In Search Of the Origins Of Vietnamese: The Unfinished Work

Toward a Sino‑Tibetan Reframing Of Vietnamese History

by dchph

While the trajectory of Vietnamese within the Sino‑Tibetan framework is clear, the research remains incomplete: mapping across all divisions is ongoing, Shafer's Sino-Tibetan data needs refinement, and the work points toward future directions in orthography, lexicons, and cultural‑historical synthesis.

I) The incomplete mapping

• Comprehensive alignment of Vietnamese across all six Sino‑Tibetan divisions (Sinitic, Daic, Bodic, Burmic, Baric, Karenic) is still in progress.

• Shafer’s framework provides a foundation, but requires refinement with modern phonological and etymological reconstructions.

• Emphasize polysyllabicity and tonal correspondences as diagnostic features.

Etymologically, it is reasonable to speculate that differences in sound change among basic words, much like the numerals, may have resulted from phonemic transitions traceable either to Sino‑Tibetan sources or even to a form of proto‑Chinese at a later stage. For instance, 臥 wò (SV "ngoạ") developed into both VS "ngủ" (sleep) and "nằm" (lie down). Each of these corresponds, respectively, to Old Bodish *snyid and Classical Tibetan Groma nyiʾ‑, rather than to the later Chinese variants 睡 shuì (sleep) and 躺 tăng (lie down). The latter forms, marked by radicals such as 目 (eye) and 身 (body), reflect later semantic specialization. Phonemically, the shifts are notable: /sh‑/ alternates with /th‑/, /sh‑/ with Vietnamese /ng‑/, and /th‑/ with /n‑/. All of these can be postulated as divergent reflexes of a common ancestral root.

In 1912, Maspero (BEFEO XII) proposed a reclassification of Vietnamese, grouping it with Thai (T'ai, Tai, Dai, Tay) and other Daic languages within the Sino‑Tibetan family. His reasoning rested in part on the tonal system of Vietnamese, which he considered structurally parallel to Chinese models.

In 1953, however, Haudricourt challenged this view in La Place du Vietnamien dans les Langues Austro‑Asiatiques (BSLP 49, pp. 122, 128). He argued that the Daic‑Vietnamese cognates identified in Thai and Laotian were in fact loanwords from Mon‑Khmer. In his subsequent work on the origin of Vietnamese (1954), he demonstrated that tonal development in Vietnamese arose through the evolution of word‑final consonants between the 6th and 12th centuries. On this basis, he concluded that Vietnamese belongs not to the Daic division of Sino‑Tibetan but rather to the Mon‑Khmer branch of the larger Austroasiatic family.

However, Maspero's viewpoint was upheld by Forrest (1958) who inserted that

“before the Chinese conquest, Annam [Vietnam] and Kwangtung [Guangdong or Canton] were long under one rule; but everything points to that rule having been T'ai [Dai] rather than Mon-Khmer.[...] When first recorded by European Missionaries in the seventeenth century, Annamese [Vietnamese] still had compound initial groups of consonants in cases where they are now reduced to simple sounds. The phonetic history in this respect is parallel to that of T'ai and Chinese, and this fact, so far as it goes, tells in favour of a T'ai basis for the language rather than a Mon-Khmer” (p.102).

Until the late twentieth century, a few linguists such as Peng Chu'nan (1984) still maintained that Vietnamese belonged to the Sino‑Tibetan family. In the Chinese-Vietnamese context, Pulleyblank (1984) observed that "Vietnamese is typologically closer to Chinese than are either Japanese or Korean and, in many ways, even Tibetan, in spite of the fact that Chinese and Tibetan are genetically related while Chinese and Vietnamese are not (unless the relationship is an exceedingly remote one)" (p. 91). In the same spirit of uncertainty, Roland J‑L Breton and Harold F. Schiffman, in their Geolinguistics (1991), chose not to classify Vietnamese within the Austroasiatic family.

Even so, the prevailing consensus for more than a century has been that Vietnamese belongs to the Mon‑Khmer branch of Austroasiatic. This view continues to dominate historical linguistics. Yet theories evolve as new evidence emerges. Given the overwhelming presence of Sinitic elements that permeate every layer of Vietnamese, the author of this study proposes a new designation: a Sinitic‑Yue (SY) subgroup. Here, "Sinitic" is understood on par with the Sinitic division of Sino‑Tibetan, though the latter encompasses all Chinese dialects, or lects, the term is used when it is not possible or desirable to decide whether something is a distinct language or only a dialect of a language, to be exact, including Cantonese and Minnan (Hokkien). These dialects themselves have accumulated more than two millennia of Han and Tang Chinese layered upon Yue substrata, and thus could historically be situated within the same Sinitic‑Yue framework. Mapping Sino‑Tibetan regions against the history of Vietnamese territorial expansion reveals that Yue linguistic pockets once extended across southern China – Fujian, Guangdong, Hunan, Guangxi, Guizhou, Yunnan – and even overlapped with Jiangxi, Henan, and Hebei, where the ancient state of Chu flourished centuries before the Common Era.

