Indo-European "cow" and Old Sinitic Reconstructions: awesome

« previous post | next post »

For at least four decades, I have suspected that IE gwou- ("cow") and Sinitic /*[ŋ]ʷə/ (< uvular? [Baxter-Sagart]) ("cow") are related.  Some new scientific research makes this surmise all the more believable.

More than three decades ago, Tsung-tung Chang already published on this idea in his "Indo-European Vocabulary in Old Chinese", Sino-Platonic Papers, 7 (January, 1988), p. 18 (of i, 56), citing Pokorny 482 gʷou and giving "gou" as his OS reconstruction.

Looks pretty simple and straightforward, doesn't it?  Well, it isn't simple at all

ŋ is a velar nasal, so it is basically a velar.  Most recent Old Sinitic reconstructions of 牛 that I know of begin with ŋ, and I think they mostly follow that with a labio-velar approximant glide, w.

This initial complex of consonantal sounds is realized in contemporary topolects as follows:

Mandarin n

Chengdu ny

Cantonese ng

Gan ny

Hakka ng

Jin n

Minbei n

Mindong ng

Minnan g

Teochew gh

Wu ny

Xiang ny

I'm especially struck by Minnan, which is usually thought to be one of the "oldest", if not the oldest, Sinitic topolects, since it has a flat out "g", seemingly no trace of a nasal.

So I'm wondering, dàodǐ 到底 ("in the end"), how nasally conditioned the Old Sinitic pronunciation of 牛 actually was.

Just a tinge / tad?

Chau Wu replied to my question thus:

I have a "healthy" reservation about the currently received Old Sinitic reconstruction of 牛.  This I suspect is based on the Middle Sinitic /ŋɨu/.  Where does the velar nasal ŋ come from?

When I compare Tw vernacular and literary lexicons, I often found that the literary pronunciation (which is supposedly close to the Tang court accent) has an additional nasal component. For examples (vern./lit.): 五 gō· / ngó·; 我 góa / ngó·; 賴 lōa / nāi.

Noting that the vernacular Tw is older than the literary, this nasalization seemed to occur in a later Middle Sinitic period.

If we compare the pronunciations for 牛: vern Tw gû, lit. Tw giû, and Sino-Japanese gu (Go-on), gyū (Kan-on), it suggests to me that the 牛 pronunciation had passed on to Japanese (both Go-on and Kan-on) before the nasalization innovation took place.

In my mind, the pronunciation of 牛 (vern. gû in particular) is probably one of the earliest loans from Indo-European.  PIE *gwou- (nom. sg. *gwōus) may have entered Asia, and with simplification of the initial consonant cluster to *g-, it could have become *gou, which eventually became vern. Tw gû (> Jpn Go-on gu).  With an infix of the glide -i-, the lit. giû resulted (> Jpn Kan-on gyū).

3 notes:

(1). Tw. voiced velar g- initial is in line with other older languages:  Skt. go- (Buck)/ gáu- (Mallory & Adams), Avestan gao-, OChSl. *govędo, Latv gouvs.

(2) The vowel change from -ou- (of PIE *gwou-) to -u in Tw gû is not too far-fetched. There is an example of PIE *gwōus turning into PNWGmc *kūz (> OE cū).

(3) I have shown the loss of the final -s/-z comparing the European and Taiwanese lexicons in my SPP paper (please see Pattern of Sound Correspondence #9).  Similarly, the Old English cū also shows the loss of the final -s.

So I humbly submit that the borrowing of PIE *gwōus (> Proto-Holó *gous > *gou) > vern. Tw gû is not too unreasonable.

South Coblin comments:

Common Min nasals underwent a form of denasalization in Southern Min in syllables that did *not* contain *nasalized* vowels. This happened to *ŋ-, which became modern [g], as in "cow" and to *m- which became modern [b], as in "horse" [be3]. Common Min *n- became Southern Min [l] under the same conditions. Southern Min g- and b- usually retain a certain amount of nasalization in the pronunciation of most speakers. It is generally very subtle, but sometimes more prominent, from area to area. For example, in JIng's language it is very slight. But a student of mine from Changhua, not far from Jing's home in Touliu, had much more noticeable nasalization. Denasalization occurred after the Common Min stage but is reconstructed in Proto-Southern Min by Bit Chee Kwok. It is clearly reflected in romanized Southern Min texts of the sixteenth century Spanish missionaries.

