Phonetic Mnemonics vs Heisig Style Stories?

[simc]

I'm an upper intermediate learner who is thinking of getting back into learning how to write characters by hand after years of mostly reading and writing them using Pinyin input on a computer or phone. I had previously had a interest in Heisig style mnemonics, and I think that identifying and learning the components is clearly a very good thing. But I am beginning to have doubts about assigning English soundwords and making up stories.

 

I've been thinking, why not just use the pronunciation the components already have and make up a nonsense word for each character, for example:

痛 = 病甬 = bìng yǒng

究 = 穴九 = xué jiǔ

营 = 草冖吕 = Cǎo mì lǚ

It seems like a viable approach. Has anyone tried to learn characters like this?

[stapler]

When I first started learning I'd use all sorts of elaborate mnemonics. However, after getting to know 1000's of characters I mainly do what you do and think of them in terms of their components (though not as drastic as remembering things like the "mi4" in your "cao mi lv". 

 

For obvious semantic/phonetic characters like 痛 I just remember the "sickness" semantic component and the phonetic component "yong3" used in the others 通 誦 etc.

For less helpful characters like 營 I would just remember is a bit like 榮 or 勞 and note which bit which is different (in this case 呂).

 

It's all very minimal cognitive effort. I can't really describe "the method" any better. But it just seems that once you're very familiar with 1000's of character it gets easier and easier to remember new ones as you can piece them into your existing "neural network" by how they relate to different characters you already know rather than by constructing detailed breakdowns of each character from scratch.

 

The only time I make mnemonics now are for characters with lots of rare components. EG 竊 or ones with lots of components like 贛 (which I remember by noting that the left side is 章 (not 立 and 早) then a mnemonic about workers getting up to go make money to remember the right)

[Shelley]

I have always been doubtful about these sorts of methods because it seems like learning a whole new layer of stuff. I agree with learning the component parts for writing characters correctly, but for remembering characters I try to just remember it as a whole.

 

Its a bit like in English I know the word memorize consists of separate letters but I see it as a whole word and the meaning is embodied in that set of letters.

 

They did some tests recently that showed that as long as the first and last letters were in the correct place it didn't matter what order the rest are in ie:mrzeiome.

it should still be understandable.

 

If I am going to use any kind of aid de memoir I like to try and use the etymology. of the character.

 

In the end though it really comes down to what works for you.

 

 

 

[Guest realmayo]

simc, it doesn't sound like a bad approach at all. You'll be learning pronunciation for some components that don't really exist now as standalone characters but that's not a massive amount of redundant work. Also I'm not sure why you'd need to think up a nonsense word for all these components since have some meaning more obviously associated with them already.

I would add that, in my experience and in that of at least some others, silly stories were not only extremely helpful to learn a load of new characters, but also that after a while (maybe a year?) those silly stories just kind of vanish from your head while the memory of how to put the characters together remains.

[lechuan]

I was/am a big fan of the Heisig approach, but I think a major drawback is his total glossing over of character component sounds. With the exception of Heisig's "secondary" keywords, most of his keywords are a correct meaning for the character (though just one of many meanings in some cases). I went through 1500 simplified characters purely using the meaning mnemonics, but I wish I had paid more attention to sound components from the beginning.

 

These days I tend to stick with whatever I happen to remember for that character. Usually when I learn a new character I do the following:

 

1) Say the meanings (keywords) of the character components

2) Say the pronunciations of the components (which I also use to describe characters to native speakers)

3) Use the pronunciation components if it sounds like the Mandarin pronunciation

4) Make up a Matthew's style sound mnemonic (which I'll drop once I know the character better) if #3 isn't applicable.

