Lantian on ‘parsing’

Lantian has an interesting post, Did that click for you? on his blog. He looks at the subject of ‘parsing’. I’ll let him explain.

When native speakers of a language speak, hear or read a sentence that is not properly formed they will often say it “just didn’t sound right.” And if the sentence is then properly re-ordered or the proper word inserted, there’s a mental “click” that happens; almost a feeling of relief, “There, that’s right.” Do other’s know what I’m talking about? (that apostrophe in ‘other’ makes you quesy doesn’t it?) You know this feeling right?

My favorite line:

I have come to realize one of the main reasons why I dislike typical classroom language instructions and materials is that there is a huge focus on what words “mean” rather than exposure to what words should come up before or after a word.

It’s well worth the read!

Ken Carroll

26 Responses to “Lantian on 'parsing'”


  1. 1 Mashhood Oct 25th, 2006 at 10:32 pm

    Yes,

    very interesting…i completely agree…

    any ideas on how to increase this component of our learning?

    I have recently felt a growing need to focus on listening/watching real media…so i have been watching a lot of mandarin soaps…in particular ‘Devil Beside You’. I do feel this is helping me….

    anyone have any ideas on how to make the transition from thinking in one’s native tongue to thinking in mandarin?

    I really think Input is the most important thing…listening to a lot of natural coversations….building up ‘lexis’…i think this is the most important thing in gaining true fluency in any language

  2. 2 chris(mandarin_student) Oct 26th, 2006 at 12:03 am

    You may guess that I may have a opinion on this.

    I have been using the ‘doesn’t sound right’ feeling for ages. As I have said before I have consistently listened to huge amounts of ‘real’ Chinese from the start (against almost all advice). I am convinced it is paying big dividends now. I am also pretty convinced that the ‘doesn’t sound right’ ability develops into a complete substitute for actually having to do any formal grammar study (YAAAAY). Don’t get me wrong grammar is great in its place but I am learning Chinese not linguistics (why do so many people seem to have a problem seperating the two).

    Almost all day at work now I listen to Chinese radio whilst working, or podcasts or whatever depends on how much I need to be thinking at work. Don’t get me wrong most of the total meaning is still beyond me but sometimes now I get a hook into the context that lets me follow on for a bit. Strangly even when I think I am not paying attention I often hear a familiar collection of words in a strange order (sometimes with a couple of new syallables with them). I have a notebook and quickly jot down the pinyin and any surrounding context. Almost always it turns out to be a new peice of juicy lexis. What dragged me back to focusing on the audio, that selction of sounds that triggered familiarity but didn’t sound quite right of course.

    I guess the best way to get this is to talk and listen face to face, followed by voip, followed by listening to media, followed by text chat, followed by reading subtitles …..
    I can gaurantee it is not present at all in a dictionary, hardly present in literate Chinese writing (’doesn’t read right’ is NOT the same skill at all, sorry Lantian I didn’t even see the apostrophe you mention, maybe because I speed read?). Lesson dialogues have very little but CPOD somewhat compensates for this by having so many you can listen to and interesting chat at the higher levels. A more restrictive course of instruction has you ‘running on the rails’ of their set piece conversations, you will pick up very little ‘dosen’t sound right’ ability from these I feel. Unfotunately judging from comments on other forums restricting yourself to one course of study of (especially a restricted one) often seems to leave the unfortuante students with an impression that all other Chinese ‘doesn’t sound right’, not what we want at all.

    Good post there from Lantian, I think CPOD already addresses more of these issues than most but it would be could to think about how that could be impoved even further.

  3. 3 mike in Jubei Oct 26th, 2006 at 12:04 am

    Mashood

    I am with you and of course with Lantian on this point. I think maybe if CPod could have the equivalent of ” Director’s Notes” or a “Teachers Handbook” it would be helpful.

    Of course I am lazy and expect CPod to give me everything. So what I mean by this is not just, ” this is very lexical” but a deeper level that helps me understand more why this is not just English translated into Chinese but what makes it Chinese and not English.

    One example is so simple but is a useful : DIRECTIONS ON THE GLOBE, unless I am mistaken in the West we all think Northwest is correct for someplace up and to the left (Aussies is this true for you guys with the world upside down?) but in China “we” say WestNorth ! even for NWA the US airline Northwest Airline in Chinese it is WestNorth Airline. I use this as the simpliest example but I would love to have a way to comprehend this deeper, faster, sooner, more from CPOD.

