systematic isolated character study

[leosmith]

Do you think systematic isolated character study will make one literate in Mandarin faster that without? This is a frequent debate on this forum, so I thought it would be nice to take a poll on it.

[leosmith]

Here is my quote from another thread:

People who learn isolated characters in a systematic way need to continue their studies after they have finished this isolated study. Nobody is under the impression that they will be able to read fluently after the isolated study. Here is a summary of the "normal" way vs the "isolated character" way.

Normal road to literacy:

1)learn words & their characters

2)read simple literature

3)repeat steps 1 & 2 many times

4)read normal literature

Isolated Character road to literacy:

1)learn characters

2)learn words

3)read simple literature

4)repeat steps 2 & 3 many times

5)read normal literature

So basically, we do everything that normal learners do. But in addition, we learn isolated characters in a very quick, organized way beforehand. We believe that taking the extra step will save us a lot of time in the other steps, and make us literate faster than normal learners.

[renzhe]

I voted yes, but it will obviously depend on the person.

The bottom line is, you have 2-3 thousand characters you'll have to know, so you might as well get started. I've met too many people who spend years doing things "the proper way",and then, years later, know all the nuances of 你好, and that's about it.

It's just a bootstrap, nothing more. You will re-learn and rediscover all these characters many times over the years, learn nuances and refine them. It also has to be coupled with vocabulary study and lots of reading.

But what learning a lot of characters and words in the beginning does, is open up lots of new reading materials for you. There are only so many texts a person who knows 200 characters can read, and only so many words. Putting in some memorisation effort will open up lots of texts to you, which become accessible and allow you to put in much more reading effort much sooner. Reading more (real-world) material will give you plenty more examples of using the words and characters you've learned, far more than a textbook.

I've found that learning characters slowly, as they come, didn't work for me, for many years. There were no texts for me to read, even the simplest ones required too much dictionary-fu. I've started rote memorisation daily (coupled with reading, listening comprehension, textbook materials and all the usual stuff), and I've progressed far faster than ever before.

[imron]

I voted no, but as renzhe mentioned, everyone has their own way of studying. Personally I find isolated bouts of systematic character study to be useful sometimes, but generally find that I learn things much faster when I'm learning them from context. Having the context helps me remember the meaning, as well as how to use the word.

[imcgraw]

I voted no.

I wouldn't discourage it. And I agree that a certain amount is necessary to find decent reading material in Chinese in particular. Handwriting recognition solves the dictionary look-up problem though. You can read authentic material after about a year of character study at the university level. (Check out Doraemon... it's fun to read and has pictures!)

In general though, the notion that one needs to learn vocabulary (let alone an individual character) outside of some meaningful context is, in my opinion, a broken view of language learning. It's the "delayed gratification" approach to studying a foreign language, and it's precisely why of every six college students taking a foreign language course in the United States only one of them will go beyond the 2nd year.

If you can stomach wrote memorization in large enough quantities, I admit, it may be faster. But most people derive more pleasure from using a foreign language than knowing it, and most learners will study in the way that gives them the most pleasure (unless otherwise forced by a curriculum). Fortunately most of the second language acquisition theory suggests that using a foreign language (reading, speaking, etc) is the best way to learn it.

[imron]

It's the "delayed gratification" approach to studying a foreign language, and it's precisely why of every six college students taking a foreign language course in the United States only one of them will go beyond the 2nd year.

It's also the reason that many Chinese learners of English suffer from 哑巴英语, and can score really well on tests, but are still be unable to hold a normal conversation. They have all the vocab, they just can't use it.

[renzhe]

I realise that this is very individual. I'm good at memorising things and I like to work with organised things where I can gauge my progress (engineering background, as I said). I like building the framework first, then filling it.

But this is the 8th or so language that I'm learning, and the 5th that I can actually currently use at a moderately useful level, so my current approach is based on previous experience as well as 6 years of experience trying to learn Chinese without character study. After 6 years, which includes numerous courses and different textbooks, as well as having a Chinese girlfriend, I could read about 200 characters, and all the others would get quickly forgotten because I simply didn't have enough material to read, and the material I had required 95% dictionary-fu to make any sense of it.

For me, the character study helped a lot with vocabulary and progressing through textbooks. In a bit over a year of doing it, I can recognise over 2300 characters, some of them better, some of them worse, and what's more important, I can learn new words far more easily because I don't need to learn a new character + a new pronunciation + the meaning of the character + the combination of the two characters + a meaning of the combination every time there is a new word. The character study made learning new words work just like it does in any language with a phonetic alphabet. So now I can watch modern TV series with subtitles and understand most of what goes on, and looking up new words is extremely easy, because I can recognise the characters used.

