Chinese Parts of Speech

20 Mar 2006

OK, this is an entry that’s likely to bore many readers to tears. You have been warned.

While I don’t find the study of Chinese grammar remarkably stimulating, there are some aspects of it that catch my interest. It’s kind of cool how Chinese parts of speech don’t fit so neatly into our Western designations. When China first starting applying Western linguistics to Chinese, Chinese syntax was forced into the Western mold. Over the years Chinese scholars have decided that this just doesn’t work.

So what’s different about Chinese grammar?

Well, for one thing, the Chinese like to divide all words into two big categories: 实词 and 虚词. I’m not completely sold on the necessity of this, but the Chinese sure seem to like it. Basically, 实词 often refer to the “real” (like things, properties, actions, ideas), while 虚词 serve only grammatical functions. 实词 can usually form meaningful utterances by themselves, and can combine with each other in meaningful ways. 虚词 only combine with 实词 and mean nothing out of context.

虚词 are much smaller in number but frequently used. They include prepositions (把, 将, 让, 比, 给, 对, 为了, 在, 从, etc.), conjunctions (和, 跟, 或, 即使, 如果, 虽然, 而且, 而, etc.), auxiliary particles (的, 得, 地, 了, 着, 过, etc.), and modal particles (了, 呢, 吗, 吧, 啊, etc.).

Chinese prepositions are special because they closely resemble verbs in many cases. In fact, most Chinese prepositions evolved from verbs. To illustrate how Chinese prepositions can be like verbs, check out the following simple sentences:

1. 车外面。 (The car is outside.)
2. 车停外面。 (Park the car outside.)
3. 不要外面停车。 (Don’t park the car outside.)

OK, so the question is: which of the above s are verbs, and which are prepositions?

Number 1 and number 3 are easy. 1 is clearly a verb; 3 is clearly a preposition. But what about number 2? Well, part of the definition of a preposition (according to the current standard view) is that when it modifies the predicate it must come before it. In this case, it clearly comes after. It’s not a preposition, but rather a verb acting as a complement to the main verb 停. Crazy!

Another special feature about Chinese parts of speech is the nouns. In Chinese you have “time word” nouns performing the function of English adverbs. For example, “today” as in the English sentence “I went shopping today” is an adverb. In Chinese 今天 is a noun. The same goes for other time words like 现在, 明年, 刚才, 平时, etc.

If you can handle all this craziness I will give you yet another example. In English we use prepositions to explain physical spatial relations, but Chinese uses “nouns of locality.” So while we would use the prepositional phrases “on top,” “in the middle,” or “to the east,” the Chinese would use the nouns 上面、中间、东边.

OK, if you made it through all that, I will reward you with an exciting chart showing how simple “nouns of locality” combine to form compounds (and which don’t combine). I like this kind of chart.

~边 ~面 ~头 以~ 之~
上边 上面 上头 以上 之上
下边 下面 下头 以下 之下
前边 前面 前头 以前 之前
后边 后面 后头 以后 之后
左边 左面
右边 右面
里边 里面 里头
外边 外面 外头 以外 之外
东边 东面 东头 以东 之东
以内 之内

Finally, I will end this entry with part of speech statistics “top 4” from my syntax professor. I tried to get the source out of him, but he just told me “structuralist lingustics.”

– In modern Chinese, “normal” nouns (i.e. not “time words” or “nouns of locality”) make up 45% of all vocabulary
Verbs make up 30% of all vocabulary
Adjectives make up 10% of all vocabulary
Measure words make up a little over 1% of all vocabulary.
– All other parts of speech make up less of a chunk than measure words.

Conclusion? Chinese grammar is wacky!


John Pasden

John is a Shanghai-based linguist and entrepreneur, founder of AllSet Learning.


  1. Any tips on how to get started learning spoken (and maybe later, written) Chinese? Unfortunately, taking a class is not an option for me right now. I’ve picked up a number of books and even some flash cards, but finding a starting point is a problem. Once you practice tones until you throat is sore, what do you move onto next?

