Bear words

« previous post | next post »

In "Dynamic stew" (10/24/13) and the comments thereto, we had a vigorous discussion of words for "bear" in Korean, Sinitic, Tibetan, and Japanese,  And now Diana Shuheng Zhang has written a densely philological study on "Three Ancient Words for Bear," Sino-Platonic Papers, 294 (November, 2019), 21 pages (free pdf).

Let's start with the basic word for "bear" in Sinitic:  xióng (MSM) 熊.

Etymology

From Proto-Sino-Tibetan *d-wam. Cognate with Tibetan དོམ (dom, "bear"), Burmese ဝံ (wam) (in ဝက်ဝံ (wak-wam, "bear")).
————————
Major languages

 

Topolectal data

less ▲

Variety Location

Mandarin Beijing /ɕyŋ³⁵/
Harbin /ɕyŋ²⁴/
Tianjin /ɕyŋ⁴⁵/
Jinan /ɕyŋ⁴²/
Qingdao /ɕiŋ⁴²/
Zhengzhou /ɕyuŋ⁴²/
Xi'an /ɕyŋ²⁴/
Xining /ɕyə̃²⁴/
Yinchuan /ɕyŋ⁵³/
Lanzhou /ɕỹn⁵³/
Ürümqi /ɕyŋ⁵¹/
Wuhan /ɕioŋ²¹³/
Chengdu /ɕyoŋ³¹/
Guiyang /ɕioŋ²¹/
Kunming /ɕiŋ³¹/
Nanjing /ɕioŋ²⁴/
Hefei /ɕiŋ⁵⁵/
Jin Taiyuan /ɕyəŋ¹¹/
Pingyao /ɕyŋ¹³/
Hohhot /ɕỹŋ³¹/
Wu Shanghai /ɦioŋ²³/
Suzhou /ɦioŋ¹³/
Hangzhou /ɦioŋ²¹³/
Wenzhou /joŋ³¹/
Hui Shexian /ɕyʌ̃⁴⁴/
Tunxi /ɕin⁴⁴/
Xiang Changsha /ɕioŋ¹³/
Xiangtan /ɕin¹²/
Gan Nanchang /ɕiuŋ⁴⁵/
Hakka Meixian /iuŋ¹¹/
Taoyuan /ʒuŋ¹¹/
Cantonese Guangzhou /hoŋ²¹/
Nanning /juŋ²¹/
Hong Kong /huŋ²¹/
Min Xiamen (Min Nan) /hiɔŋ³⁵/
/him³⁵/
Fuzhou (Min Dong) /hyŋ⁵³/
Jian'ou (Min Bei) /xœyŋ³³/
Shantou (Min Nan) /him⁵⁵/
Haikou (Min Nan) /hiɔŋ³¹/

Reconstructions

Middle Sinitic: /ɦɨuŋ/

Old Sinitic

(BaxterSagart): /*C.[ɢ]ʷ(r)əm/
(Zhengzhang): /*ɢʷlɯm/

Source (also for most of the following in Japanese, Korean, and Vietnamese)

Japanese

Readings

 

 

Pronunciation

Etymology

From Old Japanese. Probably cognate with (kuma, "inside corner; inner bend; hollow or hole in something"), perhaps from the way that bears often live in dens. Probably also cognate with Korean (gom, "bear; hole").

Korean

(ung)

  • Eumhun:
    • Sound (hangeul): ung 웅
    • Name (hangeul): gom 곰

gom 곰 ("bear")

The Sinitic morpheme xióng 熊 would be pronounced "ung" in Korean, but the Korean word gom 곰 is preferred.  The latter is homophonous (okay, homographic) with the Korean word for "a thick broth made of cooked meat".

Vietnamese

(hùng)

The following observations are mostly extracted from "Dynamic stew" (10/24/13), cited at the beginning of this post.

One commenter suggests that gom is xióng 熊; it's just a really old piece of shared vocabulary. Newer OS reconstructions give things like *wum; adding the "uvular" suggests ~*ɢʷum.

From Juha Janhunen:

My idea of Korean gom [kom] is that it is a borrowing from Para-Japonic and represents the same word as regular Japanese kuma 'bear'. The original form may also be reconstructed as *kuma. (I know of arguments presented against this reconstruction, but I do not think they are valid.) I cannot see how this item could in any regular way be connected with any early form of Chinese xiong.

The same commenter cited above responds:

I find Janhunen's comments interesting as, whatever we might suppose regarding the cause for the resemblance, *ɢʷum is certainly phonologically close to /kom/ and still closer to the Middle Korean kwom cited at Axel Schuessler, ABC Etymological Dictionary of Old Chinese, p. 542 — where, it turns out, the author has already noted the similarity between these two as well as a number of other area 'bear' words, largely TB items but also PMonic *kmum 'Himalayan black bear'.