    I think we have al seen movies where a non-English speaker has learned English from outdated TV and uses expressions that only you can hear if you watch obscure channels on cables as some one speaks dated English.

    Todays lesson had an excellent example. “How does it feel”. I would have said “ganjue zenme yang ” but the lesson used ganjue zenyang” So ok but when can I use this? All the time or do I sound like some US TV program from the 80’s if I use this . I don;t want to look like I should have long sideburns , wear cardigans and resemble Woody Allen when I speak Chinese.

    Mike in Jubei

  4. 4 海宁 / Henning Oct 26th, 2006 at 5:10 am

    What I found most intriguing about Lantian’s parsing-post is the discussion about those break-down points. That is when the flow of comprehension is cut off and everything suddenly drops.

    In most Intermediate lessons there are words and even larger chunks in the banter I do not know yet - but in many cases I do *not even notice* because the “flow” is there.

    Sometimes while relistening an old show I suddenly identify words that I just very recently learned - but that I have never been aware of before (although they must have been in that particular show unless you changed the file on my harddrive).
    That even happens when I listen to a show for the 3rd of 4th time. Those “Dark spots” obviously do not destroy general understanding.

    When the flow is gone quickly everything turns black - even the vocabulary that I should know since eons ago. Of course that happens when unknown keywords come in rapid succession (some Upper Intermediate and all zh.Chinesepod without proper preparation).

    Even worse than these natural “I just don’t know” situations are what I would call the “evil blockwords”. Those are words or phrases that I know I *should* know. They drag away my concentration from the actual meaning and I start thinking “What was that? I know that. How does it fit in here?” And because dialogue continues this accumulates: “Ups - I just missed that Chengyu…I remember…what…another one…[fade into semantical void]”. It is like a mental hiccup. I hate it.

    Only remedy I can think of is learning more and more vocab. By reading. As reading does follow your own pace (and you can cheat with your dictionary).

  5. 5 Mashhood Oct 26th, 2006 at 5:40 am

    My chinese studies are taking a new direction…

    ChinesePod’s role has reduced quite significantly…although it still has an important role. I don’t feel i am using my time as efficiently when listening to chinesepod…because its not easy to review previous lexis (takes too long to go back many many lessons and re-listen)…and i have noticed that some things i learnt before have faded slightly…so i have almost completely shifted my focus to real native speed chinese - streaming tv/radio/ and chinese podcasts. I think this is better for revising and reinforcing previously acquired language…and also for building up understanding of ‘native speed’ spoken chinese.

    My time is roughly split 3:1 75% on listening and 25% speaking (skype and other friends). I can now follow the gist of near-native speech if it’s on a topic i am familiar with. In terms of chinese radio - i recognise over 50% of the words used…and sometimes follow the gist…but yet need to continue listening and listening.

    I was re-reading the theory of how babies acquire language - the first thing they acquire is the sound pattern of the language. They don’t understand any of the language at this point…but somehow through exposure they acquire the phonetic structure of the language and at the same time develop the fine motor requirements to reproduce those sounds. Its only later that they begin to ‘decode’ the speech that they hear. I wonder what if it would be at all similar if someone who didnt speak chinese was left in china..with no access to translators or english speakers…

    so…for at least a month i am direct my efforts in this direction…and hope my stubborness pays off… :) Hopefully strength my listening ability.

    Another thing that i’ve realised…and don’t get me wrong here…i think ChinesePod is fantastic…i am a huge promotor of chinesepod (i always talk about chinese to my friends or to new people i meet and recommend it to them)…but although chinesepod is excellent…it’s difficult to actually learn how to sound completely natural..ie..to become fluent and sound completely natural in the language…one definitely needs more than chinesepod can offer at the moment. And i understand this is not something we should expect from chinesepod…its asking too much…and could actually cause difficulties. China is a huge country…people in different parts speak ‘different kinds’ of mandarin…also the youth have a slightly different language. Then there’s Taiwan…taiwanese mandarin is definitely very different…it’s difficult if not nigh on impossible for ChinesePod to address these issues. Anyway my point is that chinesepod is just the beginning (an excellent beginning) but if one’s goal is true fluency then more time and effort (particularly in the later stages) is needed on focusing on real media. And as Chris has found out - the earlier one starts to listen to real chinese media the better :)

    However, i should repeat that without ChinesePod, the afore-mentioned is very difficult (i’m trying not to say impossible). And ChinesePod is definitely the easiest and quickest path to opening the window to fluency.

