He's (very) good / well / fine in Mandarin and Cantonese

« previous post | next post »

When I first started learning Mandarin in 1967, one of the things that troubled me most about Chinese grammar was the fact that when I wanted to say "He's fine / good / well", I couldn't just say tā hǎo 他好 ("he [is] good"), I had to say tā hěn hǎo 他很好 ("he [is] very good", but without really meaning the "very".   That bothered me, because I couldn't understand the function of hěn 很 in the simple sentence tā hěn hǎo 他很好 ("he [is] very good").  My teachers told me not to worry about it, that hěn 很 in these sentences didn't really mean anything.

At least I wasn't saying *tā shì hǎo *他是好 (*"he be good") or *Tā shì hěn hǎo *他是很好 (*"he be very good") like some of my fellow students, who felt the need to insert the copular "is" shì 是, even though hǎo 好 by itself is an adjectival / stative verb, i.e., "is good".

I'm prompted to write about this now because I received the following inquiry from someone in Hong Kong.

There is a phenomenon in Chinese (both Mandarin and Cantonese) that has puzzled me for a while, and I wonder if you can discuss it in the blog at some point or direct me to the academic work that has been done on it, as I don't know where to start looking. The phenomenon concerns adjectives in predicate position. It's wrong to use shì 是 to predicate adjectives to subjects. So to say "I'm happy", it should be wǒ kāixīn 我開心. But this sentence sounds very odd if said in either Cantonese or Mandarin. To make it sound natural, I have to say wǒ hěn kāixīn 我很開心 in Mandarin or ngo5 hou2 hoi1sam1 我好開心 in Cantonese. But hěn 很 or hou2 好 are normally intensifiers. Does this mean that in Chinese you can only say "I'm very happy" but not "I'm happy"? Or do hěn 很 and hou2 好 function like a copula here?

One thing I'm fairly confident of is that, in these sentences, neither hěn 很 nor hou2 好 is functioning like a copula.  But what, then, are they doing?  What is their purpose in such sentences?

I suppose you could say tā hǎo 他好 ("he [is] good"), but I think people would feel that it is abrupt almost to the point of being ungrammatical.  Why is it so necessary to have the hěn 很 there even when it is no longer functioning as an intensifier?  What's it doing there?

I asked several colleagues these questions, and will give their replies below, but first will relay Geoff Pullum's response when I asked him what the different nuances are among these three sentences, all of which might be translations of tā hěn hǎo 他很好:

I'd use "He's good" mostly to mean "He is a morally upright person", and "He's fine" or "He's well" mostly to mean "There's nothing wrong with him healthwise." But people are tending to use "He's good" for the latter meaning as well, making all three synonymous.

From Bob Bauer:

Your correspondent has raised an interesting question regarding the use of hou2 in Cantonese.

What people have told me is that even though hou2 is there in the sentence it does not always have to function as an intensifier, and so it does not necessarily always translate as "very" or "quite".

In fact, this very same issue has been addressed on page 178 of 'the new edition (2011) of Cantonese, A Comprehensive Grammar by Stephen Matthews and Virginia Yip with the following example sentence and comment (which are quoted verbatim):

佢      呢   排        好       開心
keoi5 li1 paai4 hou2 hoi1 sam1

s/he these days very happy

"She's happy these days."

Without hou2 the sentence (keoi5 li1 paai4 hou2 hoi1 sam1) would be incomplete.

[N.B. The Yale romanization used in the original text has been converted to Jyutping]

In this example sentence we clearly see that hou2 has not been translated into the equivalent English "very". I suspect the situation in Cantonese may be just like in English with some intensifiers becoming drained of their intensive function over time. New words and expressions are always being sought to replace old ones that are felt to be inadequate. I think such words come and go, e.g., awesome, super, fantastic, divine, etc.

Having said this, however, I think what really matters is what was the speaker's intonation when uttering the sentence. If the word hou2 were particularly stressed (uttered with greater duration, more loudly and with higher pitch), then I think the translator would be justified in translating hou2 as "very".