In this light, the differences between Vietnamese and other Sinitic languages resemble those between Cantonese and Hokkien and the five major northern Chinese lects. This is not to suggest that Vietnamese was ever simply a Chinese dialect before its political separation from the Middle Kingdom more than 1,100 years ago. Rather, while Cantonese and Fukienese "lects", evolved under continuous Han rule for over 2,200 years, Vietnamese developed along a different trajectory to become another "lect". Any linguist familiar with Cantonese or Hokkien can still detect Yue lexical and syntactic residues distinct from the mainstream northern dialects that spread from Shaanxi and Shanxi eastward to Shandong and southward to Jiangxi, Sichuan, Guizhou, and Yunnan.

II) Vietnamese as central witness

• Vietnamese emerges not as marginal but as a central witness to Sino‑Tibetan history.

• Cardinal numbers, basic vocabulary, and cultural lexicon show systematic correspondences.

• Contrast with Mon‑Khmer parallels: scattered, collateral, and often misclassified.

One of the strongest reasons for rejecting the classification of Vietnamese within Austroasiatic Mon‑Khmer is that many basic words for nature, kinship, and daily life are cognate with Chinese. Proto‑Vietic may once have shared certain forms with Mon‑Khmer, but during a millennium as a Chinese prefecture and another millennium after independence, Muong subdialects likely served as conduits for Chinese‑influenced Annamese loanwords into Mon‑Khmer. This would explain why a thin substratum of basic words, such as numerals 1 to 5, remains problematic for resolving genetic affiliation.

Moreover, many Sinitic elements in Vietnamese are interwoven with Yue residues. Virtually all Vietnamese grammatical function words derive from Chinese "xūcí" (虛辭), such as prepositions, conjunctions, and the like, each spawning multiple Sinitic‑Vietnamese forms: 和 hé → VS "và" (and), 輿 yú → VS "với" (with), 雖 suī → SV "tuy" > VS "dù" ~ "dẫu" (though), 然 rán → VS "vậy" (then). At the same time, many fundamental words are also cognate with Chinese: "mắt" 目 mù (eye), "thấy" 視 shì (see). By contrast, Cantonese and Hokkien preserve native residues such as /pin5dow2/ (where), /fan1gao1/ (sleep) and /kẽ/ (child), /bat7/ (know), /soã/ (mango), respectively. Similarly, Vietnamese shares both Sinitic and native strata. While 'xūcí words' are indispensable for modern syntax, Vietnamese distinguishes itself from Chinese through features such as [noun + adjective] order that is still traceable in southern Hokkien dialects like Hainanese, Fukienese, Amoy, and Cantonese (e.g., 雞公 /kei5kong1/ "rooster", 大老 /dai2lou4/ "big brother"), as opposed to Mandarin 公雞 or 老大. This structural evidence further supports the proposal of a Sinitic‑Yue grouping. 

Even words long listed as Austroasiatic or Mon‑Khmer often resist such classification. For example, "bò" (ox, cow) contrasts with "trâu" 牛 níu (water buffalo) and clearly reflects Sino‑Tibetan origin. Other cases are ambiguous, though, for instance, while Vietnamese "bú" (suck) aligns with Chinese 哺 bǔ,  "sữa" (milk) does not plausibly connect with Chinese 乳 rǔ or Mon‑Khmer /tukdaohko/, but may echo Indonesian "susu". 

Such examples of the above and the following suggest semantic shifts, homonymic substitutions, and lost etyma. They are familiar to all, yet the less obvious Sinitic‑Vietnamese etyma – those rarely cited in wordlists – deserve closer attention, as they may hold the key to deeper historical connections.

• "đực" (male): 特 tè (SV đặc),
• "mái" (female): 母 mǔ (SV mẫu),
• "quà" (presents): 饋 kuì (SV quý),
• "gỏi" (minced meat with salad): 膾 kuài (SV khoái),
• "lạc" (in place of 'đậuphụng' 花生 huāshēng 'peanut' [Hai. /wun1dow2/]) 落 luò (SV lạc) [ ~ VS 'đậulạc' #落豆 luòdòu. Also, an etymon of VS 'rơi', 'rớt', 'rụng' <~ M 落 luò, lè, luō, lào, là < MC lak < OC *gra:g as in modern Chinese for both 落花 'fallen flowers' and 落花生 'earth-nut' (Arachis hypogaea) (attested in Qing classics) which is obviously related to Vietnamese 'lạc' (earth-nut) although the direction of borrowing is not quite clear. Cf. perhaps also (as a more archaic loan) Viet. rắc 'to sprinkle, to dredge, to sow' ('to let fall') ],
• "bánhdày" ('bánhgiầy') # 糍粑 cíbā ('glutinous rice cake'), "bánhbao" # '包餅 bāobǐng – modern M 包子 bāo​zi, 'meat bun') – and "bánhchưng" # 蒸餅 zhēngbǐng – modern M 粽子 zòngzi – (being a different kind of glutinous rice cakes) as they have been previously believed to be indigenous glosses. This is an important point since because, in the cultural context, these words are closely related to the mystic legends of 18 ancestral Hồngbàng Kings of the Vietnamese people.
• "dưahấu" (watermelon) [ M 塊瓜 kuàiguā (khốiqua) | @# M 塊瓜 kuàiguā \ @ 塊 kuài ~ 'hấu'. "Hấu" is a monosyllabic syllable not to be used alone, it must go with 瓜 guā as 'dưa' (melon) to make up the dissyllabic word "dưahấu" (M 瓜 guā < MC kwaɨ < OC *kʷra ) || Note: as apposed to modern M 西瓜 xīguā, which becomes another word in Vietnamese as 'dưatây' (literally 'western melon') to mean a different kind of non-native melons. ],

and so on so forth. (1)

