Raymond, why did you stop studying Chinese? It has no grammar!

One of my colleagues, a native Chinese speaker, asked me whether I was still learning Mandarin Chinese. I told him that I had given up. He was baffled by this.

"But Chinese is such a simple language. It has no grammar!"

Now of course, Mandarin has a grammar, because every language has a grammar.

This is one of the curses of being a native speaker of a language: You don't even realize how hard your language is. As far as you're concerned, your native language is as easy as falling off a log.¹

Now, it's true that Mandarin has almost no inflections, unlike most European languages. But that's not the same as saying it has no grammar. It's just that the grammar moves from being internal (inflection) to external (helping words and word order).

Sidebar: David from Popehat lays out some of the simplifications, but I think he oversimplifies the use of the completion marker 了. It's not strictly speaking a past-tense marker, at least not in the sense we consider it in English. Proper use of 了 is more complicated, and this page tries to explain some of the subtleties. David later tries to explain Mandarin phonology. I must admit that I have an advantage in already having a tonal language wired into my brain, so I don't have the hurdle of learning to hear and speak tones. I just have to learn to hear and speak different tones. Which is still frustrating. End sidebar

One of the consequences of "your own native language is simple" is that native speakers are sometimes the worst choices for explaining their own language, since they simply fail to recognize how weird their language is. An example I gave some time ago was that elusive third tone. If you ask a native speaker how it is pronounced, they will say one thing ("dipping tone"), but then when they themselves speak the third tone they do something completely different ("low level tone"). Native speakers are so convinced that the third tone dips that when you call them on it, they insist that the tone dipped, when in fact it barely moved at all.

That conversation with my father went something like this.

Me: "In that sentence, you said ⟨low level tone⟩."

My father: "No, I didn't. I said ⟨dipping tone⟩."

Me: "Well, sure, that time you said ⟨dipping tone⟩. But in the original sentence, you said ⟨low level tone⟩."

My father: "No, I didn't. Listen again. ⟨repeats sentence and uses low level tone⟩."

Me: "There, see? You used ⟨low level tone⟩."

My father: "No, I didn't. Here, you repeat back to me what you think I said."

Me: "⟨says sentence with low level tone⟩."

My father: "There, you got it!"

Me: "But I used the wrong tone! I should have said, ⟨says sentence with dipping tone⟩."

My father: "No, that's wrong. You exaggerated the tone too much."

That last remark from my father was what made it click for me: The low level tone and the dipping tone are complementary allophones. (My father, of course, has no idea what a complementary allophone is, but that's okay.)

Another example of native speakers not seeing the complexity in their own language is the use of the negative adverb 没. Mandarin has two main adverbs that mean "not": 不 and 没. If you ask a native speaker, they will tell you, "It's very simple. 不 is the general-purpose negation, and 没 is used only to negate the verb to have. In other words, 没 is always followed by 有." But then you will see that native speakers use 没 to negate all sorts of things that aren't 有. If you point this out, they will retcon it by saying that the phrase 没關係 ("no connection", which is an idiom that means "it doesn't matter, don't worry about it") is really a shorthand for 没有關係 ("doesn't have a connection"). Native speakers play this card whenever an out-of-place 没 shows up. "Oh, it's negating an invisible 有." If you ask them how to tell when there is an invisible 有 in a sentence, they will say "You just have to know," or sometimes the circular "Stick a 没 in front and see if it makes sense."

Sidebar: Here's a page that tries to explain the difference between 不 and 没. The way I internalize it based on limited observation is to say that 不 is not tied to a moment in time (innate or habitual property), whereas 没 refers to a particular incident (momentary property).

doesn't get wet
(it's water-resistant)
isn't wet
(he has an umbrella)
doesn't drink milk
(he's lactose intolerant)
isn't drinking milk
(he chose water)

This is similar to the distinction between English simple ("I do") and progressive ("I am doing"). Furthermore, 没 carries a sense of "yet"; you are denying that something is true now, but expecting that to change in the future. End sidebar

One downside about having such a superficially simple grammar is that it makes the language much more ambiguous. The more complex grammar of European languages acts as a checksum. If I say, "He are coming," then you know that something went wrong. The grammatical doodads act like signposts to confirm that you the listener are parsing the sentence correctly. It's like the road sign after every highway exit that reassures you, "You are still on Highway 405 Northbound." One of my colleagues told me that he missed those signs on his trip to Italy. There would be signs labeling each exit, but rarely was there a sign telling you what highway you were currently on!