When a syllable contained a *nasalized* vowel, denasalization did *not* occur. And, conversely, in this environment, older *l-  became nasalized. So you get Proto-Southern Min *nuĩ4 [卵]" egg"and *muĩ2 [門] "door". Interestingly, this principle did not affect all of Southern Min dialects where initial l- is concerned. In Xiamen/ Quanzhou/Zhangzhou/Taiwan you get [laŋ2] (儂) "person" and [lam2] (南) "south", because these words do not contain nasalized vowels in S. Min. But Kwok reconstructs  Proto-Southern Min *naŋ2 and *nam2 respectively for these words. That's because S. Min dialects of the relevant areas retain Common Min *n-, which must therefore be restored for PSM as well. Those dialects don't have nasalized vowels here, but the denasalization did not affect *n- in them. The process just had not reached them yet.

Tsu-Lin Mei comments:

My typewriter does not have a special symbol for velar nasal. So I will write the velar nasal as ng-.

Gong Hwang-cherng 2002: 281 Old Chinese 牛*ngwjEg : Proto-Tibeto-Burman ngwa.  So we can be quite sure that the ST word for 'cow' has a velar nasal initial.  Your dialect data is fine.  But I would add 温州ngau, same as Cantonese.  As to the Minnan g-in the word for "cow", it was *ng, but denasalized to become the homorganic stop g-.  A parallel case is the Minnan 1st person singular 我gua.  我&吾clearly has the velar nasal initial in Old Chinese, and yet it became gua in Xiamen and Minnan.

One of the first things I learned from Jerry Norman, in 1967-68 at Princeton, is that Old Chinese 歌部 has *-ai < *-al final, and Min is the only Chinese dialect which still preserves many –ai final words.  You can read Jerry's account in Jerry Norman, Chinese, p. 212.  One of the examples he gives is 我, Fuzhou nguai.  I think you should have no trouble in connecting it with Minnan gua "I, me".  Ting Pang-hsin, "Derivation Time of Colloquial Min from Archaic Chinese" (1983) BIHP54.4:1-4 pointed out that Min is the only Chinese dialect which can distinguish Old Chinese 之部 *-Eg from 幽部*-Egw. He gives the example 牛which is 之部 vs. 九 which is 幽部。 Thus "cow"  Amoy gu, Fuzhou ngu, Ch'ao-chou 潮州 gu. "nine" AM kau, FZ kau, CZ kau. There are more examples in Ting 1983. "E" is my way of writing upside down "e".

Diana Shuheng Zhang comments:

The velar nasal ng is perfectly fine. You may remember how gn is pronounced like ng in Latin; so having a voiced velar g with the same place of articulation and voicedness is quite smooth and natural in Minnan's case. So we could stick with ng for the time being.

Etymology of English "cow":

From Middle English cou, cu, from Old English ("cow"), from Proto-Germanic *kūz ("cow"), from Proto-Indo-European *gʷṓws ("cow"). Cognate with Sanskrit गो (go), Ancient Greek βοῦς (boûs), Persian گاو‎ (gāv)), Proto-Slavic *govędo (Serbo-Croatian govedo, Russian говядина (govjadina) ("beef")), Scots coo ("cow"), North Frisian ko, ("cow"), West Frisian ko ("cow"), Dutch koe ("cow"), Low German Koh, Koo, Kau ("cow"), German Kuh ("cow"), Swedish ko ("cow"), Norwegian ku ("cow"), Icelandic kýr ("cow"), Latin bōs ("ox, bull, cow"), Armenian կով (kov, "cow").

The plural kine is from Middle English kyne, kyn, kuin, kiin, kien ("cows"), either a double plural of Middle English ky, kye ("cows"), equivalent to modern kye +‎ -en, or inherited from Old English cȳna ("cows', of cows"), genitive plural of ("cow").



"female of a bovine animal," especially the domestic ox, Middle English cu, qu, kowh, from Old English cu "cow," from Proto-Germanic *kwon (source also of Old Frisian ku, Middle Dutch coe, Dutch koe, Old High German kuo, German Kuh, Old Norse kyr, Danish, Swedish ko), earlier *kwom, from PIE root *gwou- "ox, bull, cow."



Middle English cou, from Old English cū; see gwou- in Indo-European roots.


Ox, bull, cow.

Nominative singular form *gwōu-s.