    • Davoud Derogar Says: January 29, 2014 at 10:31 am

      try YouTube for any language , there are so many lessons for beginners or advance levels. I personally like learning Chinese with mike, but I listen to all the other teachers. For grammar, use search for any subject you want and there are many great sight to learn any language, good luck.

  2. Hello,

    I’m a native Chinese speaker and I study linguistics as an undergrad. I have to say, the catagorization you described in your post seems a bit non-standard to me.

    Maybe what your prof said is right: the most of linguistics taught around the world now focus on the Chomskyan theory of generative grammar (that’s what I learned), if he’s still using ‘structuralist grammar’, then it’s a much older (and conflicting) theory of how languages work.

    The analysis of 在 seems reasonable, though. Contrasting:

    车停在外面 vs 车在外面停

    I can see that the PP in the 1st case is more like an adjunct (of the verb), whereas the second case, it seems like it’s more like an argument (compliment) of the verb. I’m trying to come up with tests to figure this out, but it’s way too early for me to be thinking about this…..

    Thinking a little bit more on the topic, I can start to see why time-words in Chinese act differently from their counterparts in English and why they might not be analyzed as adverbs. In English, one could easily add on various PPs after the verb, some about the location of the event, some about the instrument, and the time, etc…

    i.e. I wrote a letter in my room at night with a pencil.

    the Chinese equivalent is structrurally quite different:


    It seems like there might be a restriction to the number of 在s in a single sentence (我在晚上在房间… seems wrong)

    Being in the US, I never got to do a lot of work on Chinese when studying of syntax. I got a taste of studying Chinese Syntax last summer, and it’s interesting to see Chomskyan grammar applies to Chinese.

  3. Basically, there is nothing especially wacky in what you described.
    Most Chinese prepostions can described as co-verbs, which are just some verbs in their special mission, corresponding with verb participles in Indo-European languages. Some co-verbs now rarely appear outside co-verb usage, for example, but still in some combinationary form, for example “将军 (chess mate)” or “将来 (future)”, in which 将 retains its original meaning of “take, hold”, so can often be replace with 拿, 把, which basically mean the same things; some co-verbs can both be used in co-verb or normal verbal function, e.g., 拿, 把, 用, 通过; some normal verbs used in co-verb function are rarely realized that they are not normal, e.g., 坐 in 坐电梯上楼 or 吃 in 吃药治病.
    Some Chinese particles are the results of the tendency in Chinese to sound very to sound every consonant with a vowel. When some nonsyllabic consonants are syllablized, they are devided from the original word and becomes independent. The most prominent example of this case is causative particle 使, which was in proto-Sino-Tibetan languages just a simple s- prefix and remains so in the modern Tibetan language.
    Other particles come from the words which were previously pronounced with a consonant at the ending, e.g. 鞋 or 靴. This ending consonant is retained in modern Japanese, albeit syllablized in 靴 kutzu. In modern Chinese they are written seperately with a 子, incidentally, syllablized as well.
    The same wackiness also shows up in modern English, e.g. syllablization in whater for what, or noun as adjective/adverb like way in “We do it American way”.
    The true wackiness of modern Chinese is its sentence particles, comparing the following sentences.
    他来了吧。 (note: very different from 他来了吧?)
    他来了么。(note: very different from 他来了吗?)

  4. John, personally I find grammar the most interesting of all language topics. It is just so… interesting.

    Shuan. Do you have a good textbook? I think Hanyu Kouyu isn’t bad. The introductory books can be had for around RMB100 and at a chapter once a week will have you up to a good standard within a year. At a chapter a day you’ll be there in a couple of months. Get a native speaker that doesn’t have to be an expert, just be able to pronounce reasonably, to help you along in the speaking and listening, RMB50 hour should suffice.

  5. typo.

    “实词 often refer to the “real” […] while 实词 serve only grammatical functions”

  6. Duncan:

    Your example sounds somewhat wacky to me. I heard no native speakers around me say 在晚上 or 在晚上的时候, rather 晚上 or 晚上的时候。Basically nobody I met would not say 在早上 or 在中午。我在我的房间里… sounds like you put a lot of emphasis on the point that the room is mine, a more natural equevalent would be 我在自己的房间里…

  7. “Basically nobody I met would say 在早上 or 在中午, either”


  8. Duncan,

    Which categorization did you mean? The 实词/虚词 one? Pick up any 现代汉语 textbook. It’s standard in China.

