In Search Of Sound Change Pattern
by dchph
This paper investigates how Vietnamese words can be traced through systematic sound changes, particularly in relation to Middle Chinese. Regular Sino‑Vietnamese (SV) vocabulary follows predictable correspondences: voiceless initials yield upper‑register tones, voiced initials yield lower‑register tones, and finals such as /‑wŋ/ and /‑wkp/ explain reflexes like ‑ông and ‑ục. These patterns allow scholars to distinguish between "true" Sino‑Vietnamese loans and other strata of the lexicon.
I) Sound Change as a universal process
Sound change is the most pervasive force in linguistic history. It operates across all languages, shaping phonology, semantics, and syntax. The central question is: how do sound changes come about, and what patterns can we discern?
This article demonstrates that while Sino‑Vietnamese borrowings generally follow predictable phonological correspondences rooted in Middle Chinese, vernacular Sinitic‑Vietnamese forms often diverge through irregular transmission, performance errors, or semantic reinterpretation. To understand these shifts, one must look beyond one‑to‑one glossaries and instead examine the paradigms of frequency, analogy, and child acquisition that drive change. Vietnamese, layered upon nearly a millennium of Han contact, localized Chinese roots into its own phonological and semantic system, producing doublets, contractions, and innovations that continue to define its expressive resources today.
Languages change, and sound change is a fundamental fact of linguistic history. In earlier times such changes occurred more rapidly, whereas in the modern era they have slowed considerably. The widespread reach of electronic media now disseminates speech and writing with remarkable uniformity, even as it generates new abbreviations, jargons, and neologisms in digital communication. The reality of sound change is self‑evident and beyond dispute. The challenge, then, is to explain how such changes arise and to uncover the patterns that govern them.
II. Paradigms of sound change
A. Frequency Principle (Mańczak)
Sound change takes many forms, both regular and irregular, as illustrated in the Sino‑Vietnamese and Sinitic‑Vietnamese vocabularies examined in the preceding chapters. Witold Mańczak, in his study "Irregular sound change due to frequency in German" (Recent Developments in Historical Phonology, p. 309), summarizes the principle as follows:
In brief, the theory of irregular sound change due to frequency can be presented as follows. There is a synchronic law according to which the linguistic elements which are more often used are smaller than those which are less often used. There is a kind of balance between the size of linguistic elements and their frequency. Anyhow, the size of linguistic elements is not stable. As a result, the size of words may change considerably as the comparison of some Old and New High German words [...]
There are four criteria which allow us to recognize that regular sound change due to frequency is involved:
(1) If a frequency dictionary for a given language and for a given epoch exists, we may use it, since the majority of words showing an irregular change due to frequency (about 90%) belong to the thousand words most frequently used in the given language.
(2) In addition to irregular sound change due to frequency, there are other irregular sound changes, namely assimilations, dissimilations, metatheses, and expressive and overcorrect forms. [...]
(3) If in a given language, a morpheme, word, or group of words occurs in a double form (regular or irregular), irregular sound change due to frequency is characterized by the fact that the irregular form is usually used more often than the regular ones. [...]
(4) If the irregular sound change due to frequency occurs within a paradigm, it may be recognized by the fact that only the more commonly used forms are subject to it, whereas the forms used less frequently remain regular. [...]
B. Sturtevant's paradox (Anttila)
A paradigm may take many forms, one of which is captured in Sturtevant's paradox, cited by Raimo Anttila in "The Acceptance of Sound Change by Linguistic Structure" (Recent Developments in Historical Phonology, p. 43): sound change is regular and produces irregularities, while analogy is irregular and produces regularity. Beyond this paradox, other forces also shape the emergence of paradigms.
C. Competence vs. performance (Dressler)
Competence and performance, for example, are among the factors that contribute to their formation. As Wolfgang Dressler observes in "How Much Does Performance Contribute to Phonological Change?" (Recent Developments in Historical Phonology, p. 145):
1.1. In opposition to traditional views that language change starts in performance (parole), generative grammarians have equated linguistic change with change in competence. [...]
1.2. According to Antitila (1972:128 and later studies) "most changes seem to be triggered by performance". Performance contains variation due to imperfect control, to imperfect articulatory organs, to memory restrictions, slips of the tongue or of the ear, the error such as involuntary contaminations, variation due to fluctuations in attention and to inadvertence, to confusions, to playfulness, etc., and individual "biophonetic" characteristics (see Trojan 1975), which cannot be described by rules. [...]
In performance hypotheses sound change is said to be (always or most of the time) the result of random vacillations and gradual fluctuations, to be imperceptible, to be due to ease of articulation or to individual tendencies, to result from the inability of the individual to produce exactly the sounds which he hears, to be due to stylistic fluctuations, to be of a statistical nature, etc. [...]
2. Sound change due to loans (from a substratum, superstratum, or adstratum) is probably the case where possible origins or phonological change can be more easily ascertained. Imperfect application of phonological rules of the target language by speakers of the source language (cf. Fasching 1973) is often seen as due to lack of competence in the target language, or more precisely performance errors. However, we must distinguish between confusion errors which can not be directly traced back to a model in the source language on the one hand, and transpositions of parts of the competence in the source language on the other. [...]
3.1. (Non-analogical) contaminations in speech errors are rather different from blends in language diachrony (see Dressler 1976a) [...]
According to Paul (1920:160-2) diachronic contamination nearly always occurs between words which are either etymologically related or suppletive or antonymous, which is not the case at all in speech errors (see Dressler 1976a ss 13) [...]
Most, if not (nearly) all historical blends are either of an analogical nature (and not phonological contaminations such as many errors) or are consciously coined or are due to interference between the two dialects, [...]
3.2. Non-contaminatory slips of the tongue (cf. Meringer – Mayer 1895, Meringer 1908, Gromkin 1973) are generated by syntagmatic (more rarely dissimilatory) anticipation and preservation, assimilatory increment, and dissimilatory loss. [...]
3.3. Perceptual errors are very complex and cannot possibly give rise to "sound laws". [...] Systematic mis-perception of phonemes of allophones either occur as contact phenomena (i.e. interferences of competence, see 2) or in language disturbances such as aphasia (i.e. disturbance or loss of competence). [...]
4. Child-language acquisition is thought to be the principle source of phonological change by most linguists nowadays. If a process of child language (such as A. Meillet's French example [入] ~ [ǐ] or the final devoicing of obstruents) is said to initiate phonological change, then it is a question of general and systematic substitutions, reflecting the (older) child's competence, and this even in the case of incipient lexical diffusion.
All paradigms of how sound change has been explained by theoretical experts converge to reinforce the central argument developed throughout this study. With that foundation in place, we now turn to the specifics of sound change as it appears in the transmission of Chinese loanwords into Vietnamese.
Table 1 - Comparative grid: Paradigms of sound change
| Paradigm / Principle | Scholar | Core Idea | Vietnamese Example | Notes |
|---|---|---|---|---|
| Regular vs. Irregular Change | General framework | Sound change follows systematic correspondences, but irregular shifts arise from performance or erosion | 父 fù → bố vs. 爹 diē → cha, tía | Regular reflex vs. vernacular divergence |
| Frequency Principle | Mańczak | High‑frequency words undergo irregular reduction; irregular forms often more common | 得 dé → được vs. 被 bèi → bị | Everyday verbs show contraction and simplification |
| Sturtevant's Paradox | Anttila | Sound change is regular but produces irregularities; analogy is irregular but produces regularity | 當 dāng → đang, đáng, đương, tưởng | Multiple reflexes illustrate paradoxical outcomes |
| Competence vs. Performance | Dressler | Performance errors (slips, misperceptions, ease of articulation) trigger change; child acquisition central | 中 zhōng → trung, trong vs. trúng | Misperception and semantic shift create irregular doublets |
| Assimilation / Dissimilation / Metathesis | Historical phonology | Phonetic processes reshape borrowed forms | 順 shùn → thuận → suônsẻ | Shows metathesis and semantic expansion |
| Semantic Localization | Vietnamese vernacular tradition | Borrowed roots adapted to cultural context | 周年 zhōunián ('anniversary') → thôinôi ('baby leaves cradle') | Illustrates contextual reinterpretation |
The author's final remark for this section is not an additional commentary but rather a reflection: the synopsis itself serves as a source of encouragement, a way of drawing strength from the insights of the scholars cited above, especially the formulation presented in item 3.3.
