Remembering the Hanzi trouble

[tysond]

If you are one of those many people for whom mnemonics do not work then the Heisig method is worse than any other method of learning characters. 

 

 

I would recommend spending enough time to become proficient at the techniques before dismissing mnemonics.  Mnemonics and SRS are about the only two technologies that we have available.  Throwing them out is very much like being in a boat and throwing out a paddle saying rowing isn't effective for me.

[hedwards]

@StudyChinese, mnemonics work for pretty much everybody, and no, it's not worse than any other method. Even the worst mnemonic system is better than rote memorization of characters by stroke.

 

Most of the people who claim that mnemonics don't work for them haven't bothered to learn a proper system. I remember the first time I was shown mnemonics it didn't work for me, but later on I've come into contact with good methods.

 

If mnemonics didn't work for everybody, we would have developed written systems much sooner and made quite sure that everybody knew how to use them as we'd never get anything done.

[imron]

Even the worst mnemonic system is better than rote memorization of characters by stroke.

I would guess that very few people beyond absolute beginners remember characters stroke by stroke, or that many people would recommend that as a way to memorise characters.  Shape by shape is by far the more common way, and this is certainly how native speakers conceptualise all but the simplest of characters.

[studychinese]

Looks like we are going to have a good old debate here.

 

I don't see how going to the effort to learn hundreds or thousands of stories is an efficient method of learning to read and write Chinese characters (under Heisig you don't even learn the proper pronunciation, right?). 

 

To me the best mnemonic is simply the pronunciation of the character. The way I learned the Japanese kanji, the 1980 JOYO KANJI, was through reading, looking them up, and then multiple exposures to the characters. In Japanese an individual character usually has multiple pronunciations. It was easy enough for me to add Korean and Chinese readings to these characters.

 

I wonder what the opportunity cost is for remembering all these mnemonic stories based on the Heisig method. I would surmise that a lot of people are spending an inordinate amount of time memorizing the stories when time could be more efficiently used attaching pronunciation to the character.

 

Lastly... one must ask why Japanese or Chinese people do not use this method. How can a billion people be wrong?

 

I agree with this young man.

https://www.youtube.com/watch?v=uxuPUHe1bRU

[tysond]

@studychinese do you have a technique that does not involve memorizing thousands of things?  Simply pointing out that there are thousands of things to remember does not mean phonetics suddenly reduce the learning effort from thousands to a few hundred.  It's still thousands of things you need to remember, just that you can use phonetic as part of the mnemonic.  You are assuming that people using these techniques haven't done the math.  I believe they have thought about it quite a lot.

 

As I am sure you know but didn't bring up, there are many homonyms in Chinese.  Actually it's much worse with phonetic components because it's not just "same sound".  It's "similar sound" meaning the tone, initial and sometimes the entire sound is different although the written component may be the same.  If you learn a character by phonetics, how do you remember which ONE of the similar sounds goes with the character you are trying to write?  I guess you have to remember the *meaning* as well as the *sound*, as well as what the *variation* of the sound.  So at this point, you need some way to remember all of that.  

 

楼， 数 - have the same component but how do i distinguish them?  

How do I distinguish 淘 and 桃?  桃 and 逃?  And 陶？ Exactly the same sound.  Help me out here.

 

Maybe it's just me, but I end up referring to the most common meaning of the components and making stories.  Once you get to that point, you are using meaning + meaning to help remember which is which.  So you are back to building some kind of mnemonic based on the meaning, or just brute forcing it.

Now if your spoken Chinese is great -- you can mnemonic in Chinese.   But if you don't already know the meaning of 3000 characters, maybe it's OK to know the English meaning for a while until you learn them and their readings and meanings.  

 

The phonetics are very useful, for a less familiar character (is it 淘 or 缺? both have similar components...).  Frequently it will give a clue that "yep that is pronounced similarly, probably got the right one".

And sometimes you'll think of the phonetically similar character, and need to think "gee I wonder if it's 桃 or 逃" and immediately you'll jump to meaning to figure that one out.

And then after a while you'll stop worrying and just write the correct one, and it will be ingrained.  

