cognate with verb 憤る (muzukaru, “to throw a tantrum; to get into a bad mood, to become upset”). Modern muzukaru is from older (but still current) reading mutsukaru, cited all the way back in the Nihon Shoki of 720.

Separately, I note the existence of reduplicative adverb むつむつ (mutsumutsu, “sullen”, archaic), and related forms むっつり (muttsuri, “sullen”) and むっと (mutto, “sullen; stuffy, muggy”). This may be related to reduplicative adverb ぶつぶつ (butsubutsu, “complainingly, grumblingly”).

Morphologically, mutsukaru implies older verb form mutsuku. However, that form is only cited to 1191; it may be a back-formation, or it might just not have been recorded earlier. OJP mutukaru might be derived from an unattested earlier mutuku, or it might represent a verb derived from the adverbial root mutu.

There are many -ashii adjectives in Japanese that derive from verbs and can be analyzed as "of a quality to cause [VERB]". By one analysis, these could all be a shift in usage from the 連用形 (ren'yōkei, “continuative / infinitive / noun / stem conjugation”) of the causative form of the verb. In this case, it would proceed from verb root mutsuk + -u (terminal) → mutsuk + -asu (causative in the terminal form) → mutsuk + -ashi (causative in the continuative form) → re-analysis or repurposing as an adjective in the -shi terminal form.

In fact, this might explain why the しき attributive and しく adverbial conjugation arose: if the し derived initially not as an adjectival suffix, but as a part of an underlying conjugated verb form, the し would not change to き or く as it does for regular し-terminal adjectives, but would instead require the additional suffixes.

→ To conclude, the underlying Japanese etymon here appears to be an adverbial root mutu, with derived verb mutukaru arising from that, and adverb mutukasi. I'm curious if there are any Ryukyuan cognates aside from this *motuka. ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:50, 18 September 2020 (UTC)

Module Error at 破落戸

I think this is caused by something about the main entry (ならず者) that {{ja-see}} doesn't know how to process. One thing I noticed is that there are no readings in the entry, just a pronunciation section. That may be only a coincidence, but it's something to start with. Chuck Entz (talk) 03:45, 1 July 2020 (UTC)

Never mind- I see @Suzukaze-c has already fixed it. Chuck Entz (talk) 08:33, 1 July 2020 (UTC)

At some point, ja-headword and ja-r started to accept kanji with no reading. I think it's an odd modification. (@User:Huhu9001?) —Suzukaze-c (talk) 08:36, 1 July 2020 (UTC)

@Suzukaze-c: ja-headword had always been allowing "kanji with no kana" before I modified it. So I had to make "kanji with no kana" a tracking rather than an error: Special:WhatLinksHere/Template:tracking/ja-headword/no_kana_form_is_available. I fixed the old ones like 蝶蝶, 火氣, 蛋黃, but someone seems to be adding new ones. -- Huhu9001 (talk) 08:57, 1 July 2020 (UTC)

@Chuck Entz, Suzukaze-c, thank you both. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:58, 4 July 2020 (UTC)

気持ち悪い

Did you mean Daijisen says there is a space between "kimochi" and "warui"? But it seems Daijisen does not provide romaji for these words. See this. -- Huhu9001 (talk) 18:04, 28 July 2020 (UTC)

@Huhu9001: Apologies for the confusion, no, that was not my intent -- rather, that Daijisen ultimately describes this as from the phrase 気持ちが悪い (kimochi ga warui). I think your edits to the etymology helped to clarify; thank you for that.

Regarding spacing in the romanization, our practice is to separate individual terms by spaces, except in cases of rendaku, when either we'll spell it as one thing or use a hyphen. ‑‑ Eiríkr Útlendi │^{Tala við mig} 03:16, 30 July 2020 (UTC)

Issues:

1. It is not a contraction. It is an omission of が in a clause.
2. "Passive" and "active" make sense for adjectives, see suspicious.
3. Many Wiktionary entries use semicolons for for separating synonymous senses. -- Huhu9001 (talk) 05:04, 30 July 2020 (UTC)

@Huhu9001: I'm confused by your opposition. An omission such as this is a shortening, a contraction of the fuller phrase. If "contraction" is too specific for you, would "shortening" be more acceptable? Where possible, we should avoid using the overly general "From ..." and state in more specific detail how the term derives.

