Latest comment: 1 year ago1 comment1 person in discussion
Userspace templates like what you have under User:Chuterix/Templates are great for prototyping and testing, but they really shouldn't be used in mainspace, like what you're doing here.
A basket too, bringing a pretty basket, and with a spade too, bringing a good spade
I also just discovered that ONCOJ returns zero results when grepping for \bmwo\b, despite the existence of strings like what we see on this page from MYS 8.1613, which clearly has 跡毛奈久(ato₁ mo₁ na → atwo mwo na). So it seems they omitted any mwo romanization for the particle, even though it does appear to be spelled as ⟨mo₁⟩ in various cases.
@Chuterix I will have a look, but it may take a little time as I don't know how complex a job it will be.
By the way, I have created {{kzg-head}}, {{okn-head}} etc. which are generic headword templates. They work in the same way as {{ja-pos}}, but you should use them for all parts of speech. I've also converted Okinawan to the same format with {{ryu-head}} as a replacement for {{ryu-pos}}, because that's the format headword templates take in most languages. {{ryu-head|noun}} is identical to {{ryu-noun}}, so it's easier if we have just one template. Ideally, we'd do the same for Japanese so that we can get rid of {{ja-noun}}, {{ja-verb}} etc, but that will take a large bot job.
No hits based on any searches. Also based on the limited Korean I know, I think the content as well as the title are gibberish. Hangul phonetic for Japanese is unlikely. Shen233 (talk) 13:49, 11 June 2023 (UTC)Reply
this is why this is a troll poem; spoiler is that it's converting shakkun (phonetic kun'yomi) to korean equivalents or eumhun Chuterix (talk) 14:18, 11 June 2023 (UTC)Reply
Recovered:
山鶏上畑 (Yama-(niwa)tori-ue-hatake?)
家間事荷蚊歯
ihe-ma-koto-ni-ka-ha
(please add an English translation of this usage example)
categories 'LANG terms spelled with KANJI read as KANA'edit
Latest comment: 10 months ago1 comment1 person in discussion
It was a mistake that allowed these to be created with just {{auto cat}}. The garbage display should have triggered you that something was wrong. I have deleted all the categories of this form that you created; if you re-create them you need to specify the type of reading in the {{auto cat}} param 1. In general I would not recommend that you manually create categories containing only {{auto cat}}; let my bot do it. Benwing2 (talk) 08:34, 16 June 2023 (UTC)Reply
Latest comment: 10 months ago3 comments3 people in discussion
I looked at the page "Reconstruction:Proto-Japonic/mƏmƏ", and the reconstructed Proto-Japonic word should be "məmə", not "mƏmƏ". Please fix it. Thank you! 103.43.79.10505:05, 18 June 2023 (UTC)Reply
The larger schwa is due to undetermining whether this is */ə/ or */o/. Because Old Japanese only different mo and mwo in the Kojiki, and this word is not attested there. Chuterix (talk) 05:08, 18 June 2023 (UTC)Reply
Other notations which can be considered are notations using "...(.../...)..." or "... ~ ...". Such notations are used in Wiktionary at least for Proto-Sino-Tibetan (see Proto-Sino-Tibetan lemmas). In this case: "Reconstruction:Proto-Japonic/m(ə/o)m(ə/o)". This would be my preference.
Latest comment: 9 months ago11 comments3 people in discussion
波珍飡, 海干, and 海飡 are both found in the most basic primary source, the Samguk sagi. The kana transcription is part of the kana for the Silla name 微叱己知波珍干岐 in 日本書紀. Hope this helps.--Saranamd (talk) 09:32, 29 June 2023 (UTC)Reply
No need to ping me multiple times. I don't have direct access to Japanese sources, but the kana transcription みしこちはとりかんき/ミシコチハトリカンキ is discussed in Korean-language sources (there seems to be several J hits also) and W. G. Aston's 1896 translation of the 日本書紀 explicitly says "the traditional kana has Mi-shi-ko-chi Ha-tori Kamu-ki".--Saranamd (talk) 10:12, 29 June 2023 (UTC)Reply
@Saranamd, I suspect that Chuterix is curious where the "traditional kana" came from. We have occasionally encountered words that dictionaries list with a specific reading and a citation from the 700s, but upon further digging, we find that the earlier attested spelling is only logographic, and the oldest phonemic spelling isn't until centuries later -- and thus cannot be viewed as an attestation of the actual ancient reading.
In the kanji string 微叱己知波珍干岐, the tori reading for the 珍 character is a bit odd -- I've never encountered that before.
@Eirikr Why not? My understanding was that the NHSK kana glossing tradition overall accurately preserved ancient linguistic knowledge. The Man'yoshu also did not have kana glosses until centuries later, but the said glosses nonetheless preserve the Old Japanese readings behind the obscurity of the man'yogana.
In the case of the specific name, 微叱己知波珍干岐 is clearly a Korean orthography (probably from a Korean written source) that was adopted wholesale into Japanese. 叱 is always used in Old Korean to write coda -s, for example, and the sequence 波珍 for the OK word "sea" is directly attested in the Samguk sagi as mentioned above. If 微叱己知波珍干岐 was read as Korean, it would be easily read as *miskə-ti-patər-kan-ki or such even without any knowledge of the traditional kana spelling. The fact that the kana spelling matches this perfectly shows that we can trust the traditional kana here.
The strange tori reading for 珍 only verifies the authenticity of the traditional kana. It cannot be read as such in any Japanese tradition and only makes sense if it preserves the Korean reading of 波珍 as the native word for "sea", cf. Middle Korean 바ᄅᆞᆯ〮(pàrór) with */VtV/ > /VɾV/ shift in Middle Korean.--Saranamd (talk) 00:20, 30 June 2023 (UTC)Reply
If you mean, "why would kana renderings from centuries after the Old Japanese stage not be dependable as accurate representations of Old Japanese", the simple fact is that the language changed, sometimes substantially. The most obvious shift is the loss of the 甲・乙(kō-otsu) vowel distinctions for the vowels /i/, /e/, and /o/ (which we mark in romanization with subscript "1" for 甲 and "2" for 乙).
For this particular kanji string 微叱己知波珍干岐, the kana rendering みしこちはとりかんき tells us nothing about 甲・乙 vowel distinctions, for instance.
"The Man'yoshu also did not have kana glosses until centuries later, ..."
I must assume here that you are referring to those portions of the Man'yōshū (MYS) that were written in kanbun? A very large portion of the MYS was written in phonemic man'yōgana, for which we have a decent understanding of the vowel values -- albeit with some disagreement on what exactly the 甲・乙 distinctions were realized as on the phonetic level, hence the practice of subscripting in romanization.
In addition, some of the kanbun poems may well not be kanbun at all, and might instead be written in entirely other languages. Alexander Vovin made a strong case for MYS poem #9 being half Old Korean, about which you can read more at the Wikipedia article: w:Princess_Nukata#Poem_9.
“The strange tori reading for 珍 only verifies the authenticity of the traditional kana. It cannot be read as such in any Japanese tradition and only makes sense if it preserves the Korean reading of 波珍 as the native word for "sea"”...
