On the propinquity of Vietnamese and Sinitic

« previous post | next post »

Several comments to this post raised the issue of the closeness of Vietnamese and Cantonese:

"Cantonese is not the mother tongue of Hong Kongers" (5/4/18)

Here:

I am told (I don't speak them) that Vietnamese and Cantonese are similar enough that they might be called dialects of the same language.

Here:

I'd say rather that Chinese loans in Vietnamese are much closer to Cantonese than Mandarin (which makes sense given geography). Once going over something with my Vietnamese teacher a colleague who had long ago learned some Cantonese said he could understand a lot of it. But shared vocabulary and family origin are of course very different things.

Here:

Just to make sure it doesn't remain unsaid: Vietnamese has several layers of Sinitic loanwords, amounting to even more – and even more basic – loans than Japanese has, but its basic vocabulary is very different. Rather than a Sino-Tibetan language, it's an Austroasiatic language like Khmer, Mon, and the Munda languages of India.

And here:

According to Wikipedia, Cantonese features substrate influence from Tai-Kadai, but this is a different language family from Austroasiatic.

I suspect the claim that Cantonese is related to Vietnamese is part of a "Southern Yue" identity thing. The kingdom of Nanyue was destroyed by the "peaceful" Han dynasty and became thoroughly Sinicised.

Much later, Vietnam wanted to call itself Nanyue, but given that Nanyue had covered parts of Guangxi and Guangdong, the Qing emperor refused and called it Yuenan (Vietnam) instead.

Response

From Steve O'Harrow:

Of course, as everyone knows (or should do), Vietnamese has a ton of words of Sinitic origin and they are in layers of borrowings across the centuries. Some would appear to have come in no later than about 600 CE and some as late as yesterday. But Vietnamese is basically unrelated to Chinese as far as anyone can see – in the era of the troglodytes, perhaps many languages were born out of the speech of one group that migrated out of Africa toward the northeast, but my guess is we will never quite know about how that evolved.

* * *

For what it's worth, here's a definition I like, "Two or more ways of speaking, if they are at least 51% mutually intelligible, are "dialects" of a common "language."  However, in many cases, politics wins out. Sometimes a "dialect" becomes a" language" if it has a flag and an army (pace Max Weinreich). Yugoslavia (or what it became) is an example of the latter proposition. You might be tempted to say the same in some cases in Scandinavia. Folks on one side of the border between China & Viet Nam speak varieties of "Zhuang"; folks on the other side speak "Tay-Nùng." They can understand each other, but don't tell that to the authorities in Bei-jing or Hà Nội. Or, at least, don't say it loudly in public, OK?

In the case of the Chinese "language," I would say that we might more properly speak of the Chinese "languages," since mutual intelligibility presents considerable problems to most speakers of, say, 廣州話 (Cantonese) and 國語 (Modern Standard Mandarin). But for nationalistic political reasons, the Chinese government (read "governments") finds the concept totally anathema. If you are the people in charge, those who "look south," in the national capital, your constant worry is the centrifugal nature of the Chinese polity. So it's not surprising that many Chinese with a strong sense of national pride are loath to admit the idea that Chinese is a language family, not a single national language with a bunch of dialects – it's politics, not linguistics that rules in most places.

Readings

"Similarities in Vietnamese and Cantonese can be traced back to antiquity", Wee Kek Koon, SCMP (5/2/

"Chinese Loanwords in Vietnamese", Encyclopedia of Chinese Language and Linguistics" (Brill)

Vietnamese belongs to the Mon-Khmer language family, having numerous basic Mon-Khmer etyma (Shorto 2006), but it is unlike other Mon-Khmer languages, which typically do not have complex tone systems, do have numerous polysyllabic words, and utilize morphology that includes prefixes and infixes. Instead, Vietnamese, like Thai and Hmong, belongs to the “inner Sinosphere” (Chinese loanwords in Southeast Asia).

Sino-Vietnamese vocabulary:

"…words and morphemes of the Vietnamese language borrowed from Chinese. They comprise about a third of the Vietnamese lexicon, and may account for as much as 60% of the vocabulary used in formal texts."

Vietnamese language:

Vietnamese was identified more than 150 years ago as part of the Mon–Khmer branch of the Austroasiatic language family (a family that also includes Khmer, spoken in Cambodia, as well as various tribal and regional languages, such as the Munda and Khasi languages spoken in eastern India, and others in southern China). Later, Muong was found to be more closely related to Vietnamese than other Mon–Khmer languages, and a Viet–Muong subgrouping was established, also including Thavung, Chut, Cuoi, etc. The term "Vietic" was proposed by Hayes (1992), who proposed to redefine Viet–Muong as referring to a subbranch of Vietic containing only Vietnamese and Muong. The term "Vietic" is used, among others, by Gérard Diffloth, with a slightly different proposal on subclassification, within which the term "Viet–Muong" refers to a lower subgrouping (within an eastern Vietic branch) consisting of Vietnamese dialects, Muong dialects, and Nguồn (of Quảng Bình Province).