Regarding "passive" and "active", I disagree with the usage at English suspicious, and refer you also to w:Passive voice. Further, I'll point out that most English readers will not be familiar with this use of "passive" and "active" to refer to adjectives -- this is strange and non-standard, and strikes me as jargon.

Regarding punctuation, when a sense gloss itself does not include commas, and there are multiple synonymous glosses, and moreover the entry as a whole has other terms with multiple non-synonymous glosses, differentiating the use cases by different punctuation simply makes sense. For 気持ち悪い, the first two sense lines list synonymous senses, whereas for the third, the two senses are more distinct: "feeling bad" and "in a bad mood" are disparate and not necessarily synonymous.

Separately, looking again at the entry, your use of clause is a bit confusing. See also w:Clause. If 気持ち悪い is used as a standalone statement, it could be described as a clause with an omitted and implied subject (whoever or whatever is "disgusting" or "feeling ill"). If 気持ち悪い is used as an adjective, it cannot be a clause, and is instead clearly a lexicalized term.

I'll synthesize the above into the entry. ‑‑ Eiríkr Útlendi │^{Tala við mig} 05:28, 30 July 2020 (UTC)

Your {{q|of a thing}} is wrong. The sense "disgusting" can be used on humans as well as non-humans ("気持ち悪い男"). Also {{q|of a person}} for the sense "sick" is problematic. It is not that this sense is restricted to humans but that we seldom describe the feeling of a non-human. We don't have that label in the entry sick.

The clause does not have "an omitted and implied subject", its subject is 気持ち. If translated literally, it is a clause meaning "where/in which the feeling is bad". Consider "気持ち悪い形" ("a disgusting shape", or lit. "a shape where the feeling is bad"). This is how Japanese clauses work. Japanese adjectives and clauses share the same form of syntax. They are not different by nature. -- Huhu9001 (talk) 05:57, 30 July 2020 (UTC)

先生

Thanks. But you reintroduced a number of lint errors due to the embedded bold formatting. Do you intend to repair the underlying template so it doesn't generate MALFORMD html when using wikimarkup then? ShakespeareFan00 (talk) 17:52, 13 August 2020 (UTC)

@ShakespeareFan00:

I'm not sure what you're talking about.

• The markup is not malformed. Here is the source as viewed in Firefox 68.11.0(esr), pretty-printed for (slightly) easier viewing:

<dl>
    <dd>
        <span class="Jpan" lang="ja"><ruby>学<rp>(</rp><rt>がっ</rt><rp>)</rp></ruby><ruby>校<rp>(</rp><rt>こう</rt><rp>)</rp></ruby>のみんなも<b><ruby>先<rp>(</rp><rt>せん</rt><rp>)</rp></ruby><ruby>生<rp>(</rp><rt>せい</rt><rp>)</rp></ruby></b>のことさ<ruby>好<rp>(</rp><rt>す</rt><rp>)</rp></ruby>いてる……<ruby>母<rp>(</rp><rt>か</rt><rp>)</rp></ruby>っちゃすきか⁉</span>
        <dl>
            <dd style="font-style:italic">Gakkō no minna mo <b>Sensei</b> no koto sa suiteru…… Katcha suki ka⁉/dd&gt;</dd>
            <dd>Everybody at school likes Sensei…… Ya like him, Ma⁉</dd>
        </dl>
        <span class="Jpan" lang="ja">もちろん<ruby>好<rp>(</rp><rt>す</rt><rp>)</rp></ruby>きよ‼ほんとにいい<b><ruby>先<rp>(</rp><rt>せん</rt><rp>)</rp></ruby><ruby>生<rp>(</rp><rt>せい</rt><rp>)</rp></ruby></b>ですもの…</span>
        <dl>
            <dd style="font-style:italic">Mochiron suki yo‼ Honto ni ii <b>sensei</b> desu mono…/dd&gt;</dd>
            <dd>Of course‼ He’s a very good teacher…</dd>
        </dl>
        <span class="Jpan" lang="ja">オラもすきだあ。んでも、<ruby>先<rp>(</rp><rt><b>せん</b></rt><rp><b>)</b></rp></ruby><b><ruby>生<rp>(</rp><rt>せい</rt></ruby></b><rp>)</rp>は<ruby>先<rp>(</rp><rt><b>せん</b></rt><rp><b>)</b></rp></ruby><b><ruby>生<rp>(</rp><rt>せい</rt></ruby></b><rp>)</rp>だあ…オラの<ruby>父<rp>(</rp><rt>と</rt><rp>)</rp></ruby>っちゃにはなれねえ！</span>
        <dl>
            <dd style="font-style:italic">Ora mo suki dā. N demo, <b>Sensei</b> wa <b>Sensei</b> dā… Ora no totcha ni wa narenē!/dd&gt;</dd>
            <dd>Ah like him, too. But Sensei’s only Sensei… He could never be mah pa!</dd>
        </dl>
        <span class="Jpan" lang="ja">え…⁉それ…ひょっとして、<ruby>私<rp>(</rp><rt>わたくし</rt><rp>)</rp></ruby>と<ruby>八<rp>(</rp><rt>や</rt><rp>)</rp></ruby><ruby>重<rp>(</rp><rt>え</rt><rp>)</rp></ruby><ruby>樫<rp>(</rp><rt>がし</rt><rp>)</rp></ruby><b><ruby>先<rp>(</rp><rt>せん</rt><rp>)</rp></ruby><ruby>生<rp>(</rp><rt>せい</rt><rp>)</rp></ruby></b>が<ruby>結<rp>(</rp><rt>けっ</rt><rp>)</rp></ruby><ruby>婚<rp>(</rp><rt>こん</rt><rp>)</rp></ruby>するってこと……⁉</span>
        <dl>
            <dd style="font-style:italic">E…⁉ Sore… hyotto shite, watakushi to Yaegashi <b>Sensei</b> ga kekkon surutte koto……⁉/dd&gt;</dd>
            <dd>What…⁉ Wait… You think I’m gonna marry Yaegawa Sensei……⁉</dd>
        </dl>
    </dd>
</dl>

The above parses as well-formed markup.

• The page renders correctly before your edit and after my reversion, and renders incorrectly after your edit and before my reversion.

If you had previewed the page and expanded the quotation, you would have seen two red examples of Template:bold, and one big Lua error message also in red:

Lua error in Module:ja-ruby at line 544: Separator "%" in the kanji and kana strings do not match.Lua error in Module:ja-ruby at line 544: Separator "%" in the kanji and kana strings do not match.

Both are clear indications that your edit was problematic, and unhelpful to the reader.

The appearance of the page as rendered is what our readers see. This is more important than strict adherence to technical standards. If you can find a way to resolve the lint errors without goofing up the appearance of the page, then please do so. I don't care about lint errors so long as the page renders correctly.

‑‑ Eiríkr Útlendi │^{Tala við mig} 18:24, 13 August 2020 (UTC)

Thanks for the explanation. Now to review every single 'lint-error' fix made in the last 48 hours or so to make sure there aren't any more inadvertent problems. Unless you want to review my 'cleanup' efforts yourself? ShakespeareFan00 (talk) 18:27, 13 August 2020 (UTC)

@ShakespeareFan00: Thanks, but no thanks!  :) FWIW, looking at your recent activity, Wigidonar looks identical before and after your changes. Were I in your position, I might ping other editors listed in the history for that entry to see if they have any concerns about your change in the formatting of the wikitext. I only backed out the changes you made at Chinese 乖張／乖张 (guāizhāng) because that goes against long-standing editor consensus, and should be discussed with other editors before implementing (and that pattern of including linebreaks in the template calls is used in just about every ZH entry); and at Japanese 先生 because it visibly broke the page (from a reader's perspective, not necessarily the browser's technical perspective).