To me, Chinese is difficult to learn because of its lack of guideposts that help steer you onto the right track. Without them, many sentences end up ambiguous. (In that example, the lack of any grammatical particle that distinguishes imperative from declarative mood led to the confusion.) The relative scarcity of grammatical particles makes me feel like I'm talking baby-talk. "Me want eat cookie."²

Resolving ambiguity is made even harder by the fact that every word in Mandarin has about a dozen homophones (fortunately, most of them not used in everyday speech), so you aren't even sure what word you're dealing with at the moment you hear it. You just know it's one of these two or three, and you have to wait and see which one actually makes sense when combined with the other words in the sentence (some of which may themselves also be ambiguous).

Adding to the ambiguity is that in many cases, you can omit words from a sentence if they are implied from context. So you now have to juggle the ambiguous mapping of sounds to words, the ambiguous grammatical context of those words (was that a statement or a direct order?), and choosing which implied words to insert to support your conclusion! Of course, native speakers can resolve all of these ambiguities very quickly, having been doing so since birth, and they are much better at picking up other cues (such as where the speaker speeds up and slows down) to help steer toward the correct interpretation. Indeed, in the language I learned as a young child, I can resolve these ambiguities with no difficulty at all.

Sidebar: Even native speakers sometimes have to go into explicit ambiguity-resolution mode by adding clarifying context. This happens in English occasionally: You might say, "He had a bat (the animal)" because the shorter sentence "He had a bat" would be ambiguous. Did he have an animal or an instrument for striking? End sidebar

One thing I do like to quibble about is the treatment of classifiers in Mandarin. Most people treat them as a quirk of the language, making them sound like an oddball feature that doesn't exist in European languages. An analogue in English would be the word "pair" when applied to scissors or pants. You can't say "a scissors" or "a pants"; you have to say "a pair of scissors" or "a pair of pants." (Particularly confusing because a "pair" of scissors or pants is still one article.) In Mandarin, every noun has a corresponding classifier.

You can think of classifiers as the Mandarin version of grammatical gender. The nouns in the language fall into around 170 different categories, and you just have to know which category word goes with each noun. There are patterns that help the learning process, but there are always exceptions that you simply must memorize.

For example, 条 is generally used for long, thin, flexible things, like a fish or a ribbon. But you also use it for dogs. Oh, and also for skirts and dresses. Go figure.

So the next time a native Mandarin speaker complains that English has all these arbitrary rules that serve no purpose other than making the language harder to learn, just ask them about classifiers. (They will naturally defend classifiers by saying that they are completely obvious and in no way arbitrary.)

Anyway, the bit about classifiers explains why the subway ticket vending machine asks you how many "sheets" you want: In Mandarin, it is very common to omit the noun and use only the classifier when the noun is implied from context. This happens in English, too. If you are a shop that repairs scissors, the clerk might ask, "What's wrong with this pair?" as shorthand for "What's wrong with this pair of scissors?"

The classifier word for ticket is 張 which translates as sheet. The full question is "How many sheets of tickets?" But since you are at a ticket vending machine, the noun is implied from context, and the shorter sentence "How many sheets?" is used instead.

¹ This natural tendency to think of what you do as normal reveals itself in the words that the Chinese language uses to refer to itself. The name for the country of China is 中國, which translates as the middle kingdom, because by an amazing coincidence, China happens to be right in the middle of the map. And the name for the language itself is 普通話, which translates as normal speech, because we all talk normally; it's the foreigners who talk funny by using their own words for everything.³

² In practice, the distinction between baby-talk and adult-talk in Chinese is accomplished in two ways. First, babies have a specialized vocabulary: babies say doggy instead of dog, for example. Second, adults employ modal particles which convey the attitude of the speaker. Cantonese is notorious for having a large number of these sorts of particles. I don't know most of them, so my speech tends to come off as rather rude and abrupt.