 

"How can a billion people be wrong?" is assuming a billion people are in the same situation.  If adults learned Chinese like Chinese children do, it would take over 10 years to read a newspaper.  Progress is made by looking for better ways than the status quo.

 

I watched the video.  Seriously this guy hasn't read the books and took a phone call in the middle of the video :-) .  But let's consider what he said anyway.

 

His arguments are that sure you might be able to learn all the kanjis meanings really fast but you won't know all the readings.  He says it's a flimsy (at first), and then later he says a good skeleton, but then you have to read a lot afterwards, it's not complete, etc.  And he says it's good if you are having trouble remembering the kanji.  This is all true.

 

It's just a scaffolding on which you will hang the pronunciations and vocabulary.  My recommendation is to do book 1 quickly (10+ a day), revise using SRS, simultaneously start reading and studying vocabulary/pronunciations -- you will quickly recognize and be able to write the top 1000 characters, and another 500 fairly frequent ones, which covers a lot of vocab.  Study away until you are well on your way to HSK4.  Then read book 2 more slowly, integrate it with your other studies and use it to expand your vocabulary - probably good to add example words at this point.  Book 2's character frequency is more suitable for HSK5/6 in my opinion.

 

However if your spoken Chinese is already near HSK4-ish you might just want to power through both books and then go on a reading binge.  

[imron]

Mnemonics and SRS are about the only two technologies that we have available.

Yes, how on earth did people ever learn Chinese without them?

 

Makes me wonder how I learnt Chinese, considering I don't use mnemonics and when I started learning, SRS was basically unusable for Chinese (Supermemo anyone?), and I've never really used it anyway (flashcards yes, SRS not really).

I tend to agree with studychinese and have written similar things here.

 

How do I distinguish 淘 and 桃?  桃 and 逃?  And 陶？ Exactly the same sound.  Help me out here.

淘 has the water radical followed by 匋, whereas 桃 has the wood radical followed by 兆。逃 has 兆 followed by the walking radical, and 陶 has the ear radical followed by 匋.

To help with with sounds when you are learning them, don't think of a sound and associate a character with it (you'll get far too many overlaps), rather, think of a character and associate a sound with it.  That way it doesn't matter if multiple characters have the same sound because in your head they are all being treated separately.

 

Progress is made by looking for better ways than the status quo.

I very much agree with this, and that it's always important to be looking for ways to improve.  I'm not sure mnemonics for learning Chinese help with that (see linked post above for reasons) even though they provide short term relief from the arduous task of memorisation.

 

Where I think techniques like Heisig are useful are that they teach the learner to break down characters into component parts, which is a very valuable skill.

[studychinese]

It's just a scaffolding on which you will hang the pronunciations and vocabulary.  My recommendation is to do book 1 quickly (10+ a day), revise using SRS, simultaneously start reading and studying vocabulary/pronunciations -- you will quickly recognize and be able to write the top 1000 characters, and another 500 fairly frequent ones, which covers a lot of vocab.  Study away until you are well on your way to HSK4.  Then read book 2 more slowly, integrate it with your other studies and use it to expand your vocabulary - probably good to add example words at this point.  Book 2's character frequency is more suitable for HSK5/6 in my opinion.

 

 

I don't take issue with this at all. I just don't think that Heisig can be the primary way of learning Chinese characters. In the above example you could replace Heisig with a more standard study of radicals (the radicals have names and proniciations) for a much better result.

The fact that Heisig presents no pronunciations in the book is unacceptable, in my opinion. Perhaps Heisig accounts for all the people that are able to give a rough meaning in English of a given character, but are unable to pronounce it.

[hedwards]

@imron, the sense I got from the folks I was dealing with was that they learned stroke by stroke and it wasn't until they were actually reading that they'd be going by the shapes. For all the instances I saw of them trying to remember or communicate a character without writing, they were always going stroke by stroke. Additionally some input methods for computer are purely by stroke as well for at least the first couple strokes.