  6. 6 Mashhood Oct 26th, 2006 at 5:44 am

    ouch…as my chinese improves…my english worsens…has anyone else noticed that? (And english is my mother tongue!(virtually)) - well the only language i am truly fluent in.

    Mashhood (Uk)

  7. 7 memcpy Oct 26th, 2006 at 8:41 am

    I think that this subject is a little far out of everyones grasp. But still interesting…

  8. 8 Ed Oct 26th, 2006 at 11:06 am

    Isn’t this just the point that words by themselves are poor communicative devises, it is only when they are put in a row in a sensible manner that they can communicate information. Mapping a newly acquired word back to how ‘we’ would use it in English (for me!) causes us to make assumptions that are not always valid. So test driving the word in a context that works and ones that don’t allows us to position it relative to how we instinctively use it in English.
    There is so much tacit knowledge associated with a word that a native speaker could write volumes about a single word! This is why I was suggesting previously that if the text of the podcast allowed discussion links for important words an underlying dictionary of usage, making explicit this tacit knowledge, could evolve. This would harness the power of the Big Brain in a very effective way. If as an elementary Cpodder, I have to learn a character, its stroke number, its stroke order, its tone - I want it history, it’s usage, its coolness, even what’s its favourite colour, boxers or briefs - anything to bring the character to life and make it mine.
    What do the words that you choose to communicate say about you - probably a well intentioned foreigner to a native’s ear but how can I be good enough to be a spy - a benign one!? How can I fake being a native. The words in a native’s head are like rounded pebbles on a beach made smooth by millions of waves crashing over time, how can I get to smooth with only a few hours of wave crashing a week?!
    Admittedly this idea is low tech, uncool, tedious to implement but the value contained in this knowledge is I believe immense.
    Ask Jenny, John and Connie to braintorm around a word, anything goes - be it silly, funny, stupid. I bet the results although unstructured would be informative and with the faint sound of a ‘click’ at the end.
    I envy Mike in Jubei who has neighbour’s kids to practice CHinese with, no wonder he is a star at Cpod. The learning of Chinese is very adult-like, we need to play with words as if they were lego.

    Here is an example of the English word nice:

    1. it is nice to be nice
    2. Tastes nice but would probably say “sounds good”. Could say looks nice
    3. Nice is a bit if a feminine term, as a man if you are describes as nice, it might be a veiled insult.
    4. Sugar and spice and all things nice- from a nursery rhyme.
    5. nice is a brand of biscuit with “nice” etched from the surface.
    6. “Have a nice day” - popular expression in America but pointed to by non-Americans as cliched and said without meaning.
    7. ‘Nice and easy’ is a popular expression to indicate a relaxed approach.

  9. 9 Will Oct 26th, 2006 at 11:33 am

    I get this especially with Chinese and English. Some languages (and some people) are much more relaxed about what ’sounds right’ or not. I think English and Chinese are on the end of the spectrum where things are often correct, but don’t sound right.
    So many times I’d come up with a sentence in Chinese, only to be told by my teacher “X is correct, but it doesn’t sound right. You’d be understood, but you’d sound like a foreigner. We just don’t use those words together. You should say Y.”
    It’s important, if you want to sound good in a language, to take this into account. Of course if you just want to make sure you don’t get ripped off while you’re haggling, it’s not so important.
    When I was learning Spanish, some people could understand me even when I forgot what gender a word was or mucked up the verb conjugation. Other people would misunderstand me when I had everything correct except I didn’t use the subjunctive. Then I’d be told that it’s not correct. You have to say it like this.
    As a result, my grammar’s usually pretty good in Spanish, but I haven’t got it in Chinese as much because I don’t use it enough.
    If you want to make yourself sound more Chinese, subscribe to Mike’s favourite Lexical Chunks theory, where you can learn bits as a phrase, as combinations that you know work well together. And read and listen to as many people as you can and copy them.
    Having said that, I rarely follow my own advice there. I’m a slacker.

  10. 10 Erika Oct 26th, 2006 at 1:53 pm

    You know, I think my one good access into the language at the beginning of my studies is when my husband (who has a big head) taught me a little rhyme about how if you have a big head you don’t need an umbrella when it rains. It’s funny, short and something I could remember.. really helped me practice the tones.