Excellent that this particular kana rendering has a likely external corroboration. Other kana renderings of uncertain provenance, such as those for MYS poem #9, remain problematic.
The 海等#Old_Korean entry describes the likely equivalence of 海 (logogram) and 波珍 (apparently a phonogram?) in Silla texts. The problem arises in the next bit:
波珍, Old Chinese reading */pˁaj trə[n]/, is thought to be a purely phonogramic writing of this same word.
The problem is that, by the time of any Classical Chinese texts written by either Koreanic or Japonic speakers, Chinese had already progressed to the w:Middle Chinese phase -- so reconstructions of w:Old Chinese pronunciations would seem to be irrelevant. The Middle Chinese reconstructed pronunciation for 珍#Chinese is /ʈˠiɪn/, so 波珍 doesn't seem like it could be a phonogrammatic spelling of Old Korean *patol.
→ I have no argument against the Old Korean pronunciation of the word for "sea" being something like patol. I just don't think that 波珍 would fit as a phonogrammatic spelling for this pronunciation. I wonder if this might instead be a kind of poetic ideogrammatic spelling ("waves" + "jewel / treasure" certainly seems poetic). ‑‑ Eiríkr Útlendi │Tala við mig00:09, 15 July 2023 (UTC)Reply
@Eirikr珍 does indisputably transcribe /*tVr/. Besides the "sea" example being discussed, there are also Samguk sagi toponym examples:
石山縣本百濟珍惡山縣 where 珍惡 is the same word as 石; note that the Old Korean word for "stone" is reconstructed as *tworVk, cf. Middle Korean돓〯(twǒlh) where final *-h comes from *-k. The character 惡 transcribes *-Vk, leaving the first character for *twor.
While not from the SGSG, there is also alternation between place names 武珍 and 無等. The character 等 traditionally writes ᄃᆞᆯ(tol) in Sinographic orthography because ᄃᆞᆶ(-tólh, /h/ not pronounced in isolation) is the plural-forming suffix and 等 is also a pluralizing morpheme in Chinese; this was a practice that continued into the 1800s. This leaves 珍 as transcribing *tVr also.
However, the transcription is probably not from Old Chinese as you note. According to this chapter by Lee Ki-Moon, 珍 was perhaps used to transcribe *tVr because its rough Korean equivalent is *tworVk "stone" as mentioned above. It would then be a 훈가자(訓假字)(hun'gaja). This explanation does seem neater.--Saranamd (talk) 13:50, 15 July 2023 (UTC)Reply
@Saranamd, thank you very much for the additional corroboration of the /*tVr/ pronunciation for 珍 in Koreanic contexts, that is super helpful. I am very early in my Korean studies, and based solely on the reconstructed Chinese phonetic values, /*tVr/ did not make much sense to me (as described earlier). Seeing this used in Koreanic in multiple other texts and contexts, and the crossover with the "stone" meaning, does much to clarify things. Cheers! ‑‑ Eiríkr Útlendi │Tala við mig19:23, 17 July 2023 (UTC)Reply
Ah, perhaps you did not see the block log? Copying here:
(Disruptive edits: Chuterix, please explain your edits. Unexplained removals, and unexplained reversions of restorations, are not acceptable (among other issues).)
In truth, I've been growing increasingly concerned by the disruptiveness of your edits -- edit comments mostly missing, removing content without comment or explanation, excessive moves, strange reconstructions apparently arrived at in isolation, excessive pinging (in terms of both frequency and number of people pinged at once), demanding tones in messages left on Talk pages, etc. I've commented on all of this in various ways, sometimes on multiple occasions, but I have not seen much meaningful change in your editing behavior.
Earlier today, you removed an explanatory paragraph from the 悲しい entry, without adding any edit comment. I restored it, with an edit comment explaining that: "+restoring para about possible cognate that was removed without explanation". You then partially reverted that restoration, again without any edit comment. This is disruptive editing and unacceptable behavior. I applied a one-day block with the above block log note in the hopes of getting your attention and, ideally, getting through to you that you need to change how you edit. Specifically:
Use edit comments to briefly explain what you are doing, and if appropriate, also explain why.
Be especially sure to use edit comments when removing chunks from entries, and when reverting (in whole or in part) another editor's reversion.
Slow down. Evaluate what you are doing more fully before executing. Moving and re-moving reconstruction pages multiple times is not helpful. Fully research what form a reconstruction should take, before creating an entry. Things like the Proto-Ryukyuan*pE((p/b)E)t(O)zi(C)a entry simply should not exist.
Do more to coordinate with other editors. If you are unsure about something, ask. If a particular editor is not responding, ask someone else. Better yet, use the public forum pages, like the WT:Beer parlour and the WT:Etymology scriptorium.
PR for 'goat' must be *pebeza and this connection to hitsuji (OJ *p(i/wi)tuzi) must be rejected.
There is no Proto-Japonic for 蛙(kaeru, “frog”); it's also only attested in 楓(kaede < kapyerute, “maple tree”). Ryukyuan forms are different. Chuterix (talk) 15:16, 20 July 2023 (UTC)Reply
Not sure why you bring up 蛙(kaeru)? I don't recall having anything to say about this at the proto level. Looking at it now, I note that several of the terms listed for Ryukyuan languages here at JLect raise the interesting possibility that the core term may have been something like peru -- around 10 of the related terms there contain an element piru or piri (with variations on that initial /p-/). ‑‑ Eiríkr Útlendi │Tala við mig21:44, 20 July 2023 (UTC)Reply
@Eirikr: the ryukyuan terms are cognate with 蟇(hiki, hikigaeru, “toad”). Amami (Yamatohama; from Nagata et al. 1980 p. 782, text here; Nase, Koniya, also cited in Jarosz, 2015 p. 657), biki, Amami (Nase) bikki, Okinawan (Shuri) wakubichi, Okinawan (Naha; Sakihara 2006, p. 259) wakubitchi with deriative atabichaa (< *atabEkiwa; also found in Shuri), Kunigami (Nakijin/Yonamine, Nakasone 1983 p. 300, text here) tintoohathabichaa (< *tentau-pata?-bEki-wa; the initial *tentau is also seen in descendant of PR *nozi(“rainbow”)), Amami (Shodon) biki, bikii (/bjikji(:)/) with deriative bikkya (/bjikkja/), Yoron atabiku (text here), possibly Yonaguni atahita. The Japanese term however is apparently first attested in Hozo Wanmyou of 918. I don't see any other Southern Ryukyuan cognates at the moment, but the Amami cognates point to PR *piki(“frog”). See JLect etym here. Chuterix (talk) 22:27, 20 July 2023 (UTC)Reply
As I must repeatedly stress, slow down and coordinate with other editors.
You can't base a reconstruction off a single work by a single author. Also, we are a collaborative project, meaning we strongly encourage communicating and working with other editors. For reconstructions, we strive to synthesize the academic consensus. Changing things to align with what one author said, while ignoring other works, is not synthesizing. And doing so without discussing at all with other editors is not collaborating.