 

@StudyChinese, I'd strongly suggest you try learning in that fashion rather than discarding the idea out of hand. Some characters like 大， 中 and 人 do benefit a bit from understanding them, but others like 最 or 龍 aren't intuitively obvious in their entirety and anything you tell yourself to justify it is probably not too bad. Then there's characters like 雪 where a mnemonic to connect 雨 and 山 is probably going to help. technically speaking it's a mnemonic to associate 山 with mountain as that doesn't look anything like the mountains around here.

 

There is a mnemonic system for any kind of information that will work for you, but that doesn't mean that every mnemonic system will work well for everybody, it does depend a bit on how your particular memory is laid out. Auditory systems rarely, if ever work for me, because I store my memories visually. So, when I want to store a character's sound, I store the letters and the tones primarily as a hint as to how it's supposed to be pronounced.

[tysond]

淘 has the water radical followed by 匋, whereas 桃 has the wood radical followed by 兆。逃 has 兆 followed by the walking radical, and 陶 has the ear radical followed by 匋.

To help with with sounds when you are learning them, don't think of a sound and associate a character with it (you'll get far too many overlaps), rather, think of a character and associate a sound with it.  That way it doesn't matter if multiple characters have the same sound because in your head they are all being treated separately.

 

 

I do appreciate the responses @studychinese I just don't get the technique.  I've had friends also say the same thing to me - that you CAN'T throw away the sound because it's how they learned characters.  But then when it comes to how they actually learned the characters, they won't explain a technique, just that the sound is useful.

 

However, what you said is, I associate a character with a sound.  What does "associate" mean?  My brain doesn't have an "associate" function, it just has imagination, visualization, etc, the standard functions that come with the human brain.  I sure wish it did have an associate function.

So what process do I go through to associate?  I assume at the end of the association phase I will have perfect recall, is that right?

 

For reading, I suppose you keep reading/drilling on the character's pronunciation until the sound of the character comes back to you when you read it.

I get that, it's exactly how I learn readings.  No fancy mnemonics.  Just repetition.  The brain seems quite good at this.  And the best part about phonetic characters is that even when you mis-recognize a character you might pronounce it right!  Yay!  Although you might also mis-recognize the meaning by mistake too.

 

But for writing, this doesn't work.  I need to take the sound (tao2tai4) and produce the characters.  How do I know which one it is? 陶 淘 逃 桃?  Or something else maybe?  Or any other combination like wood radical with 匋, or ear with 兆?  

[imron]

My brain doesn't have an "associate" function

 

It absolutely does!  But perhaps you haven't paid enough attention about how to consciously control it?

 

For example, if I say the word "France" a whole bunch of things you associate with France will spring to mind (or at least will come to mind with some minimal thinking) - perhaps the Eiffel tower, baguettes, red wine, onions etc.  You do that because your brain has associated those things together.

 

On a more concrete level, everytime you remember someone's name, you associate that person with their name - once again your brain's 'associate' function is working.

 

On an even more relevant level, everytime you create an mnemonic for a character, you are creating an association in your mind between one part of a character and something else, and then the next part of the character and something else, and so on, with all those things related in someway that you create a chain of associations that you can follow from one point to get to another point.  Eventually those mnemonics fade (or at least fall to the background) and all you are left with is a direct association between the bits you are interested in.  Or at least that will be the ideal situation if for example you are ever wanting to read or speak fluently because you will never be able to do that if you are following chains of association for a large number of words.

 

Throwing your own words back at you about how progress is made by looking for better ways than the status quo, an obviously better way would be to avoid needing to create a chain of associations which fade away and eventually lead to a direct association, and just start making direct associations.  I can fully appreciate if you can't do that very well at the moment, but like almost everything, it's a skill that can be trained.

 

For me, the association happens by thinking of things together and mindfully thinking of them as connected.  As I visualise the characters (see my post linked to above, and several of the follow up posts in that topic), I just think about the meaning, and mentally hear the pronunciation, all the while thinking about points of interest that connect them together, and I do that over and over until I feel that the association has stuck.  Taking 淘汰, they both contain the same radical (氵), delving further, if you look at the meaning for 淘 it means to rinse/wash clean, which provides a nice hook to the 氵.  The pronunciation hook then comes from the other part of each character and the reason I know that it's 匋 and not 兆 comes down to having visualised it dozens of time so the association becomes very clear.  This whole process can become very fast.