    A cultural element does help I think.. which is maybe why I like C-Pod so much.. I feel like I’m learning in a dynamic way, even though it’s still often me talking to myself. :-)

    Erika Lee

  11. 11 Lantian Oct 26th, 2006 at 3:12 pm

    Hi memcpy,

    “I think that this subject is a little far out of everyones grasp. But still interesting”

    I guess that’s why Steven Pinker’s book is a national best-seller and I just have a little blog in a small corner of the internet! It was really tough trying to write the post, no wonder Ken and John are such good ‘blog writers’, they have a knack for getting to the point!

    One thing about this topic ‘parsing’ that I think is also very interesting is that actually it’s a very ‘natural’ way of learning. It should be in every student’s grasp, but I think adults often let their education and ‘intellect intrude.

    For example, when one talks to kids what happens? If it’s a grown-up conversation’ they often tune out and go play on the X-box. Kinda like me listening to Chinese broadcast news.

    But if you have them ‘engaged’ in a conversation they are definitely working the ‘parsing’. As soon as they get to a point in the sentence they don’t understand, they ASK.

    For example, if you just say ‘ni kan’ (you-look) …and then wait a few seconds longer than expected, almost any Chinese kid will impatiently YELL, ‘kan shenme?!” (look-WHAT?).

    In other words, almost after every natural ‘idea’/parsed block, their mind naturally asks ‘who, what, when, why’. And they interupt, yell and talk all the time this way. We adults have gotten so ‘polite’ and intellectual that we have a tough time doing this.

    One idea that I have is that early Chinese curriculum really needs to drill these question phrases into our minds, so that we can mentally ‘think’ and ask in Chinese. I’m not talking about a full interrogative sentence that rephrases a whole sentence, but the lexical chunk that is the minds ‘mentalese’. Then we need to practice doing it. (It’s not easy — just getting the ’shenme’ to be a reflex response, might take a couple-months of “no English” to make the switch.)

    For example (we shouldn’t do this):

    The wind is blowing northwesterdly on the tarmac.
    Q: Which direction is the wind blowing on the tarmac?

    A parsing question intead would look like:

    The wind- is what? (什么)风?
    Blowing- blowing where? 吹(哪里?)
    The wind - where is this? 风那的? (那个地方的)?

    I’m curious about Mike and others, Do you ask these questions in your mind in English or Chinese?

  12. 12 Lantian Oct 26th, 2006 at 3:17 pm

    Hi John,

    Do you have some extra cycles to comment on ‘parsing’. Is it part of studies in applied linguistics? Do the currenty academics consider it an outdated concept? A valid concept? What’s the latest, greatest? I still think I wasn’t able to explain well the difference between breaking down a sentence ‘grammatically’ versus diagramming out a ‘right-branching tree’. And most importantly, is it an important exercise/tactic to learning languages?

  13. 13 Mike in Jubei Oct 26th, 2006 at 3:55 pm

    Lantian

    I practice “lexical chunks” or parsing all the time with the little kids in my apartment building. Those under 10 who are supposed to know a little English. Mostly their moms push them forward and make them speak to me.

    So after hi and what is your name it is up to me. So I speak to them in English and Chinese. asking the same questions or responding using both languages. In doing so I don’t have the brain processing power to make full sentences. And with their limited english but speaking in chunks it does seem to break down the barriers.

    If mom is not there we actually can talk more. My biggest limitation is not language its just what do you talk to a 9 yr old boy and his little sister about.

    Mike in Jubei

  14. 14 Michael Butler Oct 26th, 2006 at 4:41 pm

    Mike in Jubei,

    I don’t think that parsing and “lexical chucks” are equivalent. They do overlap to some extent but one (parsing) talks about dividing lanuage into (universal?) grammatical categories while the other (lexical chuncks) talks about how to combine multi-word units to form a single meaning (that sometimes is not explicable by looking at the separate words). One (parsing) tends to talk about dividing while the other (lexical chunck) tends to talk about combining.

    In fact, it is my understanding that the notion of chunking and parsing are at odds with each other in some academic circles in terms of how our brains create and understand language. I’ll see if I can find my references.

  15. 15 mike in Jubei Oct 26th, 2006 at 10:16 pm

    Michael Butler,

    You explanation is very clear and thanks for the clarification. No wonder I have problems I am parsing when I should be chunking and…

    I do beleive I do a bit of both in fact all in an attempt to communicate in English and Chinese with the kids when are strengths are the opposite language. And the age difference means simplifying anyways to convey a thought on a topic at their level.