For all the reasons outlined above, together with the undeniable closeness of Vietnamese to the Chinese languages, applicable across all dialects, the Vietnamese language should be placed within the newly designated Sinitic–Yue (SY) branch of the Sino‑Tibetan family. This family, as currently classified, encompasses nearly 400 languages and dialects.

As for Tibetan and Chinese, tradition holds that in prehistoric times the Tibetans and the proto‑Chinese were originally one people, speaking a common proto‑Tibetan tongue before diverging into separate languages. The formation of Archaic Chinese (ArC) – the ancestor of Old Chinese spoken in the pre‑Qin and Han eras – was the result of fusion between Tibetan and Taic‑Yue languages spoken by native populations in what is now Henan Province. Archaic Chinese thus emerged under the influence of these aboriginal tongues, including early Yue speech. Norman (1988, p. 17) speculated that "the fact that only a relatively few Chinese words have been shown to be Sino‑Tibetan may indicate that a considerable proportion of the Chinese lexicon is of foreign origin [...], languages which have since become extinct."

Those "foreign origin" languages Norman referred to may well have been of proto‑Taic stock, the same root that gave rise to Daic‑Kaida and other Yue languages, including ancestral Vietic language, Cantonese, and Hokkien and perhaps even some Austroasiatic Mon‑Khmer tongues. Speakers of these languages lived under Han rule, and the later Chinese dialects, after centuries of Sinicizing immersion, became part of the Sinitic Division, now represented by the seven major Chinese dialect groups.

The ancient Yue stratum in Vietnamese also shared basic words with Chinese, long before the Han ruled Annam from 111 B.C. until 939 A.D., when it broke away from the Southern Han State (南漢 帝國 NamHán). By that time, although Baihua (白話) in NamHán and early Annamese had already diverged, they still retained a portion of common stock of basic vocabulary and layered official Mandarin. This shared layer is visible in Buddhist texts of the Phậtthuyết tradition (short for Phậtthuyết Đạibáo Phụmẫu Ântrọng Kinh 佛說 大報 父母 恩重 經), a fifteenth‑century Nôm scripture teaching filial piety. It was explicitly written for the common people, as the preface states: "Cho người thiểnhọc nghĩ xem nghĩ nhuần" ('For the uneducated, easy to read and absorb.'). In that era, Buddhism was regarded as the national religion.

Most of the basic words in Phậtthuyết cognate with Chinese etyma are still in use today, but absent from Mon‑Khmer. What readers often encounter are Mon‑Khmer words frequently cited as "fundamentals", yet many of these may also be cognate with Sino‑Tibetan or Chinese forms. 

Below are samples of such basic words, including grammatical function words, that are rarely found in Austroasiatic Mon‑Khmer, for which specialists have yet to identify cognates. Readers may treat them as a worksheet for further exploration.

Comparable vocabulary also appears in other Vietnamese literary works: the fifteenth‑century Hồngđức Quốcâm Thitập 宏德 國音 詩集 by King Lê Thánh Tôn, the Quốcâm Thitập 國音 詩集 by Nguyễn Trãi, the seventeenth‑century Đoạntrường Tânthanh 斷腸 新聲 (or Truyện Kiều) by Nguyễn Du, and the lụcbát hexa‑octosyllabic poems of Nguyễn Bỉnh Khiêm, including Lâmtuyền Kỳngộ 林泉 奇遇. (See Nguyễn Ngọc San, 1993, pp. 95–104, 138–142.)

