Has anyone finished Heisig's Remembering the Hanzi and been disappointed?

[steveboy]

I've been studying Chinese on and off for two years and have hit a wall. My speaking improves, but I forget characters constantly.

 

I have read  about Heisig's RTH, but it always turned me off.

 

I think mnemonics are great and I don't mind that they are fictional, but I don't like that Heisig invents meaning for components. Please correct me if I am wrong about this?

 

Opinions on RTH seem divided between those who finished and say it was worth using, and those who have not read it but think it's flawed.

 

Is there anyone who read the whole book (not just part) and came away with a bad impression?

 

If nobody like that exists, then I think I will finally breakdown and submit to the gospel of RTH. 

 

I sort of hope this doesn't happen...

[li3wei1]

Whether you like it or not, and whether it's effective or not, can be two different questions. You may find it easier to remember certain characters or components because of the strength of your disregard for Heisig's invented meaning. If it's so bad it makes you vomit, by all means give it up, but if not, test yourself on the characters you've covered so far. Is it helping? There's no magic ending to the book, finishing the whole thing is no different from reading half of it, it's just more characters.

 

What else have you tried? And are you working on recognition or production?

[wibr]

Well if you want to use mnemonics in the way Heisig proposes you will end up with some components which just don't have a real meaning on their own. The logical thing Heisig did is to invent a keyword so it can be used in the mnemonics, I don't really see an alternative to that. Usually it's obvious that this is not a real meaning of the character.

You should also keep in mind that even if the keyword refers to a real definition, the character might have many other meanings depending on the context in which it appears. I see the Heisig mnemonics as a useful tool to remember the shape of the character, either for writing it or for recognition. How to use these characters is a different story...

[sparrow]

Haven't used it, but here's my two cents:

 

Just pick ten or twenty characters per day. Ten characters takes about two minutes to write if they're brand new and traditional, so do that three to five times per day. For a total of six to ten minutes, you should remember characters fairly easily, especially if you review the next day once or twice.

 

If you're learning your first 150 words, perhaps write them five to ten times each to get used to things.

 

Say the words loudly and with the correct tone each time you write it.

 

Why do I recommend brute-forcing it? Because if you're sitting around thinking of your story for 聊, for example, I feel like something is wrong. When you hear liao2 and the character should come immediately.

 

As for learning what characters mean individually, I think it's a bad idea. Learning characters doesn't translate to words known. Learning words translates to words known.

 

If you want a leg up on characters, learn extra vocabulary and the characters you need will come by themselves.

 

Use a vocab list organized by frequency. Wikipedia has them for free. Here's the one for Mandarin, up to the most frequent 10,000 words. The most frequent 5000 word get you approximately 95% general reading comprehension in English, so that's probably a good measure for any language.

 

 

Edit: I accidentally a character.

Edited December 15, 2013 at 06:14 AM by sparrow

[JenniferW]

I've been studying longer than you, but am now only a part-time student of Chinese.  I remember a stage where I felt remembering characters was impossible.  I use ANKI - for testing my writing as well as for learning to recognise characters - and for me, that was the breakthrough.  It's ruthless - ruthlessly honest!  I use etymology to help me remember characters, and in effect that way am making up my own mnemonics, and there's a strength in that.  It's slower, because you spend mental time working out your own mnemonic, but you pretty much then never forget how to talk yourself through the problem character(s).  Long-term, it also builds up into a sound background knowledge about the history of the characters which you then pull on for learning new ones with the same components in them.  As to which characters I learn, it's just the list from the current lesson, plus others I've come across and need to learn.  Nevertheless, I do still have some characters I find pretty near impossible to remember, but that happens less and less frequently the more I know.

[sparrow]

Note: I'm about to write a post containing a spreadsheet of the most common 10,000 Chinese words (that's words, not characters).

 

It took me a few minutes to rip the information from that Wikipedia page into different columns of a spreadsheet.

[Ruben von Zwack]

But no-one has answered steveboy's question yet.

I am not planning to use Heisig, as for now, I am doing just fine with the same approach like Jennifer W. But still, I'd be genuinely curious! This whole Heisig-or-not issue holds some macabre fascination for me.

 

 

I sort of hope this doesn't happen...

 

 

PS - I bought the book many years ago for Japanese, accidentally, and I was not amused when I found out it wasn't about etymology, read a few pages and sold it on ebay. So I am not qualified to comment either!