▲ Derivatives include cow1, beef, bugle1, butter.

1. cow1, kine; cowslip from Old English , , cȳe, cow, from Germanic *kōuz (> *kūz).


a. beef, bovine, bugle1 from Latin bōs (stem bov-), ox, bull, cow;

b. buccinator from Latin būcina, horn, trumpet, from *bou-kanā-, "bellower" (*-kanā-, singer; see kan-).


a. Boötes, boustrophedon, bucolic, bugloss, bulimia, bumelia, buprestid, butter, butyric from Greek bous, ox, bull, cow;

b. buffalo from Greek boubalos, buffalo, perhaps from bous;

c. boy perhaps from Old French buie, fetter, shackle, from Latin bōia, collar used to restrain a criminal (originally made from ox hide), from Greek boeiā, ox hide, from bous.

4. gayal; guar, Gurkha, kouprey, nilgai from Sanskrit gauḥ, go-, cow.

5. Suffixed form *gwou-no-. gunny from Pali goṇa-, ox.

6. Suffixed form *gwōu-ro-. gaur from Sanskrit gauraḥ, wild ox.

7. Zero-grade suffixed form *gww-ā-. hecatomb from Greek hekatombē, "sacrifice of a hundred oxen" (hekaton, hundred; see dekm̥)

[Pokorny g̒ͧou- 482.]

Examples of words with the root gʷou-: beef, Boötes, boustrophedon, bovine, boy, buccinator, bucolic, buffalo, bugle, bugloss, bulimia, bumelia, buprestid, butter, cow, cowslip, gaur, gayal, guar, gunny, Gurkha, hecatomb, kouprey, nilgai.


So, now what's the "new scientific research" that I referred to at the beginning of this post?

Thinking over all of these linguistic data, I serendipitously received the following book notice:

The Origin of Cattle in China from the Neolithic to the Early Bronze Age

Chong Yu


'Research on cattle domestication in China is in a period of great change and new revelation. Yu Chong's manuscript contributes important biometric data that complement the growing zooarchaeological and genetic evidence for overlapping exploitation of domestic and wild cattle populations during the Neolithic and Bronze Age. Dr. Yu's careful re-analysis of existing zooarchaeological collections is impressive and thorough. Her work shows how new insights can be gained by re-visiting existing zooarchaeological collections in China with new methods and new research questions.' Dr Katherine Brunson, Brown University

Cattle (Bos taurus), domesticated from the extinct aurochs (Bos primigenius), has been an important animal to many human societies since prehistoric times. Cattle provides not only meat for subsistence, but also hide, blood, dung, milk and traction that contribute to the organization of human beliefs, cultural attitudes and social complexity. This book provides the widest range of cattle bone biometrical information from the Early Neolithic period to the Early Bronze Age (10000 to 3600 years ago) and investigates the morphological variation of this animal from a biological point of view: the main indicator for tracing domestication. The results suggest that cattle in ancient China was imported* from the Near East around 4,300 years ago and made their first appearance in the Yellow River Valley. Once they had arrived in central China, these small-sized domesticated cattle soon spread and was exploited intensively from then on.

Chong Yu is an associate professor at Sun Yat-sen University. She is a zooarchaeologist researching various aspects of bio-cultural evolution including the origins and spread of animal domestication, palaeoeconomy, food processing, ritual, and cultural uses of animal bones. This includes regional and comparative studies on Neolithisation, early civilisations, hierarchy, and social complexity with a special interest in ancient Asia.

Keywords: Bronze Age,Cattle,China,Domestication,Neolithic,Zooarchaeology

RSP: £27 / €40.5 / US$54 | ISBN: 9781407316871 | Language: English | Sub Series: Archaeology of East Asia, 2 | 108 pages, Illustrated throughout in colour and black and white. 13 tables, 20 figures


And here is a relevant article on the genetics of the spread of domesticated cattle to East Asia:

Dawei Cai, Yang Sun, Zhuowei Tang, Songmei Hu, Wenying Li, Xingbo Zhao, Hai Xiang, Hui Zhou, "The origins of Chinese domestic cattle as revealed by ancient DNA analysis", Journal of Archaeological Science, 41 (january, 2014), 423-434.