Jongseong Park:

On the subject of the homonym 곰 gom [ɡ̊oːm] "bear", it is already attested in the current spelling in Seokbo sangjeol (釋譜詳節; 석보상절), 1447, one of the first works written in the Korean alphabet. The Japanese クマ/くま kuma is no doubt a cognate. The Old Chinese reconstruction *wum ~ *ɢʷum is certainly intriguing. Note that the Sino-Korean reading for 熊 is ung [uŋ], reflecting -ŋ in Middle Chinese. So if there is a connection between Chinese 熊 and Korean 곰 gom, then it must be very old, predating the establishment of Sino-Korean readings connected to the large-scale adoption of Classical Chinese as the literary language during the late Three Kingdoms / Unified Silla period. I've seen a similar suggestion that the Korean word for "wind", baram [b̥aɾam], is from Old Chinese 風 *priəm, though I'm completely out of my depth when it comes to evaluating such claims.

More from Jongseong Park:

I found some assertions that the Old Korean form of 곰 gom "bear" was 고마 goma. One evidence is that the native name of the early capital of Baekje, one of the Three Kingdoms of Korea, is given as 固麻, which is 고마 goma in Sino-Korean. This capital is better known as 熊津 웅진 ungjin, which seems to be a translation of a native name like 고마나루 goma naru "bear crossing".

It does look to me like 고마 goma is a plausible early form for Middle Korean 곰 gom and Japanese kuma. The extra vowel at the end does move it further from the Chinese 熊, though. Not sure what this means….

From an anonymous colleague:

If you want to go way out on a limb, there is modern Tibetan དོམ༌ (tom) meaning "bear". The vowel and final work nicely but the initial consonant is a stretch.

From Juha Janhunen (personal correspondence [9/15/19])

I wrote a paper on bear cult, available here (pdf).

However, in this paper I did not discuss bear words, and I do not think I have discussed them elsewhere, either. In many languages, 'bear' is referred to by a taboo word, e.g., Finnish karhu from karhea 'rough to touch', a taboo word with an age of at least 1,500 years, since it is also present in Estonian (karu). An older word for 'bear', with cognates as far as the Permic languages, was oksi : ohde- (< *okti), which is also preserved in Finnish in the derivative ohto 'bear' (< *okt-iw). Then we have languages that, probably also for taboo reasons, have borrowed the word for 'bear', like Hungarian medve (from Slavic).

I was during the summer in Tuva (Tannu-Tuva) and learnt the word adïg 'bear' – there were a lot of bears there. However, more recently, when I looked at some data of regular Turkish I noticed that they also have the same word, in the regularly evolved shape ayï (orthographically ayı). So, this is a Common Turkic word, *adïg, deriving from Proto-Turkic (without Chuvash). Erdal has *adïg 'sober' in his Old Turkic Word Formation, but I am not sure if it is the same word.

From Chris Button (personal correspondence [9/9/19])

Personally I think the Mon-Khmer, Japanese, and Korean forms are probably all ultimately somehow related to Chinese. There doesn't seem to be a likely Indo-European cognate though.

 

Reading

"Bear talk" (11/15/19)

"Horses, soma, riddles, magi, and animal style art in southern China" (11/11/19)

 

[Thanks to Chris Button for reminding me to make this post.]



37 Comments

  1. Chris Button said,

    November 19, 2019 @ 10:30 pm

    Personally I think the Mon-Khmer, Japanese, and Korean forms are probably all ultimately somehow related to Chinese. There doesn't seem to be a likely Indo-European cognate though.

    Just to add to this, 熊 *wə̀m "bear" reminds me of (the likely PIE loanword) 犬 *kʰʷə́ɲʔ "dog" purely by virtue of its interesting phonological shape. The rhymes -əm and -wəm were not distinctive in Old Chinese and so *wə̀m as a standalone syllable seems somewhat unlikely to exist without any external influence.

  2. Victor Mair said,

    November 20, 2019 @ 8:59 am

    From Guillaume Jacques:

    The etymologies of 羆 and 熊 are discussed in the following article:

    A study of cognates between Gyalrong languages and Old Chinese
    http://www.jolr.ru/index.php?article=263
    pdf also available directly here

    and this earlier unpublished paper (from a conference in Shanghai in 2005, where we met):

    嘉绒语与上古汉语
    https://halshs.archives-ouvertes.fr/halshs-00137451/document

  3. Victor Mair said,

    November 20, 2019 @ 10:33 am

    Because it is relevant here, I'm copying the following comment from "Horses, soma, riddles, magi, and animal style art in southern China" (11/11/19):

    I asked John S. Major the following questions:

    Basic notes and questions about 熊 in Chu culture:

    1. It's the clan name of the royal, founding family, right?

    2. Is it supposedly a word in Chu language? Or, if not, do we have a
    Sinographic transcription of the sound of it in Chu language?