[sparrow]

From this thread I just wrote, the word frequency to character distribution in Chinese newspapers as of 1993 looks like this:

Word Set	Characters in Set
0001–2500	1119
0001–5000	1658
0001–7500	2048
0001–10,000	2397

So, if you do decide to learn characters by themselves, know that 3500 might be over-doing it.

 

The Routledge frequency dictionary, which goes up to 5000 words includes just over 2100 characters, and it's corpus draws from a wide range of modern materials, not just (somewhat outdated) newspapers.

[querido]

I also thought like JenniferW, taking the etymology from Wenlin as I went.

 

I didn't realize that number was so small sparrow. Thank you.

 

Edit: According to the lists at hskhsk.com, it looks like those ~5000 words include about 2668 characters.

[HerrPetersen]

I finished the Japanese version and then adapted the method before RSH came out. After a 2 year or so break of studying Chinese I restarted by just going through the Chinese version (RSH). At no point did I regret doing it.

Even though the biggest advantage seemed occur as an absolut beginner.

There are some prominent examples of people who used Heisig for Chinese. John Pasden from Chinese Pod used it up to a certain stage. He felt it was no longer necessary after a few thousand or so.

I personally do not follow it as strictly as before.

As for people who are disappointed after finishing it - there were some who quit dissappointed at the 700 or so kanji mark. There were also some who finished and did not know what to do afterwards. I remember those examples from the kanji.koohii.com page which is dedicated to the kanji version of the Heisig books.

[tysond]

Is there anyone who read the whole book (not just part) and came away with a bad impression?

 

It's a negative question but I'll just answer with my experience.

 

I did both RTH books (Simplified), because of exactly the problem you outlined - couldn't remember the characters.

I found it helpful to have a book where they had 1. already picked the most frequent characters, 2. already decomposed them all into pieces, 3. ordered and grouped them to introduce pieces one at a time, 4. suggested mnemonics and keywords for me.  

I finished in July and now I am reading comics and enjoying it.

 

If I learn Japanese it will be the first book I purchase.  If I did Chinese again here's what I'd change:

- Just do the first book (1500 characters including the 1000 most common), using Skritter.  Do it as fast as possible.

- Then I would learn the pronunciation by learning sentences using this 1500.

- Then I would learn the next 1500 (book 2) slowly, learning a sentence for each one.

 

For me the arguments about "purity" are meaningless.  I would like to read about etymology, and character development and history... in Chinese.  Authenticity of the method by which I get myself to that point is not important to me.  

[gato]

The key to not forgetting is continuous practice. Native Chinese don't use mnemonics when learning characters, but they do practice repeatedly and everyday. Yet, native Chinese start forget how to write characters, once they stop writing by hand on a regular basis.

One thing that is often not mentioned by those who praise the Heisig method is that even with the Heisig mnemonics method, you have to continue to practice after you have learned the characters. Once you stop practicing, you will start forgetting. Heisig isn't the magic formula to "study once and remember forever".

Thus, continuous practice, whether through flashcards on your smartphone or on paper, is key. The method is secondary, though some methods may be more efficient than others.

Are you practicing everyday (writing or reading or both)? If not, then that's the problem. If you are practicing everyday and finding your current method inadequate, then that's the time to seek out other techniques.

[Demonic_Duck]

Why do I recommend brute-forcing it? Because if you're sitting around thinking of your story for 卿, for example, I feel like something is wrong. When you hear liao2 and the character should come immediately.

 

That's 卿卿我我的卿 not 聊天的聊 (for what it's worth, I wouldn't have noticed if I hadn't used perapera popup dictionary).

 

Maybe this is one character where a mnemonic really would help?

[偉爾豆仁]

I'm speaking as someone who finished both book one and two of Remembering the Traditional Hanzi. 

 

I more or less went along the path of AJATT, so I was doing a lot of immersion while going through these books, which I think is the best way to use the books. As you learn more and more characters, you start to recognize the ones you've learned through Heisig, providing reviews all day long.

 

First off, I think it's worth saying that the true strength of the method is really in the order of the characters. When you memorize ten or so characters that all have 青 on the side (請 清 情) in short succession, you are 1. Practicing writing that component and 2. Learning to see these characters as simply 言＋青. More than anything I think the strength of the book is in teaching people to see characters like the chinese do.

 

When I showed my father-in-law (native Taiwanese) the character 療, as in 治療, as part of my reviews, he told that most Taiwanese wouldn't learn that character until high school. However, it's easy for me to see it as "sickness + pup tent." Of course he doesn't think of it as "pup tent," but he still sees it as "sickness" + the component from 療，僚，瞭, etc.