It may not be for everyone, but it has given me good results, and I intend to continue. I see it as exactly the opposite of "delayed gratification". I want to understand books, comics and TV shows now, not in 7 years.

[renzhe]

It's also the reason that many Chinese learners of English suffer from 哑巴英语, and can score really well on tests, but are still be unable to hold a normal conversation. They have all the vocab, they just can't use it.

I think the reason for many Chinese learners of English not being able to hold a normal conversation is that they haven't practiced conversational skills and did vocabulary study exclusively.

Nobody is advocating this. I spend at least four times more time on listening comprehension and reading than I do on vocabulary study.

[character]

[...] so my current approach is based on previous experience as well as 6 years of experience trying to learn Chinese without character study. After 6 years, which includes numerous courses and different textbooks, as well as having a Chinese girlfriend, I could read about 200 characters, and all the others would get quickly forgotten because I simply didn't have enough material to read, and the material I had required 95% dictionary-fu to make any sense of it.

Just my opinion:

I'm inclined to think your situation is rather rare, or perhaps you remembered the hard part of characters, which is not how they look, but their meaning when used in different kinds of sentences, and/or as part of different words. Knowing the correct meaning of characters such as 就, 一, 做, etc. is very context dependent.

I tried isolated character study and don't think it's a good use of time for most learners. Probably a mixed approach is best for most, including recognizing the character, knowing what it means in sentences and in different words, writing the character, etc.

[renzhe]

I think it also depends on your learning environment.

For someone like me, who is extremely busy, and squeezing Chinese study on top of everything else in a non-Chinese environment, it will be different than for a person attending a Chinese university in China. These people have such ridiculous amount of exposure to Chinese language, written and spoken, that they basically don't have to worry about memorising anything.

For people who can do at most 1 hour a day of individual study in Europe, and this often at the end of a hard day, there is far less exposure.

I'll just re-iterate that things like rote memorisation are nothing but tools which can help the REAL sources of learning to be more useful more quickly, not an end unto itself. The real learning comes from conversation, listening, reading, etc. As long as you understand this, and can stomach the effort, it can help you get more productive more quickly, IMHO.

[anonymoose]

This is an interesting issue, and I think I'm pretty much on the fence in the middle here. I rarely ever learn isolated characters, but often I'll learn a two (or more) character word just for the sake of learning one of the characters in it. I find I am able to retain memory of the character for longer if I can attach a more precise meaning to it (which is often easier to do when it is a component in a word rather than in isolation). Having said that, I wouldn't class my study as systematic either way, so...

[imron]

The real learning comes from conversation, listening, reading, etc

I agree with what you are saying, but it seems to me that this doesn't seem to match with what leosmith is saying. Going by the steps he listed for isolated study, it would seem that the idea is first to learn a large chunk of characters, and then once you have that to begin applying that to learning words/reading literature.

[renzhe]

It could be that leosmith and I are talking about slightly different things.

I am talking about learning characters and even vocabulary ahead, but always reinforcing that with other learning, through reading progressively difficult literature (well, comic books right now ), watching progressively more difficult TV shows, etc. You learn a character/word expecting to see/hear it soon in context from a different source.

For example, when starting a new volume of New Practical Chinese Reader, I will first go through all the vocabulary and take out all the characters I don't know, put them in a flashcard program, and start memorising them. So I will learn the characters ahead of time, but I will follow that up a month or two later by seeing them in context, in the words and in the sentences. This makes the actual learning and reading process far more enjoyable for me because I don't have to stop each lesson countless times to look up the new characters. Similar thing with TV-shows, I've memorised all characters and words from the lower HSK levels, and when watching TV-shows, I will remember seeing them before and their rough meaning. I can concentrate on the pronunciation and context, without having to bother with dictionary lookup at that very moment.

There are people who advocate more of a Heisig-style approach: learn characters ONLY, then do everything else later, and I don't know about that. I've never tried it. I do it all in parallel, only I memorise characters and words ahead to make these other tasks easier.

When I hit 3000 characters, I'll stop learning them individually and concentrate only on reading/listening/conversation, like everyone else seems to be doing.

[Lu]

I learned a lot of vocab (pure words) in my second year, and while it must have helped, I still often enough come across words that I know I learned during that time but just forgot afterwards for lack of use. (Like the word for arrest, the first time I saw it again I knew I had learned it, but had no idea what it was; I can't remember how to write it but I can read it now, after coming across it a couple more times.)