III. Vietnamese case studies
A. Sino‑Vietnamese vs. Sinitic‑Vietnamese strata
To understand the phenomenon of sound change, both synchronically and diachronically, readers must first distinguish the paradigms that separate Sino‑Vietnamese from Sinitic‑Vietnamese lexemes. Although both may derive from the same Chinese sources, their induced drifts follow different models, especially in cases where imperfect transmission produced the most drastic shifts. In many instances, Chinese loanwords entered Vietnamese layered upon older strata of Ancient Chinese vocabulary, including basic items that likely evolved from the same roots, or cognates, as seen in Shafer's basic Sino‑Tibetan wordlists. This principle rests on two pillars: the historical record of nearly a millennium of Han occupation in Annam, and the abundant linguistic evidence of Chinese loanwords that became embedded in Vietnamese. The phenomenon of sound change thus unfolded in multiple dimensions, drifting and shifting semantically and syntactically, leaping forward synchronically while unfolding diachronically.
Another guiding principle is that Chinese – Vietnamese etymological work should be approached as a matter of semantic translation rather than one‑to‑one lexical correspondence. Sound changes are the products of associations between sound and concept, and what we are dealing with are equivalent concepts, not simple glossaries. A Chinese character appearing in a Vietnamese word may also serve as a syllabic stem in other formations, functioning as an etymon across multiple derivatives. For example, 順 shùn (SV thuận) in 順便 shùnbiàn (sẵntiện, luôntiện) conveys the notion of "conveniently," and serves as the etymon for both sẵn "available" and luôn "conveniently." Likewise, 順 in 順利 shùnlì (suônsẻ "smoothly") and 孝順 xiàoshùn (hiếuthảo "filial piety") gives rise to suôn "smoothly" and thảo "devotion." Each can stand independently as a morpheme, though in compounds like suônsẻ and hiếuthảo only hiếu functions as a free word‑syllable. We will return to this morpheme-syllable-word relationship, known in Vietnamese as tiếng, later.
When tracing Vietnamese words of Chinese origin, restricting ourselves to rigid one‑to‑one correspondences will obscure many roots. Sinitic‑Vietnamese words often appear in multiple forms and guises. For instance, insisting that 成 chéng (SV thành) corresponds only to sẵn "ready" would cause us to overlook its other reflexes, such as vâng "yes" or xong "finished." In 現成 xiànchéng (sẵnsàng "ready"), 現 xiàn yields sẵn while 成 chéng becomes sàng, a reflective duplicate morpheme that cannot stand alone. Similarly, vâng and xong may also derive from other sources, such as 行 xíng "okay" and 完 wán "done."
The following expanded examples of disyllabic words, some combining Sino‑Vietnamese and Sinitic‑Vietnamese element, illustrate these processes. Such forms best demonstrate phonetic discrepancies and homo‑organic synonyms, akin to the Chinese concept of 諧聲 xiéshēng, where characters like 見 and 現 in the Han period could both mean "appear." Vietnamese usage of these words has often shifted under the influence of contraction, syncope, metathesis, localization, innovation, derivation, association, dissimilation, corruption, contamination, adaptation, and other mechanisms. In short, sound change not only altered existing forms but also generated new words and meanings, sometimes diverging far from their originals, though often maintaining concurrent phonetic and semantic correspondences with Chinese, especially in the case of later loanwords.
IV. Comparative perspective
Parallels with Latin in English highlight the coexistence of learned and vernacular registers. In a similar way, Cantonese and Vietnamese both preserve Tang‑era tonal matrices, unlike Mandarin, which underwent later tonal simplification. These cases demonstrate that sound change is at once systematic and creative, producing regular correspondences while allowing for irregular innovations across languages.
Table 1 - Comparative grid of selected Vietnamese forms of Chinese origin
| Viet. | Chin. | Pinyin | Sino‑Viet. (SV) | Meaning | Type of Change | Semantic Notes |
|---|---|---|---|---|---|---|
| suônsẻ | 順利 | shùnlì | thuậnlợi | "smoothly" | Derivation | Narrowed to everyday sense of "smooth, uneventful"; cf. trótlọt |
| hiếuthảo | 孝順 | xiàoshùn | hiếuthuận | "filial piety" | Metathesis | Split into hiếu "filial" + thảo "devotion"; thảo gains independent moral nuance |
| tốttính | 德性 | déxìng | đứctính | "virtue" | Innovation | Semantic shift: đức "virtue" reinterpreted as tốt "good" |
| rác | 垃圾 | lāji | lạpcấp | "garbage" | Syncope | Reduced to monosyllable; generalized to all refuse |
| mai | 明兒 | míngr | minhnhi | "tomorrow" | Contraction | Retained temporal meaning; simplified form stabilized |
| tìnhyêu | 愛情 | àiqíng | áitình | "love" | Localization | Order reversed; yêu becomes core vernacular verb "to love" |
| ưathích | 愛戴 | àidài | áiđái | "like" | Association | Blended with 疼愛 téng'ài; broadened to "fondness, affection" |
| ănnói | 言語 | yányǔ | ngônngữ | "speech" | Association/assimilation | Recast as vernacular ăn + nói; semantic reinterpretation of "speech" as "eating and speaking" |
| nóinăng | 語言 | yǔyán | ngônngữ | "speech" | Localization by reduplication | Reduplicative năng adds emphasis; colloquialized register |
| lờilẽ | 語辭 | yǔcí | ngữtừ | "speech" | Innovation + assimilation | Split into lời "words" + lẽ "reason"; semantic bifurcation |
| thơtừ | 書辭 | shūcí | thưtừ | "letters" | Innovation | Extended to broader concept sáchvở "books, literature" |
| thợmộc | 木匠 | mùjiāng | – | "carpenter" | Corruption | Hybridized VS thợ + SV mộc; semantic stability retained |
| một căn gáctrọ | 一間 + 閣 + 宿 | yī jiān + gé + sù | nhất gian, các, túc | "lodging room" | Adaptation | Mirrors cănphòng 房間; semantic calque of "room" |
| bọnngười | 一幫人 | yī bāng rén | nhất bang nhân | "the gang" | Adaptation | Generalized to "group of people"; neutral/pejorative shift |
| giấcmộng | 一場夢 | yī chǎng mèng | nhất trường mộng | "a dream" | Adaptation | Reduced to binom giấcmộng; retains poetic register |
Table 2 - Comparative grid* of selected cuộc/việc expression forms
| Vietnamese | Chin. | Pinyin | Sino‑Viet. (SV) | Meaning | Type of Change | Semantic Notes |
|---|---|---|---|---|---|---|
| cuộctình | 情場 | qíngchǎng | tìnhtrường | "love story" | Adaptation | 場 chǎng reinterpreted as cuộc; narrowed to "romantic affair" |
| cuộcđời | 世局 | shìjú | thếcuộc | "world, life" | Adaptation | cuộc generalized to denote "life" as an unfolding course |
| bỏcuộc | 破局 (cf. 放棄) | pòjú (fàngqì) | phácục (phóngkhí) | "to give up" | Adaptation | cuộc reanalyzed as "undertaking"; semantic shift to "abandon" |
| rốtcuộc | 結局 | jiéjú | kếtcuộc | "final outcome" | Adaptation | rốt adds finality; narrowed to "in the end" |
| cuộcsống | 生活 | shēnghuó | sinhhoạt | "life" | Adaptation | cuộc mapped onto "life" as lived experience |
| côngcuộc | 工作 | gōngzuò | côngtác | "task" | Adaptation | công + cuộc = "undertaking"; broadened to "enterprise, cause" |
| côngviệc | 公務 | gōngwù | côngvụ | "official business" | Adaptation | Semantic generalization to "work, job" |
| làmviệc | 幹活 | gànhuó | cánhoạt | "to work" | Adaptation | Vernacularized into everyday collocation làm việc |
| bậnviệc | 忙活 | mánghuó | manghoạt | "to be busy" | Adaptation | Narrowed to "busy with work" |
| nóidối | 假話 | jiǎhuà | giảthoại | "a lie" | Adaptation | nói + dối replaces SV giảthoại; semantic equivalence preserved |
| phảitrái | 是非 | shìfēi | thịphi | "right and wrong" | Adaptation | Binary opposition re‑expressed as phải/trái "right/wrong" |
*This grid shows how cuộc and việc became highly productive morphemes in Vietnamese, absorbing multiple Chinese inputs and extending their semantic range.