Japanese te (standalone) and ta (compounding) are clearly 手 "hand", and this term was used to refer to ways and directions more than places. We do this some in English too: see lefthand, righthand. Location-related uses appear to be outgrowths of the underlying "hand" sense, or extensions of the "way, direction" sense. This te / ta is a distinct morpheme from any tô / tö (to use Vovin's diacritics) meaning "place, location".
As a side-note, and response to part of that same text of Vovin's about "place" -- I cannot agree with Vovin's contention that genitive な did not exist in Old Japanese. This is widely documented in various Japanese reference works, and appears in many terms traceable to the OJP stage, such as the 掌(tanagokoro) you mention, as well as 手な末(tanasue), 港(minato), 目交(manakai), etc.
Regarding ato1/2, this term is also sparsely attested in phonetic spelling. Of the 9 hits in ONCOJ for \batwo (the \b ensures this atwo is at the beginning of a word), only two are phonetic spellings, and only one is clearly this same word 後・跡 (see MYS 15.3625) -- the other (in the NSK) is even spelled 阿都, which is atu, not ato1. Meanwhile, ONCOJ has 17 hits for \bato in 16 texts, although all of the relevant hits are from the same Buddha stepstone poems (the w:ja:仏足石化) -- the two MYS hits are 1) a typo or copy error (see the text at MYS 9.1740, apparently a typo in ONCOJ: 相誂良比 should be あひとぶらひ, no あと anywhere) and 2) actually part of the word あとり in reference to a kind of bird (see MYS 20.4339, requires a bit more digging to clarify that the stanza あとりかまけり is the bird あとり + verb かまく + stative ending ～り).
The Buddha stepstone is dated to 753, older than the w:Shin'yaku Kegonkyō Ongi Shiki (SKOS) dated to 794. Moreover, the SKOS text we have today is not from 794, but is a later copy, and it is recognized as containing copyist errors -- meanwhile, the Buddha stepstone was carved in 753 as the original, still-accessible text. This makes the Buddha stepstone more trustworthy for OJP kana use.
In comparing the Buddha stepstone texts and the MYS, I do note that the one phonetic spelling in 15.3625 appears in the stanza 安刀毛奈吉. We know that this is poetry, which raises the possibility that the use of 刀 (to1) may have been phonetically influenced by the following particle 毛 (mo1).
At any rate, we have some limited textual evidence for ato1, and more textual evidence for ato2. On balance then, this to element should be treated as to2 -- which happens to match what we see phonetically for 所.
In terms of erratic editing, your recent posts (in the past few days) have been noticeably harder to understand than in the past, and other editors have also voiced their inability to understand your writing (such as in [Grease pit thread about romanizations for Ryukyuan languages]).
And if I remember, I had counterargued that how else are we supposed to interpret atu as? Definitely not hot (no pun intended). Anyways Vovin (2020)'s Man'yognaa list has 都 as to1 in both variety A and B, and even JDB has 都 in the list of man'yogana usage for to1 (see page 897 or look in Discord for the picture of this page), where it's attested in Shosoin documents, the Nihon Shoki, the Manyoshu, the Fudoki, the Shoku Nihongi and it's Senmyogaki.
Anyways the to2 in to2ko2ro2 was assimilated from to1. And logically, this violates Arisaka's law, where a is found in same morpheme (possibly even lexicalized compounds) as o2, so assimilation occurs, as seen in the quote. Chuterix (talk) 19:24, 9 December 2023 (UTC)Reply
跡 was used as man'yōgana for /to₂/, adding circumstantial evidence that the "o" vowel in ato would have been /o₂/.
Arisaka's law as described at w:Old_Japanese#Phonotactics states that /o₂/ is the unstable one -- /o₂/ would assimilate in the presence of /o₁/, not the other way around. So if the initial mora in tokoro were originally /to₁/, that would presumably have forced any following /o₂/ to shift. Yet we see /to₂ko₂/ (MYS 14.3554) and /to₂ko₂ro₂/ (MYS 19.4288), not /to₁ko₁/ or /to₁ko₁ro₁/.
That said, Arisaka's law appears to apply to morae in the same morpheme. If tokoro is composed of noun to + suffix ko + suffix ro, and likewise if ato is a compound of otherwise-independent nouns a + to, then Arisaka's law might not apply anyway, as these elements would be separate morphemes.
Latest comment: 9 months ago6 comments3 people in discussion
@Eirikr You blocked me once again, this time for 3 days, for "Disruptive edits"; even though I gave a clear reasoning for undoing accent pronunciation removals. Was it really that nessecary? Chuterix (talk) 20:54, 26 July 2023 (UTC)Reply
This is not an insult, but rather an honest description of my impression of your behavior.
I have told you multiple times that you must coordinate with other editors, discuss, and achieve agreement, before engaging in various potentially disruptive types of editing -- including restructuring entries and reverting other editors, particularly reverting other reversions.
Earlier today, you reverted my reversions of you, relating to entry restructuring, and again without discussion.
I must assume that you do not understand my advice to you? Or are you ignoring that advice?
To your credit, you are using edit comments more than in the past -- thank you for that, and please continue using edit comments.
That said, you must do more to coordinate with other editors. I am blocking you for a short period to hopefully give time for discussion and consensus. I have configured your block so you should still be able to post on Wiktionary and Talk pages. Please use this time to discuss your ideas with other editors before implementing them. ‑‑ Eiríkr Útlendi │Tala við mig21:12, 26 July 2023 (UTC)Reply
Given the unsettled nature of this issue (my concerns about including these sections, Aramaki-san's concerns about not including them), I think we need to discuss more with the broader community.
Might I add that pinging someone on their own talk page is just silly. Merely editing their talk page triggers a notification, and if they don't get that, they're not going to get the ping, either. Chuck Entz (talk) 05:51, 27 July 2023 (UTC)Reply
Latest comment: 8 months ago3 comments3 people in discussion
Hi @Chuck Entz! Chuterix recently showed up on the #wiktionary IRC channel, asking to be allowed to edit their own talk page. While I see that they've been disruptive outside of the main namespace, it seems reasonable to at least grant them the ability to request an appeal. If nothing else, I assume that since the block was relatively short-term, that the offense is not bad enough yet to warrant such a measure.
For what it's worth, Wikipedia's policy on blocks (Wiktionary's is less developed) mentions that removing the ability to edit the talk page should be reserved for cases where that ability itself is abused, which from a brief search I haven't quite seen evidence of yet in this case. — ObſequiousNewt — Geſpꝛaͤch — Beÿtraͤge01:16, 3 August 2023 (UTC)Reply
I'll let Chuck respond to this; when you block someone you have to deliberately click on a checkbox to prevent them from being able to edit their talk page, and Chuck did this so I assume it was intentional. Benwing2 (talk) 01:14, 4 August 2023 (UTC)Reply
That does not address the issues with unilaterally renaming and moving a widely used template.