Table 3 - Some ancient Nôm basic vocabularies and their Chinese correspondences

English	Ancient Vietnamese	Mon- Khmer	Sino- Vietnamese	Vietnamese	Mandarin	Example
already	đà	(?)	dĩ	đà, đã	已 yǐ	"Khách đà lênngựa người còn ngoáitheo." (Kiều)
ask	han	(?)	vấn	hỏi	問 wèn	"Vội han ditrú nơinao." (Kiều)
attend to	chực	(?)	chờ	tý	伺 sì	"Mộtmình chực mộ nhàđường." (Phậtthuyết)
brothers	anhtam	(?)	huynhđệ	anhem	兄弟 xiōngdì	"Tin ngườixa làm anhtam" (Phậtthuyết)
classifier (time period of the day)	ban	(?)	phạn	bữa, buổi	飯 fàn	"Cáo kêo eoéo ban trờitối." (Hồngđức)
don't	mựa	(?)	mạc	chớ	莫 mò	"Nghĩa, nhân, lễ, trị mựa cho khuây." (Quốcâm Thitập)
dumb	nghê	(?)	ngu	ngô	愚 yú	"Nghêdại chẳng hay cóc." (Phậtthuyết)
easy	nghĩ	(?)	dĩ	dễ	易 yì	"Cho người thiểnhọc nghĩ xem nghĩ nhuần." (Phậtthuyết)
endure	khắng	(?)	khẳn	khẳng	肯 kěn	"Mảy chút trầnai chi khắng luỵ." (Lâmtuyền Kỳngộ)
even	liễn	(?)	liên	lẫn	連 lián	"Đạo làm con liễn đạo làm tôi." (Quốcâm Thitập)
evening	hôm	(?)	vãn	hôm	晚 wăn	"Hômdao lòng chẳng với." (Phậtthuyết)
hear	mắng	(?)	văn	nghe	聞 wén	"Mắngtin xiết nỗi ngạingùng." (Kiều)
how	há	(?)	hà	há	何 hé	"Lộc nặng há quên hơn chúa nặng." (Nguyễn Bỉnh Khiêm)
if	nhược	(?)	nhược	nếu	若 nuò	"Nhược có người nào ngheđược." (Phậtthuyết)
large	cả	(?)	đại	to	大 dà	"cảthẹn", "bểcả"
love	dấu	(?)	ái	yêu	愛 ài	"Ángná lòng thực dấu." (Phậtthuyết)
mix, and	hoà	(?)	hoà	và	和 hé	"Ắtlà khôn hết cả hoà hai." (Nguyễn Bỉnh Khiêm)
morning	dao	(?)	hiểu	sớm	曉 xiáo	"Hômdao đủ bữa bátcơm." (Quốcâm Thitập)
mother	ná	(?)	nương	nạ	娘 niáng	"Chẳng biếtơn áng ná." (Phậtthuyết)
no	nọ	(?)	nã	nào	哪 ná	"Dânghương nọ kẻ nện chàykình." (Hồngđức)
only	chỉn	(?)	tận	chỉ	儘 jǐn	"Chỉn thực quannhân dạ sắtvàng." (Lâm Tuyền Kỳngộ)
private	tây	(?)	tư	tư	私 sī	"Mựa nghe sàmnịnh có lòng tây." (Quốcâm Thitập)
purchase	chác	(?)	thụ	tậu	售 shòu	"Chác được của rẻ." (Phậtthuyết)
quiet	ắng	(?)	yên	im	安 ān	"Từ chưng ấy ắng mất." (Phậtthuyết)
respect	vì	(?)	uý	nể	畏 wèi	"Nhà ngặt bằng ta ai kẻ vỉ." (Quốcâm Thitập)
s/he	nghĩ	(?)	y	nó	伊 yī	"Gia tư nghĩ cũng thườngthường bậctrung." (Kiều)
solely	bui	(?)	duy	duy	維 wéi	"Bui có một niền chăng nỡ trễ." (Quốcâm Thitập)
stupid	dại	(?)	ngai	ngốc	呆 dài	"Làm những kế dại." (Phậtthuyết)
that	của	(?)	sở	sự	所 suǒ	"Vợcon cầmbắt mỗi của nươngđòi." (Phậtthuyết)
then	chưng	(?)	đang	đang	當 dāng	"Già này chưng thuở mặttrời tà." (Lâmtuyền Kỳngộ)
there	nào	(?)	na	đó	那 nà	"Nào hồn ômã lạcloài đâu." (Hồngđức)
though	là	(?)	la	là	啦 là	"Đêm chia nửa, khéo hay là." (Hồngđức)
to be	thì	(?)	thị	thị	是 shì	"Vô sự thì hơn kẻo phải lo." (Nguyên Bỉnh Khiêm)
what	sở	(?)	xá	gì	啥 shé	"Trong thếgian chẳng sở nào tày." (Phậtthuyết)
when	thuở	(?)	thời	thuở	時 shí	"Gối mác nằm sương thuở Tấn Tần." (Nguyễn Bỉnh Khiêm)
wish	mòng	(?)	vọng	mong	望 wàng	"Cảnhcũ non quê nhặt chốcmòng." (Quốcâm Thitập)
worry	âu	(?)	ưu	âu	憂 yōu	"Chẳng âu ngặt chẳng âu già." (Quốcâm Thitập)
would like	sính	(?)	tưởng	trông	想 xiăng	"Sính làm con yênlành." (Phậtthuyết)
wrong	thác	(?)	thác	sai	錯 cuò	"Sámhối tội thác." (Phậtthuyết)

Taking into account the historical circumstances of China, the emergence of what later came to be called the "Han" can be understood as the fusion of ancient proto‑Tibetan and proto‑Taic peoples, the original indigenes of southern China (see Lacouperie 1966 [1887]). Their descendants spread across the mainland and eventually became speakers of the various Chinese "lects". This deduction rests on the fact that Chinese contains a large number of loanwords from the Southern Yue (南越 NamViệt) aboriginals, whose languages were spoken in the seven major states of the pre‑Qin era. Many of these forms were later collected in the Kangxi Dictionary and designated as 方言 (dialects).