    • Ancient DNA reveals that Chinese domestic cattle originated from the Near East.
    • In the early Bronze Age, cattle found in Northern China belong to taurine cattle.
    • Zebu cattle did not spread into the Central Plains until at least 1500 BC.
    • Ancient cattle made an important contribution to modern East Asian taurine cattle.
    • Haplogroup distribution has genetic continuity from the Bonze Age to present day.


Recent ancient DNA analyses have revealed the origins of European and Near Eastern domestic cattle. In East Asia, however, only a few ancient cattle remains from Korea have been studied. The origins of East Asian domestic cattle and the genetic contribution by ancient cattle to modern cattle are still unclear. To provide new insight into the early history of East Asian domestic cattle, we analyzed mitochondrial DNA (mtDNA) from 53 cattle remains, aged between 4500 and 2300 years, excavated from five archaeological sites in Northern China. All ancient Chinese cattle were identified as belonging to taurine cattle. On the one hand, the results support the previous hypothesis that taurine cattle spread into Northern China between 3000 and 2000 BC; on the other hand, the results suggest that zebu cattle did not spread into the Central Plains until at least 1500 BC. Three haplogroups T2, T3, and T4 were present in the ancient Chinese cattle, of which T3 was predominant (79.3%), while T2 and T4 were less common (9.4% and 11.3% respectively). Considering the geographic origin and estimated age of mtDNA haplogroups and the archaeological record of cattle remains in China, our results suggest that Chinese domestic cattle originated from the Near East and were already introduced into the Central Plains around 2500–1900 BC. Furthermore, phylogenetic network analysis indicates that the haplogroup distribution pattern of ancient Chinese cattle is similar to that of modern East Asian taurine cattle, suggesting a genetic continuity from the Bronze Age to present day. Lastly, population pairwise FST distance analysis and multidimensional scaling analysis also support close genetic relationship between ancient Chinese cattle and modern East Asian taurine cattle. All these results suggest that ancient Chinese cattle made an important contribution to the gene pool of modern East Asian taurine cattle.

Comment by Pita Kelekna, author of The Horse in Human History:

I am not particularly surprised that cattle raising reached central China 4,300 years ago. The western Afanasievo culture introduced metallurgy, wheeled vehicles, pottery, wheat, barley, and stock raising of horses, cattle, and ovicaprids to the eastern steppes and highlands between the Altai mountains and the Minusinsk Basin of the Yenisei River c 3600 BC.  So it took stock raising a thousand odd years to diffuse southeast to Yellow River.  Makes sense.

From Rod Campbell (Institute for the Study of the Ancient World, NYU), a specialist on ancient cattle remains in China:

It is pretty much consensus now that Bos taurus in China all derived from the west as part of a Eurasian Bronze Age exchange (along with sheep, goats, wheat, barley, and bronze). The dates might be a little conservative, but on the other hand, I think more and more evidence is pointing to around this period.

There are claims for earlier dates, but late 3rd millennium BCE is when the package is undeniably present.

A seminal paper for understanding the overall Eurasian context of this transmission is Andrew Sherratt's remarkable posthumous "The Trans-Eurasian Exchange: The Prehistory of Chinese Relations with the West," in Victor H. Mair, ed., Contact and Exchange in the Ancient World (Honolulu:  University of Hawaii Press, 2006), pp. 30-61.

To this Bronze Age Eurasian "package", I would add a specific type of burial for exalted personages under tumuli with a central chamber having sacrificial victims, chariots, horses, etc.  It is also possible that writing was part of this package.  We do not have enough specific data to demonstrate that writing was part of the Bronze Age Eurasian cultural "package", but it sprang forth full blown without indigenous precursors at just the right time to fit into the package.  And we do have some scattered, but very intriguing, bits of evidence.  See:

Victor H. Mair, "West Eurasian and North African Influences on the Origins of Chinese Writing," in Bernard H. K. Luk, ed., Contacts Between Cultures:  Eastern Asia:  Literature and Humanities, Volume 3 (Edwin Mellen Press, 1992), pp. 335-338.  I have a three hundred page monograph on this subject that I finished around three decades ago, but have never had a chance to publish it.

_________,"Old Sinitic *myag, Old Persian maguš, and English 'magician'," Early China, 15 (1990), 27–47.  On the origin of the Sinograph ☩ (U+2629) > 巫.