My understanding of the "2" in the name was simply that this template was the second one created to reference the Nihon Kokugo Daijiten. See also Template:R:Kokugo Dai Jiten.
@Chuterix, do you have any intention of bringing up your move for broader discussion? Do you have any plans for resolving the issue with Wikidata, or of cleanly updating all the existing entries?
I don't remember but looking at it, the 2 seems to refer to the edition number. I do not have the interest to read about what has happened to the template. —Fish bowl (talk) 21:53, 16 November 2023 (UTC)Reply
Latest comment: 5 months ago13 comments4 people in discussion
I wanted to request if I can add *umo for Proto-Japonic. "umo" can refer to any tubers like taros and yams. However, cassavas and potatoes weren't introduced in Japan until the Europeans came. There's also an Okinawan word not borrowed but cognate with Japanese called " 'nmu ". Slimsilkyweave (talk) 03:28, 20 November 2023 (UTC)Reply
Possibly. The Nihon Shoki mentions only one usage of "yam", which occurs in a record in 501 where the Emperor plucked out the men's mails and made them dig for yam, which apparently is called Dioscorea japonica. This word is attested in Ryukyuan with the meanings 'sweet potato' (サツマイモ(satsuma-imo, literally “potato [from] Satsuma [Province]”) or 'taro' 里芋(sato-imo, literally “village potato”), such as from the following quick checks: Northern Amami-Oshima ʔumu, (the Toku-no-Shima words are loans such as satuimu) Oki-No-Erabu ʔumuu, Kunigami ʔumuu, Okinawan ʔNmu, Miyako m, Yaeyama un, possibly Yonaguni unti. According to the history records on the Nihon Kokugo Daijiten (second edition):
The sweet potato was apparently borrowed to Ryukyu Islands in 1597. However, it appears that the OJ word umo disappeared after the 新撰字鏡, so it cannot be a straight borrowing (with the exception of Toku-no-Shima). Thus, I classify this as a Proto-Japonic word. Notifying @Eirikr and @荒巻モロゾフ.
The so-called yama imo or literally "mountain tuber" is native to Japan and the Ryukyu islands -- see also [[w:Dioscorea japonica]]. As a plant native to both the Ryukyu and Japanese archipelagos, we would expect any proto-language terms for this to be found in both areas, and it seems we have that here. ‑‑ Eiríkr Útlendi │Tala við mig01:35, 21 November 2023 (UTC)Reply
My name is Jude. I'm from the Philippines. I'm interested in putting reconstructed proto-language words in Wiktionary as well as marking some words inherited from proto-languages. I'm also the same person who wanted to compare the Proto-Japonic word *poy with Proto-Austronesian *Sapuy.
The OJ word umo disappeared so no accent for that. It has been replaced by the fronted version imo, which has LL accent in the Heian period, LH(L) accent in Tokyo, HL(L) accent in Kyoto, according to the Nihon Kokugo Daijiten 2. Also Kagoshima B accent [LH(L)] (from Zenkoku Akusento Jiten) Those accent corresponds to accent class 2.3. The Shuri Okinawan form 'nmu has accent number 0, indicating no accent (M?). The Ishigaki Yaeyama form is actually un. It has HH accent. This is a regular accent correspondence as seen in Ishigaki Yaeyama in which has HH accent. Usually a Ryukyuan tone class B accentuates the the whole word in Ishigaki Yaeyama, as opposed to LH as seen in pana 'nose'. Thus the tone class for potato is B, which strengthens the existence of the pJ accent class 2.3 because this is the regular tone correspondence. I want to add it but I'll have to wait a week since I got banned for a week because I undid a Middle Chinese module without asking. Chuterix (talk) 14:33, 22 November 2023 (UTC)Reply
As a late post-script, it looks like the sato imo (literally "village tuber") plant is native to Southeast Asia, and is thought to have first been imported into Japan during the w:Jōmon period, long ago enough that we would again expect to see any related terms at the proto-Japonic level.
The reason why you interpreted that as ute is because various Old Japanese texts do not distinguish voiced and unvoiced stops. See MYS.4.485, for example:
Well the problem is where his proof that Ryukyuan tone class C indicates tone rising in first mora of monosyllables. So I got fixated on that. See なば, where reconstructs a RH pitch and not a LH pitch on pJ *naNpa 2.4a. Chuterix (talk) 18:32, 27 November 2023 (UTC)Reply
Latest comment: 4 months ago3 comments3 people in discussion
@Ultimateria: This word is not attested in Ryukyuan, except for a literary Shuri term chimi which is obviously a borrowing from Japanese. And it's inappropriate for Proto-Japonic to borrow from Old Korean, so this is reason for {{d}}ing there. Please delete that page. Chuterix (talk) 09:15, 27 November 2023 (UTC)Reply
That's what I thought, too. It should be deleted because the word "kimi" is possibly from an Old Korean word "kum". Also, I don't think it's a Proto-Japonic word. Slimsilkyweave (talk) 11:26, 2 December 2023 (UTC)Reply
The 続日本紀 is written in 漢文, but we also know that it was written by Japanese people, and that 漢文訓読 was a thing. Given the date, this could not be considered modern Japanese, so it must be Old Japanese.
@Eirikr: The problem is that it would be spelled semantically like Classical Chinese (or similar), but the actual Japanese pronunciation is not there (I recall reading a paper about a Mokkan which has Man'yogana transcription of Sino-Japanese words, but I need to find it sometime). Chuterix (talk) 22:35, 27 November 2023 (UTC)Reply
While I do not have access to any 訓読 versions of the 続日本紀, I have seen Google hits that suggest such a thing exists. Considering again that multiple Japanese resources cite the 続日本紀 as the first appearance of the word in Japanese contexts, and that those editors likely did have access to such 訓読 renderings, and considering also that the bulk of 音読み vocabulary in Japanese traces back to when written Chinese terms were borrowed into Old Japanese, everything points to 武士(bushi) being one of these terms -- originating in Middle Chinese, and borrowed into Old Japanese, then inherited into modern Japanese.
@Eirikr: And I forgot to say the existence of Sino-Japanese loanwords in Old Japanese might be supported by words such as 昆布(konbu, “kelp”), attested in the Shoku Nihongi and has Ryukyuan cognates that must've came from pR *kombu C, c.f. Yamatohama xubu [kʰubu] Yoron kubu HH, Nakijin kʰubuu LLH (irregular accent), Shuri kuubu MMM (M means middle pitch), Yonaguni kʰubu C (C in Yonaguni means first mora low pitch, and all syllables until the end are high, then the final mora falls). There are more Ryukyuan cognates I listed; the cognate I listed are minimal accent comparisons crucial (except Yamatohama of Amami doesn't have accent, but it's useful for comparison). This is further supported by Japanese kobu, attested in 1275, and the Tokyo accent of HL-L and modern Kyoto accent LH-L shows that this is belongs to accent 2.5(a; if you take Ryukyuan tone class C). It is worth noting that Middle Chinese /kwon puH/ (Baxter) has a level tone + departing tone, which could explain the phenomenon.