  9. Leo,

    I still contend that Chinese grammar is, in fact, quite wacky.

    Did someone say… thesis topic??

  10. Tuur,

    Thanks for catching that typo. Fixed. I also fixed one where I wrote “adjectives” instead of “prepositions.” I actually fixed a lot of typos before I published. I am just full of typos today.

  11. “John Says:

    I still contend that Chinese grammar is, in fact, quite wacky.”

    The usage of coverbs is quite widespread in Sino-Tibetan and Dai languages. A lot of languages do not differentiate verb/noun, noun/adjective, adjective/verb, the latter two of which is true for modern Chinese. Some things in English/Germanic languages are also wacky from the point of view of Latin languages. The verb participles in English is actually more similar to coverbs in Chinese than their false friends in Latin.

  12. “John Says:

    Did someone say… thesis topic??”

    I am sorry that my words are sounding rambling mumbling. Sometimes my brain and logic would give a better performance.

  13. Leo, I frequently hear “在晚上的时候” and variations of it in Taiwan. That doesn’t mean much, though.

    John, if the 在 in “车在外面” is a verb, then how are verbs defined? If this 在 is classified as a verb, then it seems to me that words my text books would call adjectives, such as the 熱 in “今天很熱” can also be called “stative verbs” or something like that. Can you elaborate a bit?

    Also the whole time word/noun issue seems due to the topic/comment structure of Chinese sentences. At least that’s what it looks like to an intermediate speaker such as myself…

  14. Leo –

    I don’t see anything wrong with 在早上, or 在中午。
    To me, it’s okay to say, 在早上,我们工作 or 在中午,我们吃飯。

    As for 我的房间 vs. 自己的房间,they don’t feel all that different to me.

    Also, thanks for mentioning co-verbs. I didn’t know about them. I took a class on some of the verb-resultative constructions in Chinese, but we didn’t talk about co-verbs.

    John –

    I guess what happened was this: initially I thought what you wrote was wacky, and felt compelled to reply. But as I started writing my response, what you wrote started to make more sense, and towards the end, I didn’t really have anything to disagree with your post.

  15. The 实词/虚词 terminology is actually kinda cute, I wonder if it maps nicely into lexical/functional categories in syntax.

  16. Sorry for writing so many replies all of a sudden.

    Going over that chart, I see that there’s 下头 and 前头. Also, my understanding of 上头 refers to one’s boss, and not anything locational.

    They just don’t seem right to me, what do you guys think?

  17. Hi John, this is Paulo from Florianopolis (an island in southern Brazil). I seldom do it, but i felt compelled to stop by and say your site is a gem. So, please, receive this warm thanks for your work. Whishing you luck and happiness, from afar.

  18. While I don’t think Chinese grammar fits neatly into structural linguistics, I’m not sure any language does, especially in colloquial form. The issue with #2 above in 车停在外面。is not so much in the “在” as it is in the “车”. 车 is a noun which serves as the object of the sentence, yet nothing is actually moving it, as in “把车停在外面”. So is the “把” implied or is this an entirely different meaning? If it is an entirely different meaning, how can we justify the placement of “车” at the beginning of the sentence? I’m not convinced that this sentence is any different than “车,停在外面”

  19. DD,

    As far as I know, the sentence is to be understood as “车,停在外面”.

    车 is not actually an object. It’s the topic of the topic-comment structure.

  20. Duncan,

    They’re right. They were taken directly out of my 现代汉语 textbook (written by Chinese for Chinese).

    Click on these links and scroll down to see the definitions: 上头, 下头. 前头 doesn’t run up much, but it’s listed in my good dictionary as meaning “in front; at the head; ahead.”

  21. Is there a reason why the 现代汉语 text gives the imperative interpretation for 车停在外面 rather than, say, “The car parks/is parked outside”?