__________, "The Earliest Identifiable Written Chinese Character", Archaeology and Language: Indo-European Studies Presented to James P. Mallory, ed. Martin E. Huld, Karlene Jones-Bley, and Dean Miller. JIES Monograph Series No. 60 (Washington, D.C.: Institute for the Study of Man, 2012), 265–279.  On the northwest origin of wàn 卍 ("ten thousand").

While I'm at it, here's a curious coincidence:

gnu (also called the wildebeest)

It belongs to the family Bovidae, which includes antelopes, cattle, goats, sheep, and other even-toed horned ungulates.

Some sources claim the name "gnu" originates from the Khoikhoi name for these animals, t'gnu.  Others contend the name and its pronunciation in English go back to the word !nu: used for the black wildebeest by the San people


Taken from Nama,* likely onomatopoeic of the grunt-type noise of the gnu.


*The Khoekhoe language (/ˈkɔɪkɔɪ/; Khoekhoegowab), also known by the ethnic term Nama /ˈnɑːmə/[5] and formerly as Hottentot, is the most widespread of the non-Bantu languages of southern Africa that contain "click" sounds and have therefore been loosely classified as Khoisan.


And here's an odd fact about the ubiquitous usage of niú 牛 ("cow") (in the expression niú 牛B ("cow c*nt" — there are various unadulterated and bowdlerized versions for writing it) in contemporary China.  It vies with China's "national swear word" (guómà 国骂), tāmāde 他妈的 ("his mother's"), for most frequently uttered vulgarity, and is euphemistically translated into English as "awesome".  Moreover, both niú 牛B ("cow c*nt") and tāmāde 他妈的 ("his mother's") refer to the same part of the female anatomy, the former bovine and the latter human.


  1. It is important for linguists to take into account the findings of archeology and genetics.
  2. Whither / whence the nasal?

Selected readings

[Thanks to Karen Rubinson and Andrew Robinson]


  1. Chris Button said,

    January 17, 2020 @ 12:01 am

    For at least four decades, I have suspected that IE gwou- ("cow") and Sinitic /*[ŋ]ʷə/ (< uvular? [Baxter-Sagart]) ("cow") are related.

    I'm curious as to why they would suggest a possible uvular here. Surely they aren't suggesting an actual uvular nasal? That would be even more unlikely than their already unlikely voiced uvular stop (to which perhaps they are suggesting adding one of their nasal prefixes?)

    As for the OC onset *ŋʷ- of 牛 and its possible relationship with *PIE *gʷ-, I made the following suggestion elsewhere on Language Log regarding a possible reanalysis of the PIE voiced obstruent as a nasal in OC:

    …it is notoriously difficult to retain the voicing of a velar obstruent (as indeed to a lesser degree with any obstruent) such that pre-nasalisation would have been very possible here such that *gʷ > *ᵑgʷ > *ŋʷ. (Incidentally, this is why I am very suspicious of theories regarding a nasal voicing prefix in Old Chinese since the nasalisation of voiced obstruents being cited as evidence in loanwords from OC seems far more likely to result from the basic use of nasalisation as an articulatory mechanism to retain voicing)

  2. Christian Weisgerber said,

    January 17, 2020 @ 5:01 pm

    Given the timing and geography, I expect the IE cow word would have come from Indo-Iranian, so instead of PIE *gʷou- the form would have been *gau-.

  3. Jan Michalowski said,

    January 17, 2020 @ 5:19 pm

    With regards to the coincidence of !nu, what of the possibility that these would all simply be onomatopoeic words?

  4. Chris Button said,

    January 18, 2020 @ 9:19 am

    Given the timing and geography, I expect the IE cow word would have come from Indo-Iranian, so instead of PIE *gʷou- the form would have been *gau-.

    That's a good point. Although, I'd add that I don't think it affects the likelihood of the loan origin.

    The earliest underlying phonological reconstruction we have for 牛 is *ŋʷə̀ɣ. In terms of how that would have surfaced phonetically with some kind of rounded vowel and probably as an open syllable for many speakers (it would become underlying *ŋə̀w in the later OC phonological system with the labialization of the onset rounding the coda), there is really nothing that would have been more appropriate (providing we accept that the articulatory grounds noted in my earlier comment for ŋ- being used for g-).

    The other alternative *ŋʷàɣ would have been too distinct due to the lower -a- nucleus resisting the labial coloring more than its default -ə- counterpart. The evolution of phonological *ŋʷàɣ to *ŋàw (paralleling *ŋʷə̀ɣ becoming *ŋə̀w) might appear then to be closer to *gau- but that evolution seems too late for the time depth here.