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:40, 13 August 2020 (UTC)

https://en.wiktionary.org/w/index.php?title=Module:ja-headword&diff=prev&oldid=60059564 was something I'd appreciate a second opinion on though, as it seemingly resolved a 'misnesting' error on over 70.000 pages! (The Job queue hasn't caught up yet.) ShakespeareFan00 (talk) 18:50, 13 August 2020 (UTC)

@ShakespeareFan00: I think your fix on Wigidonar looked appropriate. The formatting before the fix looks odd to me. — justin(r)leung _{{ (t...) | c=› }} 19:04, 13 August 2020 (UTC)

• @ShakespeareFan00: Your edit over at Module:ja-headword looks good to me, as best I can tell.

(As you noted in another edit comment, our Module code is woefully underdocumented, and for some reason our coding conventions even recommend against adding comments to document the code inline -- something I think is a fundamental mistake, but there you go. I'm familiar with Jeff Atwood's position as explained in his blog post linked from our coding conventions page. However, his advice to reduce inline comments as much as possible only works within an idealized situation where everyone working on the code has much more context than we do -- his developers are highly familiar with the code language; all functions and parameters use obvious and self-explanatory names. We (the Wiktionary editor community) don't have either -- many of us aren't familiar with the Lua language, nor should we assume that everyone new to the project is either; our functions and parameters often don't have obvious and self-explanatory names.)

In the meantime, I was playing around with an HTML validator, and while the output of the ruby functions is well-formed, it is not valid HTML, as you noticed with the Lint checker. One key is that apparently the dl element requires a dt child before any dd children, whereas the module consistent outputs dl without any dt. I may have also spotted a nesting tag structure error, where a b tag would up in the wrong place (splitting up a ruby set of elements in a strange way).

It would be much safer to deal with markup as actual elements, rather than text strings, but I'm not sure how possible that is with our Lua setup. ‑‑ Eiríkr Útlendi │^{Tala við mig} 21:41, 13 August 2020 (UTC)

Lua has mw.html. See Module:zh-dial-map for an example. —Suzukaze-c (talk) 04:05, 15 August 2020 (UTC)

螳臂當車

As you have more competence with CJKV scripts and the templates used, Can you take a look at this and figure out why the quote in the Etymology appears to be malfromed? ShakespeareFan00 (talk) 09:10, 16 August 2020 (UTC)

ShakespeareFan00 (talk) 09:09, 16 August 2020 (UTC)

@ShakespeareFan00: The Nu HTML Checker is an online HTML validator, and this is their output for the 螳臂當車 page.

Looks like the most common issue is misuse of <dl>, an issue that came up not too long ago, in this thread over in the Beer Parlour: [[Wiktionary:Beer_parlour/2020/June#Using_"#*"_or_"#:"_with_requests_for_quotation]]

It seems like this is a fundamental design flaw in how the MediaWiki back-end interprets wikicode. Consequently, I'm not sure how easy this'll be to fix. ‑‑ Eiríkr Útlendi │^{Tala við mig} 05:03, 17 August 2020 (UTC)

水

Hello, I'm not entirely sure why my edit on 水 was reverted - elsewhere in the page it says that the historical kana for すい was すゐ. LittleWhole (talk) 18:35, 16 August 2020 (UTC)

@LittleWhole: Ya, an anon from Texas added that mistakenly back in 2017, but I can't find any sources that list this as historical. It may be ancient, as in, a reading used before the invention of kana, but we have no textual evidence for it. (For that matter, I'd recently changed that to an "ancient" reading, but I think it's safer at this point to just remove it altogether, until and unless we can get cited textual evidence.)

As an example, have a look here at the Kotobank page. The top of the page includes a list of readings for the 水 kanji headword, with historical readings in square brackets after the modern ones (where they differ, anyway). We see みず〔みづ〕 and もい〔もひ〕, but no すい〔すゐ〕.

HTH, ‑‑ Eiríkr Útlendi │^{Tala við mig} 04:47, 17 August 2020 (UTC)

That address was from me, but it wasn't from Texas at all according to your analysis. From the address, it's probably from an AT&T hospot which changes address somehow everyday. The すゐ ancient reading can be found in very old dictionaries, think we've discussed this before... ～ POKéTalker（═◉═） 08:24, 17 August 2020 (UTC)

@POKéTalker, I didn't know that was you. And for me right now, https://www.ip2location.com/demo/2602:306:8054:B620:5486:D528:A934:3C5C is still showing this IP as originating from Dallas, Texas. Curious if that URL gives you different results? I didn't think IPs were supposed to move around geographically...