    He replied:

    What I know about this is summarized in Barry Blakeley's essay in Cook & Major, Defining Chu, pp. 53-54. Blakeley doubts that Xiong was a lineage name for the Chu ruling family; he finds the most likely explanation of the term (choosing from among several competing explanations) is that it derived from an old official title. That would imply that it was a Chu word, and that the graph 熊 was borrowed to convey the sound of that Chu word (however it was pronounced 3,000 years ago). I don't know of any credible attempts to reconstruct the Chu language.

  4. Nick Kaldis said,

    November 20, 2019 @ 10:53 am

    Hi: I am curious if Modern Greek, of which I know a handful of words and phrases, is unusual for having a 3-syllable word for "bear" : "ar-kou-tha [αρκούδα]". English and Chinese, e.g. have single-syllable words.

    Nick

  5. cameron said,

    November 20, 2019 @ 11:25 am

    Is the Ainu word kamuy, which means both "bear" and "god", and which seems similar to Japanese kuma and the Shinto concept of kami, thought to be a borrowing from Japanese?

  6. David B Solnit said,

    November 20, 2019 @ 7:17 pm

    Just to throw some more ingredients into the stew: if I remember rightly, Paul Benedict reconstructed proto-Austro-Tai *krumbay for 'bear', yielding
    (a) From the 2nd syllable with influence from the 1st syllable's vowel, proto-Tai (Pittayaporn) *ʰmwɯjᴬ plus similar forms in other Kra-Dai branches.
    (b) From the 1st syllable, proto-Hmong-Mien (Ratilff) *qrep.
    (c) From the whole thing, proto-Austronesian *Cumay.
    Later of course Benedict also connected this *krumbay with Japanese kuma.

  7. Jonathan Smith said,

    November 20, 2019 @ 7:48 pm

    Re: the Blakeley remarks, Schuessler (2007: 3) (citing Pulleyblank's "The Chinese and their neighbors…" which I don't have to hand) notes that "the ruling family of the ancient state of Chu 楚 had the clan name xiong 熊 'bear', but in the Chu language the name was mi 芈 which is the KT word for 'bear'."; cf. Pittayaporn's pTai form above. This comparison arguably rises to the level of "famous" in certain historical linguistic circles so I am curious whether Blakeley had reasons to be suspicious of it or rather was not familiar with it.

  8. Chris Button said,

    November 20, 2019 @ 11:00 pm

    And now Diana Shuheng Zhang has written a densely philological study on "Three Ancient Words for Bear," Sino-Platonic Papers, 294 (November, 2019)

    Personally I think her explanation of 羆 on p.10-11 as a "striped bear" is truly brilliant. Although, following Pulleyblank, I would reconstruct 羆 with an *-l coda in Old Chinese anyway (we've discussed the alternation of OC *-l ~ *-n elsewhere on LLog) without needing to go back to the PST stage for it.

    Regarding a graphic analysis, I think 能 depicts a 熊 "bear" in the same way that 來 depicts 麥 "wheat". That 能 and 來 were used for other purposes doesn't necessarily affect that. However, the phonetic distance of 能 from 熊 (unlike 來 from 麥) does make the notion that 能 represented a different (Austroasiatic?) word for "bear" from 熊 eminently possible. I suppose it's unlikely that the regular dissimilation of the original *-m coda of 熊 to -ŋ has any connection with the rather random nasalization of the *-ɣ coda of 能 to -ŋ. In addition to the Shijing rhyme evidence, the word now represented by 能 nə́ɣʔ "ability, capable" clearly did not originally have a nasal coda since it shares word family connections with 女, 奴 and other derivatives via the ə/a ablaut. The form 女 originally depicted the slave/servant of 奴 rather than a woman, as Takashima has recently discussed, and hence a sense of "able-bodied work".

    "the ruling family of the ancient state of Chu 楚 had the clan name xiong 熊 'bear', but in the Chu language the name was mi 芈 which is the KT word for 'bear'."

    Given that 芈/咩 is supposed to be an onomatopoeic representation of a sheep bleating, it certainly does seems unlikely that it would be being used for anything but the sound for the word for "bear" which is then being translated as 熊, as Schuessler remarks.