    I say that realizing if the topic was some computer game I would be the one with no comprehension.

    Chinese is only my second language and I am very poor at it as well. Besides enjoying the entire learning experience (a selfish pleasure) I also am amazed to “observe” my own brain as it works trying to convey thoughts in a different language with natives Chinese. Its far more than just having a vocabulary in chinese available. I suppose this is what parsing aand chunking is all about.

    Lastly I am fortunate to be in Taiwan and speak to people who do speak Chinese. I am curious what others may think about the difference of speaking/practicing Chinese with fellow students who again unless the teacher is good as well as observent may ignore the fact that the communication is mainly a direct translation of say English into Chinese rather than speaking in Chinese as Ken and the CPod staff are trying to instill in us.

    Wihtout trying to promote the Premium Section , John and the academic team are really starting to do a great job of giving plenty of varied examples in this section to enhance the experience of thinking in Chinese.

    Again I want to thank you for your explanation about parsing vs chunking

    Mike in Jubei

  16. 16 chris(mandarin_student) Oct 27th, 2006 at 1:01 am

    Mike being in England probably accounts for the differences in our learning styles. Apart from that there are apparently many striking similarities.

    I am really having to try very hard now to find anyone to talk to too. Skype helps but I am not often online at the right time and SKYPE is not as good as face to face conversation. Some small part of my brain is tying to think in Chinese all the time (with varying degrees of success). Some thouhgts are now completly in Chinese (wei4shen2me? , you3 shen2me yi4si, hao3de) stuff like that. Even when I am talking in English so these simple thoughts have to be converted back to English . I dream in Chinese (which restricts converstation somewhat). I am trying to create an immersion environment by fooling my brain.

    I could probably do with some time in China or similar right now but it isn’t going to happen so I guess I just have find new ways to be inventive :).

  17. 17 Lantian Oct 27th, 2006 at 3:52 am

    THOSE EUROPEANS - I’m very much a mono-linguistic imperialistic American English language snob, and as such I’m not part of the world-majority that can casually switch between 2, 3, 4, 5 languages/dialects. I think the reason though that some people can do this is because there is similar parsing between certain language families, it allows one to parse at the right places, even when words are unknown. I think it’s what allowed one of my Danish friends to read some original philosophy in Russian…for fun!

    It’s also what makes some languages harder for those from a certain linguistic background (for example a right branching language). Take for example Japanese which is typically a left-branching tree with head-last, that final verb in a Japanese sentence that tells you what the whole sentence was about. This is very different from English sentences.

    My hope is that after some explicit going over of parsing in a target language, then some exercises and practice,it opens up a flood-gate of language acquisition. This would beat memorizing/forgetting 10,000 vocabulary flash cards.

    Any multi-linguals out there that can describe how it is they can speak so many languages?

    Mike and Michael,

    Here’s an example of ‘parsing’ versus ‘grammar’ or ‘lexis’. It’s what makes the following sentence easy to read.

    Ex: This is the cow with the crumpled horn that tossed the dog that worried the cat that killed the rat that ate the malt that lay in the house that Jack built.

    Ex: This is the cow/ with the crumpled horn/ that tossed the dog/ that worried the cat/ that killed the rat/ that ate the malt/ that lay in the house/ that Jack built.

    My belief is that ‘conversations’ are also 80% pretty much like this. So this tells me that one way to get my Chinese to 80% there is to learn how to generate all these little chuncks (is that lexis?) to convey each like a train carriage chain of thought.

  18. 18 Bob Mrotek Oct 27th, 2006 at 5:26 am

    Lantian,

    I think when people can speak another language like Jenny speaks English, or John speaks Chinese, and Ken Speaks German and Chinese, I think that they have a rightful claim to call themselves bilingual or multilingual. On the otherhand, I think that there are a certain people who only “parrot” a number of languages and only in their own minds are they truly “multilingual”. Having no standard to judge by we can only speculate. In the end, however, I think your 10,000 flashcards is the best way to go. That, and about ten years of study and patience. In that way, in about fifty years you too will be able to claim five or six languages so that amaze your friends and terrify your enemies :)

  19. 19 Michael Butler Oct 27th, 2006 at 1:11 pm

    Lantian,

    Insofar as parsing is concerned it appears you are almost there (I think).But does this help us learn a language? I’m not sure. I would however like to point out something quite interesting about your example.