    In short, the phonological likelihood that 牛 *ŋʷə̀ɣ came from an Indo-European source is strong. When combined with the other evidence adduced by Prof. Mair above, I see little reason not to treat it as a loan.

    It is pretty much consensus now that Bos taurus in China all derived from the west as part of a Eurasian Bronze Age exchange (along with sheep, goats, wheat, barley, and bronze).

    羊 *ɣàŋ "sheep" is an interesting one since, unlike an isolate like 牛 *ŋʷə̀ɣ, it has a word family of its own with words like 養 *ɣàŋʔ (cf. the relationship of Old Irish dinu "lamb" with Sanskrit dhāpáyate "suckle, nourish" and, via the same PIE root, "felix" which can then bring in words like 祥…)

  5. Matt said,

    January 19, 2020 @ 6:37 pm

    Sagart entered the follow commentary in "Dated language phylogenies shed light on the ancestry of Sino-Tibetan" –

    Cattle – The cognate set for "cattle" in 5.1 presents an interesting riddle: it has all the appearances of phonological regularity, implying knowledge of cattle by PST speakers, as proposed by Bradley (2016); yet currently the first archaeological occurrence of domesticated cattle in East Asia is in far western Yangshao or Majiayao area at 5400-4700 BP, too late for PST by our dates." (7200 B.P.)

    Supposing that non-Sinitic speakers first encountered domesticated cattle in that north-westerly region, and that their term for it was later transmitted to Sinitic through contact will not work either, because the Chinese loanword should then have the vowel *a, not *ə (while the correspondence between non-Sinitic *a and Old Chinese *ə is regular in cognate words). Zooarchaeology provides a solution: there is evidence that morphologically wild cattle was managed by humans in early Holocene northern China (Zhang et al. 2013). Presumably the cognate set in 5.1 is the PST term for early East Asian managed cattle; it was later applied by westward-expanding Sino-Tibetan groups to west Eurasian domesticated cattle that they encountered as they reached the western end of the loess plateau." DNA reveals evidence of abundant aurochs (Bos primigenius) in Neolithic Northeast ChinaAncient DNA analysis of 24 archaeological bovid remains recovered from large Neolithic (6300 BP to 5000 calBP) pit and ditch features at Houtaomuga, Northeast China, identified 23 of these samples as aurochs (Bos primigenius). These DNA-based identifications contrast with the morphological analysis of the remains, which identified them as Bison exiguous. The abundance of auroch remains at this site contradicts the general assumption that this species was not present in large numbers in Neolithic China. It also suggests archaeologists need to revise the notion that wild aurochs played an insignificant role in the lifeways of Neolithic peoples in China.

  6. AG said,

    January 20, 2020 @ 12:06 am

    I also wondered about Jan Michalowski's (to borrow a phrase from Friends' Joey) moo point.

  7. Chris Button said,

    January 20, 2020 @ 1:13 pm

    what of the possibility that these would all simply be onomatopoeic words?

    Even if the Proto-Indo-European form were onomatopoeic in origin (who knows?), that still wouldn't preclude it traveling as a loanword.

    …because the Chinese loanword should then have the vowel *a, not *ə (while the correspondence between non-Sinitic *a and Old Chinese *ə is regular in cognate words).

    To be clear, that refers to the relationship between Tibeto-Burman and Old Chinese and has no bearing on a possible Indo-European source. It also doesn't always occur–particularly in the environment here, such that Burmese နွား nwa² "cow" is unlikely to be related to 牛 *ŋʷə̀ɣ in spite of superficial similarities since the onset and rhyme are both off.

    Ting Pang-hsin, "Derivation Time of Colloquial Min from Archaic Chinese" (1983) BIHP54.4:1-4 pointed out that Min is the only Chinese dialect which can distinguish Old Chinese 之部 *-Eg from 幽部*-Egw.

    I haven't read Ting's article (I'd love a pdf if anyone has one), but its worth noting that this is probably more to do with the vacillation between *-ʷəɣ and *-əw (the ultimate source being *-ʷəɣ), with individual words settling in individual ways in Min (as opposed to a complete shift to *-əw in other Sinitic languages), than any mutually exclusive distinction. It comes back to the need to look at OC as a living language rather than some rigid algebraic formulation.

RSS feed for comments on this post