In reality, Chinese is not a single speech but a constellation of lects, their dialects and sub‑dialects. Their differences lie chiefly in having (1) a whole unique set of fundamentals, and (2) pronunciation of the same written Chinese characters, derived from the common base of Mandarin (now called 普通話 putonghua), though usage also varies. Word choice and grammar in Wu, Kegan (客贛), Minnan, and Cantonese are strikingly distinct. For example:

• A Beijinger says, "你 先 說 吧!" ("You speak first then!"), while a Cantonese speaker in Guangzhou says, "你 講 先 喇!" ("You talk first then!").

• Beijing: "我 給 你 這個." ("I give you this."); Cantonese: "我 卑 呢個 你." ("I have this one you.").

• Beijing: "我 說, 你 聽得懂, 你 說, 我 聽不懂." ("I speak, you understand; you speak, I do not understand."); Hokkien: "咱 講 汝 聽 有,  汝 講 咱 聽無." ("I talk, you hear yes; you talk, I hear no.").

• Beijing: "給 我 點 水." ("Give me a little water."); Shanghainese: "撥 點 水 我." ("Pass a little water me.").

• Beijing: "你 有沒有 錢?" ("Do you have money?"); Shaoxing Wu: "你 啊 有 銅鈿?" ("You have copper coin?").

Except for the four major dialects, namely, Cantonese, Minnan, Kegan, and Wu, the northern sub‑dialects of Shaanxi, Shanxi, Shandong, Sichuan, and elsewhere appear largely as variations of one another (Zhou Zumo 周祖謨 in Wang Li, et al., 1956, Hanzu de Gongtongyu he Piaozhunyin, p. 63). It is not difficult to see that the southern dialects were originally Yue languages, now grouped under the Sinitic branch.

By the same token, the lexical cognacy between Chinese and Vietnamese suggests that the formation of Vietnamese repeated the same developmental process undergone by southern Chinese lects and their dialects. Like Cantonese, Minnan, Kegan, and Wu, Vietnamese absorbed a vast amount of northern Mandarin vocabulary layered upon its aboriginal Yue substratum.

After the Han conquest of the NamViệt Kingdom in 111 B.C., the Vietic language emerged as an admixture of the speech of Han colonists and Yue inhabitants, possibly including early Mon‑Khmer speakers from the southwest (today’s Lower Laos). This helps explain the presence of Mon‑Khmer words dating back as early as two millennia B.C. These groups descended from a mixed stock of Yue natives in northern Vietnam before Han infantry advanced further south into Annam, a process that continued for centuries. The long‑marching Han soldiers themselves were descendants of populations from across southern China, including subjects of the Chu State (楚國) and NamViệt. In Giaochâu (交州) prefecture, those who had remained in metropolitan centers and cooperated with the Han gave rise to the early "Kinh" people. Han colonists, soldiers, and displaced migrants intermarried with local women and settled permanently; their descendants would later rule the independent state of the Southern Viets, or Việtnam, another way of writing 南越 NánYuè.

The historical development of Vietnam thus helps explain the deep linguistic commonalities shared by Chinese and Vietnamese, both rooted in the ancient substrata of what Lacouperie (1887) called "China before the Chinese". Archaeological findings by Zhang Zengqi (1990) in Zhongguo Xinan Minzu Kaogu (中國 西南 民族 考古, Archaeology of Ethnic Minorities in China’s Southwestern Regions) lend support to Lacouperie’s theory of Chinese as a fusion language, including elements of what we now call the Taic‑Yue family. Specifically, ancient Annamese took shape through the fusion of several linguistic streams:

1. Proto‑Vietic with Taic substrata – Early proto‑Vietic forms blended with basic elements of ancient Taic languages spoken by the native indigenes south of the Yangtze River. These groups were later collectively classified as Yue. Their basic lexicons survive in the Tai‑Kadai languages, also known as the "Tày" language, with attested comparanda cited in Chapter 8.
   
   
2. Archaic Chinese of the pre‑Qin era – Forms of early Chinese spoken by the northwestern Qin people before 250 B.C. contributed another layer to the developing Annamese speech.
   
   
3. Old Chinese of the Han – The Old Chinese spoken by the Han (漢族 Hànzú) since the dynasty founded by Liu Bang (劉邦) and his generals, themselves former Chu subjects (楚國人) of Daic origin. Vietnamese author Bình Nguyên Lộc (1972) even referred to them as "the Malay people".
   
   
4. Southern immigrant admixture – The cumulative influence of southern Chinese immigrants who resettled in what is now northern Vietnam beginning some 3,000 years ago, a process that has continued into modern times.  (2).

As for the Austroasiatic Mon‑Khmer elements in Vietnamese: the history of Vietnam’s formation is one of ethnic and linguistic intermixture. Along their southward migrations, ancient Annamese speakers absorbed Chamic and Mon‑Khmer words, adding Austroasiatic and Austronesian elements atop the older Taic‑Yue substratum. Remnants of these layers are preserved in the Muong languages.

