How Wu Chinese packages meaning
Wu Chinese grammar at a glance
At a glance
| Information type | Plain question | How Wu Chinese encodes it |
|---|---|---|
| Actor | Who does the action? | SEPARATE WORDCONTEXT |
| Recipient | Who or what receives the action? | WORD ORDERSEPARATE WORD |
| Time | When does the action happen? | CONTEXTSEPARATE WORD |
| Aspect | Finished or still happening? | SEPARATE WORD |
| Certainty / mood | Real fact, or desired / possible / hypothetical? | SEPARATE WORD |
| Negation | Did it not happen? | SEPARATE WORD |
| Voice | Is the subject doing, being done to, or causing? | SEPARATE WORD |
| Evidentiality | How does the speaker know this? | SEPARATE WORD |
| Specificity | This particular one, or any one? | CONTEXTSEPARATE WORD |
| Quantity | How many? | CONTEXTSEPARATE WORD |
| Possession | Whose is it? | SEPARATE WORD |
| Location / Space | Where? In which direction? | SEPARATE WORDSERIAL VERB |
| Information structure | What is this sentence about? What is new? | WORD ORDERSEPARATE WORD |
| Register | What social relationship does the speaker encode? | CONTEXT |
In detail
Actor in Wu Chinese — Who does the action?
A separate pronoun identifies who acts. The verb never changes — 讲 gang (speak) stays 讲 whether the subject is I, you, or they. The pronouns themselves are completely different from Mandarin: 我 ngu (I), 侬 nong (you), 伊 yi (he/she/they/it).
The verb 讲 is identical in all three sentences — no suffix, no prefix, no change of any kind. The pronoun alone distinguishes who acts.
Wu pronouns are completely different from Mandarin: 我 is pronounced ngu (not wǒ), the second-person pronoun is 侬 nong (not 你 nǐ), and the third-person pronoun is 伊 yi (not 他/她 tā). The first-person plural is 阿拉 aq-laq (not 我们 wǒmen) — a form originally from the Ningbo dialect.
The third-person pronoun 伊 yi makes no gender distinction at all — not even in writing. Mandarin distinguishes 他 (he), 她 (she), 它 (it) in writing though they sound identical. Wu uses 伊 for all genders with no distinction anywhere.
Pronouns are frequently dropped when clear from context. "讲上海闲话伐?" (Speak Shanghainese?) is natural with no pronoun at all.
Compare: Mandarin uses 你 nǐ / 他 tā / 我们 wǒmen — entirely different words. Despite sharing a writing system, Wu pronouns are as different from Mandarin as Portuguese pronouns are from Spanish.
Recipient in Wu Chinese — Who or what receives the action?
Word order carries the basic subject–object distinction (SVO). The verb 拨 pəʔ handles dative constructions ("give to") — the same verb also does duty as passive marker and disposal marker, a triple function unique to Wu.
The verb 拨 pəʔ is Wu's most versatile grammatical tool — it serves as the main "give" verb AND as a passive marker AND as a disposal marker (like Mandarin 把). One word does the work of three Mandarin words (给, 被, 把). This polyfunctionality of GIVE verbs is a hallmark of Southern Sinitic languages.
Double-object constructions reveal a generational split: older speakers say "拨 DO IO" (give paper me — direct object before indirect), while younger speakers say "拨 IO DO" (give me paper — Mandarin word order). This shift directly measures Mandarin influence on Shanghainese grammar.
Disposal constructions (moving the object before the verb to signal it is "dealt with") also use 拨 — unlike Mandarin, which uses the separate word 把 bǎ.
Nouns carry no case markers — position alone determines grammatical role, same as Mandarin.
Compare: Mandarin uses three separate function words — 给 gěi (give), 被 bèi (passive), 把 bǎ (disposal) — each specialized for one grammatical role. Wu collapses all three into the single verb 拨 pəʔ.
Time in Wu Chinese — When does the action happen?
The verb has no tense forms — it never changes for past, present, or future. Time words or context carry the temporal meaning, same as Mandarin, but Wu uses different time vocabulary: 昨日 (yesterday), 明朝 (tomorrow).
The verb 看 is absolutely identical in all three sentences. No suffix, no auxiliary, no tone change — nothing on the verb signals time.
Time words appear before the verb, after the subject — same rigid word order rule as Mandarin. But the time words themselves differ: 昨日 (yesterday, vs. Mandarin 昨天), 明朝 (tomorrow, literally "bright morning," vs. Mandarin 明天).