Just as the evolution of Chinese dialects illustrates the trajectory of a living language shaped over more than two millennia since their speakers became part of the later Han Chinese (漢人), the emergence of modern Vietnamese reflects a comparable path. Its present form is the cumulative outcome of continuous, gradual change, driven by inevitable interactions between an ancient aboriginal substrate and the various Chinese dialects introduced into the Annamese region. These influences arrived not only through conquest and military presence, but also via successive waves of emigrants, exiled officials, local administrators, scholars, and families of mixed background. Filtered through patterns of competence and frequency, such forces contributed to the irregularities of sound change.
Evidence of shared cognates in the basic lexical stock of both Chinese and Vietnamese suggests that many common words were already in use within the Viet‑Muong group before further differentiation from Old Chinese. Later, massive layers of Middle Chinese vocabulary entered Vietnamese up to the end of the Tang dynasty, forming the core of what is now recognized as Sino‑Vietnamese (Karlgren, Compendium of Phonetics in Ancient and in Archaic Chinese, 1954).
The hypothesis that ancient Vietnamese, the Vietic branch of the Viet‑Muong subfamily, descended from the same Yue linguistic source as Cantonese remains contested. At first glance, Vietnamese and Cantonese share a substantial portion of Middle Chinese glosses, yet far fewer basic substratum words. Strikingly, Northern Mandarin often fills lexical gaps where Cantonese does not, lending weight to the argument that ancient Annamese belonged within the broader Sino‑Tibetan family – a position earlier chapters have described under the label "Sinitic‑Vietnamese".
The proto‑Viet‑Muong language, spoken by the LạcViệt (雒越 LuoYue) and later the Âulạc (歐雒 Ouluo), had already existed alongside early Cantonese (Jyutwa 粵話) long before Han colonization of Annam (安南, then known as Giao Chỉ 交趾 Jiaozhi). Situated southwest of Phiênngung (番禺 Fanyu), the capital of the NamViệt Kingdom (南越王國 NanYue), Annam developed in relative isolation from the Cantonese heartland. This helps explain why Cantonese, though also a Yue language, shares only a limited substratum with Vietnamese. Illustrative examples include xơi 食 /shik8/ ('eat'), uống 飲 /jaam3/ ('drink'), ốm and ỉa 屙 /o5/ ('illness, excretion'), and đéo or đụ 屌 diǎo (SV điệu) /tjew3/ ('copulate').
Archaeological parallels reinforce this picture. Đông Sơn artifacts from Thanh Hoá Province, dated between 400 B.C. and 100 A.D. (Karlgren, 1945), reveal cultural continuity. In similar fashion, the early Sinitic‑Vietnamese vocabulary acquired under Han rule eventually circulated back into southern China. Re‑packaged in new phonological and semantic forms, these words became part of the broader repertoire of Old Chinese variants – many of which still coexist in Vietnamese today, for instance,
Table 3 - Comparative grid** of selected lexical doublets
| Viet. | Chin. | Pinyin | Sino‑Viet. (SV) | Vernacular (VS) | Meaning | Notes |
|---|---|---|---|---|---|---|
| mèo | 猫 | māo | mẹo | mẹo | "cat" | Direct borrowing; VS later replaced by mèo |
| vi / voi | 為 / 豫 / 象 | wèi / yú / xiàng | vi ("do, for") | voi ("elephant") | "do, for"; "elephant" | Doublet: archaic 豫 yú "elephant" vs. modern 象 xiàng |
| hổ / cọp | 虎 | hǔ | hổ | cọp | "tiger" | Parallel SV and VS reflexes |
| hùm | 甝 | hán | hàm | hùm | "white tiger" | Archaic variant; VS specialization |
| đường | 唐 | táng | đường | đàng ("road") | "hall, Tang" → "road" | Replaced 道 dào (SV đạo); doublet with 途 tú (SV đồ) |
| xuyên / dòng | 川 | chuān | xuyên | dòng | "stream" | SV vs. VS differentiation |
| giang / sông | 江 | jiāng | giang | sông | "river" | Doublet; VS sông generalized |
| phòng / buồng | 房 | fáng | phòng | buồng | "room" | SV vs. VS doublet; semantic overlap |
**This grid shows how SV forms (literary, learned borrowings) and VS forms (vernacular, often older or colloquial) coexist, sometimes as doublets, sometimes as semantic splits.
The latter forms may already have existed before the Han Empire annexed the Nam Việt kingdom in 111 B.C. This polity had been ruled by King Triệu Đà (趙佗 Zhào Tuó), a former general of the Qin state (秦國) whom some scholars suggest was of Yue origin (1). His heirs governed a vast territory that encompassed ancient Fujian, Guangdong, and Guangxi, extending southward into the Annamese land of today's Red River basin in northern Vietnam.
With the onset of Han domination, the Viet‑Muong group divided into two branches Viet and Muong, whose speakers dispersed along different migratory routes. Some retreated into mountainous refuges, while others established settlements along the coastal plains (2). This divergence gave rise to the early Vietic sub‑family, from which the language later known as Annamese emerged. Annamese was an admixture of the indigenous substrate and Western Han (西漢) elements, and its basic vocabulary, as attested in Han‑dynasty classics, already reveals this blend. Examples include:
-
劉 líu ('Liu', 'axe', 'conquer') → SV Lưu (surname), VS rìu, lấy
-
戉, 鉞, 粵 Yuè ('Yue', 'axe') → SV Việt ('越 Yuè'), VS rựa
-
銅 tóng ('copper', 'bronze') → SV đồng, VS thau
-
車 chē ('wheeled machine', 'chariot', 'carriage') → SV xa, VS cỗ, cộ, xe
-
藍 lán ('indigo') → SV lam, VS chàm
-
落 luò ('earthnut', 'fall') → SV lạc, VS rơi
-
羅 luó ('net fishing', 'fishing net') → SV la, VS chài, lưới
-
籮 luó ('bamboo basket', 'fish trap') → VS rỗ, rọ
The similarity of basic vocabulary between Vietnamese and Chinese has been a central subject of this study, with evidence of cognateness presented throughout. Yet no historical records written in a native "Vietnamese script" have been found prior to the fifteenth century, let alone before 939 A.D. The early history of Annam is therefore reconstructed largely from Chinese sources, which governed the region from 111 B.C. to 939 A.D (3). Within this framework, Sinitic‑Vietnamese etymology can be surveyed by drawing on ancient Chinese linguistic records, including the thirteenth‑century Annam Dịchngữ dictionary (Vương Lộc 1995), to approximate how the archaic Vietic language may have sounded before its deep blending with Ancient Chinese more than two millennia ago.
Equally important is the recognition that thousands of additional cultural and scholarly Sino‑Vietnamese loanwords entered the daily speech of ordinary Vietnamese. Their presence is not speculative: Sino‑Vietnamese vocabulary has become inseparable from modern Vietnamese, to the point that no complete sentence can be spoken today without it. This popularity is a crucial dimension of Vietnamese historical linguistics. Politically and linguistically, scholars and officials in Annam, like their counterparts elsewhere in the empire, were required to master the official court language: quanthoại (官話 'Mandarin'). This explains why Middle Chinese, particularly Tang‑era Mandarin, entered Vietnamese en masse in the form we now call Sino‑Vietnamese. (4)
Discrepancies between Vietnamese usage and that of major southern Chinese dialects – Cantonese, Fukienese, and Wu (e.g., Shanghainese, Shaoxingnese) – are also revealing. Tang and Song rhyme books show that each language developed minor phonetic divergences that rendered them mutually unintelligible. Each character often carried both colloquial and literary readings, producing at least two distinct pronunciations. These southern dialects, originally Yue languages south of the Yangtze, retained many substratum words identified by Austroasiatic specialists in the early twentieth century. Yet all underwent Sinicization, their territories having remained within the Chinese cultural sphere since the Han dynasty. This historical continuity institutionally placed them within the Sino‑Tibetan family. (5)
Nevertheless, despite lexical similarities with Chinese dialects, neither ancient Annamese nor modern Vietnamese should be considered a Chinese dialect – before or after 939 A.D., when the Vietnamese threw off Chinese colonial rule. Phonetically, the modern Vietnamese sound system evolved gradually over centuries, shaped by successive waves of northern vernacular forms brought by emigrants during each colonial stage. This explains why Vietnamese today contains many expressions strikingly similar to Mandarin, the living northern language of China. Such parallels can be verified in classical Chinese novels – Romance of the Three Kingdoms, Journey to the West, Water Margin, Dream of the Red Chamber – all written in the vernacular style (báihuàwén 白話文) from the Yuan through the Ming dynasties. These texts reflect a northern dialectal base that resonates with many Vietnamese forms (Wang Li et al. 1956: 46, 82, 98).
