Talk:毛冬

Derivation?

@Karaeng Matoaya, in past discussions with retired User:Wyang, he'd brought up the possibility that this is a nativized borrowing ultimately from Reconstruction:Proto-Sino-Tibetan/ma. For Old Korean /*mwotol/, I'm guessing possibly via Old Chinese 勿 (*mɯd) /*mut/. See also User_talk:Wyang/Archive2#Question_re:_copula_in_Korean for the related discussion thread.

Curious as to your thoughts? — This unsigned comment was added by Eirikr (talk • contribs) at 16:44, 14 September 2020‎.

@Eirikr It's certainly a good match on both phonetic and semantic grounds, especially because *-ol in *mwotol is probably a suffix of some sort: see 不冬 (*ANtol) with apparently the same suffix. But in the end it will probably neither be proven nor disproven definitively.

About the wider points your link brings up, if you're still interested in Korean negation diachronically, you might be interested in the new entry at 不知 (*ANti) which to me suggests that the original root of 아니 (ani, “not”) was *ant-, not *an-. This is quite the important new finding which is the definitive nail in the coffin for the perennial attempts to link 아니 (ani) to Japanese negation involving /*n/.

Instead there's an exciting new Korean-Tungusic link which currently seems very difficult to dispute. And what's really thrilling about it is that it's not with Manchu, as you would expect, but with the Northern Tungusic languages like Evenki, spoken deep in the forests of Siberia. Now that I think about it, I should add that to the *ANti article now.

(P.S. Vovin thinks that the original root was *an- after all, but absent more evidence of the correct OK reading of 喩, I think *ant- should be seen as the base root still. He compares *ti in *ANti to the Middle Korean adverbializer 디 (Yale: -ti), but it behaves as a noun per Korean-language academic sources, so the better connection would be the noun-deriving suffix -i.)--Karaeng Matoaya (talk) 14:13, 15 September 2020 (UTC)

Very interesting indeed, thank you!

• A question about the 不冬#Usage_notes. You stated that "-ti was a verbal suffix and not a nominalizer". There are two ways I could interpret this -- 1) that -ti is a suffix that attaches to verbs, but I don't know how it functions; or 2) that -ti is a suffix that creates verbs, but I don't know what it attaches to. Could you clarify?
• Curious how this 不知 (*ANti) behaves as a noun. Negation as a grammatical function would suggest and adverb. Negation as a noun seems odd. Would this have meant something like "an absence of [rest of the clause]"? I can't wrap my head around this otherwise.
• Regarding possible alternative form 不喻, any chance this was a pure logogramic spelling? I note that Chinese 喻 lists a sense of analogy, like, and I wonder if the Old Korean spelling was intended to mean "not like, unlike, different from". Much as noted at 不知#Usage_notes, it is hard to see how 喻 could have been used as a phonogram for ti -- but at the same time, it could not have been used as a phonogram for Vovin's suggested koy, either.
• (The rest of this might belong on another page, but given the Siberian connection, here goes...)

Separately, I've bumped into the odd word here or there that make me wonder about any potential PIE roots floating around in Koreanic.

• This reconstructed *ANti and *ANtol and PIE *h₂énti, particularly the against sense
• More generally, Korean negatives (and even JA negatives) and PIE *n̥- -- although there are so many widely divergent languages that use /Vn/ or /nV/ as a negative marker that this might be something more closely tied to biology somehow.
• Distant similarities between 배 (bae) and a cluster in Japanese including 腹 (hara, “belly”), 張る (haru, “to become or make taut”), 腫る (haru, “to swell”, modern form 腫れる (hareru)), 孕む (haramu, “to become pregnant”), among others, and also PIE *bʰelǵʰ- (“to swell”). Possible connections to KO 불다 (bulda, “to blow”), 부풀다 (bupulda, “to swell”), 부르다 (bureuda, “to be full; to be pregnant”)?

Perhaps more likely, I've also run into loose threads that might suggest distant connections to more areal languages than PIE, such as Hungarian and Navajo. Both may sound far-fetched, but both can be traced back to the w:Xiongnu Confederacy in roughly modern-day Mongolia, not too far from where we might find the ancestors of the Koreanic languages. For Navajo, for instance, see also w:Dené–Yeniseian languages, where a linguist by the name of Vajda has built a pretty convincing case linking the North American Na-Dene languages with the Yeniseian languages of Siberia. Yeniseian speakers are mentioned as possible members of the Xiongnu.

• Hungarian and Navajo both have a word "to become" with an /l/ initial that is used for certain future aspects, Hungarian lenni and Navajo -LĮ́Į́ʼ.
• Many Navajo verb stems manifest a future aspect with an -l suffix, broadly similar to Korean.
• Hungarian derives certain adverbs with an -l suffix, also broadly similar to Korean.
• Navajo expresses a completed aspect of verbs with a modal prefix ni- (our entry is incomplete; see an example of the prefix at nilį́), which some authors think is related to Navajo niʼ (“the earth, the ground”). This completion aspect is probably also present in the /n/ or nasalization morpheme that appears in the past tense of many Navajo verb stems. This echoes the past determiner -n suffix in Korean, the nu / ni ancient copular verb or suffix reconstructed for Proto-Japonic and the ぬ (nu) perfective marker in Old and Classical Japanese, and the ancient and now-rare Japanese term 土・丹 (ni, “earth, soil”).

I'll wrap up here rather than grossly overload this thread.  :) ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:54, 15 September 2020 (UTC)

1) The conventional view of MK -ti is that it's a connective suffix that attaches to the end of main verb stems in order to connect it to the negated auxiliary ᄒᆞ다 (Yale: hota), or to inherently negative verbs like 말다 (Yale: malta). See Modern Korean 지 (ji) for the contemporary reflex with the identical function.

The issue is actually somewhat more complicated. -ti sometimes appears to act as a nominalizer, e.g. from Seokbo sangjeol (1447):

부텨 맛나디 어려ᄫᅳ며 법 듣디 어려ᄫᅳ니

Pwuthye mas-na-ti elyeW-umye Pep tut-ti elyeWuni

Meeting the Buddha is difficult and listening to Dharma is difficult

But it's an unusual type of nominalizer which can never be followed by a case marker. What's even more perplexing is that the two uses of -ti went completely different ways in the Early Modern period. The connective suffix remained as such and underwent a regular sound shift to become Modern Korean 지 (ji), while the nominalizer underwent an irregular and extremely difficult-to-explain sound change to 기 (gi). This raises the question of whether MK -ti was ever perceived as a single morpheme at all by Korean speakers, or were simply two homophonous morphemes.

2) 不知 (*ANti) is perhaps best literally translated as "something which is not; something counterfactual". It's considered a noun because it regularly takes the copula 이다 (ida, “to be”), which shouldn't happen to non-nouns. It also seems to have behaved as a determiner too, which is not unattested: for example, Middle Korean 새 (Yale: say) "new" meant both "new" as a determiner, and "something new" as a noun. This probably reflects a more widespread phenomenon in Old Korean where nouns could more freely attribute each other without intervening verbs.

3) My impression was that 喻 now means "analogy" only because of retroactive influence from words like 比喻 (“comparison, lit. 'understanding side-to-side'”), 暗喻 (“metaphor, lit. 'understanding secretly'”), so it wasn't one of the original meanings of the character. I could well be wrong about this, however. The Korean pedagogical tradition for Chinese characters to this day only uses the two definitions cited by Vovin, "to inform" and "to understand".

As for the long-range connections, I'll have to think about them some more. And thanks for the discussion, this has been very interesting :) --Karaeng Matoaya (talk) 08:03, 16 September 2020 (UTC)