From Maiheng Dietrich:

I don't know if I am a good person to answer this question as I am not a trained linguist. In teaching, we generally don't explain this phenomenon at first, and just have the students memorize that in Chinese "I am good" is wǒ hěn hǎo 我很好, not *wǒ shì hǎo *我是好. The hěn 很 does not function as an intensifier but serves a grammatical function. Later when structures of comparison are introduced, we'll bring this up again and give more explanation. Wǒ hǎo 我好 is grammatical and acceptable when used in comparative context, e.g. wǒ hǎo, nǐ huài 我好,你坏 ("I'm good, you're bad").  Jiějiě piàoliang, mèimei bù piàoliang 姐姐漂亮,妹妹不漂亮 ("older sister is good looking, younger sister is not").  Xiāngzào piányí, mùyùyè guì 香皂便宜,沐浴液贵 ("the soap is cheap, but the bath gel is expensive").  A sentence with an adjective as the predicate without the presence of hěn 很 always implies a comparison.  For pedagogical purposes, this explanation is sufficient.

This phenomenon has long been studied in the Chinese linguistics field.  The first time I heard a presentation on this topic was at a conference about 20 years ago.  The presenter was Féng Shènglì 冯胜利, a Penn-trained linguist….

My response to Maiheng:

It's interesting that one can say dōngfāng hóng 東方紅 ("the east is red").  One doesn't have to say dōngfāng hěn hóng 東方很紅 ("the east is very red").  Maybe wǒ hǎo 我好 and tā hǎo 他好 are just too short unless they're situated in a comparative context.

To which Maiheng replied:

Dōngfāng hóng 東方紅 ("the east is red") is poetic, so I think the practices of pure colloquial form don't apply fully.  And I think it could be understood as implying comparison; that's the point of eulogizing Mao, no?

Perry Link:

In my first-year Chinese courses, I have always taught that noun-plus-stative-verb is implicitly comparative.  Zhuōzi gāo 桌子高 ("the table is high") is an implied answer to questions such as zhuōzi gāo háishì yǐzi gāo? 桌子高还是椅子高? ("which is taller, the table or the chair?"), and therefore should be thought of, rigorously, as implicitly comparative: not "the desk is tall" but "the desk is taller."   In order to say just plain "the desk is tall," without the implied comparison, you insert an "unstressed hěn 很".  I call it an "unstressed hěn 很" because it is neutral tone; if you stress it, using full tone, then it adds the meaning of "very".  In sum:  zhuōzi hen gāo 桌子很高 means "the desk is tall"; zhuōzi hěn gāo 桌子高 means "the desk is very tall"; zhuōzi gāo 桌子高 means "the desker is taller (than something else)."

Tom Bartlett:

The question is interesting and significant.  Y. R. Chao's Mandarin Primer, to my recollection, does mention that an adjective as a solitary predicate can imply comparison, but it doesn't explicitly highlight the role of "hen" as indicating non-comparison; or, if that is mentioned, I somehow overlooked it during my very thorough reading of Chao's text over two years' time when learning it and later when teaching it several times.

Rather, my first memory of seeing this non-comparative implication of "hen", before adjective, stated in a succinct form for teaching Mandarin was in early 1989, when I began preparing to use the textbook College Chinese, written by Harvard's Mrs. Lin, based very much on her consultations with people in Beijing.

A propos the remark above about Feng Shengli's awareness of this matter, with all due respect for his very insightful understanding of the grammar and rhetoric of colloquial Chinese, I hypothesize that he may also have had access to the same analyses made by applied grammarians and pedagogues in Beijing when he was a graduate student at Penn.  I met him no earlier than summer 1990, if I remember correctly, and I enjoyed a couple of very pleasant conversations with him in those years, although we didn't touch on this subject, I think.

I haven't taught from Mrs. Lin's book in about 20 years now; however, as I remember, that book makes three points:   1)  As noted below, "Ta hao", regularly (but maybe not absolutely always) implies a comparison:  "he's better (than someone else in some regard.)"   One should always be alert to this implication.  2)  But the question form "Ta hao ma?" and 3)  the negative form "ta buhao" do not require "hen" in the same way and are regularly used without "hen".  My observation is that "hen" is sometimes used in 2) and maybe less so in 3), but if so, it is likely to be the emphasized "hen" which implies "very", i.e. a slightly intensive form of the adjective but w/o implied comparison to someone else (?).

I don't know Cantonese at all, I regret to say, and don't have any compelling reason right now to learn it, while my attention is drawn to other things.  In Taiwanese Mandarin, "hao" is often used in place of "hen"; does that come from a Minnan influence?   When I was preparing to travel to BJ for the first time, a young man from BJ working at Princeton suggested I use "ting" instead of "hao"; he said it would sound more authentically local than "hao", as I remember.  With regard to the comparative question, I suppose "hao" or "ting" has a similar effect to "hen", in that it indicates absence of comparison (?).