    Also, what’s the spread for English vocabulary, just for comparison?

  22. zhwj,

    The translation was my own. At the time I considered the non-topicalized interpretation but thought it seemed strange (which is why I made comment #19), but now that I’m thinking more clearly I realize it’s a valid option as well.

  23. 车停在外面 is just an abbreviation of 车是停在外面, which is awkward.

    车 是 停 在外面 -> 车 停 在外面 -> 车 在外面
    她 是 住 在二樓 -> 她 住 在二樓 -> 她 在二樓 / 她 住 二樓

    chinese has structure, it’s just abbreviated a lot in colloquial form, and there’s many ways to abbreviate. the choice of abbreviation depends on emphasis.

  24. Hi John,

    Been a lurker on your website (1st comment ever), and appreciate all your interesting posts. Just a few comments, which will inevitable come off as know-it-all (sorry!):

    1. “Parts of speech” doesn’t do for the lingo of a(n) (applied) linguist… word CLASSES?

    2. “Locative expressions” would be more of a catch-all term to describe Chinese prepositions, especially since examples like the ZAI of “ting zai waimian” are adverb-like. (You have prepositions that can serve double duty as “sentence particles” — i.e. OUT in “go out of the house” vs. “go out”)

    (I’m writing a semantics paper on English prepositions right now).

  25. Shaun, I’m learning to speak Mandarin myself, a good website to have a look at is and also Hope that helps!

  26. after thoroughly going over your English Blog Entries,I just simply can’t believe my eyes on what you’ve written. It is definitely true that you have the right to comment on whatever you want to but seems like it’s better to keep a little bit consistent in attitudes between your english blog and chinese blog and most who ‘ve been logging on your blog read english well.

    Regarding to this entry,you really donot have to change the words that you are frequently using and just simply by using one instead of another all the time,like 上面(书面)instead of 上头(口语).Good luck!

  27. Ens,

    I avoid the lingo when I can. Some of my readers may have no interest in a post about categorization of words, but at least when I use “parts of speech” almost everyone knows what the entry is about.

    Thanks for the info.

  28. Marco,

    What are you talking about? You make it sound like I’m trash talking China on my English blog or something. If you’re upset about my use of the word “wacky,” let me suggest that perhaps understanding every word doesn’t necessarily mean you have fully understood my intent. (Don’t worry, I think a lot of my native English-speaking readers don’t get my sense of humor either.)

  29. zhwj,

    Oh, I forgot to mention… I tried to get the spread for English out of my professor, but he just told me “about the same.”

  30. John –

    Did you read the entries for 上头 and 下头 from that site?

    The only mention of 上头 came from a verse written in 文言,which makes its use exceptional today,

    as for:下头, it says :下头(低头), which is not locational at all.

    I have no doubt that you can find examples of these words in standard Chinese textbooks. They could very well be written in China, by Chinese scholars. In fact, a friend of mine is learning Chinese right now, and I’ve repeatedly found phrases / examples in his textbook (written by Chinese) that are problematic, or as your put it, wacky.

    Maybe it’s because I’m from Hong Kong, and I speak both Cantonese as well as Mandarin.There may also be all sorts of regionalisms and what not, but 上头 and 下头 as directional / locational “nouns of locality” just do not work for me.

    Come to think of it, 东头 feels problematic too. (If there’s 东头, why isn’t there 西头?)I googled, and most of the references are either names, or they refer to 村 or 院, i.e.村东头,院西头。 I guess it might be an exception in those cases.

    P.S. 请不要误解我的心意,我不是来踢馆的 😉

  31. Duncan,

    Yes, I read them. Still, I think that the examples given don’t necessarily correspond to the only ways to use the words.

    Nevertheless, I readily admit that I never hear anyone say 上头, 下头, 前头, or 东头. All these came from the textbook, which also listed the same combinations for all four cardinal directions (including 西头). Out of laziness, I only listed the combinations with 东.