It must be certain that editors have access to many literary text, so they should have various manuscripts, particularly for the Nihon Shoki and sometimes the Kojiki. There are even Chinese loanwords in Manyoshu, such as 力土儛 (RIKI1-ZI-MAP-I1). Chuterix (talk) 22:54, 27 November 2023 (UTC)Reply
Latest comment: 5 months ago5 comments2 people in discussion
@Eirikr: I've analyzed words with the Heian period accent LLL, extracted from NKD2 and ZAJ:
LLL accent analysis; irregularities are underlined
Heian Kyoto
Tokyo
Modern Kyoto
Kagoshima
Gloss
Pronunciation
LLL
LHH-L
HLL
LLH-L
trousers
hakama < fakama
LLL
LHH-H
HLL
LLH-L
fish scales
uroko < iroko (but actually an ablaut)
LLL
LHL
HLL
LLH-L
sword
katana
LLL
LHH-L
HLL
LLH-L
mirror
kagami
LLL
LHH-L
HLL
LLH-L
word
kotoba
LLL
LHH-H / LHH-L
HHH / HLL
LLH-L
Buddha
hotoke < fotoke
LLL
LHH-L
HHL / HHH
LLH-L
scissors
hasami < fasami
Typically the pattern would be Heian Kyoto LLL, Tokyo LHH-L, Modern Kyoto HLL, Kagoshima LLH-L. But in pJ *arasi, we see Tokyo HLL, Modern Kyoto HLL (the only expected accent), and Kagoshima LHL. Thus, there is yet to be explained why this accent irregularity is occurring.
The accent of oroshi
For oroshi, it seems to be derived from 下ろす(orosu, “lower down”), and the following occurs:
＜標ア＞［シ］ (4)は［オ］ (10)(12)は［シ］［オ］＜京ア＞ ［オ］ (1)は［０］も
Accent guide: oroshi-ga; sense 4 is oroshi-ga, senses 10 and 12 are oroshi-ga or oroshi-ga. Kyoto accent is oroshi-ga, but sense 1 is 0 (oroshi-ga).
The accent HHL for Shuri arashun and the falling accent in Hateruma arasun allows a reconstruction of pR *aras- A., which may suggest that this is derived from 荒らす. Chuterix (talk) 00:32, 28 November 2023 (UTC)Reply
Hmm, hmm, interesting, thank you!
Japanese causatives tend to be 中高 in 標準語, like こわす or かます or とます. あらす here is 平板 from what I can find, but then this might be unsurprising if we view this not as the causative of some verbal root *ar- but rather from stem ara- + su -- not a causative conjugation, but rather a different word formation pattern.
As あらす is 平板, I think we'd expect a simple deverbal nominalized form to also be 平板? There's 晒す + 晒し, both 平板... I haven't read as much about pitch accent patterns, and I'm struggling at the moment to come up with more examples. ‑‑ Eiríkr Útlendi │Tala við mig01:11, 28 November 2023 (UTC)Reply
Latest comment: 4 months ago3 comments3 people in discussion
According to "The Culture of Japanese Chestnuts", it says "Chestnut cultivation began in Japan during the Jomon Period more than 5,500 years ago. With the 1992 discovery of the Sannai Maruyama Site in northern Aomori Prefecture, it became clear that people at the time had planted chestnut trees around their villages and used the nuts as a stable food source." The Japanese chestnuts are also native to the Korean Peninsula, but the Korean word from chestnut is "bam". There's a Korean word that's similar to the Japanee word "kuri". I think one of the Korean word is from a Middle Korean word "kắrái", possibly I think is "karae". However, the Korean word means a "wild walnut", so I don't think this can be comparable. Are there any Ryukyuan words cognate with Japanese "kuri"? If so, I'll make an article for it. Also, please put accent classes, too. Thank you! Slimsilkyweave (talk) 11:21, 2 December 2023 (UTC)Reply
@Slimsilkyweave: According to the JapanKnowledge Nihon Kokugo Daijiten 2, the Heian Kyoto accent is LL, the Tokyo accent is LH-L, and Kyoto accent is HL-L. This corresponds to the 2.3 tone class. However, there appears to be no Ryukyuan cognates. This word is possibly connected with 涅(kuri, “black soil in water”), which had LH accent in the Heian period, and 黒(kuro, “black”) or 黒い(kuroi, “black”); the former noun 'black' seems to be a Late Middle Japanese formation, while kuroi has existed since the kojiki (as kuro1si). Anyways kurosi 'black' had LLF accent in the Heian period.
Apparently the Castanea crenata plant doesn't grow further south than the tail end of Kyushu, per the article at w:ja:クリ. As such, it is unsurprising that this word would have no reflexes in the Ryukyuan languages.
Given the limits of this plant's distribution range, I don't think we can take the lack of reflexes in the Ryukyuan branch as solid proof of this term's existence or non-existence at the Proto-Japonic stage -- absence of evidence is not evidence of absence. 😄 ‑‑ Eiríkr Útlendi │Tala við mig21:33, 4 December 2023 (UTC)Reply
Latest comment: 4 months ago2 comments2 people in discussion
When editing pages in other users' userspace, as you did earlier today here, leave an edit comment explaining what you're doing. Without any such explanation, such edits could be viewed as disruptive editing. ‑‑ Eiríkr Útlendi │Tala við mig20:23, 5 December 2023 (UTC)Reply
It is not quite certain; it is only attested mostly in the Northern Ryukyus: Northern Amami mo/muu, Toku-no-Shima [muu, Oki-no-Erabu [muu, Kunigami [moo, Okinawan mu]u, Yaeyama mo]o.
Pinging @Eirikr; the accentual correspondence is regular but why do we have a mid back vowel, not a high one?
Latest comment: 2 months ago5 comments2 people in discussion
Chuterix, it is inappropriate for any user to create user pages for other users. Even more so when the creating user has been harassing the other user.
I have warned you about harassment numerous times. You persist in your harassment.
I am left with no choice but to block you.
When your block expires, I sincerely hope that you will desist from any further harassment. I value your passion and energy, and I would hate to lose that due to inappropriate behavior. Please reevaluate your approach. ‑‑ Eiríkr Útlendi │Tala við mig18:35, 29 January 2024 (UTC)Reply
Note to self: this same harassment I've done on JA WT had a JA WP admin "I carry Enokida/it's a hackberry tree" (enokidatamotsu) to havr blocked me with an expiration date of never/infinite (無期限) and a reason of 暴言または嫌がらせ (inappropriate language and/or harassment), since I have said お前(omae) to him multiple times om his talk; my talk page and email is also blocked Chuterix (talk) 00:50, 13 February 2024 (UTC)Reply
Latest comment: 3 months ago4 comments2 people in discussion
@Eirikr: Is there any evidence that the Harima Fudoki actually preserve mo1 vs mo2? Also see the Wiktionary Discord (Japonic). There's no ko/otsu marked for simo in JDB 370. Chuterix (talk) 02:55, 30 January 2024 (UTC)Reply
According to Vovin, he said that Bentley found out that Manyoshu book 5 statistically preserves mo1 and mo2 (with exception of nouns). In book 5, we see 斯毛 in MYS.5.804 (both variants). Chuterix (talk) 02:57, 30 January 2024 (UTC)Reply
I just did a survey of the version of the MYS made available via the University of Virginia's https://jti.lib.virginia.edu/japanese/manyoshu/ section. (Currently offline, but I have a local cache.)