 

 

I assume at the end of the association phase I will have perfect recall, is that right?

 

Depends on your definition of perfect so I'm not going to claim that.  What I will say is that you get good at the things you train.  The more you train yourself to be able to visualise characters and associate sound and meaning with them, the better your ability to do that will get.  I certainly couldn't do it very well when I first started learning Chinese, but I'm considerably better at it now.  Are there times when I forget things?  Yes, of course, but not in any way that has a meaningful effect on my studies.

 

 

the sense I got from the folks I was dealing with was that they learned stroke by stroke and it wasn't until they were actually reading that they'd be going by the shapes

 

Everyone starts learning stroke by stroke.  Characters like 一 十 人 大 etc.  Once you get the basics down you very quickly (after a hundred or so characters) move on to shapes.  Ask a person how to write a character e.g. 草, and they won't start going horizontal, vertical, vertical, vertical, and so on… they will say 艹 radical on top and 早 on the bottom, and the will do the same for all but the simplest of characters.

 

 

Additionally some input methods for computer are purely by stroke as well for at least the first couple strokes.

 

Stroke by stroke input methods exist, but they are horribly inefficient.  Far more common are shaped based input methods like Wubi (do a search on the forums and you'll find dozens of topics discussing it).

[renzhe]

It's amazing how the simplest things can evolve into heated arguments

Mnemonics are simply silly stories that are easy to remember. You will memorise large amounts of data anyway, it's just that some information is easier to remember than other. If you have a nice trick that can help you remember something you're struggling with, you'd be a fool not to use it.

At the same time, I agree with imron. It's impossible to learn 3000 characters stroke-by-stroke. Your brain will figure out components in a way that's easier to remember, that is what our brain does, this is what we evolved to do, and this is why a complex system such as Chinese writing is possible in the first place. Even if you TRIED to only memorise strokes, you couldn't. Your brain sees patterns everywhere. It is the basis for imagination and all of our cultures.

We all use mnemonics -- whether we know it or not. We all decompose characters into components once we're past the first few hundred. The thing about books like Heisig is that they offer a complete, consistent programme based on this idea, which may (or may not) increase the speed of learning.

Lastly... one must ask why Japanese or Chinese people do not use this method. How can a billion people be wrong?

Illiterate Japanese and Chinese people learning characters as adults learn them exactly the same way as the rest of us, with all the associated problems.

[hedwards]

@imron, thanks for the info, I seem to have made the mistake of assuming that since the locals were encouraging me to go stroke by stroke that they were doing that for their characters. But, this does make sensel

 

@renzhe: mnemonics are more than just stories, that is one form of mnemonic, but there's quite a few different systems out there that I encourage everybody to take a look at. Stories work well for some things, but they may be rather poor for others. I personally don't bother going to a story unless lesser methods aren't cutting it.

 

And you're absolutely right about mnemonics, it's impossible to draw a definite line between the use of mnemonics and natural memory as ones natural memory involves creating these connections, it's just that people are usually not aware that they're doing it.

[tysond]

Hope this isn't considered too heated :-)  I am really interested in alternative methods to learn characters efficiently.  

 

Heisig's decomposition and imaginative memory technique took a bit of time to learn and I think are definitely a vast improvement on the brute force I was previously using.

It's easy to criticize some of the decisions he's made, but I'd prefer to understand what people are using instead.  

 

The pronunciation hook then comes from the other part of each character and the reason I know that it's 匋 and not 兆 comes down to having visualised it dozens of time so the association becomes very clear.

 

 

So this is the nub of the discussion for me.  I find it hard to believe that an abstract visualization has enough staying power.  However going back to your older post I come across something very interesting.

 

 so typically I'll visualise the scene/story/article at the point related to the character.

 

 

 

So you actually associate the meaning of the character with a narrative where you found it.  I get the staying power of this.

 

But, do you associate any of the component meanings with this meaning (e.g. with the context)?  Or that has just become pure visualization after the top layer?