    Based on your example I’ve come up with a very close cousin to your original sentence. My sentence, as a whole, is close to “clicking” but it obviously doesn’t. But look at the separate parts of the sentence bit by bit and I think you’ll see some great similarities to your original.

    Lantian:
    Ex: This is the cow/ with the crumpled horn/ that tossed the dog/ that worried the cat/ that killed the rat/ that ate the malt/ that lay in the house/ that Jack built.

    Michael:
    Ex: This cow/ has a crumpled horn/ tossing the dog/ worrying the cat/ killing the rat/ eating the malt/ lying in the house/ that Jack built.

    The point I’m getting at is this: we can make sense in small bits but when we stick these bits together we can end up creating nonsense. Getting the parts rights (par-sing) doesn’t mean we can get the whole right.

    Now, in defense of your idea, the sentence you produced belongs in a book (or is part of a script we call poetry). Conversation, or unscripted speech, can be broken into smaller pieces than writing and still make sense. As Ken pointed out earlier we don’t always speak in complete sentences.

    Finally, I’d like to clear something up. Is your “defense” of parsing a defense of the idea that we can learn a language by taking it apart and understanding it as bits and pieces strung together?

  20. 20 Ken Carroll Oct 27th, 2006 at 2:41 pm

    I love when this happens: I try to post something interesting, then just stand back and let the Big Brain go at it. Excellente.

    Ken Carroll

  21. 21 chris(mandarin_student) Oct 27th, 2006 at 8:42 pm

    eeerk.
    This is getting too intellectual. I do like pineapple chunks though.

    Assuming that what we are looking for is what little kids pick up naturally why do we need to analyse it so much?

    My basic gut feel is simply that I need to listen to as much Chinese as possible early on, from as many different sources as possible. I can take some semi-chewed and easily digestible chunks (just like mom dolled out in my native tongue when I was very young) and I can take full speed full on language. Much of the full speed stuff will wash over me but still leave its impressions in the coastlines of my mind. Increasingly I will want to parse more and more of it to grab information that is interesting to my selfish little child brain.

    When it comes to talking I will say it just good enough for mom to understand my most pressing need, then dang I have to up-my game a little bit so some other big people understand. Some things I will say very well and some will be rubbish. Then I talk to little poeple like me for the first time and I have to improve my speaking again becasue that other stupid kid doesn’t seem to understand that the toy he is playing with is MINE. I will keep developing in this manner, the more differences I have to adjust to the more flexible I will be. If I get too lazy as I am growing then mom will make me go to a speech therapist but only because those other stupid little people can say something much better than me. Throughout this I do not have say anything perfectly before I move onto the next word.

    I will learn to read and write only the spoken language I already know. What did you do at the weekend? “I played” very good. Later: you should write who you played with, “I played with Ben” very good. Later: who is Ben, “I play with my friend Ben” very good but remember that should be played not play.
    Eventually I start seeing some strange written words I never really use but I get used to them becasue mom reads them to me from a “big boy” book that I can’t quite read yet. If I am lucky I will eventually get an education and learn lots of new words and patterns that I never really use day to day unless I am writing or sometimes talking to other ‘educated people’. Sometimes I am scared talking to big people I don’t know becasue it is important, and I am scared that they won’t understand me or think I am stupid, but most of the time I can run around saying what I like without being worried about being made to look stupid.

    So I am trying to grow a Chinese child alter ego in my mind. Sometimes it is confusing becasue he is growing in a mind that already contains a lot of sophistaced but rigid adult structures . Because of this I may fool myself in the early stages by thinking that I am dealing with adult issues but really the Chinese child ego is just very precocious and occaisionally says something very sophisticated. The Child is also very backward and sometimes realises he can’t say simple thinks yet like “where is the toilet” (even though he can count to a million, strange little child but still a child).

    How do I apply this, well I didn’t bother with characters until I was ready, I listen to all levels of Chinese, I stopped worring about sounding like an idiot, I don’t stress about anything I can’t pronounce (just re-visit it from time to time). The talky bit is a problem at the moment but only becasue I don’t have the right time or access.

    So what to do, linguistics people quite rightly are going to go off on one about this topic, but that is their job. Instead of trying to build some sophistacated theoretical lexical excercise generation algorithmic post-parsing machine, why not keep it simple and try something.