"Why does Vietnamese language seem to be so similar to Mandarin Chinese", Linguistics Stack Exchange

"The Origin and Nature of the Vietnamese Language"

"Linguistic Research on the Origins of the Vietnamese Language: An Overview", Journal of Vietnamese Studies 1(1-2):104-130 · February 2006



9 Comments »

  1. Bathrobe said,

    May 11, 2018 @ 5:32 pm

    John Phan (Columbia Univ.) has put forward the idea that Vietnamese was originally a locally-spoken dialect of Middle Chinese, as opposed to a local language that was heavily influenced by Chinese. I suspect that it isn't the sort of thing that Vietnamese nationalists would embrace.

    http://chl-old.anu.edu.au/publications/csds/csds2010/03-1_Phan_2010.pdf

  2. Eidolon said,

    May 11, 2018 @ 6:25 pm

    It's occurred to me that single descent models of language organization, in which all languages can only be genetically related to one parent, while all other sources of influence are considered loans or substrates, may not be as hegemonic as they appear in current linguistics scholarship.

  3. Jenny Chu said,

    May 12, 2018 @ 12:17 am

    I always considered the role of Cantonese loan words in Vietnamese to be more like Latin/Greek/French loan words in English: use 'em for the fancy vocabulary, but not for regular stuff.

    Hence: "a long time ago" (ngày xưa) is not a loanword, but "historical" (lịch xư) is.

    Well, in general, loanwords in Vietnamese are good fun because you can watch history happening through the lens of which language the words are borrowed from:

    University / đại học –> from Cantonese, quite some time ago.
    Screwdriver/tu vít and brake/phanh –> from French, 19th/20th century.
    Modernization / hiện đại hoá –> from Mandarin; definitely a 1950s invention.
    Kulak / cu-lắc –> three guesses where & when that's from
    Photocopy/phô tô cóp py, microchip/chíp vi xử lý –> English, it would seem, 1980s-1990s

  4. Chas Belov said,

    May 12, 2018 @ 2:39 pm

    I noticed đặc biệt and, according to Wiktionary, its from Chinese 特別. In Cantonese, it's pronounced approximately "duck beat" which is how I recognized that the two are related (in its use on menus). But I see it's not specifically a Cantonese word.

  5. Chas Belov said,

    May 12, 2018 @ 2:49 pm

    That said, I'd guess the borrowing is from Cantonese or its antecedents based on the pronunciation.

  6. Bathrobe said,

    May 12, 2018 @ 4:43 pm

    The source of Vietnamese pronunciations of Sinitic morphemes in the language does not appear to be settled. John Phan, in the introduction to his thesis, remarks as follows:

    "Mantaro Hashimoto (1978) proposed the possibility of a "southern koine", spoken during the Tang dynasty, as the source for most Sino-Vietnamese words, and provided a brief but suggestive comparison with various modern Sinitic languages spoken in the vicinity of the Vietnamese border. Similarly, Mark Miyake (2003) would later propose an affiliation with Cantonese. However, as will be discussed in Chapter 1, the details of Hashimoto's scenario are problematic, and Miyake's proposal of a Cantonese or proto-Cantonese source for Sino-Vietnamese lexica is conclusively refuted by the data. Thus a unified account for the history of Sino-Vietic contact has yet to emerge in modern scholarship."

    Phan argues that the bulk of Sinitic loanwords in Vietnamese resulted from bilingual contact between a form of Sinitic native to the region of northern Vietnam, which he terms 'Annamese Middle Chinese' (AMC), and contemporary forms of Vietic language. He claims that the northern river plains of Vietnam were home to a "rooted and thriving" community of AMC speakers for most of the first millennium. In this, Vietnamese is claimed to be quite different from Korean and Japanese, where Sinitic vocabulary was largely acquired through reading and writing practices rather than direct contact with native speakers of Chinese. In these languages, the pronunciation of Chinese characters was acquired from "limited bilingual contact of a few specialists, mostly Buddhist clergy".

  7. Michael Watts said,

    May 13, 2018 @ 4:41 am

    Re: "single descent models of language organization", wikipedia suggests this model is formally known as the tree model, in contrast to the wave model. It seems pretty clear to me that both models have value — the tree model is certainly the right way to explain why the languages spoken by Americans and Australians have so much in common; the wave model is usually invoked to explain why many Asian languages have so much in common despite apparently not being related by descent at all.

    The tree model already requires language change; all you really need for the wave model is the idea that languages spoken in proximity to each other may experience the same change simultaneously, possibly because they are being spoken by the same people.

  8. David Marjanović said,

    May 14, 2018 @ 5:15 pm

    John Phan (Columbia Univ.) has put forward the idea that Vietnamese was originally a locally-spoken dialect of Middle Chinese, as opposed to a local language that was heavily influenced by Chinese.

    I recommend the paper, but what he's saying is it's a Vietic language with a big fat AMC substrate which formed when the empire withdrew and the local AMC speakers shifted to Vietic. "Mường" turns out to be not a branch on the tree, but a cover term for all Vietic varieties that lack this AMC substrate.

    It's occurred to me that single descent models of language organization, in which all languages can only be genetically related to one parent, while all other sources of influence are considered loans or substrates, may not be as hegemonic as they appear in current linguistics scholarship.

    That has occurred to many people, but – apart, probably, from Michif and very few other cases – it continues to yield more detailed insights than all alternatives.

  9. Bathrobe said,

    May 14, 2018 @ 6:56 pm

    what he's saying is it's a Vietic language with a big fat AMC substrate

    Yes, I misrepresented that in my initial comment. As a substrate, would it be reasonable to liken the role of AMC in Vietnamese to that of Norman French as a substrate for Middle English?

RSS feed for comments on this post

Leave a Comment