When you recall a character (especially the first few times, or after a long break) do you get a bit of context coming back to you as you try to remember it?

[Gozer]

Hi there,

 

Sorry to just wade in, I'm new here, first post, handshakes all around, sorry no cake..

 

What qualifies me to chuck in my 2 cents?

Well, I completed "Remembering the Kanji" 1 in about 7 months and am currently halfway through "Remembering the Simplified Hanzi 1" since I just moved from Japan to mainland China.

 

Admittedly, doing Hanzi after Kanji gives you a small edge as you are already used to the process, making up mnemonics and -most importantly- the handwriting.

It has also given me some troule when I confuse similar traditional/simplified/chinese/japanese characters - must make very distinct mnemonics to avoid confusion.

 

Because I had fun and becasue it worked for me, I recommeded the method to others, who then ended up having similar issues as the OP.

I have found that usually, if the mnemonics do not work for you, if you are either:

- Focussing too much on the key words rather than the 'movie in your head'

- Not making your mnemonics memorable enough (oh, the irony...)

 

The proof that mnemonics work is -imho- simple. If you go and see a movie (some blockbuster) you can still recall exact scenes, dialogues, quotes, sequences etc. months after you saw the movie ONCE.

You didn't go into the theatre with a notepad and study the movie notes with flashcrads afterwards, did you? Yet you can recall/visualise TONS of information about a movie you saw once, a long time ago.

Even complete dialogues and the clothes they wore. Why?

Because that stuff was interesting enough for your brain to remember. Your brain loves that. It WANTS to remember it So what does that mean for your mnemonics?

 

Go crazy! Go Tarantino/Abrams/Spielberg/Coppola on them. Make them outrageous, hilarious, disgusting, sexy, repulsive. Or even all of the above in one. And you must ADD SOUND FX in your mind while you do this. The more vivid, shocking, entertaining or exciting you make them little movies, the better they stick. Every single one should be Oscar-worthy.

Don't be shy. Go mental. Delve down into the deepest abyss of your imagination. You don't have to tell anyone what your mnemonics really are. Trust me, if you are too emberassed to share them, they probably work.

That's how I did it. Make the colors bright and garish, add a stonking soundtrack, outrageous characters and make MOVIES. Don't try to remember keywords and think that is the same as creating working mnemonics.

A lot of readers don't quite get the importance of this, especially if they skip reading the introduction.

(It may have helped that I've read Derren Brown's 'Trick of the Mind', which has a thing or two to say about mnemonics and memory. The '20 word' exercise he asks you to do convinced me that mnemonics really work if you apply them correctly. I can still recall all 20 perfectly - in both directions- 3 years later, without ever studying them.))

 

As for StudyChinese who wondered: "Lastly... one must ask why Japanese or Chinese people do not use this method. How can a billion people be wrong?"

Well, that is becasue you need to be fluent in a second language to use this method to learn your native language. You really don't see the issue with that? A billion people are not wrong or right, they simply have no choice.

From my discussions with Chinese and Japanese teachers, I've learned that their education system is strained because such an unusually large part (compared to, say, countries using a romanized alphabet) of their time in education is spent learning their own laguage. A professor at Keio admitted that using a Heisig-esque method, focussed on efficient memorisation rather than use would be the most efficient way to learn Kanji, but practically not doable in Japan because of the need for a second language in order to do this. Furthermore, if you had read even the introduction, you would have understood that generally a human needs to be older than 12-13 before they can produce effective and memorable mnemonics by themselves. You can't simply delay native language teaching until that age,

 

And no, I'm not a blind follower of Heisig, I believe there are flaws in the process and execution. But too may people expect it to do something it is not designed to do. I see it as basically learning the alphabet. You don't ask what B or T or Z means, right? You use it as building blocks to learn how to read. Especially where Chinese is concerned, where so many 'words' are actually bigrams, making a 'single' character not that useful. The plus points are that once you have learned the 'alphabet', you can use dictionaries to look up words far more quickly, so you can use 'native' language teaching materials (Working through Chinese primary school language books).