The only thread I'm aware of that touches on すゐ is [[User_talk:Poketalker/2019#Japanese_kanji_historical_reading_{{d}}_requests]]. That doesn't go into すゐ or related ～ゐ kana spellings in any detail, however. And sadly the Heibonsha link you gave there is currently not working for me, with the site giving the following message:

この資料は、著作権の保護期間中か著作権の確認が済んでいない資料のためインターネット公開していません。閲覧を希望される場合は、国立国会図書館へご来館ください。

Re-reading [[w:ja:ゐ#江戸時代]] and the immediately following section [[w:ja:ゐ#明治時代以降]], I'm inclined to view すゐ etc. not as historical kana, so much as an academic tangent that didn't go anywhere. Considering that no modern sources I've seen include すゐ as an historical kana spelling, I don't think we should either. ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:33, 17 August 2020 (UTC)

The website that our main geolocation link goes to (www.ip2location.com) has some real holes in its database when it comes to IPv6. Most Comcast addresses are shown as Mt Laurel, NJ, which happens to be Comcast corporate HQ, and AT&T shows more often than not as "Redmond, WA". Because of this, I added an alternate link that Wikipedia was using on their IP contributions pages: whatismyipaddress.com. If you click on the alternate geolocation link for the IP address in question, you'll find that it shows the IP as AT&T U-verse in Montebello, CA (given the way ISPs operate in large urban areas, I suspect "Montebello" isn't just Montebello itself). The main geolocation link is somewhat better for IPv4 and it has much more information about proxies and VPNs, so I haven't changed it. I would advise using the alternate link for IPv6 and the main one for IPv4, but checking both if you need to be certain. Chuck Entz (talk) 03:31, 18 August 2020 (UTC)

Are katakana small forms strictly Ainu?

Please look at the contributions of 2600:1:B122:B069:45EE:40EE:8873:10F4 (talk • contribs • whois • deleted contribs • nuke • abuse filter log • block • block log • active blocks • global blocks), who has systematically replaced all of the translingual sectios for small-form katakana with Ainu sections. This would only make sense if these were specific to Ainu and not used for any other languages. I started to revert these, but realized I didn't know enough on the subject to be mass-reverting someone's entire contributions without doing a lot of research. Thanks! Chuck Entz (talk) 02:10, 24 August 2020 (UTC)

@Chuck Entz: Thank you for the heads up!

Hmm. Some of the small katakana forms might currently be limited in use to just Ainu, or at least in materials we editors are aware of, but 1) some of these have also definitely been used to transcribe Taiwanese (see w:ja:台湾語仮名 ("Taiwanese kana"), and 2) I see more and more inventive use of kana in dialectology and other Japanese linguistic pursuits, so defining these as only Ainu would be a mistake, I believe.

In other glyph entries, standard MO as I've understood it has been to start with the ==Translingual== section and then list language sections under that -- even if there's currently only one language, as at ά, or even no languages, as at Ϋ. Does that match with your understanding?

Separate but related, {{mul-kanadef}} needs updating -- it makes no reference to Taiwanese kana, or other usage by academic linguists in Japan. I'll see if I can do anything useful with the template. ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:58, 24 August 2020 (UTC)

Deletion of Acela Republican redirect in error

I believe your deletion was in error. The redirect is acceptable under the redirection guidelines, "Sum-of-part terms that are likely to be searched, to the part that the meaning mainly derived from" Purplebackpack89 16:44, 25 August 2020 (UTC)

Do you know what "sum-of-part terms" means? —Μετάknowledge^{discuss/deeds} 18:45, 25 August 2020 (UTC)

An "Acela Republican" could easily be defined in a sum-of-parts manner: a Republican who lives in the Acela region of the country (def. #2 for Acela). Purplebackpack89 19:29, 25 August 2020 (UTC)

Were I to search for [[Acela Republican]] and land on the [[Acela]] page, I would be confused. In fact, that's what happened, which prompted me to delete the redirect.