    Maybe ChinesePod could select some audio, that is appropriate and post it somewhere. Just the audio, no text. We are advised to go off and listen to it, glean what we can but basically get familiar with it (whether we can fathom it out or not). A week later there is a lesson disscussing the audio. The lesson could still be instructional even to those that didn’t do the ‘homework’. Only then do we get the transcript so we can’t cheat and pretend we are educated in Chinese. Grab something real like part of a chat show dialogue. The cpod lessons are a bit like this anyway but I think haveing adsorbed the test dialogue for a week or so beforehand will provide some interesting feedback.

    I am your Child, I have been singing from this hymn sheet from almost day one, literal translation, don’t learn from characters, listen to Chinese and listen to more Chinese, don’t get overstressy about the pronounciation early on (if you form a permanant bad habit that is your fault, sort it out, it is not inevitable). Maybe I have got somethings wrong, but I am learning Chinese this way and it feels good.

    Pass me a pineapple chunk somebody.

  22. 22 Richard Sharpe Oct 28th, 2006 at 5:32 am

    Mike Butler said:


    I don’t think that parsing and “lexical chucks” are equivalent. They do overlap to some extent but one (parsing) talks about dividing lanuage into (universal?) grammatical categories while the other (lexical chuncks) talks about how to combine multi-word units to form a single meaning (that sometimes is not explicable by looking at the separate words). One (parsing) tends to talk about dividing while the other (lexical chunck) tends to talk about combining.

    Or, to put it in more explicit computer terms, parsing is the process of building up the parse tree from the input, while lexical chunks seems to be the process of generating a well-formed tree from prestored chunks for output.

    I have found that after much listening to a language like Cantonese in a setting where you are called upon to use it from time to time (eg, having to speak with your 岳母) it gets to the point where you don’t have to think about it and the expressions come to mind.

    One of the other things I have noticed is the pronunciation differences between older and younger Cantonese speakers. Eg, ngoh3 vs oh3 for 我 and ngap6 vs ap6 for 鴨 and so on. The same applies to English. Having grown up in an Australian-English community but now living in California, I am very aware of the differences in pronunciation. For example, words that contain or at the beginning are pronounced differently in most of the US to how they would be in Australia or the UK. It seems to be that in the US, orange is pronounced more like or-ange while it is pronounced more like O-range in Australia and the UK (perhaps only in BBC English though).

    At some point I would like to have that level of appreciation of Mandarin — but there is a long way to go.

  23. 23 Richard Sharpe Oct 28th, 2006 at 5:45 am

    Lantian says on his blog (which requires signon before you can comment):


    When we are given grammar and translations of words, we lose ‘the lexis’, we lose the exposure to what should come after or before. Words in any language come with a whole lot of expectations about the words that should come before or after it, how they should be used, and the underlying grammar. This is OFTEN different or completely disassociated from the same linkages in the word of the other language. Giving me the meaning of a Chinese word in English completely disassociates me from the lexis and appendages of the other Chinese words that inform that term.

    I actually disagree somewhat here.

    When we learn our native language we learn the meaning of most of the words in our vocabularies through context. Either because of their use in the context of other words or the context in which the utterance is made (your mother is telling you not the touch the vase or she will kill you and you have heard her going on about how much she loves that vase).

    As we go along we (or some of us at least) build up a richly interconnected vocabulary where we have meanings and shades of meanings and appropriate usages all stored in there and interrelated.

    When learning a new language it is important to hear many examples of complete sentences that demonstrate the usage of words, but I do not think you can get away from having to deal with individual words as a way to bootstrap the process of building a large vocab in that new language. As you go along you also begin appreciate the flexibility of some words and phrases and can start to use them more appropriately as well.

  24. 24 Lantian Oct 29th, 2006 at 2:07 pm

    PLAYING D - Hi Michael Butler, you asked,

    “Finally, I’d like to clear something up. Is your “defense” of parsing a defense of the idea that we can learn a language by taking it apart and understanding it as bits and pieces strung together?”

    Yes, and no. There are many ways to convey a meaning, for example these two sentences (Pinker, Talking Heads chapter p 200).

    Version 1: He sent the poisoned candy that had received in the mail from one of his business rivals connected with the Mafia to the police.

    Version 2: He sent to the police the poisoned candy that he had received in the mail from one of his business rivals connected with the Mafia.

    Both versions are ‘grammatically’ fine and both express the same thing. Version 2 however is easier to ‘parse’.

    Why is it easier? In the parsing-speak, the first sentence is ‘top-heavy’ and you have to hold some of the bits of the sentence in memory until it’s connected to something else that’s said.