Actually, English can have similar implied comparative meaning.  A propos a recent online article about the South China Sea situation, produced by someone writing fluent English from a Beijing research institute, which claimed that the US needed to be seen as supporting a "rules-based order" in the seas, rather than on naval superiority, I found myself ruminating on the famous old saying that Teddy Roosevelt made famous (whether or not he invented it):  "speak softly and carry a big stick".  It implies "carry a bigger stick", which is not insignificant now that China wants to challenge traditional US naval superiority, at least in those waters.

Knowing what I do now, nearly half a century after encountering tā hěn hǎo 他很好 ("he [is] very good", I'm much clearer about the function of hěn hǎo 很 in that superficially simple sentence.  But, already long ago, my initial puzzlement over tā hěn hǎo 他很好 ("he [is] very good" soon gave way to delight when I learned how to say things like:

tā fēicháng hǎo 他非常好

tā hǎo jíle 她好極了

tā hǎo dé bùdéliǎo 她好得不得了

tā hǎo dé hěn 他好得很

— all of which mean "he / she is really / extremely / extraordinarily  good", etc.  After rattling off half a dozen or so of these different forms with great gusto, I would forget all about why tā hěn hǎo 他很好 ("he [is] very good") normally means "He's good / well / fine."


"Prosodic Structure and Bare Adjectives in Chinese"

Shengli Feng

Journal of Beijing Normal University (2003)


In Chinese, bare adjectives in [NP A] structures usually imply a comparison. However, when they occur in parallel sentences the comparative meaning disappears. To date, there is no adequate explanation for why this is so. This paper argues, following Hale and Keyser (2002), that the [NP A] sentences are generated by a conflation construction and that the comparative degree morpheme is nonovert in Chinese; thus, the comparative structure is obtained by conflation of a bare adjective with a zero morpheme. It is further argued that there is a second type of phonologically nonovert head that can only be licensed by prosody; hence, only in prosodically licensed environments can adjectives be in the absolute degree. Given this analysis, the zero copular and semi-copular in Archaic Chinese can also be explained.

[Thanks to Haitao Tang]


  1. James Bradbury said,

    June 15, 2015 @ 2:03 am

    I don't remember exactly where I came across this hypothesis (it was a few years ago, possibly in a minimalist-framework analysis of Mandarin verb constructions…or in an old students' grammar I read around the same time) but I think one possibility is that 很 is only necessary for scalar adjectives, and its presence turns them into proper stative verbs.
    Basically 好 or 高 each denote a scale and a direction, but only 很好 or 很高 specify (vaguely) a point along that scale and are thus actual predications with truth-conditional meaning over nouns. This seeming nitpicking about semantic categories would then parallel Mandarin's unusual requirements of classifiers on all nouns and location suffixes on complements of 在.
    (And now for total speculation: in most other languages, these grammaticalized morphemes 很 个 上/里 would probably be described as clitics or even incipient grammatical affixes–what's special about Mandarin?)

  2. E-Ping Rau said,

    June 15, 2015 @ 3:31 am

    The "hǎo" used in Taiwanese Mandarin is probably not a Minnan influence: the Taiwanese/Hokkien words which serve the same function are "tsiok (足) / tsin (真) / kài (介)", but never "hó (好)". And is usage of "hǎo" really specific to Taiwan only? In Taiwanese Mandarin (and I suspect in other variations too), there are actually quite a few options of the same construction, but they have somewhat different nuances:
    "tā hěn gāo" (他很高)
    "ta zhēn gāo" (他真高) – undertone of "He is really tall!"
    "ta hǎo gāo" (他好高) – undertone of "He is so tall!"
    "ta pǒ gāo" (他頗高) – undertone of "He is quite tall!" *
    * I've heard some discussions on its exact meaning: some think it means "a lot", others think it means "a little", and yet others think it falls somewhere in between…at any rate, in Taiwan this one is not used as often as the others, which might explain why a clear consensus is lacking.

    I wasn't aware of the comparison implication of the "noun + adjective" construction before, but it does make sense. initially, the examples I have come up with where this construction would work grammatically were all more or less comparative in nature. However, as also pointed out in the article, this is not its sole use: sentences like "zhuōzi gāo, suǒyǐ shìhé zhèlǐ (桌子高,所以適合這裡)" ("the table is tall, so it fits here") and "píngguǒ piányí, īnwèi hěn duō rén zhòng (蘋果便宜,因為很多人種)" ("apple is cheap because many people plant it") are acceptable to me. I'm not sure if "comparison or causal relationship" covers all of its possible uses?