If a user searches for a term and finds nothing, that's a clear indicator that we don't have an entry. I strongly believe that this is preferable behavior, as opposed to the user landing on a separate page that does not include the sought-for term. ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:22, 26 August 2020 (UTC)

Don't you think that deletion out of confusion is a bad approach? Shouldn't you have ascertained why it was the way it was before deleting? Purplebackpack89 02:33, 26 August 2020 (UTC)

In answer to your questions:

• Don't you think that deletion out of confusion is a bad approach? → No, I don't. If I thought it were a bad approach, I would not have deleted the redirect.
• Shouldn't you have ascertained why it was the way it was before deleting? → Yes, I did. My initial confusion prompted me to investigate further, and upon comprehending the circumstances and details, I ascertained that this redirect was more likely to confuse other users than to serve any productive use. And, as I stated above, I strongly believe that users are better served by landing on a Wiktionary does not yet have an entry message, rather than landing on a separate page that does not include the sought-after term. In software design contexts, I have heard this described as "the principle of least surprise".

You appear to assume bad faith and malfeasance on my part. Your accusatory tone is inappropriate.

Even as a redirect for a sum-of-parts term, the phrase Acela Republican is composed of the term Acela modifying the head noun of Republican. Put another way, an Acela Republican is a type of Republican. Redirecting to Acela instead makes it seem that the term Acela is somehow the head noun of this phrase, which is both incorrect and confusing. ‑‑ Eiríkr Útlendi │^{Tala við mig} 22:03, 26 August 2020 (UTC)

Wiktionary:Usernames and user pages

If it's a "draft" doesn't it logically follow that it can still be edited? And why do you object to that particular part of the policy? If people want to view previous revisions of a user's talk page, you can always check the history. Purplebackpack89 01:36, 28 August 2020 (UTC)

Editing the page is not an issue. I do note that other editors have discussed proposed changes to policy and draft policy pages, and I did not see any such discussion thread regarding this page. Perhaps I missed it? Your particular edit did not appear to add anything substantive. Rather, it seemed to be an attempt to shore up your contentions in other threads about editorial discretion regarding content on your own talk page -- it appeared to be an attempt at post hoc wikilawyering, and thus appropriate to revert.

I realize that this might be upsetting to you. I do not intend to upset you. I am simply describing what things look like to me. ‑‑ Eiríkr Útlendi │^{Tala við mig} 07:09, 28 August 2020 (UTC)

Whether it's post hoc or not, it's something that I've felt that needed to be in those guidelines for years. There is a discussion open now. Purplebackpack89 12:40, 28 August 2020 (UTC)

And it DOES bother me that I seem to be the ONLY person who's ever had revisions to his OWN talk page reverted recently. Equinox deletes comments from his page all the time; most recently within the last day or two. Nobody stops him, nor should they. Why me? Purplebackpack89 13:38, 28 August 2020 (UTC)

@Purplebackpack89: Re: reversions to user Talk pages, I've seen it happen before, albeit rarely; it's not just you. I've generally witnessed reversions if another editor feels like the thread has been altered or deleted in a way that misrepresents or hides content under discussion elsewhere, or if the other editor is party to that thread and takes issue with the deletion. I'm not sure what happened with your Talk page recently.

Thank you for starting the discussion thread. I'll have a look there. ‑‑ Eiríkr Útlendi │^{Tala við mig} 02:48, 29 August 2020 (UTC)

Category on pronunciation with different forms of pitch accent

I’m a little surprised that for the Japanese entries there aren’t categories by the pitch accents, atamadaka, nakadaka, odaka, heiban, etc... the only ones I see is “Japanese terms with IPA” which is alright, but not particularly useful IMO. Shen233 (talk) 01:47, 3 October 2020 (UTC)

@Shen233: My subjective impression is that categorization has not been much of a priority for most of the Japanese editors of late. I agree that categories like what you describe would be useful. Considering our current pronunciation infrastructure -- specifically, {{ja-pron}} -- I don't think it would be too difficult to add that in, in an automated fashion, such that all existing entries that include {{ja-pron}} would be added to the relevant categories without editors having to alter each individual entry.