³ Someone said that a neighbor of his grandmother complained, "I don't understand why people in foreign countries bother to learn a second language. Why don't they just talk normal?"

Comments (68)
  1. Joshua says:

    My understanding of the origin of middle kingdom is it was the middle one of the BV era multi-kingdoms that managed to conquer all the others in the area of fertile land.

  2. Frank says:

    普通話 actually means something other than "Normal Speech". It is short for "普遍通用話" (take the first, third, and the fifth characters). "普遍" means "General", "通用" means "Common", and "話" means "Speech". Combine all that together, it means a language that can/should be used by all Chinese people (since a lot people in China only speak their dialect).

    [Then I was faked out by cognates. In the dialect I grew up speaking, ",普通" means "normal"; it looks like in Mandarin, it means "ordinary". Still, even with the sense of "ordinary", it gets the point across: This is the ordinary way of talking. Anybody who talks otherwise is going out of their way to be different. -Raymond]
  3. Liquorice says:

    To be fair, many languages have ambiguity and depends on context. Also the category word in Chinese is similar to collocation in English: you don't have a general rule but only convention.

    For pronunciation, ⟨low level tone⟩ and ⟨dipping tone⟩ have small difference. Some may say one is "more correct" than the other. This reminds me the case of pʰ, p and b in English.

    And 普通話 does not translate as normal speech. While 普通 means normal, 普 means popular (普遍) and 通 means common (共通, 通用).

  4. Danny Moules says:

    "so my speech tends to come off as rather rude and abrupt"

    In addition to having the more general issue of the social skills of the thermonuclear device? Or would this qualify as falling under that categorisation (no Z, because I'm inadvertently a Latin English revisionist) in a limited context? Or is THAT dependant on the language we're discussing it in?

  5. Chris B says:

    A short while back, I was attempting to learn some German for a vacation my wife and I had planned. In many ways, German was easy to learn because of it's similarities to English.  There were definitely things that tripped me up, like having to learn the gender of nouns and the declension of the various cases (both of which I'm still quite horrible at).  Throughout the process, my programmer brain kept wondering what an engineered language would look like so that it would be easy to learn, easy to use, and communicate effectively.  A few of the things that came to mind were:

    1. Do not assign gender to inanimate objects. It is not obvious that doors are boys and windows are girls.

    2. Minimize homophones; they create ambiguity.

    3. Keep subject/verb agreement. As you said, this helps ensure listener understands the speaker.

    4. Minimize variance in sentence structure. In German, word order is much more relaxed than English due to the flexibility provided by the various articles in each case. So you can distinguish the subject from the direct object by the article. For example, in English the sentence "The dog bites the man" is very different from "The man bites the dog." due to the word order. In German, you can say either "Der Hund beißt den Mann" or "Den Mann beißt der Hund.", and both would translate as "The dog bites the man" since "den" puts "dog" in the accusative case (direct object), and "der" puts "man" in nominative case (subject).

    The fourth one is especially interesting to me.  A language which attempted to conform to that restriction would likely not be a good candidate for poetry (or other forms of creative writing) as it can depend on wrangling sentence structure to match a meter or rhyme pattern.

    [You may want to check out lojban, one of whose primary goals is to eliminate ambiguity. -Raymond]
  6. Nicky C says:

    AFAIK, Chinese officials prefers calling the language "Putonghua" in English, which is how they pronounce 「普通話」 in Putonghua.

  7. Liquorice says:

    >Chris B

    I don't think such unique language exists. Think about TOOWTDI vs TMTOWTDI.

  8. Mordachai says:

    That was an awesome discussion.  Thanks for sharing that.  Helps to know that everyone everywhere is convinced their way is the normal way, duh!  ;)

  9. Joshua says:

    > doors are boys and windows are girls.

    At least in German, Spanish, etc. the wrong gender just sounds wrong (non-extant word). In Arabic it's a different word with a different meaning.