I guess it depends on the person, but for me it doesn't really work. What does work: see a word, make an effort to learn it (look it up, write it down) and then use it once or twice. I think I still remember (at least read) most of the more useful words I posted in the 'random word of the day' thread, simply because I stopped and thought about them.

[leosmith]

The real learning comes from conversation, listening, reading, etc

I agree with what you are saying, but it seems to me that this doesn't seem to match with what leosmith is saying. Going by the steps he listed for isolated study, it would seem that the idea is first to learn a large chunk of characters, and then once you have that to begin applying that to learning words/reading literature.

I'm just talking about reading and writing here. I'm making no attempt to show the "big picture", which should definitely include listening and conversation.

[imcgraw]

In a bit over a year of doing it, I can recognise over 2300 characters

Wow, that's crazy impressive. I have a sneaky suspicion your ability to retain what would normally be a short-term memory is well above average. I have an engineering background as well, and I too like being able to gauge my progress. I even built my own flash card web-site, as seen in my signature. The classes I take force me to do isolated vocabulary study (not individual characters) in order to perform well on a 听写 every other day in class. For me though, I forget most of the vocabulary the day after the test.

...and all the others would get quickly forgotten because I simply didn't have enough material to read, and the material I had required 95% dictionary-fu to make any sense of it

I think that the "lack of reading material" at an appropriate level doesn't apply after about one year of study, if you are willing to invest in a decent handwriting recognition dictionary. There are 45 books of Doraemon (not to mention a spin off series). With the aid of the pen-input dictionary, you can quickly get to a stage where you could understand and enjoy the stories without it (if you had to).

It may not be for everyone, but it has given me good results, and I intend to continue. I see it as exactly the opposite of "delayed gratification". I want to understand books, comics and TV shows now, not in 7 years.

I would wager that you are a special case. In fact, I bet that anyone who is particularly good at memorization would benefit from this method as you do. An extreme example are savants like

who memorized most of Icelandic in a week. And actually, it is possible that long-term studiers of Chinese are a self-selecting group of good-memorizers. Perhaps statistically the average-memorizer, uninformed about alternative methods of acquisition, might just give up at the prospect of learning so many characters.

Obviously there are many variables to consider, but for the average person, relying heavily on wrote memorization is likely going to discourage more than it helps. Some amount of memorization is unavoidable, especially if you are a pure beginner or if your focus is on writing by hand. In general, however, my opinion is that language learning is too often viewed as a memorization task.

For me, the definition of "delayed gratification" is doing wrote memorization for two hours in preparation for a 听写, and then (hopefully) feeling the satisfaction of having understood the single sentence that was dictated to me. Instead of spending those two hours studying, I would much rather be reading my Chinese comic books: which is something I can enjoy now (at 2.5 years). In fact... that is often how I procrastinate my formal studies!

[renzhe]

Wow, that's crazy impressive. I have a sneaky suspicion your ability to retain what would normally be a short-term memory is well above average.

It is. But it doesn't mean that regular problems don't apply. I can only retain them and learn so many because I spend at least half an hour every single day with a spaced repetition program. I'd forget much of that if I stopped revising for a few months. I can't use many of these characters properly, or in proper context, but I can definitely use more than half. I'd wager to say that this is still far more than I would have learned otherwise, and the remaining half or so will fall into place as I read more and encounter words using them more often.

This is where the reading, talking and listening comes in -- this is, IMHO, where the real understanding and remembering comes from. All the isolated character study is not a purpose in itself, but a tool to help me get more out of my reading and listening and conversation later. Even if I have a list of words where I only vaguely know the meaning, I can try using them in conversation to say something I've never said, and I'll get corrected and told how to use it properly. THEN I'll remember it for a long time.

But maybe my memory does help. I can pull things out of my memory like from a dictionary in mid-conversation. Think of it as doing the dictionary lookup ahead of time, so I don't have to do it while talking or reading.

I think that the "lack of reading material" at an appropriate level doesn't apply after about one year of study, if you are willing to invest in a decent handwriting recognition dictionary. There are 45 books of Doraemon (not to mention a spin off series).

I find that getting interesting and suitable material is much more difficult outside of China. I got my hands on a number of texts, comics, etc. and have enjoyed reading them, but I have to order them from China basically, which means waiting for an acquaintance to go there, explaining what you want, hoping you get something that's useful, etc. Or, even worse, having my girlfriend explain an acquaintance of HERS what stuff to buy.

I can spend weeks getting my hands on a TV series only to find out that it's boring as hell or not suitable for my level.