  9. Victor Mair said,

    November 21, 2019 @ 6:34 am

    From Juha Janhunen:

    Thanks for the stimulating data concerning bear words. I like the idea that Japanese kuma could be a compositum from *ku+ma 'corner place', used as a taboo word for 'bear'. I know that Sasha may have a different opinion, but there are several other words in Japanese/Japonic, which seem to contain +ma 'place' as the second element: ya+ma 'mountain', (*)pa+ma 'shore', si+ma 'island', perhaps also mi+ma+na 'Kaya' and ko+ma 'Koguryeo'. On the other hand, Korean kom <*kum(V) 'bear' has a structural analogy in *pem(V) 'tiger' - the two beasts are intimately interconnected in Manchurian folklore.

  10. Victor Mair said,

    November 21, 2019 @ 8:54 am

    From John Whitman:

    The pJK word was disyllabic, as Juha says, but the protoform was almost certainly *koma, not *kuma. See the discussion on pp. 10-11 in my forthcoming CUP book with Sungdai Cho, Korean: a Linguistic Introduction (maybe already out). Pre-OJ had regular mid vowel raising in nonfinal syllables, explaining pJ *kuma.

    That doesn't eliminate the likelihood that the word is a loan, certainly from one of the protolanguages to the other, nor the possibility that one or both borrowed it form OC (or a precursor or relative of OC). This possibility has already been suggested by Ho-min Sohn in his book The Korean Language (CUP).

  11. Victor Mair said,

    November 21, 2019 @ 12:49 pm

    From Bob Ramsey:

    I've only taken a cursory look at this thread about 'bear', so I'm not going to give much in the way of well-considered remarks. But I do have to say that I find the attempts to link the Chinese forms to languages outside of Sinitic to be a bit forced.

    I'm thinking in particular (of course!) of Japanese and Korean. To imagine that these cultures would borrow a word and concept so long and deeply embedded in their respective cultures from Sinitic sounds like too big a stretch to me. What I found much more interesting in the thread were the suggestions about linking the Japanese and Korean words. That comparison represents a very, very old hypothesis, of course–and not only because of the striking phonological similarity between Japanese kuma and Korean kom. I'm thinking in particular of the serious and knowledgeable posts by Jongseong Park. What he says about the name of the old Baekje place name, 'bear crossing', is especially worth noting. What Jongseong Park doesn't explicitly mention, though, is the famous citation in the 1447 text Yongbi och'on ka 용비어천가 ('Songs of Dragons Flying to Heaven') that Korean philologists always point to. There, in that earliest of Middle Korean texts, the Baekje place name 熊津 is spelled out in Hangul as ko ·ma* nolo (I'm transcribing the form here in Martin's Middle Korean romanization system–I don't know how to render it in Hangul, because the MK vowel known as 'lower a', which Martin writes as ⟨o⟩ , isn't in the Hangul font I'm using!) In any case, note that the MK tones here are Low-High Low-Low, and when the final vowel of ko ·ma* 'bear' elides, the resulting form is :kom, with a rising tone, which is a regular and well documented tonal development that in Korean today is realized as a syllable with a long vowel. In other words (as Jongseong Park points out), the comparison with Japanese kuma is almost perfect. What that matching means is not altogether clear, of course: Is the phonological similarity simply an accident (unlikely!), or does it represent some kind of cultural loan, or perhaps a cognate shared by Japanese with Baekje and Sillan (the direct ancestor of modern Korean)? Whatever we conclude about that similarity, though, it seems to me that this side story strongly suggests that we should probably dismiss the idea of a Chinese connection as fanciful.

    *[VHM: I had to separate the two syllables of this word. Otherwise the interpunct turned into something strange.]

  12. Chris Button said,

    November 21, 2019 @ 5:13 pm

    @ Bob Ramsey

    It's been a while–I hope all is well! You may well be correct that there is no association between the Japanese/Korean forms and the Sinitic form. However, from a purely internal phonological perspective, Old Chinese 熊 *wə̀m sticks out as bit of an anomaly (always a good indication of a possible external association). And Shorto's Proto-Mon-Khmer form *[k]mum surely fits into it all somewhere.

  13. Chris Button said,

    November 21, 2019 @ 5:17 pm

    And Shorto's Proto-Mon-Khmer form *[k]mum surely fits into it all somewhere.

    Might be worth noting that we already have a precedent here in 虎 "tiger" which I don't think anyone disputes came from Mon-Khmer.

  14. Chris Button said,

    November 21, 2019 @ 5:51 pm

    Regarding Mon-Khmer *[k]mum and Old Chinese *wə̀m, it might be worth noting that the onsets km- and w- are both labiovelar.

    Incidentally, in John Ohala's "The Story of [w]" some interesting comparisons are made with the behavior of [k͡p] in some West African languages with [w]

  15. James R. Chamberlain said,

    November 21, 2019 @ 8:32 pm

    With respect to the water association and softshell turtles addressed by Shuheng Zhang, I thought immediately of Eberhard's "yellow bear" that I recently reread for a paper I am writing on giant softshells. Here is the secton:

    Carr's (p.c.) 'giant salamander' forms from the Erh-Ya.