    In the second setence you can let go of each bit as you go along, making things less taxing on the memory.(Note to Chinese language students, Chinese LOVE ’simple’ sentences)

    I bring this up because I think of ‘parsing’ more as a memory device rather than a ‘decoder’.

    To truely understand something said requires that I understand some grammar, the meaning of the word, the lexical associations, the culture meanings of the word, the speaker’s intent, etc., so I don’t think parsing is this.

    To rephrase: parsing is passing the ball, it’s not the game.

  25. 25 Lantian Oct 29th, 2006 at 2:49 pm

    BIG VOCAB - How to get there?

    Hi Richard Sharpe,

    “but I do not think you can get away from having to deal with individual words as a way to bootstrap the process of building a large vocab in that new language.”

    I think our goals are the same, how to build up a large vocab in the new language. But to me the facts seem to show that typical vocab exercises like flashcards, definitions, being given a definition, etc., are pretty slow…and ineffective.

    Vocab is great for testing vocab.

    I think back to my learning of English, at most I’d say I learned 1,000-2,000 words via spelling test vocab I had in high-school. The rest, and I don’t think I’m boasting, of the 10,000 plus words came via some other route.

    I figure this route is available to me in second language learning as well. There’s no argument from me that at the beginning a little rote memory pays huge dividends in allowing one to at least speak and understand a little in the target language. One-to-one definitions and explanations work. “Wo” — that means “I”. “Chi” — that means “eat.”

    But after that…

    Here’s a word I looked up recently, “质 zhi4″. It means:

    1. nature, character
    2. quality, as in high-quality good
    3. matter; substance
    4. question

    I think I’m just going to for now, remember it as “xing zhi” and that it meant that my friend thought ‘his xing zhi’ was crap. It’s kinda a fun association.

    There are two ‘types’ of words.

    (1) Those that can be built out of parts by the rules of the language (Pinker, 141) and this is where I’m emphasizing there needs to be a different kind of exposure.

    (2) The second is a ‘listeme’, a word that refers to an arbitrary association of meaning. It’s like learning a name in English, I’m sure there’s some deep history behind the word “John”, but I really only know it as j-o-h-n and the sound jh-on, and that it means the big guy at Cpod.

    I think in second language teaching, teachers tend to treat more words than not as the second case, when they really should be handled as the first case. On the one hand, it’s amazing that I actually do now probably thousands of English names, but on the other hand, I know a lot more other words that aren’t ‘names’.

    A last comment because it seems somewhat related and I’ve been meaning to say it for a long time. I’ve noticed that John often says that words mean the same thing…Jenny too. They’re WRONG.

    If two words meant the “exact” same thing, there wouldn’t be two words. And certainly those two words don’t do a backwards match to English and the same English word. If they said, the words are “close”…we’ll talk about it later, I can agree with that.

    “Most common words have many meanings, few meanings have more than one word” (Pinker, 151)

    Speaking only words is tiring, frustrating and un-human. There is beauty, depth and magic to meaning; and it is why we crave to speak to another human in a shared language of meaning. (Lantian, pp 1,233)

    - - -
    (Regarding sign-on on my blog. I started to get hit with a lot of spam. Unfortunately my staff of ‘one’ couldn’t handle it. Why not sign-on and comment though! It’s just an email and me hitting the ‘ok’ button)

  26. 26 Richard Sharpe Oct 29th, 2006 at 3:56 pm

    Lantian says:


    I think our goals are the same, how to build up a large vocab in the new language. But to me the facts seem to show that typical vocab exercises like flashcards, definitions, being given a definition, etc., are pretty slow…and ineffective.

    Ahhh, then perhaps I misunderstood. I don’t use flashcards, although I do look at the definitions of words in the NPCR textbooks I have.

    What I try to do is to read the material, and when there is a word I can’t remember, I figure it out from context. I also have copies of a children’s magazine called 國語 which has Bopomofo markings against the characters so I can figure out pronunciation of those I can’t recognize.

    However, since languages are primarily spoken things, I try to get as much listening in as possible which is where ChinesePOD and local TV channels are useful (as are the subtitles on the Cantonese programs on the local channel that has Cantonese, Mandarin, Japanese and Korean).

Leave a Reply




Learn More

Ken Carroll discusses issues concerning learning generally, and learning Mandarin in particular. With technology as the driver, he believes the most effective learning combines elements of collaboration with self-direction. If that seems like a contradiction, then you need to read the blog.