In terms of competence and performance, older forms eventually had to compete with newer and more prestigious ones, namely the scholarly Sino‑Vietnamese derived from Middle Chinese and cultivated in learned circles. Among the general population, northern Vietnamese speakers tended to employ more Sino‑Vietnamese vocabulary, while southern speakers relied more heavily on the Sinitic‑Vietnamese stock. Yet this regional distinction did not prevent the Sinitic‑Vietnamese domain as a whole from absorbing additional layers of vernacular Mandarin. In the contemporary setting, both regions continue to incorporate modern Chinese terms, adapted with Sino‑Vietnamese pronunciation – for example, thịphạm (示範 shìfàn, 'demonstrate'), soáica (帥哥 shuàigē, 'handsome man'), namthần (男神 nánshén, 'Mr. Perfect'), and đạocụ (道具 dàojù, 'theatrical props').
Colloquially, such new Chinese lexicons in Vietnamese may be regarded as later loans or as modified continuations of earlier words reshaped by phonological shifts, vernacular usage, and semantic adaptation. This phenomenon is not new: it has been occurring since antiquity. Sinitic‑Vietnamese forms often evolve into words that appear deceptively familiar yet differ in origin or nuance. The key point is that Vietnamese has consistently drawn on Chinese material in ways unmatched by Austroasiatic languages, aside from a handful of questionable Mon‑Khmer items.
V) Creative localization of vernacular Mandarin
When compared with modern Chinese Pǔtōnghuà (普通話), many common words in Vietnamese reveal multifunctional usage shaped by sound change – for instance, nouns functioning as verbs, verbs as adverbs, or particles serving multiple roles. Significantly, as in Chinese, many grammatical function words (虛詞) in Vietnamese can be traced back to originally concrete lexical items (實詞). These functional particles appear to have entered Vietnamese through northern Chinese dialects during at least two stages: Early Mandarin and Late Mandarin, continuing into contemporary Mandarin (Guānhuà 官話; SV Quanthoại).
Let us now examine some of these derived variants of shared Chinese roots, both as notional words and as grammatical particles.
A) 來 lái: lai (SV)
- lại, này: "come"; variations: 來來來! Láilái lái: "Lại đây này!" or "Tới đây!" ('Come over here!'),
- sau: 未來 wèilái "maisau" ('in the future'),
- lai: "racially mixed"; 外來 wàilái: for SV "ngoạilai" [ Notionally, to mean 'foreign origin', and with the dropped syllabic-morpheme 外 wài, VS "lai", originally functioning as an adverbal to connote the notion of 'in' or 'into', has become "racially mutated", of which the syllabic-morpheme 來 lai cannot used with the same notion by it self. Modern Chinese = 野種 yězhǒng (SV dãchủng) ],
- lại (a grammatical particle indicating upcoming or to-be-complete action): Màn lái! 慢來! "Chậm lại!" ('Slow down!'),
- lại: "again" semantically as a doublet of 再 zài: 'lại', e.g., 再來 Zàilái! "Làm lại!" ('Do it again!') (# 再來 zàilái can also be: "lặplại" ('repeat!') or "trởlại" ('return')
- là: 本來 běnlái "vốnlà" ('originally'), 原來 yuánlái "nguyênlà" ('originally'), 一來... yìlái "mộtlà" ('firstly')... 二來... èrlái "hailà" ('secondly')...,
- vậy: "thus" (a grammatical particle indicating resultant completion implying an adverbal 'then' or 'thus'): 你 去 那裏 來? Nǐ qù nálǐ lái? "Mầy điđâu vậy?" (Where did you go?), 你放才說甚麼來呢? Nǐ fāngcái shuō shěnme lái nè? "Mầy vừarồi nói gì vậy nhỉ?" (What have you just said?)
- đây: 上來 Shànglái! "Lênđây!" ('Come up here!'), 過來! Guòlái! "Quađây!" ('Come here!')
- nầy: 後來 hòulái "saunầy" ('later on'),
- nay: 素來 sùlái "xưanay" ('from the start'),
- làm: 來唄 Láibei! "Làmđi!" ('Go ahead!'), 亂來 luànlái "làmcàn" ('do things carelessly'), 來不及. Láibùjí. "Làm khôngkịp." (It can't be done (on time)), 來 一 盃! Lái yī bēi! "Làm một ly!" (Let's have a drink!),
- tới: "act", "perform a to be done action", e.g., 來不及! Láibùjí! "Tới khôngkịp!" '(We, I, He..) cannot make it on (on time)', 來吧! láiba! "Tớiđi!" ('Come!'),
- nổi: 起不來 qǐbùlái "dậyđâunổi" ('Unable to get up.'), etc.
B) 了 lē, liăo: liễu (SV)
- liễu: "finish, complete" → 了結 liǎojié → "kếtliễu" ("to finish, end")
- lấy: completed action → 他 抱緊 了 我. Tā bàojǐn le wǒ. → "Nó ômchặt lấy tôi."
- mất: completed action → 吃 完 飯 他 走了. Chī wán fàn tā zǒu le → "Ăn cơm xong nó đimất."
- nổi: capacity → 忘不了 Wàng bùliǎo → "Quênkhôngnổi."
- nữa: future certainty → 我不再回來了 Wǒ bù zài huílái le → "Tôi sẽ không về lại nữa."
- rồi: "already" → 忘了! Wàng le! → "Quênrồi!"
- rõ: 了解 liǎojiě → "hiểurõ" ("understand clearly")
- ra: adverbial "out" → 累壞了 Lèihuài le → "Mệtnhoàira."
- là: "then" → 小偷 看見 了 警察 後 拔腿 就 跑. Xiǎotōu kànjiàn le jǐngchá hòu bátuǐ jiù pǎo. → "Têntrộm thấy cảnhsát là chạy."
- luôn: 冷死了 Lěngsǐ le → "Lạnh chết luôn!", etc.
C) 打 dă: đả (SV)
- đánh, quánh, đập: 打字 dǎzì → "đánhchữ" ("to type")
- đòn: 挨打 áidǎ → "ănđòn" ("to be beaten")
- tá: 一打 yī dǎ → "mộttá" ("a dozen")
- từ: 自打 zìdǎ → "từđó" ("since")
- đánh: 打劫 dǎjié → "đánhcướp" ("to rob")
- tính: 打算 dǎsuàn → "tínhtoán" ("to plan")
- lậpcập: 打抖 dǎdǒu → "lậpcập" ("to tremble")
- cá: 打賭 dǎdǔ → "cáđộ" ("to bet"), etc.
D) 開 kāi: khai (SV)
- khai: 開張 kāizhāng "khaitrương" ("grand opening"); 開幕 kāimù "khaimạc" ("inaugurate"); 開講 kāijiǎng "khaigiảng" ("school opening")
- khui: 打開 dǎkāi → "khuira" ("to open up")
- khởi: 開始 kāishǐ → "khởisự" ("to begin")
- khỏi: 躲開 duǒkāi → "tránhkhỏi" ("to escape")
- mở: 開燈 kāidēng → "mởđèn" ("turn on the light"); 開花 kāihuā → "nởhoa" ("flowers bloom")
- nở: 花開 huākāi → "hoanở" ("flowers bloom")
- lái: 開車 kāichē → "láixe" ("to drive"); 開飛機. kāifēijī. → "lái máybay" ("to fly a plane")
- hài: 開心 kāixīn → "hàilòng" ("pleased")
- coi: 讓開 ràngkāi → "tránhcoi" ("get out of the way"), etc.