    16/ 41a *ŋieg / ŋiei / yi
    16/41b *g'a / ɣa / hia (Karlgren 33i *g'å)
    BaxterSagart *[g]ʕra 'hia-ma frog' 蝦

    Schafer (1967:111) identifies the first form as a Yu from the literature of the Zhou Dynasty, but not described until the Han period. It was called variously "water crossbow; sand mother; shadow shooter; shooting artist; and short fox." Alternatively it could be,

    Karlgren 929r *g'wək / ɣwək / huo , or, *gi̯wək / ji̯wək / yü
    'a fabulous nocuous water animal, water demon'

    Schafer treats it as an insect, but Eberhard (1968: 193ff) writes that the yu creature is born from the sperm of the "speckled mountain people" who have intercourse out of doors like animals, a sin in the eyes of the Chinese. The yu he says looks like a "yellow bear", but he assumes it to be an error because in his sources the two words hsiung 'bear' and neng 'turtle (pieh [=softshell]) with three legs' look similar.

    Chang Heng in the 2nd century A.D. reported such a turtle living in the river I (Tung-ching-fu). Chien-hu-chi (Part 9, chapt. 4, 12a) proposes a pronunciation of nai or t'ai for this neng, describes it as a three-legged turtle, the animal which always, together with the snake, is regarded as a cosmic symbol. Its meat is supposed to be poisonous. (ibid)

    The semantic variation in 41b may result from applying an inland mountain stream term to a more coastal environment as the character has come to be used in compounds meaning 'shrimp, prawn, lobster.' In any event, we do not have good lexical data for 'giant salamander' which should be easily available if linguists were to explore such terms, especially in Kra, Kam-Sui and Chinese as it is a highly conspicuous animal, especially in Guizhou. It is the world's largest amphibian attaining lengths of up to six feet. So far linguists working on Kra-Dai languages seem to have been unaware of its existence with the sole exception of the Mahidol University team that worked on Sui as cited below.

  16. James R. Chamberlain said,

    November 21, 2019 @ 9:04 pm

    On 'bear in AA: At one time Gérard Diffloth reconstructed PMK *cg__w 'bear'. Based on well-known reflexes such as Việtnamese gấu. And in other Kri-Mol (Vietic) languages we find forms like Ahao tăkuu, Maleng săkuṳ, and so on, usually with the presyllable. But there is also Thémarou and Atel rɤɤm , Mlengbrou cămok, and Ahlao ʃăduul, just to add to the mix.

  17. Chris Button said,

    November 21, 2019 @ 9:42 pm

    @ James R. Chamberlain

    With respect to the water association and softshell turtles addressed by Shuheng Zhang, I thought immediately of Eberhard's "yellow bear"… Schafer treats it as an insect, but Eberhard (1968: 193ff) writes that the yu creature is born from the sperm of the "speckled mountain people" who have intercourse out of doors like animals, a sin in the eyes of the Chinese. The yu he says looks like a "yellow bear", but he assumes it to be an error because in his sources the two words hsiung 'bear' and neng 'turtle (pieh [=softshell]) with three legs' look similar

    Your comment, and re-reading Zhang's fascinating paper more closely, has me retracting my comment above about 能 depicting a 熊 "bear" and instead thinking about the relationship between 夒 and 夔 and how it compares with 能 and 熊. Both (mythical) animal pairs are graphically easily confused and phonologically unrelated.

    So may we follow Zhang's suggestion that 能 might indeed be the "soft shell turtle" (aka "yellow bear") and 熊 is the "bear" ? Perhaps, after all, 能 has as little to do with bears as 夔 does with monkeys!

    (incidentally, I suspect an association of 夔 *ʁrə̀ɣ with 夏 *gráɣʔ as we discussed elsewhere on LLog)

  18. Chris Button said,

    November 21, 2019 @ 9:58 pm

    I also wonder if Zhang's additional comment on p.20 that the 灬 (火) at the base of 熊 was originally part of the graph (we might compare the way that the very top component of 夔 was part of the graph rather than an addition to 夒) might account for the bizarre alternate reading of 炎 *làm as *wàm. That is to say, perhaps it was the association of 炎 *làm with 熊 *wə̀m as its supposed phonetic component (although actually resulting from graphic distortion) that led to the emergence of the aberrant *wàm reading.