    Don’t worry, I don’t mind the comments. This is what I should expect when I write about grammar… 🙂

  32. Duncan: Kingsoft tends to pull its examples pretty exclusively from pre-modern texts, with the sporadic inclusion of Lu Xun or Mao Zedong, regardless of whether the word is archaic or contemporary.

    上头 etc: I’ve most often heard these used to indicate the location someone is facing – 朝着上头, 望着下头 and so forth. Random sentence: “两人惊喜相望,立即朝上头大喊. ‘我们在这儿!'” The 头 is there to move you around, so to speak…

  33. IRT the –头 word debate: What about shanghai dialect? They always say “shee-doh” (前头)for 前面 and “sang-doh” (上头) for 上面。

  34. Penfold,

    I’ll have to check out the book you mentioned. I found it I do have a handful of “teach yourself” books, but I’ve found they have enough errors to make them annoying.


    Thanks for the sites. I’ll check them out.

  35. My whole experience with learning Chinese is that it has a lot to do with just accepting things and learning conventions, rather than the “big picture” of the language, especially in terms of grammar. In my experience it is generally more useful to learn example sentences than dissect the grammatical nature of each character in a sentence.

    This is in contrast to a language like German, which I have also studied. Its relatively predictable grammar and modular verb and noun stem forms make it easy to intuitively put together new compound words and structures.

  36. reading ur blog is a pleasant thing cause it is interesting to see how a foreigner think about the daily Chinese life. It’s funny. haha

  37. JASON LIU Says: January 18, 2008 at 4:43 am

    Hi, I’m also a native chinese, studies in linguistics in Delhi University. Well, i agree with Duncun on some points but i can’t agree with Leo. Chinese has a lot variations in sentence strucure and phrase strucure. For example, 车停在外面 vs 车,在外面停 are the same in syntax. Or, 我中午吃饭 VS 在中午,我吃饭 (中午的时候,我吃饭). they may vary a little in connotation but not in denotation.
    But in “将军 (chess mate)” or “将来 (future)”, even if 将 retains its original meaning of “take, hold”, it cannot be replace with 拿, 把, (which basically mean the same things). 将来 is totally different with 拿来.

  38. There are other languages which use shici, xuci categories. Arabic, for instance, divides words into nouns, verbs, and particles. But since most Arabic nouns derive from verbal roots, nouns and verbs are very fluid categories. So effectively, you have noun/verb as one category, and particles as another category.

    This is why shici, xuci were used to analyse Classical Chinese. In Classical Chinese, virtually all shici can be used as either nouns or verbs. This is why it made no sense to divide them into separate categories.

    Furthermore, Classical Chinese is based on topic-comment. This is why shici are easily used as time words and place words.

    Conversely, Modern Chinese is much closer to European languages. But it is not entirely like European languages. Whereas Classical Chinese is succinct, elegant, beautifully simple, Modern Chinese is prolix, wordy, and unwieldily complex.

    (The same problem affects Hebrew and Arabic. Both Modern Hebrew and Modern Arabic have Europeanising tendencies.)

    Personally, I ascribe this to a corporate conspiracy to corrupt the noble characteristics of the Chinese language. Or perhaps it reflects the spirit of the modern age, where style trumps substance, and quantity conquers quality. Where standardisation and mass-production fabricate rows after rows of identical schools, houses, and furnitures.

  39. Now that I think of it, Tibetan also uses the shici, xuci categories. But sometimes people analyse Tibetan using Sanskrit categories, giving the impression that Tibetan has noun declensions. This is like how some Japanese textbooks describe Japanese as having nominative, accusative, and locative cases.

    They say that Pre-Classical Chinese particles had rather defined grammatical functions. For instance, different personal pronouns were used to differentiate subjects, objects, and possessives. By the time of Mencius, these particles were already confused.

    But even in Classical Chinese, particles did a lot more work than they do in Modern Chinese. Therefore, Modern Chinese has a lot less flexibility in word order than Classical Chinese, even though Modern Chinese is far more wordy. Easy nominalisation was Classical Chinese’s greatest asset. The loss thereof resulted in the substanceless dribble which typifies modern Chinese literature.

Leave a Reply

Your email address will not be published. Required fields are marked *