Historically, the Việt‑Mường group (300 B.C. to 100 A.D.) gave rise to the Kinh ethnicity in ancient Annam. Around a thousand years ago, this group split into distinct Mường and "Annamese" branches. Modern Vietnamese evolved from the latter, after its break from the common Việt‑Mường base. Meanwhile, Mon‑Khmer languages spoken by minorities in the western highlands along the Cambodian and Laotian borders contributed additional native words, including basic vocabulary still visible in Muong dialects.

Beneath these layers lies a deeper foundation: an admixture of Proto‑Chinese, Old Chinese, Middle Chinese, Early Mandarin, and even modern Mandarin (Putonghua). Traces of ancient Yue speech remain detectable in other Daic, Miao‑Yao, and Mon languages of southern China and northern Vietnam. Yet over time, these elements have become far more foreign to Vietnamese than the Chinese strata, which remain the dominant influence.

Table 4 - Overview of proportional Vietnamese linguistic "foreign strata"

Sinitic components		Vietnamese linguistic strata
Proto-Chinese (to 1028 B.C.)	«««««« o »»»»»	Sinitic-Vietnamese e.g., nạ, mắt, giò, đi, chạy, lá, cá, gà, gió, cộ, chài, cửa, etc.
Old Chinese (ca. 600 B.C.)	«««««« o »»»»»	Sinitic-Vietnamese e.g., tía, mẹ, buồng, ngủ, bếp, tủ, đũa, sông, buồm, tàu, etc.
Ancient Chinese (from the second century A.D..)	«««««« o »»»»»	Sinitic-Vietnamese e.g., giông, gió, biển, khơi, ngoài, giấy, viết, etc.
Middle Chinese (from 601 A.D.)	»»»»»»»»»»»»	Sino-Vietnamese e.g., mục, kê, hổ, giangsơn, quốcgia, sơnhà, etc.
Early Mandarin (from 1324 A.D.) and modern Mandarin	«««««« o »»»»»	"Sinicized-Vietnamese" e.g., nonsông, nướcnhà, ytá, bácsĩ, tửtế, lịchsự, bồihồi, langbạt, etc.
Sino-Tibetan and other foreign elements	««««« o »»»»»	Daic, Mao-Yao, and other Austroasiatic Mon-Khmer elements e.g., mũi, ngón, ổi, cốc, nem, ớt, bươmbướm, etc.

Beyond the phonological and semantic resemblances noted in Shafer’s Sino‑Tibetan listings, the case for Vietnamese inclusion is further reinforced by the classic example of tonality, which sharply distinguishes it from the Mon‑Khmer sub‑family. It is worth emphasizing that Mon‑Khmer languages are toneless, whereas Vietnamese, like other members of the Sino‑Tibetan family, is tonal, one mirroring the other, so to speak.

In modern orthography, Vietnamese marks tone visually with five diacritical signs placed on vowels, yielding eight distinct tones. For example, "nhược" /ɲɨək8/ ('weak') and "ngược" /ŋɨək8/ ('reverse'). More precisely, tone is determined not only by diacritics but also by final consonants such as ‑p, ‑t, and ‑k, which together complete the eight‑tone system. This arrangement corresponds to the traditional Sino‑Vietnamese and Chinese phonological framework, in which four pitches, each split into upper and lower registers, form the canonical eight‑tone scheme attested in Ancient Chinese (Norman 1988, p. 55). (2) 

Just as Henri Maspero (1912) proposed, tone is an inherent feature of language and cannot be derived from non‑tonal elements. A corollary of this view, as Norman (1988, p. 54) observed, is that tonal languages cannot be genetically related to languages that lack tone.

Chinese cursive scripts were still in use in Korean (K) texts as late as the 1960s. Yet unlike tonal Sino‑Vietnamese romanized words, modern Korean writings rely heavily on context to interpret the meaning of transliterated, toneless Chinese loanwords. When written in Hangul without Hanja, these words are pronounced nearly the same, with no tonal distinction to separate meanings. Classic examples include:

• 防火 fánghuǒ: K /banghwa/ ~ SV 'phònghoả' (to prevent fire) 
  vs. 放火 fànghuǒ: K /banghwa/ ~ SV 'phónghoả' (to set fire)

• 水稻 shuǐdào: K /sudo/ ~ SV 'thuỷđạo' (aquatic rice) 
  vs. 水道 shuǐdào: K /sudo/ ~ SV 'thuỷđạo' (aquatic duct)
  vs. 首都 shǒudū: K /sudo/ ~ SV 'thủđô' (capital), etc.

To appreciate the idea of non‑inherent tonality, consider how Chinese and Vietnamese loanwords are pronounced in English: "chowmein," "kowtow," "taipoon," "sampan," "vietcong," "pho," "banhmi," "aodai." Written without diacritics, they are spoken with stress or intonation, but not with the original tonal contours.