Evidence indicates that because Chinese and Vietnamese phonological interchanges arose from different sources and evolved across distinct periods, their phonetic forms were often transformed beyond easy recognition. This is especially true of polysyllabic compounds: the longer the form, the greater the divergence.
E) Corpus segment: Middle Chinese → Sino‑Vietnamese
What, then, justifies the claim that Sinitic‑Vietnamese developed such a rich array of variant sounds for each Chinese root? The answer lies in the observable patterns of sound change, even in monosyllabic correspondences between Middle Chinese and Sino‑Vietnamese. In many cases, Mandarin reflexes show the loss of final consonants, producing forms that, to the untrained eye, may appear unrelated or even unrecognizable. Yet these shifts follow systematic trajectories, as the following examples illustrate.
- 必 bì /pi⁵¹/ → SV tất [tʌt7] 'inevitable'
- 屁 pì /pʰi⁵¹/ → SV tý [ti7] 'buttocks'
- 俾 bèi /pei⁵¹/ → SV tì [ti2] 'low class'
- 乘 chèng /ʈʂʰəŋ⁵¹/ → SV thừa [tʰɨə2] 'make use of, avail oneself'
- 吃 chī /ʈʂʰɨ⁵⁵/ → SV ngật [ŋʌt8] 'eat'
- 額 é /ɤ³⁵/ → SV ngạch [ŋɛk8] 'amount'
- 而 ěr /ɚ²¹⁴/ → SV nhi [ɲi1] 'but'
- 激 jī /tɕi⁵⁵/ → SV khích [kʰik7] 'incite'
- 季 jì /tɕi⁵¹/ → SV quý [wi5] 'season'
- 節 jié /tɕiɛ³⁵/ → SV tiết [tjet7] 'festival'
- 津 jīn /tɕin⁵⁵/ → SV tân [tʌn1] 'ford'
- 熱 rè /ʐɤ⁵¹/ → SV nhiệt [ɲjet8] 'hot'
- 日 rì /ʐɨ⁵¹/ → SV nhật [ɲət8] 'sun'
- 起 qǐ /tɕʰi²¹⁴/ → SV khởi [kʰɤj3] 'rise'
- 攝 shè /ʂɤ⁵¹/ → SV nhiếp [ɲjep7] 'act for'
- 溪 xī /ɕi⁵⁵/ → SV khê [kʰe1] 'brook'
- 吸 xī /ɕi⁵⁵/ → SV hấp [hʌp7] 'inhale'
- 係 xì /ɕi⁵¹/ → SV hệ [he6] 'related'
- 習 xí /ɕi³⁵/ → SV tập [tʌp8] 'practice'
- 洗 xǐ /ɕi²¹⁴/ → SV tẩy [tɤj3] 'wash'
- 惜 xì /ɕi⁵¹/ → SV tích [tik7] 'cherish'
- 赥 xì /ɕi⁵¹/ → SV hích [hik7] 'giggle'
- 需 xū /ɕy⁵⁵/ → SV nhu [ɲu1] 'need'
- 學 xué /ɕyɛ³⁵/ → SV học [xok8] 'study' [cf. Cantonese /hok4/]
- 業 yè /je⁵¹/ → SV nghiệp [ŋjep8] 'profession'
- 一 yī /i⁵⁵/ → SV nhất [ɲət7] 'one'
- 義 yì /i⁵¹/ → SV nghĩa [ŋja4] 'righteous'
- 譯 yì /i⁵¹/ → SV dịch [jik8] 'translate'
- 玉 yù /y⁵¹/ → SV ngọc [ŋok8] 'jade' [M 玉 yù < MC ŋi̯ok < OC *ŋok | FQ 魚欲]
- 禺 yú /y³⁵/ → SV ngung [ŋuŋ1] 'a kind of monkey' [ex. 番禺 Fānyú → SV Phiênngung]
- 月 yuè /yɛ⁵¹/ → SV nguyệt [ŋwjet8] 'moon'
- 樂 yuè /yɛ⁵¹/ → SV nhạc [ɲak8] 'music'
Strictly speaking, the phonological correspondences between Middle Chinese and Sino‑Vietnamese followed well‑defined patterns of diachronic sound change. These developments were scholarly in origin, having evolved from the official language of the imperial court and the literati, most notably the prestige speech of Chang'an (長安) from the sixth century onward. Such patterns can be identified and categorized through the systematic phonological method known as fǎnqiè 反切 (FQ), or phonetic "spelling." It is through this traditional system that the pronunciation of Chinese characters can be reconstructed and aligned with their Sino‑Vietnamese equivalents. For example, An Chi (2016) suggested that Vietnamese trứng 'egg' may be linked to 種 zhǒng (SV chủng) rather than 蛋 dàn (SV đản), basing the reconstruction solely on the fǎnqiè tradition.
As with all historical phonological phenomena, exceptions inevitably occur. These reflect not only dialectal variation but also historical contexts in which certain pronunciations became taboo under particular rulers, prompting the adoption of euphemisms. Nevertheless, in most cases the variants correspond closely to what is recorded in ancient rhyme books, later compiled and cited in the Kāngxī Zìdiǎn (康熙字典). Scholars such as Bernhard Karlgren, Nguyễn Tài Cẩn, Wang Li, and Li Fang‑Kuei have drawn extensively on these sources to demonstrate the systematic nature of such sound changes.
- phước, phúc: 福 fú 'good fortune'
- phụng, phượng: 鳳 fèng 'phoenix'
- cuộc, cục: 局 jú 'status, situation'
- oai, uy: 威 wēi 'power'
- chú, chua: 注 zhù 'annotate'
- chúa, chủ: 主 zhǔ 'master, lord'
- bá, bách: 百 bǎi 'hundred'
- triều, trào: 朝 cháo 'dynasty'
- dị, dịch: 易 yí 'easy, change'
- nghĩa, ngãi: 義 yì 'righteous'
- hoàng, huỳnh: 黃 huáng 'yellow'
- nguyên, ngươn: 元 yuán 'origin'
- nhân, nhơn: 仁 rén 'benevolent'
- quới, quý: 貴 guì 'precious'
- tý, tứ: 伺 sì 'attend'
- tý, tử: 子 zǐ 'offspring'
- vũ, võ: 武 wǔ 'martial'
- đông, thặng: 疼 téng, tóng 'pain, love'
- thừa, thặng: 乘 chéng, shèng 'make use of, avail oneself'
- trọng, trùng: 重 zhòng, chóng 'heavy, respect'
- đường, đàng: 堂 táng 'hall'
- tràng, trường, trưởng, trướng: 長 cháng, zhǎng 'long, grow, senior'
- lợi, lãi: 利 lì 'benefit'
There are also irregularities in modern Mandarin reflexes that produce unexpected or even amusing phonetic outcomes, deviating from otherwise regular patterns. These shifts are the result of internal sound change, loss, and reanalysis from Middle Chinese into Mandarin. Representative cases include:
- luật, suất: 率 lǜ 'lead, rate'. Standard Sino‑Vietnamese readings are suất (from MC ʂwit) and luật (from MC lwit). In Vietnamese usage, suất often means 'portion, part', while luật is used in the sense of 'rule, law'. Colloquial forms such as sốt 'at all' and suốt 'throughout' likely trace back to the same root. Cf. 比率 bǐlǜ → SV tỷsuất, VS tỷ lệ, xác suất 'rate, ratio'.
- cáp, hạp, hợp, hiệp: 合 hé 'join, unite, close'. Multiple Sino‑Vietnamese reflexes (cáp, hạp, hợp, hiệp) reflect different layers of borrowing. Cf. 哈爾濱 Hā'ěrbīn → SV Cápnhĩtân 'Harbin'.
- tịch: 席 xí 'mat, banquet'. VS tiệc 'banquet', chiếu 'mat', and chủxị 'host' (from 主席 zhǔxí, SV chủtịch 'chairman') all derive from this root.
- thoại: 話 huà 'speech'. Standard SV is thoại, while VS nói 'speak' represents a divergent reflex.
- thuyết: 說 shuō, shuì 'speak, explain'. SV thuyết corresponds to VS thốt 'utter'. The root also developed extended meanings such as 'to halt, rest overnight' (Mand. shuì) and 'to delight in' (Mand. yuè 悅).