  19. Chris Button said,

    November 21, 2019 @ 10:36 pm

    …the rather random nasalization of the *-ɣ coda of 能 to -ŋ

    Schuessler compares Burmese နိုင် *nɨŋ "can, be able". The semantics and phonology are near perfect matches. The issue is that the spelling shows the Burmese form to be a Sanskrit/Pali/Shan/Mon loanword. I'll have to look around to find the source, but perhaps that is also what lies behind the addition of the -ŋ to the Chinese form.

  20. Chris Button said,

    November 21, 2019 @ 10:43 pm

    Gordon Luce suggests a Shan origin.

  21. Jonathan Smith said,

    November 21, 2019 @ 11:02 pm

    This problem is too hard, or is too many problems :( looking through early dictionaries from Shuowen forward, it seems clear that some word or words written "能" must have ended in -m… just as one/some must have begun with a voiceless hn- or something, as already suggested by reference to 態…

  22. Chris Button said,

    November 22, 2019 @ 6:36 am

    @ jonathan Smith

    On the contrary, I think it's all becoming much clearer.

    熊 *wə̀m "bear" is unrelated to 能 nə́ɣ "soft-shell turtle" (they just became graphically confused in the same way as 夔 and 夒). It is at least related to Mon-Khmer forms (cf. 虎 "tiger) where proto-MK labiovelar *km- compares with OC labiovelar *w-.

    能 nə́ɣ(ʔ) "can, able" (later *nə́ŋ possibly due to some external influence) is an Old Chinese word etymologically related to 耐 nə́ɣs "endure". It has a xiesheng (phonetic speller) derivative 態 ʰnə́ɣs "manner, form, condition" that is etymologically related to 而 nə̀ɣ (I'll have to get into the semantics of that one separately). In a perfectly logical (and idealistic) spelling system 耐 would be written with 能 as phonetic and 態 would be written with 而 as phonetic, but alas they are essentially flipped)

  23. Jonathan Smith said,

    November 22, 2019 @ 9:20 am

    Could be :) I mean there are lots of interesting indications sometimes pointing in conflicting directions. E.g., while not necessarily meaningful, medieval dictionaries speak of two animals named "能", assigning the noj/pingsheng reading to 'tortoise or sth' and the nong/pingsheng reading to 'bear or sth'…

  24. Chris Button said,

    November 22, 2019 @ 9:52 pm

    Regarding the *-ɣ ~ *-ŋ coda alternation in 能…

    Original *-ǝŋʔ is considered to have largely merged with *-ǝɣʔ in Old Chinese. However, if one looks at Proto-Sino-Tibetan, *-ǝŋʔ appears to have merged with Old Chinese *-ǝŋ instead, and in some Tibeto-Burman languages hardened to *-ǝk (e.g., 薪 *sə̀ɲ, Old Burmese *sɐc, Kuki-Chin *tʰɪŋ², which would regularly go back to *sjəŋ, *sjək, *sjəŋʔ –the last form representing the original PST form).

    As a result, it seems original *-ǝŋʔ could have merged with *-ǝɣʔ or *-ǝŋ in Old Chinese, and 能 just happens to have retained both variants.

    So we have two words:
    1. 能 *nə́ŋʔ (becoming *nə́ɣʔ and *nə́ŋ) "(aquatic) animal–possibly soft-shelled turtle" loaned for sense of "ability"
    2. 熊 *wə̀m "bear"
    Neither are related to each other, but graphic convergence has meant they have become very confused.

    (as for sporadic evidence of -ŋ in Sinitic forms of 耳 *nə̀ɣʔ "ear", I wonder if that just comes down to some sort of analogical hyper-correction?)

  25. Chris Button said,

    November 22, 2019 @ 11:38 pm

    On the other hand, in Baxter & Sagart, there's also a suggestion about some kind of nasal spreading that might have influenced it. We do also have examples like 乃 *nə́ɣʔ and 仍 *nə̀ŋ. In that case, perhaps 能 could go back to *nə́ɣʔ as its earliest form after all… I wonder if the -ɣʔ combination might be particularly susceptible to the nasal assimilation since [ŋ] combines the occlusion of [ʔ] with the continuation of [ɣ]. That then also allows doesn't contradict the broader word family connections with 女/奴, etc.

  26. James R. Chamberlain said,

    November 23, 2019 @ 2:48 am

    Just to clarify the semanatics, I submit that the original meaning of 能 was 'giant salamander' (Andrias davidianus) and this was later confused with softshell or giant softshell. I base this mostly on an earlier work of the zoologists Liu, Ch'eng-chao. (et al) 1950. The Amphibians of Western China.

    Liu (1950: 69-70) writes that the common Chinese name is "nei-yu," and that the Shan-hai-ching ("Documents of the sea and mountains") written in 600 BCE describes it as follows:

    In the streams of Ille-tzu, Shansi Province, there are found numerous Nei-yu; their form is similar to that of a serpent; the largest one is called Sha-yu.