 

And you can 'embellish' your mnemonics with clues to pronunciation and tone, like Tuttle and Easy Chinesey try to do. On the one hand I feel Heisig 'omitted' to do this, but on the other hand I also feel that focussing on one task only might actually help with retention more.

 

Hope this helps.

 

Gozer (yeah, that IS actually my real name...)

[imron]

So you actually associate the meaning of the character with a narrative where you found it.  I get the staying power of this.

 

My reason for doing this is that I want to visualise the meaning rather than remember a translation to English.

 

For example if I was learning 茄子 I wouldn't be thinking of the English word 'eggplant', I would imagine an eggplant itself.  Then in the same way I know that aubergine is another word for eggplant, it then gets stuck in my head that 茄子 is also another word for this object.

 

Not all words can be visualised or conceptualised so easily and so for complicated or abstract items associating the meaning of the character with the context where I first encountered it comes in useful because it gives me a clear specific thing to think about.

 

 

do you get a bit of context coming back to you as you try to remember it?

 

Regularly but not for everything.  淘汰 for example I can't remember where I saw it first, but for example other 兆 words I still remember such as 逃跑 from a Tintin comic.  I can even remember the panel pretty clearly (Tintin was escaping from someone on a boat) - but perhaps ironically I've forgotten which Tintin comic it was!

 

I also get the context coming back when I see that word elsewhere or in other stories, where I will think 'oh, that's a word I learnt when reading X'.  When that happens I might also think about it a bit more, associating the current context with that word also so it strengthens the association in my mind of the generic concept.

 

But, do you associate any of the component meanings with this meaning

 

If the component is related to the meaning in a relatively direct way (e.g. 氵in 淘), or using an example from the other thread 马 with 驹 (still remember where I learnt this - 书剑恩仇录, and although I forget the specific context it was talking about a horse (or one of the horses) the heroes were riding).  Then I'll make a note of that because it does help things stick, but I don't try to exaggerate this or make it a core part of the visualisation.  I'm not opposed to this sort of hooking/associating.  It's more the associating unknown or unrelated things with a word, or creating a chain of associations that needs to be walked before getting the answer.  I'm also especially opposed to incorporating English words in any sort of mnemonic or story to 'help' with pronunciation because it creates a false association and is a recipe for bad pronunciation.

[Gozer]

Just as an illustration, this is what I used to explain to another Laowai how the mnemonic method worked in principle.

One night, we're eating 羊肉串by the side of the road and we we were discussing this topic.

 

He said he could remember 'chuan' because it is visually similar to something being skewered (which he thought was the actual meaning, rather than 'string together')

And Yang he could recall because he could vaguely picture a sheep's head (ram) with 2 horns in this Hanzi, but he did often mistake it for Xin or even Ju.

 

(Those following Heisig; this is NOT a mnemonic using Heisig keywords, skip if you don't want to get confused or are very sensitive )

So I used 肉 (Rou, meat) as an example of creating a good mnemonic, (good as in memorable, not pleasant..) I asked him if he knew Ren 人。 He did.

So I told him. Imagine TWO fat bald men. They're sweaty. They're greasy. They're oily. They are naked. They are spooning. In a box. A wooden shipping crate, with the lid off. The crate has 'MEAT" stamped on the side in big red stenciled words. The inside of the crate is lined with plastic. To preserve all the sweaty, oily, greasy goodness. Their eyes are closed, but if you get close enough to the crate, you can hear one of them whisper contently: Rou... Rou... Roooouuuuu.... and the other echoes him by murmuring 'meat... meat... meeeeaaaat...."

Now let this little video clip play in your mind for a minute or so. Make it clear, bright, smell the sweat and baby oil, hear the whispers, see how rough the wooden crate is and wonder, wonder about possible splinters.

 

Don't even worry about trying to remember...  worry about trying to forget this image.

[imron]

(It may have helped that I've read Derren Brown's 'Trick of the Mind', which has a thing or two to say about mnemonics and memory. The '20 word' exercise he asks you to do convinced me that mnemonics really work if you apply them correctly. I can still recall all 20 perfectly - in both directions- 3 years later, without ever studying them.))