- thừa, thừng: 承 chéng 'hold, undertake, receive'. SV thừa and VS thừng coexist, alongside VS nhận 'receive' and dâng, nâng 'lift up'. Related compounds include 承受 chéngshòu → VS chịuđựng 'endure', 丞相 chéngxiàng → SV thừatướng 'minister', 剩餘 shèngyú → SV thặngdư 'surplus', VS dư thừa 'leftover', and 大乘 Dàchéng → SV Đạithừa 'Mahayana'.
In fact, many Vietnamese words are formed in a similar way and, since the majority are Chinese loanwords, an analytic approach can be applied to trace the etymology of Sinitic‑Vietnamese lexicons back to both the Sino-Vietnamese <~ Middle Chinese. This involves associating their meanings with synonymous lexemes that are close both in phonological shape and semantic value, for example,
- thiệnlương: 善良 shànliáng (VS hiềnlương > hiềnlành, 'good character'), associated with 賢良 xiánliáng (VS hiềnlương).
- caothượng: 高尚 gāoshàng (VS caosang, 'high-class'), in concurrent usage with 清高 qīnggāo (thanhcao) 'noble'.
- cốkế: 估計 gùjì (VS dòchừng, 'estimate').
- châuniên: 周年 zhōunián (VS đầynăm) for thôinôi 'baby's first birthday shower', in addition to chuniên 'anniversary'. (Not to be confused with 停(搖)籃 tíng(yáo)lán; in any case, there was no 'cradle' in ancient times, and 籃 lán 'basket' or VS rỗ is a later development for the same reason.)
- toạnguyệttử: 坐月子 zuòyuèzi (SV toạnguyệttử) for nằmđầytháng. In Vietnamese, the connotation is equally applied to 'baby's one‑month shower'. Etymologically, the compound means 'puerperium', i.e., the traditional one‑month confinement period following childbirth, a major cultural ritual in Chinese society; in the West, by contrast, it corresponds only to 'maternity leave'.
- thuộcthử: 屬鼠 shúShǔ (VS tuổiChuột, 'born in the year of the rat'). Cf. 屬羊 shǔYáng (tuổiDê, 'year of the Goat'), 屬雞 shǔJī (tuổiGà, 'year of the Rooster'). Here 屬 shǔ is , in association with 歲 suì (SV tuế > VS tuổi 'age').
- hỗnđản: 渾蛋 húndàn (VS khốnnạn, 'son of a bitch'), in association with 困難 kùnnán (SV khốnnạn) vs. VS khókhăn 'difficulty'.
- quánghiện: 過癮 guòyǐn for đãghiền, đãcơn, dãcơn 'satisfy a craving'. Also, VS 'đãquá'. 'quáđã'.
- yêuchiết: 夭折 yāozhé (VS chếtyểu, 'die young').
- yếuphạn: 要飯 yàofàn (VS ănmày, 'beggar'), in contrast of giởcơm 'scoop rice'. Hence, the latter hives rise to VS xúccơm, xớicơm, bớicơm, múccơm.
- tạcnhục: 炸肉 zhàròu (VS chảlụa, 'cooked meat cake'), instead of chảgiò 'fried meatloaf'.
- phìnhục: 肥肉 féiròu (VS bachỉ, 'bacon'). Cf. barọi (五花肉 wǔhuāròu 'fatty pork belly').
The presence of irregularly derived syllabic interchanges allows us to expand and reinforce the sandhi‑association principle when revising the etymology of doublets or etyma that share common roots across numerous cases. This demonstrates that the rigid one‑to‑one correspondences once insisted upon by earlier Sinological schools are no longer tenable.
Such revised reconstruction must be grounded in the distinctive vocalism and articulation of Vietnamese initials and finals, which conform to the phonotactic patterns of native speakers. The validity of any derived paradigm must then be tested against the competence and performance attested in Ancient Chinese phonological and rhyming schemes, reconstructed from Old Chinese materials and refined through the work of modern linguists on Proto‑Chinese.
One of the most striking features is the peculiar labiovelar vocalism in modern Vietnamese, especially in finals ‑wc and ‑wng, preceded by rounded vowels (ɔ‑, o‑, u‑) or a medial ‑w‑ in orthography, i.e. [‑uwk, ‑uwŋ, ‑owk, ‑owŋ]. These are characterized by rounded, labialized codas and closely resemble Old Chinese finals ending in labiovelars *‑kw [kʷ], *‑gw [gʷ], and the labiovelar nasal *‑ngw [‑ŋʷ]. Li Fang‑Kuei, Edwin Pulleyblank, and others independently reached this conclusion.
Illustrative examples include:
- 風 fēng → VS gió 'wind'
- 心 xīn → VS lòng 'heart'
- 痛 tòng → VS đau 'pain'
- 彤 tóng → VS đỏ 'red'
- 生 shēng (further illustration of the same pattern)
Table 4 - Labiovelar finals and Vietnamese reflexes
| Chin. | Pinyin | SV | VS | Gloss | OC reconstruction* |
|---|---|---|---|---|---|
| 風 | fēng | phong | gió | 'wind' | OC *prəm / *pʷəŋ; labiovelar nasal ‑ŋʷ reflected in VS rounded final |
| 心 | xīn | tâm | lòng | 'heart' | OC *səm; nasal coda parallels VS ‑oŋ with rounding |
| 痛 | tòng | thống | đau | 'pain' | OC *l̥ʰoŋs; rounded nasal ‑oŋ > VS ‑au |
| 彤 | tóng | đồng | đỏ | 'red' | OC *doŋ; labiovelar nasal ‑ŋʷ > VS ‑o |
| 生 | shēng | sinh | sống | 'to live, life' | OC *sreŋ; nasal ‑ŋ > VS ‑oŋ with rounding |
| 國 | guó | quốc | nước | 'country' | OC *kʷək; labiovelar ‑kwək > VS ‑ươc |
| 玉 | yù | ngọc | đá quý | 'jade' | OC *ŋok; labiovelar ‑ok > VS ‑á in semantic substitution |
| 光 | guāng | quang | sáng | 'light' | OC *kʷaŋ; labiovelar ‑waŋ > VS ‑áng |
| 黃 | huáng | hoàng | vàng | 'yellow' | OC *gʷaŋ; labiovelar ‑waŋ > VS ‑àng |
| 江 | jiāng | giang | sông | 'river' | OC *kroŋ; nasal ‑ŋ with rounding > VS ‑ông |
*Notes:
-
The Sino-Vietnamese layer (phong, tâm, thống, đồng, sinh, quốc, ngọc, quang, hoàng, giang) preserves the more regularized Sino‑Vietnamese readings.
-
The Sititic-Vietnamese layer (gió, lòng, đau, đỏ, sống, nước, đá quý, sáng, vàng, sông) shows native phonotactic adaptation, often with rounded finals or semantic substitution.
-
The Old Chinese reconstructions (‑kw, ‑gw, ‑ŋw) explain the rounded Vietnamese reflexes, confirming the labiovelar connection.
Table 5 - A case study of Sinitic-Vietnamese neologism
formed
with Chinese lexemes
The Vietnamese term 'côngcuộc' – now familiar in modern discourse as a formal compound meaning 'cause', 'process', or 'undertaking' – is a persistent source of lexical confusion and scholarly intrigue. While often misinterpreted as a Sino-Vietnamese compound mapping straight onto Chinese 公局 or 工局 (Mandarin gōngjú 'public bureau', 'work office'), its correct etymological genesis instead lies in 工作 (gōngzuò, 'task', 'work'), with the element 'cuộc' emerging not from 局 (jú) but from 作 (zuò). The fact that 'cuộc' in Vietnamese phonologically and semantically diverges from both its Sino-Vietnamese dictionary reading 'tác' and its expected Mandarin reflex /zuò/ reflects a network of historical sound change, sandhi assimilation, and semantic-phonetic association – processes that collectively illuminate the complex history of Chinese lexical influence in Vietnam.
The Vietnamese word 'côngcuộc' functions in modern written and spoken Vietnamese to denote a significant collective undertaking – 'project', 'cause', 'the course of' – especially in governmental or historical phrasing (e.g., "côngcuộc khángchiến" 'resistance war', "côngcuộc đổimới" 'the undertaking of renovation/reform'). It is a compound of 'công' (from 工 'work; labor') and 'cuộc'.