    Later in the Han Dynasty, second century BCE Sima Qian noted that this salamander was found in Shensi Province where it was named Jen-yu 'man-fish.' The Erh-Ya of Koo Po (c. 300 CE) wrote that the nei-yu made cries similar to those of a small infant, and also that the larger ones, 8-9 feet long, were called Sha-yu. (ibid).

    The vocalizations being similar to a baby crying are a common feature of descriptions of this animal.

    I have no idea what the sha-yu etymology might be.

  27. Chris Button said,

    November 23, 2019 @ 3:04 pm

    @ James R. Chamberlain

    Just to clarify the semanatics, I submit that the original meaning of 能 was 'giant salamander' (Andrias davidianus) and this was later confused with softshell or giant softshell.

    Could you possibly elucidate this a little more (beyond what you already posted)?

    …wrote that the nei-yu made cries similar to those of a small infant

    That makes it interesting that the phonetic in the 鯢 of 鯢魚 is 兒.

  28. James R. Chamberlain said,

    November 24, 2019 @ 4:17 am

    I am suggesting that 'salamander' MC ŋei OC *ŋeː (Zhengzhang) is the same word as *nə́ɣ(ʔ) which might just as well have been nəɯ or nəɰ , retaining the retraction and conditioning the raising of the the vowel. If so, the 'bear' element could have originally been a modifier to distinguish the 'giant' salamander from other ones, and hence the confusion. This is a common naming practice in Kra-Dai languages, using one animal to distinguish the size in another, thus in Tai mii khwaay 'buffalo bear U. thibetanus" from mii maa 'dog bear Helarctanos' or lɯŋ muu (pig) 'hog badger' from lɯŋ maa (dog) 'feret badger'… etc.

    That having been said, the softshell totem idea, especially giant softshell such as used to be found in the Yangtze basin, should still be considered given its mythological importance in China and Vietnam. The two "giants" are ecologically distinct, the softshell only inhabiting large rivers in the lowlands, while the giant salamanders prefer higher elevations. In the Chu state the latter might indicate Kra speaking areas.

  29. Chris Button said,

    November 24, 2019 @ 7:57 am

    So 鯢 *ŋáj (the "e" in Zhengzhang's *ŋeː represents a likely surface reflex rather than the underlying phonology) would ultimately go back to **ŋjáɣ. The -j- palatalized the *-ɣ (possibly [ɰ]) coda to *-j (as happened with the other velars *-k and *-ŋ shifting to *-c and *-ɲ). In abstract terms, a shift of *ŋj- to *n- is entirely reasonable. That would give náɣ which corresponds well with *nə́ɣ (or possibly rather *nə́ɣʔ) via the ə/a ablaut (and possibly also helped by the palatal environment raising the lower -a- vowel). However, the question would then be why in this single word did the palatal feature not regularly spread to the coda and instead affect the onset? From an Old Chinese perspective, it seems entirely arbitrary (compare the nasal spreading I was talking about above where at least there are multiple examples and the environment is better controlled) unless something fishy :) was going on.

    Separately, I thought Zhang's observation on p.16 of 能 being transcribed in one instance as (a long-bodied animal) was interesting, although again the phonology precludes any obvious relationship.

  30. Chris Button said,

    November 24, 2019 @ 7:59 am

    Separately, I thought Zhang's observation on p.16 of 能 being transcribed in one instance as XX (a long-bodied animal) was interesting, although again the phonology precludes any obvious relationship.

    XX above should have appeared as 贏 minus the 貝 component in the bottom middle

  31. Chris Button said,

    November 25, 2019 @ 11:16 pm

    In abstract terms, a shift of *ŋj- to *n- is entirely reasonable

    Actually, while we do have cases like the one here of 鯢 *ŋáj giving modern Mandarin , the above comment is probably too lenient. It's a stretch to justify going beyond *ŋj- shifting to a palatal nasal *ɲ- (paralleling the shift of *-ŋʲ to -ɲ in coda position) without a significant degree of conditioning or significant restrictions on the phonological inventory. Needless to say, I doubt any relationship between 鯢 *ŋáj and 能 *nə́ɣʔ can really be established.

    Just to throw some more ingredients into the stew: if I remember rightly, Paul Benedict reconstructed proto-Austro-Tai *krumbay for 'bear'…

    The following is from Vovin (1994) "Is Japanese Related to Austronesian" commenting on Benedict's proposal (PAn, PK, OK, MK, PJ, OJ represent Proto-Austronesian, Proto Korean, Old Korean, Middle Korean, Proto Japanese, Old Japanese; LH and R denote pitch/tone):

    "PAn *krumay < *krumbay and PK *kumayᴬ 'bear'; OJ kuma 'bear' *r- > n-; if PAn and PJ forms were related, we could expect something like *numey in OJ < PJ **numai. There is an alternative Altaic etymology: PK *kumV LH (MK kwom R) 'bear'."