From what I've found, 霜(shimo) is only spelled phonetically in a few places.
MYS 20.4431: 志毛(⟨simo₁⟩) — maybe Eastern Old Japanese?
The ones in MYS 15 are clustered near each other and might be by the same author. All three have the same heading note:
遣新羅使,天平８年,年紀,広島,倉橋島,出発,羈旅,船出,叙景
The phrase 遣新羅使 seems to indicate authorship by, or a topic about, an emissary to or from the Korean-peninsula kingdom of Silla. The foreign connection suggests a cultural context close to the imperial government, so I would infer from this that the language used would be high-register and close to the prestige lect of the time.
The year 天平８年(Tenpyō hachi-nen) works out to 736 CE. Later than authorship of the Kojiki in 712, but within a generation or so.
The only outlier in the MYS phonetics is the ⟨simo₂⟩ spelling in MYS 14. This book is specifically 東歌(aduma uta), or songs from the eastern provinces, reflecting eastern dialect. I'm not sure if the ⟨mo₂⟩ invalid IPA characters (₂) here reflects a difference in pronunciation, or if it might instead be an indication that the ⟨mo₁⟩ invalid IPA characters (₁) and ⟨mo₂⟩ invalid IPA characters (₂) morae had collapsed in Eastern Old Japanese.
I think MYS 20 is also supposed to be Eastern Old Japanese, but I can't recall if it's the whole book, or only certain sections. If it's the whole book, the use of the ⟨mo₁⟩ invalid IPA characters (₁) spelling there might be taken to mean that EOJ didn't differentiate; that EOJ did differentiate, but had multiple sub-types; or even that there are simply scribal errors. ‑‑ Eiríkr Útlendi │Tala við mig06:34, 30 January 2024 (UTC)Reply
It can be difficult to search, and you have to get used to Portuguese orthography of that time -- like using ⟨x⟩ for /ʃ/, and being aware of sound changes like modern せ(se) manifesting at that time as /ʃe/ and thus appearing in the Nippo Jisho with the spelling ⟨xe⟩. If memory serves, there are also a few pages missing (due either to the pages being physically missing from that copy, or perhaps to the Google employees goofing when digitizing). Getting to see the spellings of the time, and thus the differences in pronunciation between then and now, can be quite illuminating.
One thing I'm uncertain about -- how much of the Nippo Jisho reflects "standard" Japanese of the time, and how much reflects the local dialect of Nagasaki, where the Portuguese were based? ‑‑ Eiríkr Útlendi │Tala við mig19:41, 7 February 2024 (UTC)Reply
Many words in Standard Japanese of the time (called 上(cami)), although there are some dialectial words. For instance, Naba in Nippo Jisho is labeled as a word in Kyushu (下(ximo)). Chuterix (talk) 20:10, 7 February 2024 (UTC)Reply
This is at least in translated edition. For ease of reading and because @Bendono quotes this, I like this version of Nippo Jisho (more). The original is still a treasure. Chuterix (talk) 20:14, 7 February 2024 (UTC)Reply
Latest comment: 2 months ago4 comments2 people in discussion
It occurs to me that 荒巻モロゾフ's user name has a particularly Slavic flair. Which brings to mind the possibility that this person might be caught up in the Russia-Ukraine war. Their absence from Wiktionary might be due to real-world difficulties. (I have no concrete information about this, I am only guessing here.)
Native speaker or not, we don't know Aramaki-san's actual citizenship. He (assuming masculinity here) may well be a Russian or Ukrainian citizen who happened to grow up in a Japanese-speaking environment. "Traveling a long ways" may be simply a matter of visiting family, for all we know.
My basic point: we have no solid idea of what Aramaki-san's circumstances are. Events in real life can often cause Wiktionary editors to stop editing, sometimes temporarily and sometimes permanently.
I'd agree that #3 seems the most likely, assuming though that the -su is instead the ancient honorific auxiliary (possibly cognate with the -s- honorific infix in Korean verb forms?), which had similar syntax to the regular causative -su and attached to the mizenkei. In which case, this would be the -o- shifted form of the mizenkei stem, as we see with certain other verbs. (And the exact reasons for this -o- shift remain obscure...) ‑‑ Eiríkr Útlendi │Tala við mig19:16, 12 February 2024 (UTC)Reply
Latest comment: 2 months ago6 comments3 people in discussion
@Eirikr: Ohokami could either come from *opo-kami2 'great-god' or *opo-kam-i1 'great-eat-NOM; great eater'. This word is not attested phonetically in Old Japanese, so unless we find an Old Japanese source that records this word phonetically, this etymology is still to debate. The references you used are (kind of) outdated; the NKD2 does not give any etymology, but lists the following theories:
(1)オホカミ（大神）の義〔名語記・和字正濫鈔・東雅・燕石雑志・俚言集覧・大言海〕。
(2)オホカミ（大噛）の義〔和句解・和字正濫鈔・日本釈名・俚言集覧・名言通〕。
(3)カミは「犬」の別音Kam から〔日本語原考＝与謝野寛〕。
The first etymology would parallel a semantic shift of Ainu カムイ(kamuy, “god; bear”). Meanwhile, I don't know if wolves frequently eat their prey; carnivores and herbivores obviously would hunt other animals for meat, but I don't want to dive into the ocean of knowledge about animals. Chuterix (talk) 23:59, 11 February 2024 (UTC)Reply
Semantically, it's worth noting that Ainu culture reveres the bear, and treats the bear as a god avatar. My knowledge about Japanese cultural views on wolves is admittedly limited, but my impression is that they were not revered to the same extent, and have historically been treated more as pests than deities, to the point that they have been rendered extinct. That said, apparently there are Shinto shrines honoring the wolf, so the connection to 神 is within reasonability.
We do know that Old Japanese phonemics make it clear that 神 = ⟨kami₂⟩ and 噛み = ⟨kami₁⟩, so it has to be one or the other, not both. 😊
Yes, my mistake. I mistakenly included a date when copying/pasting some code from another template. Before I edited the template, there was no date included so I've now removed the date from the edited version. If 938 is correct, you can replace |nodate=1 with |year=938 on Template:RQ:Wamyō_Ruijushō. JeffDoozan (talk) 02:58, 13 February 2024 (UTC)Reply
Fox (wolf). Fanqie: hu (扈). The Japanese call this kitune. Also the fox has been lied to be a beast demon that resides in three virtues, the fox has been suspected to be unstable, the reading of fox in Chinese is rau (良); this reading is for dogs; the Japanese had called it OPO KAMI2
What is this from? If it's talking about fanqie, I am led to believe that this is a Chinese source, in which case 大神 would presumably be pronounced using the Chinese sound values, much like 岐都禰 is earlier in this same chunk of text -- which would give us something like dajin or dashin. ‑‑ Eiríkr Útlendi │Tala við mig20:13, 13 February 2024 (UTC)Reply
Latest comment: 2 months ago2 comments2 people in discussion
One interesting problem we have then, if 薬 (kusu) is from Proto-Japonic */kuso/, is that this becomes phonologically identical to Japanese 糞 (kuso) ← OJP ⟨kuso₁⟩ ← Proto-Japonic */kuso/.The semantics become immediately problematic. How do we have both "mystical substance" (positive) and "shit" (negative) coming from the same root?