I'm on my way to China in a week, and I'll stock up with easy literature and comics for sure!

Obviously there are many variables to consider, but for the average person, relying heavily on wrote memorization is likely going to discourage more than it helps. Some amount of memorization is unavoidable, especially if you are a pure beginner or if your focus is on writing by hand. In general, however, my opinion is that language learning is too often viewed as a memorization task.

I hate memorisation, I love understanding things. In my particular example, I was being held back for years by my refusal to do any memorisation and expecting my brilliant genius bold unstoppable mind to figure out the logic behind it through osmosis. When it didn't happen, I bit the bullet and started memorising. For a guy as lazy as I am, this wasn't easy, but it helped me a lot. A year ago, it took me about 5 hours to get through a 40-minute episode of a soap opera, because I had to stop and look up things all the time. Now, I can finish it in 40 minutes. Which means I get 6x more exposure to REAL Chinese speech, real Chinese sentence structures, etc. For me, it's a huge gain.

I agree that there is too much memorisation in language teaching. But I also think that there are too many people who claim to know the "only" proper way to learn a language, and who construct complicated system for perfect language acquisition. In my view of language learning, you have to overload your brain with information (this is what immersion basically does) and shoot for way above your level, and your brain will figure it out. Give me some basic vocabulary, and then let's USE it! Read, talk, listen, as much as possible,all the time. Not listening to carefully doctored tapes and carefully prepared texts which give it to you in the "proper" order. I still use NPCR for guidance, but I'm spending more time on watching TV shows, reading simple books and comics, and talking to my girlfriend.

BTW, I see that you're also an AI guy. My focus is vision, though, so it might explaing why I'm so set on learning written characters ;)

[imcgraw]

But maybe my memory does help. I can pull things out of my memory like from a dictionary in mid-conversation.

I'm jealous. As a side question, what spaced-repetition program do you use? Is it free?

I find that getting interesting and suitable material is much more difficult outside of China.

I'm sure this is the case. It's a bit of a shame, though, as you would think that any University with a Chinese language course would be stocked with authentic, level-appropriate material. It's simply not the case though.

But I also think that there are too many people who claim to know the "only" proper way to learn a language, and who construct complicated system for perfect language acquisition.

Agreed. I think that the single thing that is consistent across "good" language learners is that they are constantly looking for new ways of acquiring the language, and never discounting any particular method entirely or assuming they have happened upon "the only" proper way of learning a language. I think my distaste for the wrote memorization in particular comes from courses which emphasize it over everything else. I still take these courses, however, because they do provide me with some structure that I wouldn't get just doing the free reading / conversation on my own.

With regards to this poll, I'm starting to think that everyone who answered is probably right, at least with respect to what is good for their own learning strategies. If you answered "yes", chances are you are a good memorizer, and this serves you well when you are reading, listening, etc. If you answered "no", perhaps wrote memorization doesn't stick well for you. Even so, I would never abandon it completely. In fact, just from this conversation I'm tempted to give the spaced-repetition approach another serious try.

My focus is vision, though, so it might explaining why I'm so set on learning written characters ;)

I worked on a high-level object recognition project for a while, but now my focus is on speech recognition... which could explain why I don't focus on writing much :-). I do think reading is essential, though, because it gives you access to vast quantities of input that you can take in at your own pace... even if most of the good materials are half a world away.

[renzhe]

As a side question, what spaced-repetition program do you use? Is it free?

I use mnemosyne, some people prefer anki, which is a bit more polished, but they are very similar. Both are free.

I try to learn only so many new characters that the program schedules between 50 and 100 characters for review every day. So the review doesn't take that long, and is managable. I've learned very early that you can't review 300 characters every single day, or your brain will explode.

. It's a bit of a shame, though, as you would think that any University with a Chinese language course would be stocked with authentic, level-appropriate material.

Unfortunately, I'm not studying Chinese at a university, and I can't attend lectures either as I work full-time. I've done all the available Chinese courses in the city, and I'm too advanced for any of them (lots of beginners, very few people who continue). So I'm basically on my own when it comes to working my way through textbooks, finding reading materials, watching TV-shows, etc.

This forum has been a great find, I've found so many good resources here!

With regards to this poll, I'm starting to think that everyone who answered is probably right, at least with respect to what is good for their own learning strategies.

I think that the overwhelming respons has been that it depends on you and that it can work for some people and not for others.

I worked on a high-level object recognition project for a while...

Who'd have thought. I work with high-level vision daily. Very rare field nowadays.