When no time word is present, the time frame is inferred from context or aspect particles — the perfective marker 了 leq can suggest a completed event, but it marks completion, not past tense.
Some scholars have argued that certain sentence-final particles in Shanghainese carry tense-like meanings — the SFP 了 leq may signal "current state of affairs" in ways that go beyond Mandarin's 了. This remains debated.
Compare: Both Wu and Mandarin are tenseless — the verb never changes for time. This contrasts with Spanish, which has distinct verb suffixes for past (hablé), present (hablo), and future (hablaré). Even the time words differ between the two Sinitic languages: Wu 昨日/明朝 vs. Mandarin 昨天/明天.
Aspect in Wu Chinese — Finished or still happening?
Aspect particles mark the internal structure of events: 了 leq (completion), 辣 laq before the verb (in progress), 过 gu (past experience). Shanghainese is "richer in aspect particles than Mandarin," with six or more distinct markers.
了 leq functions both as a post-verbal completion marker and as a sentence-final particle — two distinct morphemes sharing the same form, just as in Mandarin. As an SFP, it can signal "current state of affairs" or "change of state."
辣 laq is the most distinctively Wu aspect marker. It appears PRE-verbally to mark progressive action (like Mandarin 在 zài). But the same morpheme also appears POST-verbally as a resultative marker. This positional polyfunctionality — same word, opposite meaning depending on position — is a remarkable feature of Shanghainese grammar.
辣 laq has extended variants: 辣辣 laq-laq and 辣海 laq-he (literally "at-sea/surface"), which are more emphatic or locative progressive forms. The 辣海 form likely derives from a locative construction that grammaticalized into an aspect marker — the same path Mandarin 在 followed.
"Shanghainese is richer in aspect particles than Mandarin, and has some very precise particles that convey tense when placed in isolation." Zhu (2006) identifies at least six aspect categories. The system includes perfective, progressive, durative, experiential, resultative, and directional aspects — more fine-grained than Mandarin's four-particle core.
Compare: Mandarin uses 在 zài (pre-verbal progressive), 了 le (completion), 着 zhe (durative state), 过 guo (experience). Wu uses 辣 laq instead of 在, with the bonus that 辣 does double duty as a resultative marker when placed after the verb — a positional trick Mandarin's 在 cannot do.
Certainty / mood in Wu Chinese — Real fact, or desired / possible / hypothetical?
Sentence-final particles (SFPs) carry nuance: 伐 va (yes/no question), 啊 aa (exclamation — with up to 17 attested meanings), 哪 nao (certainty / "I told you so"), 了 leq (change of state). Shanghainese has 10–12 SFPs that can stack in fixed order.
伐 va is the primary yes/no question particle — it replaces Mandarin 吗 ma entirely. It derives from a contraction of 勿啊 "or not," making its etymology transparent: "is it or isn't it?"
Shanghainese SFPs can stack in fixed order to combine meanings. "了伐" (leq va) stacks completion + question: "has X happened?" The ordering is rigid — 了啊 is grammatical but 啊了 is not. Mandarin allows less stacking than Shanghainese.
The particle 啊 aa is "highly versatile, with as many as seventeen attested meanings" — exclamation, appraisal, asking for choices, rhetorical questions, denoting suspicions, and confirmations. It is the most functionally loaded particle in the system.
哪 nao signals certainty, reminder, or warning — "I told you so" or "as you should know." This quasi-evidential function has no direct equivalent in Mandarin.
Wh-questions place the interrogative word in situ (same position as the answer would occupy), same as Mandarin: 侬个钥匙辣啥地方 (your keys are at what-place) — no movement. Shanghainese "what" is 啥 sa, "where" is 啥地方 sa-difang, "who" is 啥人 sa-ning.
Compare: Mandarin uses 吗 ma for yes/no questions and has about 6 core SFPs. Shanghainese uses 伐 va and has 10–12 SFPs with richer stacking possibilities. Japanese takes sentence-final particles even further, with dozens that also encode register.
Negation in Wu Chinese — Did it not happen?
Four negation words, each with its own domain: 勿 vəʔ (general/habitual), 嘸没 m̩-məʔ (existential/perfective), 勿曾 vəʔ-zəŋ (not yet), and 覅 viəʔ (don't! — negative imperative). This four-way split is much richer than Mandarin's two-way 不/没 system.
勿 vəʔ is the general negator — it negates states, habits, intentions, and the future. It corresponds historically to Mandarin 不 bù but the pronunciation is completely different. It can negate the copula 是 "to be" but CANNOT negate 有 "to have" — for that you need 嘸没.