[tooironic]

@simc I think you're on the right track. I myself never used mnemonics or created stories to memorise characters, beyond the very basic ideograms like 日, 月, etc. I don't think Chinese children do either when they are faced with the task. To me it always seemed like more effort than simply memorising the characters by breaking them down into smaller, more memorable components. This, of course, can be challenging, especially if you're learning traditional characters. The key, in my opinion, is to ask what native speakers actually call the components and remember them that way - don't just learn what the pinyin readings are, because in many causes no one actually knows what they are. For example, you mentioned 營 being composed of "cǎo mì lǚ", but we actually refer to these by their nicknames - 草字頭 and 禿寶蓋 (or 平寶蓋), and 呂 we would clarify as 呂姓的呂 to distinguish from the other Lǚ's (旅, 屢, 履, etc.). Plus, it's always good to learn about the various 筆畫 - have a look on Baidu and you can find the relevant charts. I found all this foundational knowledge extremely helpful when I was starting out learning how to write hanzi. I'm going to compile all my notes and write a blog entry on it, because I think it might help other learners too, and I've never seen it covered in any of the various textbooks written for foreign learners of Chinese.

[simc]

Thanks for the replies.

 

It seems to me the most help thing about a Heisig-style scheme is that you can split the character into pieces and give each piece a name. Then you can say this character is A then B then C, where A, B and C are names of specific bits.

 

The problem with Heisig names is the names are all in English. Obviously, this is another layer of things to learn. But if you already have been studying Chinese for a while the "meanings" he has assigned might be different from what you feel they ought to be. If you use its Chinese pronunciation as the name instead avoids these problems.

 

I've found that the component information is available in Pleco under the Chars tab which was quite useful - before I found that I was thinking of typing them all in and creating a custom dictionary!

 

I'll continue my experiment - but I think trying to remember words like "cǎomìlǚ may not be any easier than remembering cǎo then mì then lǚ.

[tysond]

I went through 3000 characters with Heisig 2 years ago.

 

Before I did Heisig I was struggling to get past 100 characters through brute force memorization.  It was very frustrating and draining.  

 

However, now I can read about 2000 characters pretty well (and write most of them).  But not 3000... although I did study both books.  

 

I think learning component based decomposition is essential.  Radicals are not enough... and phonetics are only part of the solution.  Heisig gives a nice structure for total decomposition and remembering all the components.  Heisig also teaches a nice mnemonic technique for memorizing the parts and how they make up the whole.  As you get better with this technique it becomes less and less cumbersome.  Frankly it made me realize in all my years of schooling nobody every taught me how to really memorize things.

 

The English keywords are a touchy point.  The idea is that they melt away and the more natural meaning of the character (in context of its many words and uses and readings) takes their place.  But they are an artificial construct, and a crutch.  Even with the best memorization technique, there's a forgetting curve over time.  So you are learning all these keywords in English, and spending time to memorize them, but you could be spending the same time learning the Chinese readings of the characters.  [ There are other more arguments about the keywords not being historically accurate, ordering of learning etc - but I think those tradeoffs are more acceptable, at least to me ].  Readings and phonetic information are completely ignored in the books (although in this day and age you can of course look it up easily).  

 

I am working full time, and learning part time, so the forgetting curve is brutal (intensive study is much better for this, and the tradeoffs are different).  So what I found is that memorizing Heisig english keywords robbed time from other learning.  

 

Therefore, if I had my time again, here's what I'd do to get the most advantage - at intermediate level, study the first book (1500 characters, 1000 are most frequent) in parallel with other studies (Skritter is a nice app for this, but Anki and paper is also OK).  Then go through it again and make sure you learn readings for each character, most common words, and sample sentences etc, add to your study routine.

 

Buy the second book but only use it for lookup at first (lookup new characters you encounter and need to learn, go backwards through the roots).  When finished the first book, slowly go through the second book in parallel with other studies, learning each character, reading, words, sentences all at once.  I think this minimizes the time you spend memorizing English keywords, while getting the benefits of the decomposition, mnemonics, etc.  

[pross]

I am still on Book 1. Around character #500, I began adding pinyin syllables to the front side of my flash cards in addition to the English keyword, and added example words to the back side. This works for me.

[Yang Chuanzhang]

after a while (maybe a year?) those silly stories just kind of vanish from your head while the memory of how to put the characters together remains.

 

This was also my experience. The stories and the sometimes slightly weird keywords are really only scaffolding that exist to make the character stick in your head. I don't remember a single story for the 1500 characters that I learnt with his method now, a couple of years after I finished the first book.