 

The mnemonics are helpful, but they aren't really the point. Yes, they increase short-term memory recall, but the real practice is, as Gato said, seeing the characters used in books and subtitles and writing notes, etc. I still do my reviews in the "Repititions" app (an SRS like Anki) to this day. 

 

Yes, Heisig makes up component meanings that may confuse native speakers at times, but who cares? Considering I finished the books almost a year ago, the keywords and english names for components are fading, being replaced by a general sense of the character as a whole (which is why I forget them sometimes, like the Chinese). So even if you have to use silly means to learn the characters, it's only temporary if that bothers you.

 

Finally, don't use the books if you don't want to. Nothing holds back learning like anger, shame, embarrassment, or whatever. If you want, just use the order of the characters and brute force your way through. I still think the order of the characters is good information and the most efficient way to demystify the writing system. I went from 0 to recognizing around 3035 characters in around ten months, learning words and sounds through my immersion environment. I still don't know the pronunciation of all of them, but I learn that as I encounter the characters in real life. 

[mtpastille]

I finished the first book in twelve weeks I think, starting off with just a few characters a day and then slowly ramping up to about 50/day, with daily flashcarding. In hindsight, I don't think it was good use of my time. It felt rewarding to "learn" all these characters in such a short period of time, but I didn't learn pronunciations and I didn't get a good understanding of what the characters meant either.

 

Later on, I started studying the HSK1-4 word lists instead which I found much more useful. They gave me the foundations of being able to read, which is ultimately what has helped my knowledge of characters more than anything else.

[alanmd]

I gave up with Heisig quite early on, I felt that I wanted to learn the pronunciation and usage of the characters, and Heising didn't help me with that. Some of the early characters were very oscure and sometimes were unusable in isolation, or had multiple meanings, so the single meaning assigned by the Heisig system wasn't useful in the long run. I can see how it could be useful for learning to recognise and reproduce a whole load of hanzi that you don't really 'understand', but it wasn't very useful for me.

[sparrow]

@Demonic_Duck #13:

Thanks for catching the mistake! I'm not sure how I came up with that character as I was trying to write 聊！If i was writing instead of typing, that wouldn't have happened.

 

@querido #9:

According to the lists at hskhsk.com, it looks like those ~5000 words include about 2668 characters.

 

It all depends on how that list of 5000 words is generated. The list I provided used newspapers from before 1993 or before. Routledge created their own corpus from a variety of resources. From the Routledge dictionary:

Register	Word Tokens (Millions)
Spoken		3.4
News		16.1
Fiction		15.2
Non-Fiction	15.1
Total		49.7

HSK's corpus might be composed completely differently. I've also heard that HSK tends to include more obscure words, so their corpus might not be strictly based on Frequency. Perhaps, they also use a lot more 成语, which might change things.

[alanmd]

There's no concept of an "HSK corpus", the HSK syllabus includes whatever words Hanban decides to put in it, and they don't promise that they will be the 5,000 highest frequency words.

 

By my calculations there are 5,001 HSK words in levels 1-6, which are made up of 2610 characters - some people quote slightly different numbers as there are a few small but boring ambiguities about how to count words at each level.

 

I did a frequency analysis of the words at each HSK Level: http://www.hskhsk.com/2/post/2013/04/do-hsk-levels-teach-the-most-frequent-words-first.html - there are some more unusual words after about HSK3 but there really aren't that many truly obscure words in the HSK. By HSK 6 you will understand around 85-95% of speech according to SUBTLEX-CH (would require quite a bit of work to get a more accurate figure due to the problems with all frequency lists and how they classify words): http://www.hskhsk.com/2/post/2013/04/how-much-will-i-understand-at-each-hsk-level.html

[sparrow]

@alanmd #18:

Thanks a lot for the SUBTLEX-CH by the way. A speech-based corpus for frequency is freaking awesome.

 

I'm trying to figure out how to find the discrepancies between the Nciku and SUBTLEX-CH list for 5000 words, but I'm unsure how to do this with a spreadsheet. I think I may know how to do it with Google Spreadsheets FILTER() function though, so I'll get back to you on it, as we both seem to be interested in these kinds of analyses. ;-)

 

 

Edit: I was able to do it. I'll make a post about it later.

[roddy]

Could we try and stay halfway on topic folks, thanks...