嘸没 (sometimes shortened to 没 məʔ by younger speakers) negates existence and completed actions — "there is no X" and "X did not happen." It historically relates to 無 wú (classical Chinese "not have"). This is the semantic equivalent of Mandarin 没有 méiyǒu but the form is distinctly Wu.
勿曾 vəʔ-zəŋ (literally "not-once/formerly") is the "not yet" marker — a compound of the general negator 勿 with the adverb 曾. It corresponds to Mandarin 还没有. This form is "apparently rarely used by" younger speakers — a piece of Wu grammar being lost to Mandarin influence.
覅 viəʔ is a fused form of 勿 + 要 (not + want) that has become a single morpheme meaning "don't!" The character 覅 is a Wu-specific creation. Mandarin uses 别 bié or 不要 búyào for the same function, but they remain two syllables.
勿 also appears inside potential complement constructions: 困得着 (able to sleep) vs. 困勿着 (unable to sleep) — the negative form inserts 勿 where Mandarin would use 不. This V-得/V-勿 pattern is productive in Shanghainese.
Compare: Mandarin splits negation two ways — 不 bù (general) vs. 没 méi (perfective/existential). Shanghainese splits it four ways, adding dedicated markers for "not yet" (勿曾) and prohibition (覅). Turkish, by contrast, uses a single negative suffix -me/-ma for all contexts.
Voice in Wu Chinese — Is the subject doing, being done to, or causing?
Passive voice uses the GIVE verb 拨 pəʔ before the agent — the same verb that means "give" and marks disposal. This triple-duty polyfunctionality is a defining feature of Wu and other Southern Sinitic languages.
The 拨 passive, like the Mandarin 被 passive, traditionally carries a negative or adversative meaning — something undesirable happened to the subject. "Adversity passives are widespread in Sinitic languages." This give-verb-as-passive pattern is found across Southern Sinitic: Cantonese uses 畀 bei² in the same way.
The agent can be omitted in a 拨 passive, creating a short passive where the doer is left unsaid.
拨 as disposal marker moves the object before the verb to signal it is being "dealt with" — the exact function of Mandarin 把 bǎ, but performed by the same word as the passive marker. In Mandarin, 被 (passive) and 把 (disposal) are completely separate words. In Wu, they are one and the same verb 拨.
This polyfunctionality — one GIVE verb serving as lexical verb, passive marker, AND disposal marker — represents a major grammaticalization pathway: give → benefactive → passive/disposal.
Compare: Mandarin keeps passive (被 bèi) and disposal (把 bǎ) as separate function words, each specialized for one task. Wu collapses both into 拨 pəʔ, which also remains a full lexical verb meaning "give." This is like having one word serve as both "by" and a preposition for affected objects.
Evidentiality in Wu Chinese — How does the speaker know this?
Not grammatically obligatory, but several sentence-final particles carry quasi-evidential meanings — 哪 nao ("I told you so" / reminder of shared knowledge), 佬 lao (assertion based on common ground), 哇 wa (surprise at newly perceived information).
Shanghainese does not require speakers to mark how they know something — evidentiality is never obligatory. But the rich SFP inventory provides an optional quasi-evidential system that goes beyond what Mandarin offers.
哪 nao signals certainty based on prior knowledge — "as you should already know" or "I told you so." It functions as a reminder that the information was previously shared, close to evidential marking for "established fact."
佬 lao asserts something based on the speaker's personal knowledge, rumors, or past experiences — a claim of epistemic authority.
哇 wa marks mirativity — the speaker's surprise at newly perceived or unexpected information. Mirative markers are considered a category related to but distinct from evidentiality.
Compare: Mandarin has no SFPs with these quasi-evidential meanings — 吧 ba (suggestion/guess) is the closest. Turkish grammatically requires evidential marking: -di (witnessed) vs. -miş (inferred/hearsay). Shanghainese sits between: richer than Mandarin, but still optional.
Specificity in Wu Chinese — This particular one, or any one?
No articles — bare nouns are ambiguous between "a book" and "the book." But the demonstrative 搿 geq is evolving toward a definite article: "搿搿 noun" means "the noun" and is felicitous in definite contexts even without prior mention.
Like Mandarin, Shanghainese has no equivalent of English "the" or "a." But the demonstrative 搿 geq may be evolving toward a definite article: "搿搿 noun" is felicitous in definite contexts even without prior mention of the referent, as long as it is "uniquely available" — a function closer to "the" than to "this."