  10. Tom P says:

    Growing up for some years as a child in Japan, I never realized that Japanese has tones.  To my American ear, the tones sounded like accents – in English, when one accents a syllable, one's pitch tends to rise.  It was only many years later that I learned, with surprise, that my "accents" were actually tones.

  11. Wear says:

    I've ran into the Native speakers issue from the other side a few times. I'd like to learn a second language but my attempts tend to be hindered by not having a very good conceptual/technical understanding of English. I know when to use "He" and when to use "Him" but if someone randomly asked me for the English masculine subjective pronoun I'd have to think about it a bit.

  12. Andy says:

    > At least in German, Spanish, etc. the wrong gender just sounds wrong. In Arabic it's a different word with a different meaning.

    That happens in Spanish too: el papa (the Pope) and la papa (the potato) is one of the more interesting pairs.

  13. Mark says:

    That retcon issue reminds me of how "than I" is justified as more correct than "than me", because "than I" is short for "than I (am)". Sure, you can do that, but that doesn't make it more correct.

  14. Gabe says:

    Chris B: Have you seen Esperanto? I don't know much about it, but I believe one of the goals in its construction was to make it easy to learn, so it may meet some of your requirements.

    See en.wikipedia.org/…/Esperanto

  15. Joshua says:

    @Andy: That's rotten. Condider "no mí papa", does it mean "not my pope" or "not my potato"? (I've actually used "not my pope" before). What I learned from a native speaker is you really can have "la gato" though, with the obvious meaning (the particular cat is known by the speaker to be female).

  16. Mark says:

    Wear: don't worry, only ancient Latin teachers would ask you something pointless like that.

  17. John says:

    The homophone ambiguity reminds me of quantum computing.

    Quantum computer arranges a set qubits and the solution comes from how they collapse together.

    Language speaker a arranges a group of homophones and the meaning comes from how they sound together.

  18. pmbAustin says:

    I find stuff like this endlessly fascinating… but purely at a layman's level, as I'm no linguist and am essentially monolingual (I dabbled in French but never got beyond French 103 in college, and it's grown quite rusty since… I also dabbled ever so briefly in Russian).  But I love hearing these perspectives, insights, and discussions.  I could read stuff like this for hours and never grow bored.  I dated a native French speaker and got a lot of insights into my own language as well, which I share as often as I can to help remind people just how difficult English can be :-)

  19. Pseudo-Anonymous says:


    I suppose that a quantum computer would be much better at processing and understanding language than any classical computer.

  20. Wear says:

    @Mark Yes, but in a book about a language there is likely to be a paragraph that goes something like "These are the subjective pronouns in this language…". In order for that paragraph to be meaningful you need to understand what the subjective pronouns are in your language so that you can make a connection.

  21. Jon Forrest says:

    From my days as a Linguistics student I remember hearing the claim that there's some evidence that languages are constantly evolving between word order languages and case languages. Obviously this claim requires some objective evidence. The example I remember is Chinese. Apparently there are very old Chinese texts that show this. There's no evidence for other languages, but the thinking is that this is mainly because there aren't any texts in other languages that are old enough. (I learned this ~40 years ago so I have no idea what the current thinking is).

    Given the movement in spoken English away from the who/whom distinction, and other similar changes in "sub-standard" English, it's easy to see the change in the case to word order direction. I'd enjoy hearing about any examples of languages going in the other direction.

  22. jader3rd says:

    The occasional confusion with English, like the bat example, could be fixed by creating another noun for one of the objects. There's plenty of room to expand vocabulary in English.

  23. Muzer_ says:

    @Chris B: In our spare time, my friend and I have been working on our own language with much the same goals as you state (though not 3). It is indeed terrible for poetry ;)

    Our intention was for it to be really easy to parse by a computer – so every linguistic construct is based on the same idea, but taken to an extreme – prepositions and conjunctions act like verbs, for example. We haven't worked on it in quite a while and it's quite insane ;).

  24. Azarien says:

    @Jon Forrest: ancient Latin had a lot of inflections. Most of them are lost in its modern descendants, like Italian, Spanish or French.

  25. Mark says:

    Wear: you don't (generally) learn a language from a book.