A comparison with French or English loanwords in Khmer and Vietnamese makes the contrast sharper. Khmer, being non‑tonal, adapts such words without tonal marking. Vietnamese, by contrast, inherently tonal, naturally imposes tonal contours on borrowed words, making them feel native. This explains why loanwords from non‑tonal languages acquire tones in Vietnamese: they are eventually reshaped with tonal patterns to integrate into the system. Examples illustrate the process clearly:

• "caobồi" (cowboy)
• "quánhtùtì" (one‑two‑three)
• "súpquay" (subway)
• "mêtrô" (métro)
• "buộcboa" (pourboire)
• "phẹcmatuya" (fermature)

Each is accented with tones, showing how Vietnamese adapts foreign vocabulary by embedding it into its tonal framework. (See  What Makes Chinese So Vietnamese? - Wordlists in APPENDIX A).

In spatial distribution, correspondences between Mon‑Khmer and Vietnamese basic words are uneven and scattered across multiple Mon‑Khmer isoglosses. In effect, their cognates align on collateral lines, much like the numerals. The broader picture parallels Vietnamese cognates appearing across different Sino‑Tibetan etymologies as discussed earlier, and resembles how many Sinitic‑Vietnamese etyma affiliate with Old Chinese (上古漢語) of the pre‑Han period, commonly attested in Shijing Odes (詩經) and Chuci (楚辭). By contrast, Sino‑Vietnamese correspondences with Middle Chinese tend to align on a straighter axis, similar to Ancient Chinese (古漢語) into Middle Chinese. Taken together with other linguistic factors, setting aside the syntactic module where the modified word precedes its modifier, the overall affiliation of Vietnamese points strongly toward Chinese.

Why, then, is Vietnamese not grouped within Sino‑Tibetan in current classifications, but placed under Austroasiatic Mon‑Khmer? In brief, many have been content with a few dozen basic cognates and have overlooked that more than half of the surveyed items (based on lists by Thomson, Taylor, and Nguyễn Ngọc San) also align with Sino‑Tibetan etyma of the type cataloged by Shafer. By contrast, this research adopts a holistic approach: it examines the roots of Sino‑Tibetan etymologies and explores how those strata relate to Vietnamese etyma. The focus is a living language's intrinsic affiliation – its attributes, peculiarities, and historical position as the sole nation of the Southern Yue that emerged from Chinese dominion, unlike its cousin regions in Guangxi, Hainan, Guangdong, and Fujian. In that sense, Vietnamese is characteristically more Chinese than Mon‑Khmer, regardless of how many spears are thrown at that long‑rooted affiliation. Consider, historically, a thousand years under Chinese rule – there is light at the end of that tunnel.

To further substantiate the Sinitic‑Yue theory, the following chapters will continue to present basic listings of essential Chinese etyma with Vietnamese cognates. This constitutes the central focus of the research, together with its companion database project, the Chinese-Vietnamese Dictionary. Statistically, the evidence demonstrates that more than 90 percent of Vietnamese vocabulary, including nearly all basic words, derives from Chinese origin. Many of these correspondences are documented here for the first time, made available through this study. (4)

Conclusion

The search for the origins of Vietnamese remains unfinished, yet its trajectory is increasingly clear. Vietnamese cannot be explained as an Austroasiatic anomaly; rather, its lexicon and structures reveal deep affiliations across the Sino‑Tibetan family. Shafer’s framework provides a foundation, but comprehensive mapping across all six divisions still requires refinement through modern reconstructions.

Vietnamese emerges not as a marginal case but as a central witness to the historical interplay of Sino‑Tibetan languages. Recognizing this opens pathways for future research: orthographic reform that reflects polysyllabicity, annotated bilingual lexicons that bridge Vietnamese with its Sino‑Tibetan kin, and reconstructions of a shared heritage that situate Vietnam within a wider linguistic continuum.

This foundation is not an endpoint but a platform for further inquiry. The next task is to explore how etymological correspondences manifest in structural, cultural, and historical dimensions. The question is no longer whether Vietnamese belongs within this framework, but how its unique trajectory – shaped by centuries of contact, adaptation, and resistance – produced the language we recognize today.

References

• Alves, Mark J. An Updated Overview of the Austroasiatic Components of Vietnamese. Languages 9(12), 2024.

• Benedict, Paul K. Sino‑Tibetan: A Conspectus. Cambridge University Press, 1972.

• Campbell, Lyle. Historical Linguistics: An Introduction. Edinburgh University Press, 2013.

• Chen, Shu‑Fen. Vowel Length in Middle Chinese Based on Buddhist Sanskrit Transliterations. Language and Linguistics 4(1), 2003, pp. 29–45.

• Haudricourt, André‑Georges. L’origine des tons en vietnamien. Journal Asiatique 242, 1954, pp. 69-82.

• Haudricourt, André‑Georges. Problèmes de phonétique diachronique: la nasalisation vocalique en vietnamien. Bulletin de la Société de Linguistique de Paris 49, 1954.

• Haspelmath, Martin. Comparative Linguistics and the Problem of Spurious Similarities. Linguistic Typology 9(1), 2005.

• Maspero, Henri. Études sur la phonétique historique de la langue annamite. Les initiales. Paris: Imprimerie Nationale, 1912.

• Matisoff, James A. Handbook of Proto‑Tibeto‑Burman. University of California Press, 2003.

• Mei, Tsu‑lin. Tones and Prosody in Middle Chinese and the Origin of the Rising Tone. Harvard Journal of Asiatic Studies 30(1), 1970, pp. 86-110.