That said, adding such categories would require altering the module(s) referenced by the template. @Suzukaze-c, is that something you could handle? If not, do you know who could? ‑‑ Eiríkr Útlendi │^{Tala við mig} 08:53, 5 October 2020 (UTC)

@Eirikr: I don't have the confidence or the patience to understand Module:ja-pron right now. Sorry.

From the edit history, maybe @Naggy Nagumo would be interested. —Suzukaze-c (talk) 06:33, 8 October 2020 (UTC)

✗ Naggy Nagumo (talk) 05:04, 9 October 2020 (UTC)

あまり、あまりに、あまりにも

I see how あまりに can be redundant, but contrast with just あまり、 あまりに and the more emphatic あまりにも are used in affirmative sentences, so maybe something worth mentioning? Shen233 (talk) 03:34, 22 October 2020 (UTC)

@Shen233: Ya, a usage notes section is probably called for. ‑‑ Eiríkr Útlendi │^{Tala við mig} 03:40, 22 October 2020 (UTC)

Formatting of non-English definitions

Hello Eirikr, thanks for the note about the formatting of non-English definitions that you recently left me in the editor summary of a change to 音読み. I find interest in learning about all of the intricacies of English Wiktionary norms so I wanted to ask about the one related to this situation. Can you point me to a page, if one exists, with information about the formatting of non-English definitions? The page that I usually go to for similar details, Wiktionary:Style guide#Definitions, doesn't seem to contain such information. Thanks and I hope the best for you. —The Editor's Apprentice (talk) 03:03, 27 October 2020 (UTC)

寒

Hey, did you receive the ping regarding 寒? ←₰-→ Lingo ^Bingo _Dingo (talk) 19:16, 2 November 2020 (UTC)

Special:Contributions/Haimounten

This user seems to be relying on JLect to produce dialectal entries. You might want to review it. —Suzukaze-c (talk) 06:17, 7 November 2020 (UTC)

@Suzukaze-c: Hmm, ya, we also don't have any clear consensus on how to handle dialect / 方言 entries. For some of these, like Ryūkyūan, we treat them as separate languages, in which case they could possibly qualify as LDLs, and we have much laxer constraints. But if 方言 are treated as non-standard variations of mainstream Japanese, then we need three cites as backing.

Beyond that, simply copypasta-ing JLect is poor form. I'll have a look as time allows, but my time is much more limited of late... Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:42, 10 November 2020 (UTC)

also: Special:Contributions/Haimaunten —Suzukaze-c (talk) 00:08, 27 November 2020 (UTC)

Inserting % in the Japanese templates

Hi,

If I remember correctly, inserting % in the Japanese templates is now discouraged or not recommended. This IP - Special:Contributions/86.153.213.49 keeps doing it in a large number of entries as in diff. Should all those be reverted? --Anatoli T. ^{(обсудить}/^вклад) 00:31, 8 November 2020 (UTC)

• @Atitarev: I've lost track of the current state of the infrastructure with regard to % marking and ruby. I think Huhu9001 has been the main person working on the modules of late. For the anon's edits, I haven't seen anything to make me worry, like module errors or incorrect placement, so I've largely been ignoring them.

@Huhu9001, do you have any insight into % marking and ruby in headword templates like {{ja-noun}} and {{ja-pos}}? Have this anon's edits been safe, or is this aggressive insertion of % marks causing problems? ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:46, 10 November 2020 (UTC)

@Atitarev, Eirikr: It is safe. However, as I can remember, it was User:Nyarukoseijin that opposed this usage, because he had been advocating for using the input of t:ja-kanjitab to separate kana, thus these % were unwanted. I am not quite sure which approach is better and no conclusion has been reached since then as he left Wiktionary later. -- Huhu9001 (talk) 04:25, 11 November 2020 (UTC)

Etymology of ぜ

"Appears in the Edo Period as alteration from older zoe, itself likely rendaku (連濁) applied to the stem soe of verb 添える, 副える (soeru, “to add or attach to something else”). See also そえに (soe ni, “what's more, furthermore, in addition”), さえ (sae, “also”)."