[Altair]

I voted yes. In my view, learning characters is like learning spelling, except that it has some extreme advantages and disadvantages. The advantage is that you can learn many meanings as a bonus, and the disadvantage is that it is more complex than almost all other "spelling" systems.

I have learned a little of many languages. In each language, it has been more or less an elementary step to learn the basic aspects of the spelling system first, in order to be able to be able to read enough to be able to understand explanations. In every case, however, there are always aspects of the spelling system that are best to learn later: for instance, in Spanish, the difference between "cual" and "cuál"; in French, the difference between "ce" and "ceux"; in German, the difference between "Weg" and "weg"; and in Arabic, the difference between the various "hamzas." These are examples of things better learned after you have developed some familiarity with the languages.

In Chinese, I think it is an excellent idea to acquire some familiarity with characters and with frequently encountered radicals and components early on. Learning all or substantially all of them early on, however, is probably a mistake. What are you going to do with the three pronunciations of 龟 at the beginning of your study? I have since encountered all three in my reading and now have some context for them, but early on, the three readings would have just scared me away.

A similary point can be made about English. For example, in the U.S., kids learn various spelling "rules" that are very helpful early on, such as the various ways to spell the sound represented by "site," "cite," and "sight." Adults unconsciously draw on other "rules" to keep these spellings separate, such as "site" ~ "situation," "cite" ~ "recite" or "incite," and "sight" ~ "sigh" or "light." I use them every time I need to remember whether it is "websight," "webcite," or "website." These "rules," however, are too complicated for beginners.

Trying to learning all instances of a "rule" in a systematic way may also require too much wasted effort. "Sight," "light," and "fight" might be useful to learn together, but would you really want to add "bight" and "wight" or even "wright" near the beginning of your study of English? Many native speakers could not define these three words. Non-native speakers could learn the meanings, but would be mislead about the frequency and nature of the usage.

Whenever I have tried to learn a language with a difficult "spelling" system like Chinese (e.g., Gaelic, Middle Egyptian, and Akkadian), I have found that I learn best by learning a little "spelling" first, then engaging with the language a little. Then I learn more of the spelling rules and engage with the language some more. At some point, I reach a point where I just want to read for pleasure without any encumbrances and so try to learn all the "rules." At that point, you usually will have a good idea of what system of learning will work best for you.

That said, mastering all aspects of Chinese is so difficult that I think the overriding rule must be: "Go with the flow," at least until the flow starts to stagnate. For some people who like to memorize things or who, at least, have nothing better to do, memorizing isolated characters is probably the way to go early on. At times, the last thing I have wanted to do is to struggle through passages of Chinese and have to look up unfamiliar character after unfamiliar character. At such times, I have sometimes simply thumbed through the dictionary or looked up every character in Wenlin with the pronunciation "an." What could I possible learn that I couldn't learn better reading in context? Well, at least that phonetic complements are not reliable even for new scientific words: for instance, 氨 (ammonia) is ān, 铵 (ammonium) is ǎn, and 胺 (amine) is àn. Also, that gasses seem to use one radical; metals, another; and proteins, yet another. Not much, but better than nothing.

At the moment, I am almost completely ignoring learning how to recall how to write characters and concentrating on recognizing them and knowing the pronunciation in context. I have no Chinese reinforcement in home, work, or social environment. My method, like many others, is based on the etymology of the characters. On top of that, I use mnemonics that sometime incorporate the tone in some way. Im context, I would guess that I recognize 3-4 thousand characters. Out of context, I could probably give the correct meaning and pronunciations of only maybe 2-3 thousand. I first learned through a Japanese book of Kanji that taught me how to recognize about 300 or so within a week. That got me going with some momentum. I first worked diligently on stroke order and on writing from memory, but have gotten lazy, especially after learning that the Chinese stroke orders were often different.

The challenge I most have now is not how to learn a character, but whether to interrupt my reading and try to put forth the effort to do so. I have an interest in ancient Chinese history and so learned the character 狄 early on. I have since learned that it is also a surname. If you ran across a sentence that had 狄太太 in it, you can recognize that it is about Mrs. 狄; however, should you now pull out your dictionary and figure out that it is pronounced dí? After you do this, should you go on to make up a mnemonic with this sound in it? Or, should you wait until you have encountered this character several times before deciding that it is one worth learning and knowing? If you are in China or are Chinese, you may know of people named Dí and so have an interest in knowing how to spell the name. You may thus have a personal memory to associate this syllable with. I, however, have never heard of anyone named Dí and only know the character because of encountering the 戎狄蛮夷 in stories about early Chinese history.