    Also, when "hěn" is used in questions and negative forms (especially in negative forms), it nearly always carries the functions of intensifier to me.

    Finally, there are situations where "zhuōzi shì gāo (桌子是高)" is actually grammatical, and it's nearly always used with "but" to express contrast/contradiction: "zhuōzi shì gāo, dàn shì bú piàoliàng (桌子是高,但是不漂亮)" (the table is tall, but it's not pretty). This common usage of "shì", I think, can be likened to adverbs like "certes" in French.

  3. K. Chang said,

    June 15, 2015 @ 4:05 am

    Interesting. My Mandarin is mostly Taiwanese style, but I do recall another way to say it:

    他好好的!(ta1 hao3 hao1 de4) He is good/fine/doing great!

    Though that is technically a different expression altogether.

    Similarly, in Cantonese (my transliteration sucks)

    佢 呢 排 幾  好 呀   
    keoi5 li1 paai4 gai hou2 ah
    (He's pretty good these days)

    And another one ( I can't write those Cantonese words)

    Keoi yi ga ho ho dei
    (Mandarin equivalent, 他現在好好的)

    Or did my Taiwanese grammar crept into my Cantonese?

  4. John McWhorter said,

    June 15, 2015 @ 5:41 am

    My interpretation of HEN, an item which definitely threw me when I started with Mandarin because it clearly didn't usually mean VERY, has been that it is partially grammaticalized as morphology, as an adjectival marker. It seems to be one more indication that Chinese has more morphology than one is taught to expect. Such things penetrate a grammar in stages, and thus it hasn't gotten to comparative contexts (and others) yet. That poetic Chinese allows no HEN even in simple sentences is indicative that we are dealing with something about change over time. I doubt I'm remotely alone in venturing this analysis. I wouldn't be surprised if there were nonstandard dialects of Mandarin where HEN is used in a wider range of constructions with adjectives, and is perhaps even reduced to something like N'.

  5. shubert said,

    June 15, 2015 @ 8:14 am

    hen/very is in a higher degree than 還算 fairly. great vs. OK.

  6. Mr Punch said,

    June 15, 2015 @ 8:23 am

    In English, when we may say "he is quite well" or "she is just fine," it seems to me that "quite" and "just" are not exactly intensifiers (they imply definiteness more than extent), and that they move the statement away from any comparative implication.

  7. Observation said,

    June 15, 2015 @ 9:06 am

    FWIW, I think @K Chang's sentences are mostly fine.

    The first one:

    keoi5 ni1 paai4 (paai2) gei2 hou2 aa3

    Is, to me, the Cantonese equivalent of 'ting' above. I make the choice between 幾 and 好 depending my actual perception of the degree.

    As for the second:

    Keoi5 ji4 (ji1) gaa1 hou2 hou2 dei6

    Is a bit strange to my Cantonese ears. It sounds more like Putonghua to me (還好好的). I would say 好地地 (hou2 dei6 dei6), and it often implies, to me, some sort of comparison, like this example:

    Keoi5 sin1 paai4 (paai2) zung6 hou2 dei6 dei6, ji4 gaa1 m4 zi1 zou6 me1 din1 zo2.
    (He was still fine a while ago, but now he's crazy for a reason I don't know.)

    Note that the dei6 does not turn into a dei2, so it is not the same as the '-ish' in 紅紅地 (hung4 hung2 dei2, 'reddish'), unless the two 地s are allomorphs.

    As a thought experiment, I tried to think of situations where there is no copula or adverb of degree (in addition to situations mentioned in the post), and I got these:

    When the subject-pronominal construction is used as a noun phrase:
    Ngo5 hoi1 sam1 jau5 man6 tai4 me1?
    (Is there a problem with my being happy?)

    In this situation I can't explain:
    A: 你做咩搞極都搞唔掂㗎?
    B: (噉)我蠢呀嘛,唔畀呀?
    A: Nei5 zou6 me1 gaau2 gik6 dou1 gaau2 m4 dim6 gaa3?
    B: Gam2 ngo5 ceon2 aa1 maa3, m4 bei2 (j)aa4?
    A: Why can't you finish it, no matter how long you've tried?
    B: I'm stupid; don't you allow that?

    Another one I can't explain:
    Keoi5 ci1 sin3 gaa3! Gam2 ge3 caak1 jim6 dou1 ho2 ji3 mun5 fan!
    He's crazy! He can get full marks in such a test!