  26. bzakharin says:

    Am I the only one who does not find "pair of scissors" acceptable? If I heard it, I would assume two and not one, unlike pants, shoes, etc.

  27. Chasm says:

    For those who find linguistic discussions such as this fascinating, check out the Language Log: languagelog.ldc.upenn.edu/nll.

  28. Chasm says:

    @Mark: But you might learn basic grammar from a book (ie an exercise book).

    @Boris: So what do you say, "a scissors"? That doesn't make sense, because "scissors" is plural, but "a" is not. Or do you say "a scissor"? Neither of those are correct.

  29. Joshua says:

    "It has no grammar." Let's see what would happen if that were true. In which case any amalgam of words would have the same meaning. This greatly limits the complexity of sentences as "subject verb object" is hard and "subject verb object if subject verb object" is borderline impossible.

  30. Clodney says:

    Talk of scissors and pants makes me think of the old Allan Sherman song "One Hippopotami"


  31. mandarin says:

    As a native speaker, I concede that I'm probably not aware all the exceptions that you have to deal with, but I still don't understand the bit you wrote about "没".  Treating it to mean "have not" makes sense to me.  In your examples:

    没濕 = hasn't gotten wet (rather than "isn't wet")

    没飲牛奶 = hasn't drunk milk (rather than "isn't drinking milk")

    They don't translate to "isn't X" because they don't just refer to the present state.  Something could have gotten wet then dried up, and it would be wrong to describe it as 没濕.  A person could have drunk milk in the morning but is not drinking any now, and it would be wrong to say they 没飲牛奶.

    [You are exploiting the coincidence that in English, the verb "to have" has two different meanings. One is "to possess". The other is as a modal particle expressing perfect tense. Or are you saying that in Chinese, the verb "有" also serves double purposes as a verb meaning "to possess" and as a tense marker? (Thereby contracting people who claim that Mandarin has no tenses.) -Raymond]
  32. user says:

    Learning a language by translating is always going to run into issues like these.  It's very rare for grammatical constructs to have exact direct translations between two languages.  To really make sense of a language, literally all existing notions of entities, space, and time have to be revised in order to "think" like a native speaker.  It's not that native speakers are ignorant about the differences or exceptions – to them there is honestly no difference.  The reason you think they are exceptions is because you learned to think about things with a different system.  It's like going to a supermarket and wondering why eggs and milk would be in the same aisle, when you've always thought chicken-related and cow-related things are very different.

  33. Jim says:

    You are lucky that we are living in 2014. Before, in Chinese written language, there was no punctuation. So how would you read? The tonic feature of Chinese was the only help you can get. BTW, it was very poetic indeed.

  34. happens in spanish says:

    @Joshua and  @Andy

    Well, Basically if you want to refer to a Female cat, you use "La Gata".. if you refer to a male cat, use "El Gato",, "La Gato" just sounds wrong when it comes to the gender of the subject…

    the same happens with most of the subjects that have genders (example, "La Luna" (the moon") instead of "El Luna"… Luna is a female subject in spanish, El Sol (the sun) is a male subject.

    The same happens in portuguese!,  obviosuly, if you're not a native or almost native speaker, such things are really difficult to understand, specially if you come from a language that does not handle gender for subjects…

    Chinese is just plain difficult to understand

    A friend said that Korean is a way much better to learn, easier when it comes to writing, speaking it is another big deal, lol!

    Good luck!

  35. mandarin says:


    "Or are you saying that in Chinese, the verb "有" also serves double purposes as a verb meaning "to possess" and as a tense marker? (Thereby contracting people who claim that Mandarin has no tenses.)"

    有 alone isn't commonly used as a tense marker (although I've seen it used that way, more often in some dialects).  没有 is indeed used as the equivalent of "haven't" as a verb tense.

    Treating "没" as a negating adverb might not be the most productive since removing "没" from a sentence but leaving "有" does not always work to cancel the negation.  It might make more sense to treat "没有" as a single unit (with the "有" being optional) that can have multiple meanings:

    – Does not possess/include (when paired with a noun): 没(有)钱 don't have money

    – Has not happened/performed an action (when paired with a verb): 没(有)吃 haven't eaten

    – Has not become/acquired an attribute (when paired with an adjective): 没(有)湿 haven't gotten wet

    – Some others I'm undoubtedly forgetting

    I'm not saying translating it into "have not" or "not have" in multiple senses of the term is the rule by design.  But if the happy coincidence lets us simplify the way to think about it, then why not exploit it?