搿 geq shows remarkable triple homonymy — the same form serves as: (1) near demonstrative "this," (2) subordinate/possessive particle (= Mandarin 的), and (3) general classifier/measure word. Context resolves which reading is intended.
Shanghainese has two proximal demonstratives in competition: the older diəʔ (traditional "this") and the newer geq/搿 (emerging "this"). The distal demonstrative ("that") is phonologically derived from the proximal by a tone change — a pattern found across Wu dialects.
Word position signals definiteness pragmatically: nouns in subject position tend to be understood as definite; nouns after the verb as indefinite — same as Mandarin.
Compare: Mandarin demonstratives (这 zhè, 那 nà) do not show signs of developing into articles. Shanghainese 搿搿 is closer to a definite marker — an incipient grammaticalization that Mandarin has not undergone. Spanish has a full article system (el/la/un/una) with gender agreement.
Quantity in Wu Chinese — How many?
No plural suffix on nouns. Number is inferred from context or stated with a numeral + classifier. The general classifier 隻 tsəʔ is the most distinctively Wu choice — it has a wider range than Mandarin's 个 gè.
Classifiers are required between numerals and nouns — you cannot say "三书" (three-book). Each noun type has a classifier categorizing it by physical properties, same as Mandarin.
Shanghainese has two general classifiers in competition: 隻 tsəʔ (the traditional Wu default, used for humans, animals, and abstract nouns — wider range than Mandarin's 个) and 個 gəʔ (cognate with Mandarin 个 gè). Younger speakers increasingly substitute 個, "reducing dialect-specific applications by approximately 40% in informal contexts."
Over 50 specialized classifiers exist — 条 for long things, 张 for flat things, etc. But the choice of DEFAULT classifier reveals whether a speaker uses "pure" Shanghainese (隻 tsəʔ) or Mandarin-influenced Shanghainese (個 gəʔ).
Pronoun plurals use suppletive forms: 我 ngu → 阿拉 aq-laq (we), 侬 nong → 伊那 na (you all), 伊 yi → 伊拉 yi-laq (they). These are completely different words, unlike Mandarin which adds the suffix 们 -men.
Compare: Mandarin defaults to 个 gè as its general classifier. Shanghainese traditionally defaults to 隻 tsəʔ, which covers a wider semantic range. The shift toward Mandarin's 个 among younger speakers is a measurable indicator of language contact and shift.
Possession in Wu Chinese — Whose is it?
The particle 个 gheq links possessor to possessed: 我个书 ngu gheq su (my book, literally I-GEQ-book). This is the Wu equivalent of Mandarin's 的 de — same function, different form.
个 gheq serves the same linking function as Mandarin's 的 de — connecting any modifier to a noun (possessive, adjectival, relative clause). But it is also identical in form to the general classifier 個 and the demonstrative 搿 — a triple homonymy unique to Shanghainese.
For close personal relationships, 个 may be dropped: "我妈妈" (my mom), "侬弟弟" (your younger brother) — same pattern as Mandarin.
The character 个 is used for the genitive particle in Wu, while Mandarin uses 的. They have completely different etymologies — Wu 个 relates to the classifier, while Mandarin 的 has a different historical source. This is one of the clearest grammatical differences between the two languages.
Like Mandarin, Shanghainese makes no alienable/inalienable possession distinction — the same 个 construction is used for "my book" and "my hand."
Compare: Mandarin uses 的 de as its general linker. Shanghainese uses 个 gheq. Despite sharing a writing system, these are etymologically different words — a reminder that Wu and Mandarin are genuinely different languages, not just "dialects."
Location / Space in Wu Chinese — Where? In which direction?
The coverb 辣 laq marks static location — Wu's equivalent of Mandarin 在 zài. This is the same morpheme that also serves as a progressive aspect marker and resultative marker, making it one of the most polyfunctional words in Shanghainese.
辣 laq is one of the most remarkable words in Shanghainese — the same morpheme functions as: (1) locative coverb "at/in" (= Mandarin 在), (2) progressive aspect marker when before the verb, and (3) resultative aspect marker when after the verb. This three-way polyfunctionality based on syntactic position is typologically notable.
The grammaticalization path from locative → progressive is well-documented cross-linguistically: "being at a place" → "being in the process of." Mandarin 在 followed the same path. But 辣 went further by also developing a post-verbal resultative use that Mandarin 在 never acquired.