I know this is from an 8 year-old edit, but got any source for this? Seems more likely for it to just be a monophthongised variant of ぞえ/ぞい, which itself is an emphatic form of ぞ. Alves9 (talk) 11:20, 10 December 2020 (UTC)

@Alves9, thank you for the message. You've got the right of it. It's been quite a while and I don't really recall what resources I had to hand back in 2012. I've updated the ぜ (ze) entry and added ぞえ (zoe) based on the better resources I have now. Please have a look. In particular, if you have any info on dates for when ぜ (ze) usage shifted, please add that to the {{defdate}} templates. ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:35, 10 December 2020 (UTC)

Good edit, but I'm not sure if ぜ could be considered more aggressive than ぞ. It's certainly more "rough"-sounding, but in most situations it seems to like that ぞ could sound more comparatively forceful and threatening (compare 殺すぞ vs 殺すぜ). Another thing to note is the fact that ぜ is often accompanied by a rising tone (sometimes marked by a question mark) and can be used as a soft confirmation kind of particle like Kanto わ (e.g. このケーキは美味いんだぜ？). ぞ could also be used in the same way in a more limited scale, but it at least proves that ぜ's usage hasn't entirely shifted from a softening particle in colloquial speech. I believe definitions 1 and 2 can be merged.

Also, for the sake of comparison, there are a lot of other terms in Japanese that follow the same construction pattern of root particle + え/い. I suggest adding them as related terms or in the See also. わえ/わい for わ, かえ/かい/け for か, だえ/だい for だ, are some I can think of immediately. Alves9 (talk) 03:15, 11 December 2020 (UTC)

@Alves9, re: aggressiveness, I was trying to account for the すごみ and ぞんざい mentions in the KDJ definition for ぜ that are missing for ぞ. If that wording sounds 言い過ぎ to you, please adjust. I'm less familiar with informal speech -- unlike many North American studiers of Japanese, manga and anime were never really my thing, whereas news and politics were more where I found things interesting. Think Aera instead of Shōnen Jump.

Re: related terms, I'm not sure it makes sense to try to account for all the え・い-suffixed or -fused forms on the individual entries -- I think it might work better to gather all of those within the え or い entries instead: centralize the list, so we only have to maintain the data in one place.

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:22, 25 December 2020 (UTC)

Thank you for the insight. I've changed the wording a bit, mostly to make it more in line with the links you posted. I've also added an entry for い (i) and え (e), so feel free to look at them if you've got the time. Alves9 (talk) 22:35, 25 December 2020 (UTC)

Request for checking historical katakana values

Hello,

Could you have a look at the twelfth-century katakana romanizations in Appendix:Historical Koreanic numerals? My sources were transcribing the katakana in modern (!) Japanese, which is obviously misleading, so I tried to fix them by transcribing ハ as <fa> instead of <ha> and エ as <ye> instead of <e> as would have been the pronunciations of the time period, but is there anything I missed?

Thanks in advance!--Karaeng Matoaya (talk) 12:07, 16 December 2020 (UTC)

@Karaeng Matoaya, re: 4 and the Nichū-reki, any chance that the glyph ナ was corrupted to ト? If the horizontal in ト is closer to actually horizontal, and if the vertical exhibits any leftward bend, it could just be that the left-hand portion of the horizontal bar in ナ was poorly inked, or was somehow otherwise affected. Similar to your note about 5 and タ vs. ハ.

The one for 7 is odd indeed, apparently using hiragana た? Any chance it was hiragana に instead?

Your romanizations look good to me.  :) Cheers! ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:28, 25 December 2020 (UTC)

On 믈

Middle Korean high-pitch monosyllables tend to take low pitch before the locative particle -ay. Meanwhile, a high-pitch syllable which is the final syllable of a multi-syllable word never takes low pitch unless demanded by tone sandhi. But some monosyllables, such as mul, always take high pitch even before the locative. So it's often suggested that these irregular monosyllables were originally bisyllabic, which is what Vovin is getting at.--Karaeng Matoaya (talk) 01:26, 17 December 2020 (UTC)