@Eirikr: pJ *kuso is only 1/3 possible reconstructions. There can also be *kusau (c.f. 臭い(kusai, “stinky”) and *kusua (unlikely). Old Japanese kuso1 simply means 'feces (without vulgarity mentioned); affix denoting an object with bad smell or rude; trash' The suffix that's usually analyzed as -ri might be same as that of 光る(hikaru, “to glow”), like how that word got nominalized into a word 光(hikari, “light”). Chuterix (talk) 21:00, 13 February 2024 (UTC)Reply
Re: vulgarity, I intend none -- I simply note that "feces" has an inherently negative connotation, regardless of vulgarity, just by dint of smell and hygiene. :)
Re: vowels, you're right to point out the possibility of Proto-Japonic diphthongs as a source of OJP ⟨o₁⟩, I wasn't thinking about that. Considering verb kusaru, the OJP noun kuso₁ may more probably be from Proto *kusau than from *kusua (which you also note as unlikely).
Re: the -ri ending, this is almost certainly the nominalized form of common spontaneous / intransitive verb auxiliary -ru, which often exhibits the lower bigrade conjugation pattern – but also sometimes manifests as upper bigrade, or as quadrigrade. I suspect that this auxiliary is in turn cognate with copular ari / aru, and I dimly recall reading that one or two academic linguists have said something similar, although I forget who (Vovin? Frellesvig? Shibatani?). ‑‑ Eiríkr Útlendi │Tala við mig22:54, 13 February 2024 (UTC)Reply
Thanks for the ping, I'll roll back those edits. Do you happen to know if it's just the |acc_ref= in {{ja-pron}} that uses refs like this or are there other templates/params I should look out for? JeffDoozan (talk) 01:27, 16 February 2024 (UTC)Reply
Latest comment: 2 months ago4 comments2 people in discussion
@Eirikr: A correspondance of Middle Korean (Yale) /ye/ to Japanese /e/ could be seen in the Paekche (Baekje) transcription セマ(*syema) (apparently mid-vowel raising in central OJ must have stopped by time OJ was written (let alone kana); as a transcription of a name in a 5th century mound contains a name like waka take1ru, where ru is rendered as 鹵 (MC luX); c.f. how 盧 (MC lu) is used to transcribed ro1 in OJ) although in the putative "Koreo-Japonic" comparisons /ye/ corresponds to J /o/, as in 星(hoshi, “star”) to Korean 별(byeol, “star”). I unfortunately have limited knowledge about Korean.
On the Japanese side, the NKD2 JapanKnowledge (the quotation is not in Kotobank) quotes a Wamyo Ruijisho attestation. Take my translation with a grain of salt, as I've translated this mostly character by character.
未醤 楊氏漢語抄云高麗醤〈美蘓 今案弁色立成説同但本義未詳俗用味醤二字〉
Unfinished beanpaste (MISAU). The Genji Kangoshō calls this a Goryeo beanpaste. ([read] miso; at the moment, the plan distinction type goes into a theory of same (?), but the meaning of the three is unknown. [Their] custom is to use two [Chinese] characters 味醤 'beanpaste with taste')
If this word was of Chinese provenance, it either must be an earlier one, and/or it was indirectly transmitted from China to Korea, and to Japan, given the earliest (actual) information about miso at the time (the geography of China, Korea, and Japan suggest that it was most likely Korean people with Chinese influence or derivations must have been more likely, but this is mere conjecture). The semantics do look striking, considering that 醤 is apparently in Old Chinese. Chuterix (talk) 02:51, 22 February 2024 (UTC)Reply
I found that confusing – Yale is focused on morphophonemics in a way that is terrible for any analysis of actual sound values. See also w:Middle_Korean#Script_and_phonology, Middle Korean ㅕ ≠ modern Korean ㅕ. As I've learned more about Korean historical phonology, I am less inclined to view OJP posi and MK byel as likely cognates; the /j-/ glide in the Korean certainly presents an obstacle.
Eta Funayama spelling: 獲□□□鹵 (three middle characters no longer legible)
Notably, the 支 character has the Middle Chinese sound value of /t͡ɕie/ (see 支#Chinese), which explains the Japanese on'yomi of shi. However, the modern Min Nan pronunciation is ki, and I've noticed over the years that Min Nan is often the closest of the modern Chinese varieties to the expected Old Japanese readings: and indeed 支 was used as a man'yōgana for ⟨ki₁⟩. This would give us Waka Taki₁ru, rather than the expected Waka Take₁ru. I see also that there is an Old Chinese reconstructed reading of /ke/, and the swords are dated to the 400s or 500s, raising the possibility that the older Chinese pronunciation might have still been applicable — Middle Chinese is generally dated from 600 or so with the publication of rime dictionaries.
Meanwhile, the final 鹵 has the Middle Chinese sound value of /luˣ/, which would match the expected Japanese pronunciation of the name. Modern Min Nan has lō͘, corresponding to the Japanese kan'on of ろ. ‑‑ Eiríkr Útlendi │Tala við mig00:44, 27 February 2024 (UTC)Reply
Latest comment: 2 months ago2 comments2 people in discussion
@Eirikr: The kana spelling is a gloss addition to the Nihon Shoki. See here for example of where there's a gloss 嶋君 as sema-kisi (セマキシ) and 主嶋 as nirim(u)-sema (ニリムセマ), presumably from other manuscripts which have this reading (too lazy to find a manuscript with kana glosses and dig through it to find evidence of the kana writing). Linguists also consulted kana manuscripts to find other Paekche words, some which are hidden behind kanji readings. For example, サシ(*sasI, “castle”) (the baekje word is on wiktionary as a Japanese entry).
The セマ entry has references in it, so ideally you should consult what is freely accessible to you (I don't know about the third citation, but an academia account should suffice for this first two references). Chuterix (talk) 23:55, 26 February 2024 (UTC)Reply
I take issue with creating any headwords under katakana spellings for languages that patently did not use katakana. I have no problem with such spellings being mentioned in the relevant entries, but given that 1) Baekje speakers themselves did not use katakana; and that 2) the Baekje kingdom was definitively beaten in 663, over a century before katakana was even invented; it stands to reason that katakana cannot be used as a main headword script for this language.
From what I can see of the referenced sources at セマ:
Vovin never explains his use of katakana. One can infer that he simply used the katakana glosses in his source, Bentley.