    To express admittance:
    Ngo5 seoi1, ngo5 m4 wui5 m4 jing6 ge3, daan6 hai6 nei5 dou1 m4 sai2 gam2 gaa3 (gaa2)!
    I'm bad, I won't deny that, but you didn't have to do that!
    我係衰, with the copula 係 but not with 嘅 at the end, would work as well.

    It seems there's a common denominator between the two contexts which I cannot identify: There is a modal particle in the clause. (I'm not sure if that's the right way to translate 語氣助詞 into English, but Google Translate gave me that. I confirmed by typing this into the English Wikipedia, and while the article there didn't mention Chinese, the French version of the article did.)

  8. Paul said,

    June 15, 2015 @ 10:00 am

    The expression "他很好" actually sounds a bit stilted as an answer to the question "他好吗?" in casual conversation; I think it's more natural to omit the subject. This of course obviates the difficulty with "他好", as one could just reply "好。". But in any case, if I wanted to say "he's ok", not necessarily good or very good, my preferred option would actually be, "(他)还好" rather than "(他)很好".

  9. shubert said,

    June 15, 2015 @ 11:19 am

    很, 甚 shen4 are equivalent, but 甚 remains only in certain terms.

  10. Observation said,

    June 15, 2015 @ 11:31 am

    I just reread my post.

    I meant:

    When the subject-predicate construction is used as a noun phrase:

    I have no idea why I put pronominal there…

  11. Akito said,

    June 15, 2015 @ 11:44 am

    I think a better term is "contrast(ive)." "Comparison/comparative" would mean "he's taller," etc., but one can be taller (than someone shorter) without being very tall. After all, comparison is, well, relative.

  12. Tom Parmenter said,

    June 15, 2015 @ 5:02 pm

    In English 'very' doesn't mean a heck of a lot. In fact, we often emphasize it or double it to turn it into an intensifier. It almost never means 'truthfully', which is its derivation. In fact, my first sentence could be "In English 'very' doesn't mean very much."

  13. K Chang said,

    June 16, 2015 @ 4:52 am

    @Observation: you are absolutely right, 好地地 sounds more Cantonese than 好好地, but I seem to pronounce them differently. I say 好好地 with the 地 being a higher tone than the low tones of 地 in 好地地.

    I wonder if 好好地 was someone moving the Mandarin expression 好好的 into Cantonese? Or did someone said the original Mandarin with a heavy accent and created a Chinese eggcorn?

  14. Wentao said,

    June 16, 2015 @ 7:33 pm

    In the past I thought it might have something to do with word length or rhythm (for example, personally, I feel 他喜气洋洋 is somehow "more grammatical" than 他好 or 他高兴). However, after reading the post and comments above, I think grammaticization and zero morpheme explains this phenomenon better.

    Like Observation says, I also notice that the "bare adjective" is allowed in noun phrases, such as 因为我开心 or 你知不知道他高兴. It is also allowed with modal particles 语气助词 at the end of the sentence. For example, 佢黐線㗎!噉嘅測驗都可以滿分!has a Mandarin equivalent: 他厉害啊!这种测验都可以满分!

    In the case of 东方红, I agree that the language is more poetic. The same applies to the next line: we wouldn't say/write simply 太阳升 in conversation or normal written language either.

  15. Lily C said,

    June 18, 2015 @ 9:55 am

    Could it be because 好 itself often serves the same function as the 很 in this case? e.g. 你好棒! 我好累啊。。

  16. Lily C said,

    June 18, 2015 @ 9:59 am

    ..so if you just say, 他好, I might find myself waiting for the adjective, as I would if you said 他很. Seems to me that 好好 in this case is actually working just like 很好: the first 好is an equivalent to 很

  17. dainichi said,

    June 18, 2015 @ 10:39 pm

    Hehe, slightly off topic, but I'm sure there are ESL students out there who have wondered why "Thank you very much" is OK, but *"Thank you much" is unnatural.

    I think it has something to do with "much" slowly becoming a negative polarity item, whereas in the positive, you have to use "a lot", with "very much" as a slightly dated/fossilized alternative.

  18. Jason Merchant said,

    June 19, 2015 @ 8:29 am

    Tom Grano published a paper on precisely this question a couple of years ago: Thomas Grano. 2011. Mandarin hen and Universal Markedness in Gradable Adjectives. Natural Language & Linguistic Theory 30:513-565. I do think it's the best general answer to Victor's question that's ever been proposed…

RSS feed for comments on this post