  32. Chris Button said,

    November 25, 2019 @ 11:21 pm

    I've changed the forward and backward arrows to ← and → since they were messing up the formatting. Here is the Vovin quote again:

    "PAn *krumay ← *krumbay and PK *kumayᴬ 'bear'; OJ kuma 'bear' ← *kumai (Benedict 1990:164). The reconstruction *kumai for PJ is ad hoc: only PJ *kuma 2.3 'bear' is reconstructable on an internal basis. One would also expect the following development for the cluster *kr- in PJ: *kr- → *r- → n-; if PAn and PJ forms were related, we could expect something like *numey in OJ ← PJ **numai. There is an alternative Altaic etymology: PK *kumV LH (MK kwom R) 'bear'."

  33. Chris Button said,

    November 26, 2019 @ 7:07 am

    Needless to say, I doubt any relationship between 鯢 *ŋáj and 能 *nə́ɣʔ can really be established.

    Or more precisely, 鯢 *ŋáj and 能 *nə́ɣ without the glottal -ʔ since the glottal presumably only existed in *nə́ɣʔ (→ *nə́ŋ via nasal spreading ??) "can, able" for which the character 能 was loaned.

  34. James R. Chamberlain said,

    November 26, 2019 @ 5:00 pm

    I liked the "entirely reasonable" comment even though you rethought it. But also your "something fishy" remark.

    I find that within animal domains there is almost always something fishy, and because they are usually subject to taxonomic hierarchies (that is preceded by terms for bird, fish, snake and the like) and a wide variety of other forces such as collocations, sound symbolism, expressive-ness, poetics, metaphor, gender, and so on, all of which may impact on the phonology of any given term. Historically what may be preserved are vestiges of these forces rather than a single monosyllable. The likelihood of two such phonologically similar morphemes in the same geographical area, both with the meaning "large mysterious creature that lives beneath the water," being lexically (and zoologically) distinct is counterintuitive, especially given the ritual or mythological significance.

    Within the geographical area of Chŭ the Sui word for giant salamander is
    /mom6 tin2 m̥a1/ 'dog-foot fish' or /mom6 va4 va4 vi4/ (apparently an expressive), both classified as a fish, as is the softshell /mom6 tjaau5/ 'fish+turtle'.

    In the case of bear, other animals do indeed intrude into the environment, a good example is provided by Badenoch in his study of faunal naming in Sida, a Loloish language, where some bear names borrowed from Chinese include dog-bear 狗熊 pig-bear 猪熊 horse-bear 大吗熊, kou-ʃù, tɕu-ʃù, and
 tɐ́ -mà-ʃù respectively.
    I don't mean to detract from the non-fishy evidence, but rather to suggest that there is probably more to the story than is represented by the brief arc of the whole we are privy to, and to provide a glimpse into the multi-linguistic Chŭ environment as Zhang has recommended in her paper. In such contexts, fishiness is the rule rather than the exception.

  35. Chris Button said,

    November 26, 2019 @ 9:21 pm

    @ James R. Chamberlain

    Well we do have 狗 *káwʔ and 犬 *kʰʷə́ɲʔ that are plausibly related, although we have a loan at two different time depths to account for it. So you never know…

    By the way, I thought your comment below was a really interesting as regards the association with "bear":

    If so, the 'bear' element could have originally been a modifier to distinguish the 'giant' salamander from other ones, and hence the confusion. This is a common naming practice in Kra-Dai languages, using one animal to distinguish the size in another, thus in Tai mii khwaay 'buffalo bear U. thibetanus" from mii maa 'dog bear Helarctanos' or lɯŋ muu (pig) 'hog badger' from lɯŋ maa (dog) 'feret badger'… etc.

  36. Jonathan Smith said,

    November 26, 2019 @ 11:27 pm

    Re: Sui /mom6 va4 va4 vi4/
    Surely related to / (partially) borrowed from Chinese 娃娃魚?

    Wang Weihui 汪维辉 mentions first occurrence of 鯢魚 in the textual record in a paper I saw recently… I think the one about 說苑. So W. Han. There could be a connection to 能 without the phonological piece…

  37. James R. Chamberlain said,

    November 27, 2019 @ 7:52 am

    Probably not related to bear, but possibly to softshell, is Schafer (1967: 293) who calls attention to the Middle Chinese word for (giant?) salamander *t'ap 鰨 found in the 11th century rime dictionary Guangyun of Lu Fa-yen. I could not find an OC reconstruction.

RSS feed for comments on this post