The confusion with 公局 or 工局 is understandable, as both 公 and 工 read 'công' in Sino-Vietnamese, and 局 (SV: cục) is a common bound morpheme for official entities. However, 'côngcuộc' is a modern compound built on the model of Chinese 工作 (gōngzuò), but adapted phonetically and semantically within the Vietnamese system. While 'côngtác' is the canonical Sino-Vietnamese reading for 工作, 'côngcuộc' emerged as a neologism where 'cuộc' operates as a native or nativized reflex of 作, rather than 局.
The emergence of Sino-Vietnamese compounds such as 'côngcuộc' reflects longstanding processes of borrowing and semantic adaptation widespread across the Sinosphere, i.e., Japan, Korea, and Vietnam, collectively referred to as the 'Sino-Xenic' realm. In these contexts, new words for modern concepts were often coined using Chinese morphemes and then mapped phonologically into the target language in a regularized, but sometimes innovative, fashion.
Middle Chinese, as represented in rime dictionaries such as Qieyun (7th century), had a richly articulated syllable template. For the character 作 (Mandarin: zuò), used in 工作 (gōngzuò), the reconstructed MC pronunciation is commonly given as */tsak/ or /tsak-s/, with the following features: initial: ts- (voiceless alveolar affricate), vowel and medial: /a/ as nucleus, sometimes with a palatal medial in some dialects, final: -k (voiceless velar stop), a classic 'entering tone' coda., and one: entering (rusheng), which has phonological and tonal correlates in Sino-Vietnamese readings for 作 are systematically 'tác', tracing the regular sound correspondences established for Chinese readings in Vietnamese. Key observations:
-
The initial [ts-] to [k-] shift is irregular (i.e., not predicted by the regular SV correspondence), suggesting non-Sino-Vietnamese, perhaps colloquial or nativized, development.
-
Labiovelar final [‑əwkpʔ] is robustly preserved in 'cuộc', with the final -k and medial -w- (from /ua/ or /uə/) mapping closely to MC -ak, and aligning phonotactically with native Vietnamese coda structure.
-
The resultant tone is nặng [˧ˀ˩ʔ], consistent with the entering (rusheng) tone category linked to -k finals in Han-Viet transmission.
-
'cục' SV: cục, Mandarin jú, MC guawk; used for administrative, governmental, and physical 'units' or 'offices'
-
'cuộc', derived via the above pathway from 作, but due to similar form and function, is often reanalyzed by speakers and writers as rooted in 局, especially in compounds
The confusion is exacerbated by the convergence of rimes and finals, both 'cục' /kʊkpʔ/ and 'cuộc' /kəwkpʔ/ conforming to the [k•w•k•p̚] structure, with heavy final closure and possible central or back rounded vowels. *
Semantic blending in compound formation: semantic overlap also drives this folk associationIn both Sinitic and Vietnamese, compounds involving 工作 (work), 局 (office), and 作 (to do/make) are semantically related to tasks, operations, or affairs, domains where 'cuộc' has come to be used.
For example, in classical Chinese, 局 (jú) denoted physical bureaus ('bureaus', 'games') and by extension 'affairs' or 'situations' and 作 (zuò) in compounds implied the 'doing', 'working, or 'citing upon' something: matching the function of 'cuộc' in in syntagms such as "côngcuộc vậnđộng" 'the campaign task'.
Consequently, the phonetic resemblance between 'cuộc' and 'cục' enables semantic-phonetic association (lexical contamination or 'folk etymology'), especially when context or classical literacy is limited.
Evidence for such reconstruction is found in the Shijing 詩經 (Book of Odes), where 風 fēng regularly rhymes with 心 xīn, 林 lín, and related items. These belong to the 侵 Middle Chinese /tshjəm1/ rhyme group and 東 MC djung [< Old Chinese *djəŋʷ], cf. SV lòng [lɔŋʷ2], Division III (with medial ‑j‑). Yu Nai‑yong (1985: xiii, 277 – 79, 286) classified them similarly, though in classical Chinese they end in /‑m/. It is noteworthy that Vietnamese words in this class ending with /‑ŋʷ/ are articulated with a wide range of initials, making the connection with lòng [lɔŋʷ2] or [lɔwŋm2] straightforward. His reconstruction of Proto‑Chinese and Old Chinese 風 as *pljom > pljəm and 嵐 as *plom > bləm rests on xiéshēng 諧聲 evidence, which shows divergent Middle Chinese initials [piuŋ] and [lam].
Other scholars have reached similar conclusions with minor variations. Bodman (1980:121) proposed Proto‑Chinese pyəm, Old Chinese pjəm, and Middle Chinese pjuŋ for 風, noting the inter‑rhyming of *‑əm, *‑əŋ, and *‑uŋ finals in OC. He suggested that *‑uŋ may have been a dialectal reflex of *‑əm. Schuessler (1987:385) revised Li's OC reconstruction of 林 as gljəm. Forrest (1958:114) observed that archaic Chinese tolerated consecutive labials (initial /P‑/ with final /‑M/), and concluded that the OC ending of 風 must have matched that of 心, i.e. /‑m/.
From this perspective, we may posit an interchange /‑jOm/ ~ /‑jOŋʷ/, based on the hypothesis that OC ‑jəm shifted to MC ‑jung through a stage of labialization, yielding /‑juŋʷ/ or /*‑juwŋm/. This phonemic feature still surfaces in Vietnamese. Pulleyblank (1984:123) similarly represented final /‑uŋ/ as /‑əŋʷ/, hypothesizing that the OC final was pronounced like Vietnamese ông [oŋʷ] and ong [əŋʷ], with double articulation (labial + velar).
For the Middle Chinese period, Forrest noted that /‑ung/ remained stable except after labials, where it dissimilated to /‑ə‑/, as in Mandarin 風 fēng (p.182). From Pulleyblank's and Forrest's views, we may infer that Vietnamese giông developed during the transitional Early Middle Chinese stage: /p‑/ was palatalized and dropped from /pjuŋʷ/, leaving a glide /j‑/. This produced [juŋʷ] > [joŋʷ], with rounded nasalized labiovelar codas – difficult for northern Chinese speakers – which then reduced to nasalized [jõ] > [jɔ5], yielding modern giông and gió. A parallel development gave rise to SV phong [fəŋʷ].
This postulation [風 *pjuŋ > giông > gió] is plausible because Vietnamese tends to resist /p‑/, substituting it with b‑, ph‑, h‑, j‑, nh‑, or palatalizing it to t‑, s‑. At times the rounded labiovelar is dropped, leaving /‑w/ or /‑o/, as seen in 痛 tòng (VS đau 'pain'), 銅 tóng (VS thau 'bronze'), and 彤 tóng (VS đỏ 'red'). Such processes occur not only in Vietnamese but also within Chinese dialects.
On this basis, we may assume that Proto‑Vietic and Old Vietnamese (ancient Annamese) giông had phonetic values close to those of Proto‑Chinese and Old Chinese. The form gió may represent a local innovation or alternation, with the labiovelar reduced to ‑ɔ. Interestingly, in SV phong [pfɔŋʷ1], the initial alternates between b‑ and f‑, both labials, which could have yielded /j‑/. Consecutive labials were a feature of Old Chinese, and Sinitic‑Vietnamese giông preserves this archaic trait, even as later Chinese developed what Forrest called a "distaste for consecutive labials."