    I have no problem with contradicting people who pretend Mandarin has no tenses.  As you said, it has no inflections but tenses certainly exist in the language.

  36. DWalker says:

    @Chasm:  (I couldn't resist.)  You meant to say "Neither of those IS correct."  :-)

  37. DWalker says:

    Here's an interesting tense:  "I would have liked to have known her."  (Followed by "But I was juuust a kiiiid….")

  38. Mac says:

    "One of my colleagues told me that he missed those signs on his trip to Italy. There would be signs labeling each exit, but rarely was there a sign telling you what highway you were currently on!"

    That's another culture thing, I guess: we tend to miss a lot of information just because we don't know what to look for.

    In Italy, the easiest way to find out on which highway you are (without using a GPS) is to look for an overpass: there'll be a pair of square brown signs, one before and another after the overpass, bearing the name of the highway and the number of the overpass.

  39. Jeffrey Bosboom says:

    Why do we say "a pair of pants"?  Because "a pant" was at one time the word for what we now call a "pant leg", and you did indeed wear two separate pants: english.stackexchange.com/…/why-is-the-word-pants-plural

  40. @Chasm: I've both said and heard people say "a scissors" in casual speech, such as "Hey, do you have a scissors lying around?"  If I think about it enough it starts sounding a little awkward, but I'd say it's pretty much as acceptable as "a pair of scissors".

  41. JDT says:

    I've only ever heard of "classifiers" under the name "measure words" previously; and I believe it conveys the meaning better. For example, in English the nouns that require measure words are generally the uncountable ones (ie. those without inherent unit); eg. "bread" requires a measure word whereas "bun" does not; "water" requires a measure word whereas "drop" does not.

    It's often interesting to me that "whom" is unpopular. The mappings "she" => "her"; "he" => "him"; and very useful and naturally used by many, but the same mapping "who" => "whom" is not and confuses people.

    The only other language I type regularly is Hebrew; and it is always interesting to see how different pieces of software cope with switching between left-to-right and right-to-left — many of the weird things I see have been previously explained on this blog — thanks :)

  42. Behodar says:

    @MNGoldenEagle: I would use "Do you have some scissors…" or possibly "any scissors". It's rare that I'd specify "a pair", but I'd certainly never say "a scissors"; that just sounds wrong to me.

  43. Timothy Byrd (ETAP) says:

    @MNGoldenEagle: I would phrase it as "Do you have some scissors?"

  44. Joshua says:

    There have been cases where I needed a scissor. It makes a good letter-opener.

  45. cheong00 says:

    Actually I'm always conscious that my mother tongue (Cantonese) is one of the "hardest to learn" language in the world.

    To be fair, Mandarin is not that hard when compared with learning the (now abandoned proposed standard) third generation of Simplified Chinese. In the current version of Simplified Chinese we see, at least a character is a character. In the 3rd version you may found a character representing something you'd ordinary say as a word. The government officials realized that they're overdoing it so the proposed standard was thankfully abandoned.

  46. Alex says:

    I think a sheet of ticket is just as (linguistically) interesting as a school of fish.

    Mandarin is a simplified form of Cantonese and is already easier to learn. The Manchurians never succeeded to pronounce the sounds of Cantonese and ended up adopted Mandarin as the official Han language of the Ching dynasty.

  47. cheong00 says:

    @Alex: Actually Cantonese is one of the older languages in China. It was established back before the Han Dynasty, used by Chu Kingdom (the southern region on the map) in the Warring States era.

    That's why most of the "Chu Ci" (a kind of essay that used to be recite with particular rhythm) still sounds right with Cantonese.

  48. John Doe says:

    THe term "normal language", "ordinary language", or better yet, "common language", is due to Chinese, or rather, Mandarin, becoming a single language.  Give or take simple vs. traditional and different phonetics depending on the zone, it's still written the same way.