Bentley explains on page 420:
... All glosses are from the critical text of Iwanami Shoten's Nihon koten bungaku taikei (NKBT), volumes 67-68. The critical text of the edition mentioned above is the Urabe Kanekata text for the first two books, the Age of the Gods section, while the remaining 28 books are from the Urabe Kanesuke manuscript. The Kanesuke manuscript is from the Kamakura era, but clearly was based on older copies in the library of the Urabe clan.9
9 For a discussion of the Kanesuke text and its relation to the rest of the Nihon shoki textual stemma, see Tenri toshokan zenhon sōsho, washo no bu (hereafter TZH), vol. 56, pp. 24-29.
I cannot figure out how to access the Sungdai Cho text without paying USD $35 for a paperback copy.
At best, we might be able to hypothesize that the katakana glosses were applied to the Baekje words shortly after the invention of katakana — but that would still be 150 years or so after the fall of the Baekje kingdom. 150 years is a lot of time for linguistic drift, particularly when a speech community is subject to rapid and traumatic change, such as being invaded and conquered: consider the massive linguistic changes seen in the century following the Norman conquest of England. Moreover, we have the simple problem that katakana were invented for the Japanese language, and research into the language used by the Baekje strongly suggests that they had vowels and glides that cannot be transcribed cleanly using Japanese phonology.
In summary:
Speakers of Baekje never used katakana to write their own documents.
Katakana was not invented until a century and a half after the fall of the Baekje kingdom.
Katakana apparently cannot represent the sounds reconstructed for the Baekje language.
The words are attested very late, and seem to be an extremely short-lived innovation. While investigating the Ryukyuan words for kokonotari, the forms are scarce; most of them appear to be obvious compounds caused by reanalyses. Even the 沖縄語辞典 says that it is a rare word and the more common word is called [kuɲiɴ].
The following cognates of kokonotari (if you want me to give Ryukyuan cognates for short-lived nanatari, or even all of the numerals listed in what you have listed (including ikutari): Yamatohama (ryn) kʰunu-tʰaːɾi, Nakazato (kzg) kú⸣nù-⸢tˀá⸣ì, Wadomari (okn) kùnù-tà⸢í, China (okn), kùnù-⸢té, Yoron Higashi-ku (yox) kunu-tai, Ie (xug) ɸúːnú-tá⸣ì, Nakijin-Yonamine (xug) kʰùnù-⸢tá⸣ì, Shuri (ryu) ⸢kúkúnú-táí, Tarama (mvi) kukunu-taːɭ, Minna (mvi) kukunuːtaiꟲ, Yonaguni (yoi) kʰùgùnù-tàì-n-tù (< *kokono-tari-N-pito).
These appear to be innovations. -tari must be more likely than not an innovation from futari, because there are some Japonic numerals that are monosyllabic. According to Vovin 2020's A Descriptive and Comparative Grammar of Western Old Japanese (page 359):
There are no apparent external parallels for the classifier -ri ‘person.’[1] Since it begins with the /r/, which is unusual even for morpheme-initial position in Old Japanese, I think that it probably has a perfect internal etymology, being a truncation of ar-i ‘being’ (lit.: exist-nml).
^ It means if a grammatical element in WOJ can be compared to a neighboring language, such as Korean, Austronesian, or even the so-called "Altaic" languages. - Chuterix
/pitori/ → /fitori/ (regular lenition) → /fituri/ (regular /o/ → /u/ vowel shift for Ryūkyūan) → /fitui/ (regular /r/ elision for Ryūkyūan) → /chui/ (weak vowel dropping out, and initial fricative and following stop combining to produce a fused fricative)
@Eirikr: You kind of have the last development kind of mixed between agreeable and disagreeable. There must be an intermediary process /fichui/, which is the original /i/ causing progressive paltalization. There is an irregular deletion of the /fi-/, possibly by weakening. The initial fricative has nothing to do with the palatalized syllable.
I'm not entirely sure I fully understand what you say here, but I think I agree with the main thrust about the Okinawan phonological development, and that there would be a progression of something like /fitui/ → /fitʃui/ → */ftʃui/ → /tʃui/.
About -ri or -tori or -tari, I am inclined to agree with Vovin's suggestion that this is simply copular ari fusing with the preceding morpheme, with an excrescential or epenthetical/-t-/ appearing in certain forms.
For "one person", I suspect that the "stronger" pronunciation of the higher-pitched /-o/ in [hìtó] "person" and [hìtóꜜ] "one" would cause a compound like pito ari to collapse into pitori (to avoid adjacent vowels).
For "two people", puta + ari becomes putari in a pretty straightforward fashion.
Notably, Proto-Japonic /-ua-/ apparently evolves into OJP /-o-/, which gives us a possible derivation for "five people": itu + ari = ituari → itori.
For the other "[number] people" terms, the base number words contain no inherent /-t-/ phone, so I suspect that these were coined by analogy, with the /-t-/ phones in pitori, putari, and itori leading to a perceived need for such an element in these other words. I suppose this might have been formed with the "generic counter" element -tu, but then -tu + ari would presumably result in -tori, no? Or maybe this -tu + ari would drop the u instead of fusing with the a?
At any rate, I'm beginning to view the -ri or -tori or -tari counter for "people" as a bit of a phantom.
"For "one person", I suspect that the "stronger" pronunciation of the higher-pitched /-o/ in [hìtó] "person" and [hìtóꜜ] "one" would cause a compound like pito ari to collapse into pitori (to avoid adjacent vowels)."
The modern standard Tokyo accent should not be assumed to be the original accent. hito 'person' clearly belongs to accent class 2.1 (the high register; traditionally HH, but de Boer argues in favor of Ramsey's theory that the interpretation of the Heian period Kyoto accent should be flipped). For hito 'one', while the Daijirin indeed has accent on second mora, typically no other source, not even the NKD2 on JapanKnowledge records accentual data for the standalone word. Someday I will investigate into when OJ contractions are derived from in terms of pitch accent.
"For "two people", puta + ari becomes putari in a pretty straightforward fashion."
Agreed.
"Notably, Proto-Japonic /-ua-/ apparently evolves into OJP /-o-/, which gives us a possible derivation for "five people": itu + ari = ituari → itori."
While this theory is certainly plausible, the problem is that this word is not attested in Old Japanese proper. It is cited to a Nihon Shoki manuscript apparently from 1375-1377, so we have no idea about the actual phonology of this word in Old Japanese, let alone its existence, and we only have 1/2 chance that this is certain.
"For the other "[number] people" terms, the base number words contain no inherent /-t-/ phone, so I suspect that these were coined by analogy, with the /-t-/ phones in pitori, putari, and itori leading to a perceived need for such an element in these other words. I suppose this might have been formed with the "generic counter" element -tu, but then -tu + ari would presumably result in -tori, no? Or maybe this -tu + ari would drop the u instead of fusing with the a?"
Of course these are analogical coinages. Although I do not if the classifier derivation is possible; I suspect this is a back-forming of -tari.
": At any rate, I'm beginning to view the -ri or -tori or -tari counter for "people" as a bit of a phantom."