The historical correlation between Vietnamese and Chinese in the case of /‑juŋʷ/ shows syncopation into /‑uw/, /‑m/, /‑Ø/, or /‑ŋ/ ~ /‑Ø/. This pattern extends to many other correspondences, for example:
- 捅 tǒng (SV thống) → VS đâm 'stab'
- 痛 tòng (SV thống) → VS đau 'pain'
- 銅 tóng (SV đồng) → VS thau 'bronze'
- 公 gōng (SV công) → VS cồ 'hen'
- 蟲 chóng (SV trùng) → VS sâu 'insect'
- 彤 tóng (SV đồng) → VS đỏ 'red'
- 夢 mèng (SV mộng) → VS mơ 'dream'
These examples illustrate the broader interchange patterns linking Old Chinese finals with Vietnamese reflexes, and similar pattern /-ŋ/, /n-/ ~ /-Ø/ such as
- 打 dǎ < đánh 'strike'
- 道 dào < đường 'road'
- 鵝 é < ngỗng < ngang 'goose'
- 而 é < mà 'that'
- 腹 fú < bụng 'belly'
- 抱 bāo < bồng 'carry in one's arms'
- 寒 hán < cóng 'freezing'
- 里 lǐ < làng 'village'
- 林 lín < rừng 'forest'
- 溜 liù < lặn 'slip away'
- 逆 nì < ngược 'contrary'
- 葩 pā < bông 'flower'
- 怯 qiè < nhát 'timid'
- 柚 yóu < bòng < bưởi 'pomelo'
- 禺 yú < ngung 'a kind of monkey'
- 心 xīn < lòng 'heart (figurative)'
These etymological interchanges highlight the deep relationship between Vietnamese and Chinese, allowing us to trace parallel lines in the historical development of both languages. In Chinese, for example, Mandarin offers numerous cases of dropped phonetic endings and tonal reductions.
From this established baseline, it becomes possible to reconstruct many Old Chinese initials and finals and to build analogies for Chinese – Vietnamese sound change patterns. Such patterns help identify additional Vietnamese etyma of Chinese origin and reveal words shaped by multiple phonemic shifts. Loanwords in the Sinitic‑Vietnamese stratum were adapted to local speech habits and phonotactic constraints, or reshaped under the influence of colloquial usage and substrate interference through imperfect learning or imitation – phenomena often described in terms of "competence and performance."
Inevitably, confusion arises in tracing etymological roots, whether through corruption, contamination, or semantic masking. Extended usages can obscure the original connection, as illustrated in the following examples:
- 吃 chī: SV ngật → VS xơi 'eat' [cf. 乙 yǐ (SV ất) < 食 shí (SV thực) 'eat'] in place of VS ăn [唵 ān: SV ảm 'eat'].
- 川 chuān: SV xuyên → VS dòng, sông 'stream' [cf. 江 jiāng: SV giang 'river'; 水 shuǐ: SV thuỷ 'water, river'. Ex. 湘水 Xiāngshuǐ → VS SôngTương (Xiang River); 渭水 Wèishuǐ (SV Vịthuỷ) → VS SôngVịthuỳ (Wei River); 漢水 Hànshuǐ → VS SôngHán (Han River). Contrast with 泉 quán: SV tuyền → VS suối 'spring, creek'].
- 煩 fán: SV phiền → VS buồn 'disturbed' [cf. 悶 mèn: SV muộn 'sad'] instead of VS bực 'annoyed'.
- 師 shī: SV sư → VS thầy 'master' [ex. 官師 guānshī → VS quanthầy 'colonialist'; 師徒 shītú → VS thầytrò 'teacher and pupils'; cf. 巫師 wūshī → VS thầymô 'wizard' ~ phùthuỷ 'shaman']. Contrast with VS sãi 'monk' for the same word 師, besides its sense 'army division'.
- 屁 pì: SV tý → VS đít 'buttocks' [cf. 腚 dìng: SV định → VS đít. Note: 屁股 pìgǔ gave rise to VS phaocâu 'chicken butt'] instead of VS địt 'fart, nonsense'.
- 鳳凰 fènghuáng: SV phượnghoàng 'phoenix' → VS phượnghồng 'flame tree (Delonix regia)'. Here the morpheme 凰 huáng was associated with 紅 hóng (SV hồng), producing a contamination that conveniently distinguishes VS phượnghồng from SV phượnghoàng.
- 地帶 dìdài: SV địađái → VS dãiđất 'stretch of land', as opposed to 土地 tǔdì: SV thổđịa → VS đấtđai 'land'.
- 太陽 tàiyáng: SV tháidương 'the sun' → VS mặttrời, instead of VS trờinắng 'sunshine'.
- 行將 xíngjiāng: SV hànhtương 'about to' → VS sắpsửa, instead of VS sẽmau (cf. 快要 kuàiyào 'be going to').
- 明年 míngnián: SV minhniên 'new year' → VS nămmới, as opposed to VS sangnăm 'next year'.
- 去年 qùnián: SV khứniên 'last year' → VS nămngoái, alongside 往年 wǎngnián 'previous year' → VS nămxưa, which may also be rendered as nămngoái.
Conclusion
Sound change operates according to regular principles, yet its outcomes are often irregular and unpredictable.
Vietnamese illustrates this dynamic vividly: borrowed roots are continually reshaped into a distinctive phonological and semantic system, producing forms that diverge from their Chinese sources while remaining deeply connected to them.
The study of these sound change patterns reveals more than linguistic mechanics – it uncovers the cultural creativity through which languages adapt, transform, and sustain their expressive power across time.
References
-
Alves, Mark J. 2017. "Identifying Early Sino‑Vietnamese Vocabulary via Linguistic, Historical, Archaeological, and Ethnological Data." Bulletin of Chinese Linguistics 10(1): 1 – 28.
-
Alves, Mark J. 2018. "Sino‑Vietnamese Grammatical Vocabulary and Sociolinguistic Conditions for Borrowing." Asia-Pacific Linguistics.
-
Bodman, Nicholas C. 1980. Proto‑Chinese and Sino‑Tibetan: Data towards Establishing the Nature of the Relationship. Journal of Chinese Linguistics Monograph Series.
-
Forrest, R.A.D. 1958. The Chinese Language. London: Humanities Press.
-
Haudricourt, André‑Georges. 1954. "De l'origine des tons en vietnamien." Journal Asiatique 242: 69 – 82.
-
Karlgren, Bernhard. 1945. The Chinese Language: An Essay on Its Nature and History. Stockholm: Museum of Far Eastern Antiquities.
-
Karlgren, Bernhard. 1954. Compendium of Phonetics in Ancient and Archaic Chinese. Stockholm: Museum of Far Eastern Antiquities.
-
Li, Fang‑Kuei. 1971. Studies on Archaic Chinese Phonology. Taipei: Academia Sinica.
-
Lanneau, Grainger. 2025. A Re‑Examination of Regular, Unique and Unusual Sino‑Vietnamese Initial Features. PhD dissertation, University of Washington.
-
Nguyễn Tài Cẩn. 1979. Nguồn gốc và quá trình hình thành cách đọc Hán‑Việt. Hà Nội: Nhà xuất bản Khoa học Xã hội.
-
Phan, Trang, Tuan‑Cuong Nguyen, and Masaaki Shimizu, eds. 2024. Studies in Vietnamese Historical Linguistics: Southeast and East Asian Contexts. Singapore: Springer.
-
Pulleyblank, Edwin G. 1984. Middle Chinese: A Study in Historical Phonology. Vancouver: University of British Columbia Press.
FOOTNOTES
(1)^ Watch "Zhao Tuo" and "Triệu Đà" on Youtu.be.
(2)^ This event was told in the Vietnamese folktale Lạclongquân 雒龍君 ("King Lac of Dragonic Descent"); it is about the origin of the Vietnamese people.
(3)^ See Bo Yang for China's history, Wang Li for OC and vernacular Vietnamese, Kargren for pioneering in reconstruction of OC historical phonology, Li Fang-Kuei for OC reconstruction, Pulleyblank for phonology of Early Mandarin, Schuessler for Qin-Han phonology reconstruction, Kangxi Zidian 康熙字典 for vernacular variants of many uncommon Chinese characters which are cognate to many Vietnamese words.
(4)^ To understand how that could possibly be, as previously discussed, compare the similar models the have made up the people and their language around the world after their countries had been colonialized for hundreds of years, such as Inner Mongolia, Tibet, Uyghur, Taiwan, Hainan, Guangdong, Hong Kong, Guam, the Philippines, Singapore, Malaysia, Australia, New Zealand, Scotland, Ireland, South Africa, Morocco, Hawaii, North and South America, Cuba, Haiti, etc.
(5)^ If Vietnam had still remained as a colony of China like Guangdong, Fujian, and even Zhejiang, Jiangsu provinces, the 'Viet dialect" would have been certainly classed as of Sino-Tibetan linguistic family like all of the above for sure. If it were so, for the needs of bartering, selling, or trading, the Annamese speakers would have pretty well adapted the Chinese numbers if they had not already done so. Form this rationalization, we could see that cognacy in numerals is not a big deal as previously discussed, that is, to use them to determine