• Norman, Jerry. Chinese. Cambridge University Press, 1988.

• Pulleyblank, Edwin G. Middle Chinese: A Study in Historical Phonology. Vancouver: University of British Columbia Press, 1984.

• Sa, Quoc Hoang. Study on the Understanding and Use of Sino‑Vietnamese Words: Perspectives from Secondary School Students in Ho Chi Minh City. Sprin Journal of Arts, Humanities and Social Sciences 4(5), 2025.

• Shafer, Robert. Introduction to Sino‑Tibetan. Wiesbaden: Otto Harrassowitz, 1966-1974.

• Sidwell, Paul. Austroasiatic Dataset for Phylogenetic Analysis: 2015 Version. Mon-Khmer Studies 44, Mahidol University / SIL International.

• Starostin, Sergei. Sino‑Tibetan Etymological Dictionary and Thesaurus. Moscow, 1991.

• Thomas, David D. Basic Vocabulary in Some Mon-Khmer Languages. Mon-Khmer Studies, 1960.

• Thurgood, Graham. Vietnamese and Tonogenesis: Revisiting Haudricourt. California State University, Chico, 2002.

FOOTNOTES

---

(1)^ To recall, several loanwords in the Chinese zodiac appear to have been borrowed from the ancient Yue language. Examples include 子 zǐ (SV 'tý', cf. VS 'chuột') for ‘rat’, 丑 chǒu (SV 'sửu', cf. VS 'trâu') for ‘ox’—more precisely, the water buffalo—午 wǔ (SV 'ngọ', cf. VS 'ngựa') for ‘horse’, 未 wèi (SV 'mùi', cf. VS 'dê') for ‘goat’, and 亥 hài (SV 'hợi', cf. VS 'heo') for ‘boar’. These form part of the twelve animal cycle, with one notable exception: 兔 tù (SV 'thố', VS 'thỏ') ‘hare’. For reasons perhaps tied to taboo, 卯 mǎo (SV 'mão', cf. VS 'mèo') was substituted, though its sound clearly derives from /mew2/, a form for ‘cat’ that appears consistently across Daic and Mon‑Khmer languages.

Other cases, however, do not follow the same pattern. Words such as 江 jiāng: V 'sông' ‘river’, 虎 hǔ: V 'cọp' ‘tiger’, 狗 gǒu: V 'chó' ‘dog’, 犬 quán: V 'cún' ‘puppy’, and 牙 yá: V 'ngà' ‘ivory’ also trace back to Yue origins, yet they can be regarded as having developed from the same root stock as Sinitic‑Vietnamese. This does not imply genetic affiliation in the strict sense, as reflected in the Chinese characters recorded in the Kangxi Dictionary. Nor does it parallel the situation of Vietnamese numerals when compared with Mon‑Khmer.

Consider Vietnamese 'răngkhểnh' (canine tooth, cuspid), which aligns with 犬齒 quánchǐ (= 犬牙 quányá, cf. VS 'răng' 齡 líng), or 'ngồixổm' and 'chồmhổm' (to squat like a dog), which correspond to 犬坐 quánzuò (犬 quán = ‘dog’). Likewise, 'nonsông' (country) and 'hùnghổ' (gung‑ho) are culturally accented loanwords from 江山 jiāngshān (SV 'giangsan') and 猛虎 měnghǔ (SV 'mãnhhổ'). These entered Vietnamese only after the language had already broken away from the Việt‑Mường group and developed into a distinct Vietic‑Annamese form.

Today, as a language characterized by disyllabicity, Vietnamese continues to build many disyllabic words from Chinese materials, though often in reverse order, following the [modified + modifier] pattern.

(2)^ Much as the Chinese languages themselves emerged, the fusion of "pre‑Chinese" migrants from the upper reaches of the Yellow River with the aboriginal peoples long settled along both banks of the Yangtze gave rise to what became the Han and, in time, the later Chinese tongues. A parallel process unfolded in the south: Chinese linguistic elements entered an early form of Annamese, spoken by the mixed population of Giaochi Prefecture (交州 Jiaozhou, SV 'Giaochâu' < 交趾 Jiaozhi) around the turn of the first century B.C.

(3)^ Let us pause briefly to consider an intriguing point about tonality. Over the past few decades, linguists and language educators in the United States, Canada, Australia, France, and elsewhere have studied how Vietnamese and Chinese children acquire their mother tongues. A recurring question has been: At what age do children acquire tones? Do they master them when learning their very first words, or only later, as if tones were a secondary layer?

The answer is clear. Toddlers acquire tones simultaneously with their first words; tonal control is part of the initial process of language acquisition, not a later addition. By contrast, children who are not exposed to their mother tongue until adulthood typically speak Vietnamese with the intonation of Western foreigners, that is, monotonous, flattened, and lacking the natural tonal contours.

(4)^ The percentage figure was obtained by arbitrarily sampling pages from a standard Vietnamese dictionary and counting the entries. Provide any sample page, and the same statistical proportion will emerge. Alternatively, by examining the etyma cited throughout this study, one can readily reconcile with that figure.