    In software, we have an example, called Common Lisp, an attempt to standardize and unify lisp, which it did, but it was so expensive that it was fossilized.

  49. cheong00 says:

    And than regarding Mandarin, it's language origin was the "Northern Officials' Language" (北方官話), which evolved from "Officials' Language" (官話) that established somewhere in latter Han Dynasty, so it's not correct to say "Mandarin is a simplified form of Cantonese".

  50. Cesar says:

    @user: that reminds me of one thing that trips me *all the time* in English… The prepositions "on", "in", and "at" map to the same preposition ("em") on my native language. When writing in English, I never know whether I used the correct one. Except for the very obvious cases (on the table, in a box), I end up going by "pattern matching" with everything I've ever read in English; the correct case should sound more natural since it's the one I've read more often, but there's always a nagging doubt.

  51. Quirk says:

    @Raymond: In English, ambiguity is frequently and gleefully exploited to humorous effect. This is especially true for homophones. Is the same true of the ambiguities in Mandarin? Do Mandarin speakers make more puns than English speakers?

    [Puns exist in Chinese, but homophones are in such abundance that ambiguity resolution is run automatically. If you had to consciously resolve ambiguity, you would be overwhelmed by the possibilities. You might not even realize that there's a pun in there until somebody points it out, and then you have to backtrack and run your parser a second time after disabling the ambiguity resolver. By analogy: Consider a chess puzzle, and the answer is "Rb5, checkmate!" You say "No, that isn't checkmate." And then you have to stop and realize, "Oh, he meant to move the other rook, the one on a2, which would be checkmate, except that it's also an illegal move." You had to rewind and turn off the ambiguity resolver ("Was that Rab5 or Rcb5?") in order to let you see the illegal move. (Side note: In Chinese, puns are sometimes used to get around censors.) (Second side note: My ambiguity resolver for Cantonese is not well-developed, so I see puns where others don't. I'll make a pun, and people will just look at me funny until I explain it. Their ambiguity resolver filtered out my joke.) -Raymond]
  52. Dave Harris says:

    According to Iain M Banks, before his Culture settled on calling itself The Culture, there was a strong movement to call itself The Aliens.

  53. @Quirk says:

    Take, for instance, "He had a bat". Until recently I wouldn't have laughed at that. I'd have thought 'he had a go'. Good on him.

  54. Marcel says:

    That was really interesting to read, thanks a lot! By the way, as a native German speaker I have never assumed that it's easy to learn my language, quite the opposite actually :-)

  55. HollerWorld says:

    Maybe the next time somebody says that Chinese has no grammar you can point out about word order, which *is* grammar.

    (I'm not very knowledgeable in Chinese, so correct me where I'm wrong)

    Like if you say "我爱你", this can only mean "I love you". It can't never mean "You love me", unlike, say, German and Latin where you can switch the subject and object and the sentences still have the same meaning but translating the sentences word for word (without regard to case) will give the wrong meaning.

    Also if you want to say "I love you", it has to be "我爱你". Or maybe (I'm just guessing here) "爱你我". But in either case the word order grammar rule will give only 1 unambiguous meaning.

    Perhaps they don't realize that word order is grammar because all languages they know have the same word order. If they also know Korean or Japanese (SOV), or Irish or Tagalog (VSO), or Klingon (OVS), or German, Latin, or Russian (somewhat free in order), or French and Spanish (which can be SVO or SOV depending on whether the object is a pronoun) they would have thought differently.

    Or perhaps their understanding of "Grammar" is the overly simplistic, "inflected by person and number".

  56. Random832 says:

    > (They will naturally defend classifiers by saying that they are completely obvious and in no way arbitrary.)

    Maybe _because_ they're native speakers, their mind imagines a commonality between ribbons and dogs and skirts that doesn't actually objectively exist. Like how different languages' color names allegedly affects how people see color – Russian having two different kinds of blue [one of which includes most dark blues, and one of which includes the color of the daytime sky] which they don't see as any more similar than blue and green. Or, on the other hand, Japanese having a category "blue" which _includes_ green.