Localizers (里 "inside," 上 "on," 旁边 "beside") follow the place noun to specify precise location, same as Mandarin: 牢监里 (jail-inside = inside the jail). The preposition 辣 is head-initial, but the localizer is head-final — the same mixed headedness found in Mandarin.
Directional complements after the verb encode path and perspective: 进 (enter) + 来 (come toward speaker) or 去 (go away from speaker). This serial-verb-like chaining system is shared across Sinitic.
Compare: Mandarin uses 在 zài for location and progressive aspect. Wu uses 辣 laq for both — PLUS a third resultative function post-verbally. The same grammaticalization pathway produced different downstream results in the two languages.
Information structure in Wu Chinese — What is this sentence about? What is new?
Wu is topic-prominent: the topic comes first and the rest comments on it. Unlike Mandarin, Shanghainese has a dedicated topicalizer particle 末 mo that explicitly marks the topic boundary — "speaking of X, ..."
The topicalizer particle 末 mo is a distinctive Shanghainese feature with no Mandarin equivalent. It explicitly marks the boundary between the topic and the comment — "speaking of X, ..." — providing a grammaticalized signal that Mandarin expresses only through intonation and position.
末 mo is not a sentence-final particle — it occurs medially, after the topicalized phrase and before the comment. This is an unusual position for a discourse particle in Sinitic languages.
Topic-comment structure is fundamental to both Wu and Mandarin, but Shanghainese provides more explicit grammatical tools for it. The combination of 末 mo and the rich SFP system means Shanghainese has more dedicated information-packaging morphemes than Mandarin.
Like Mandarin, the default topic-marking strategy is simply sentence-initial position — whatever comes first is the topic. 末 mo is optional and adds explicit emphasis that "the following is about this."
Compare: Mandarin marks topics purely through sentence-initial position and intonation — no particle. Shanghainese adds the topicalizer 末 mo for explicit marking. Japanese goes furthest with the dedicated topic particle は (wa) that can mark any noun phrase as the topic.
Register in Wu Chinese — What social relationship does the speaker encode?
No grammaticalized register system at all — no polite pronoun, no honorific verb forms, no speech levels. Politeness is expressed entirely through vocabulary choice, indirectness, and address terms.
Shanghainese has no polite second-person pronoun — unlike Mandarin, which has 您 nín (formal "you"). The pronoun 侬 nong is used with everyone: friends, elders, superiors, strangers. Politeness is never encoded in the pronoun system.
This is a striking absence compared to other major East Asian languages: Mandarin has 你/您, Japanese has an elaborate five-level register system reshaping verb forms and vocabulary, Korean has seven speech levels. Shanghainese has none of this.
Politeness is expressed entirely through pragmatic means: titles and address terms (先生 "Mr./sir," 老爷 "sir/master," the familiarizing prefix 阿 ā-), indirectness, and the choice of certain sentence-final particles that can carry softening or assertive force.
There is a strong generational dimension — younger speakers use a more "Mandarinized" variety (新派 xīn pài) while older speakers maintain more conservative Shanghainese (老派 lǎo pài). But this is sociolinguistic variation, not a grammaticalized register system.
Compare: Mandarin has a minimal register system (你/您 pronoun swap, verb unchanged). Japanese has five or more levels that change verb forms, vocabulary, and sentence-final expressions. Shanghainese has zero grammaticalized register — the flattest politeness system among major East Asian languages.
Common questions about Wu Chinese
How is Wu Chinese different from Mandarin?
What does it mean that Wu preserves voiced consonants?
Is Shanghainese the same as Wu Chinese?
How many tones does Wu have?
Why is tone sandhi a bigger deal in Wu than Mandarin?
Sources for Wu Chinese
The grammatical descriptions on this page are informed by the following published reference and descriptive grammars. Grammatical facts themselves are not subject to copyright; the scholars who documented them deserve attribution.
- Chappell, Hilary & Peyraube, Alain (2016). "A Typological Study of Negation in Sinitic Languages: Synchronic and Diachronic Views." In New Horizons in the Study of Chinese: Dialectology, Grammar, and Philology, pp. 483–534. Hong Kong: CUHK Press.
- Chappell, Hilary & Li, Lan (2016). "Mandarin and Other Sinitic Languages." In The Routledge Encyclopedia of the Chinese Language. London: Routledge.
- Myers, Ethan C. (2015). "Sentence Final Particles in Shanghainese: Navigating the Left Periphery." M.A. Thesis, Purdue University.