  57. Klimax says:

    @Marcel: OR Czech language. We already have enough problems with it… (Complex grammar and free form in one package)

  58. John Ludlow says:

    This reminds me of every time someone tries to explain F# to me. They say "look, it's simple". Then they say something about currying (which is apparently nothing to do with a Lamb Balti) and the rest becomes gibberish.

  59. George says:

    In his autobiography, _Little Wilson and Big God_, Anthony Burgess writes that

    "[Malay] was a revelation. It had learned something that the more conservative tongues of the Indo-European family did not wish to learn–that properties like gender and word inflection were needless luxuries, that the strength of a language lay in semantic subtleties and not syntactic complexities, that the rigid taxonomy of 'parts of speech' mean nothing. The Malay language, and later the Chinese, changed not just my attitude to communication in general but the whole shape of my mind."

  60. Muzer_ says:

    @John Ludlow: I actually read that as "Lambda Balti" and thought you were making a pun :p

  61. alexi says:

    Regarding 普通话, I recall a Shanghai cab driver complimenting my friend (from Inner Mongolia) on his Mandarin using a phrase that ended with 很普通的. He translated it to me as "your speech is very standard." (As an American I know we have dialects here too, but the idea that a Cajun could compliment someone from Maine on the "standardness" of his midwest accent is just really funny.) To compliment another Chinese native on his standard Mandarin definitely does not feel like "everyone talks like this" attitude to me.

    Also a lot of older Shanghai people, especially older folks, instead of 普通话 would say 国语 which I took to mean "national language".

    Finally, regarding  有 as a "perfective" particle. It seems like it is not a coincidence that it's also "to possess" because it does the same double duty in a lot of modern European languages like modern Greek, German, Albanian and English. And in Portuguese it even shows up twice — there are "ter" and "haver" which both mean to have but both serve as auxiliaries for perfect constructions. Seeing it do the same thing in both Indo-European languages and Chinese is kind of amazing.

    What a fun blog you have!

  62. stephen says:

    As a native Chinese speaker, English is too hard to me,I hate grammar :-(

  63. Neil says:

    @Boris One pair of anything is still singular.

  64. cheong00 says:

    @alexi: Actually it's called "國語" in Taiwan too. I think I've seen it on timetable of my cousin like 20 years ago.

  65. Pseudo-Anonymous says:


    Not to nitpick but isn't the Spanish preposition that covers English "in", "on", or "at" written "en" and not "em"?

    Going back to the article, I do find it funny though that everyone assumes what they experience or know to be "normal" and that everyone else is just going out of their way to be non-normal.

  66. Myria says:

    I'm a native speaker of Western U.S. English, yet I recognized by learning Spanish just how hard English is.  English is just a huge mess.  How to inflect words tends to vary depending on whether the word is German or French/Latin in origin.  Spelling is ridiculous because the printing press came to England at the worst possible time in the history of English.

    The way ambiguities in English are resolved is annoyingly subtle: in the last sentence of my first paragraph, if I had used "English history" instead of "history of English", it would sound like I was referring to the history of England rather than the English language.

    Of all the languages I've seen, which, granted, is not many, Spanish seems the simplest among the natural languages.

  67. Guest 2014 says:

    Why do you use Traditional Chinese characters for the most part in the post, but use 条, which is Simplified?

    [Because I'm not a native speaker and mess up a lot. -Raymond]
  68. Anonymous says:

    > 1. Do not assign gender to inanimate objects. It is not obvious that doors are boys and windows are girls.

    That is just to your English speaking brain.  To someone speaking a language with gender this is a goofy statement.  Various phrases would "sound wrong" if you don't have proper agreement of gender.

    Even your exception ("inanimate" objects) shows your English speaking bias.  If we are being objective, why exclude animate objects at all?  English used to have genders for inanimate objects, too.  It happened to drop them for inanimate objects and retain them for a few pronouns.  But that is just the path it took, not all possible paths.  Someone who speaks a language with no gender inflection whatsoever would find the distinction between "he" and "she" odd and confusing.  I know people for whom this is a problem in their English learning.

Comments are closed.

Skip to main content