These sort of things are great for recalling random, unrelated things where there is a high tolerance for recall time, e.g. taking 1-2 second to recall a word isn't going to cause any serious issues.

 

I'm less convinced of its utility for language learning, because I don't like the introduction of any intermediate steps required to process or generate language as this slows things down, and anything less than instant is going to cause issues to some degree.  Also, for example do you want these things:

 

 

Go crazy! Go Tarantino/Abrams/Spielberg/Coppola on them. Make them outrageous, hilarious, disgusting, sexy, repulsive. Or even all of the above in one. And you must ADD SOUND FX in your mind while you do this. The more vivid, shocking, entertaining or exciting you make them little movies, the better they stick. Every single one should be Oscar-worthy.

Happening in your mind when you are trying to read a newspaper?  If you build up strong associations between a word and these other vivid, shocking movies with sound fx, then once you start using the word you are going to have to do work to disassociate them or have random unrelated thoughts going through your mind at the same time (e.g. bald, naked men, dripping in baby oil).  If you don't have the associations going on and are just going directly from recognising the word and understanding it, why not just try to get to that step first without all the unrelated associations.  I can accept how creating a story might be useful for a particularly tricky word that you just can't seem to make stick in your mind, but I wouldn't want to be trying it for every word.

[imron]

Don't even worry about trying to remember...  worry about trying to forget this image.

I'm not sure whether to upvote your post for providing such a vivid example of what you were talking about, or downvote your post for providing such a vivid example of what you were talking about.

[Gozer]

Hi Imron, (Can't figure out how to do a "quote on this forum?!")

 

I think a distinction should be made between someone who starts Heisig as a "non-immersed/true beginner" and an "immersed/false beginner"

When I studied Japanese, I did it at home in England, starting from scratch with zero knowledge. I found that the stronger, the more vivid the mnemonic was, the faster the recollection/process.
As Heisig also states in his book, after a while the

Keyword > Mnemonic > Character sequence gets shorter and shorter until you find yourself going effortlessly from English>Kanji and Kanji>English, without using the menmonic. Your brain can skip that step.

Especially if you read/write Kanji every day and/or review daily.

 

As for Chinese, I picked some up years ago (very basic) and since I have a 'grounding' in Kanji and I am actually learning Chinese while in China, the whole process is much faster and more efficient and often I do not need to come up with elaborate mnemonics like this anymore, because I can alreay clearly separate radicals and hanzi by sight. Is you correclty guessed, they are used for the tough ones.

 

But as a true beginner, a little overkill can really pay off in the long run, especially if you don't review efficiently (say, without SRS) or often enough. In that case those exaggerated mnemonics jump right back in and put you back on track.

Of course, everyone's brain works differently and what works for me may not work for someone else. The key point was that in, my experience, when mnemonics fail to do their job, they were not 'strong' enough.

 

The mnemonics/movies do not pop up in your mind when reading a newspaper or subtitles (althoug with the programming on CCTV, it would be an improvement...), because by they are merely tools to anchor the characters in your memory and are not used for recall after a while. (I hope that makes sense... See them as training wheels on your first bicycle)

[imron]

Can't figure out how to do a "quote on this forum?!"

Manually.

 

after a while the

Keyword > Mnemonic > Character sequence gets shorter and shorter until you find yourself going effortlessly from English>Kanji and Kanji>English, without using the menmonic. Your brain can skip that step.

And the point I'm trying to make is that if your brain can skip that step, why not try to train it so that it can always just skip that step, especially if you read/write Kanji every day and/or review daily.

 

As an aside, I would also argue that going Kanji>English is also sub-optimal.  Ideally you would want to go to Kanji>Meaning (see above example re: eggplant).

 

 

 

The mnemonics/movies do not pop up in your mind when reading a newspaper or subtitles

I'm sure they don't when you are competent at reading and once you have already got to the stage where the mnemonic doesn't play a role.

 

This plays in to my whole point though of not wanting to spend time and effort remembering things whose only purpose is to be forgotten.  In the time it takes to come up and visualise scenes like the one you described above, I can breakdown and visualise a character several dozen times all the while thinking about meaning and pronunciation and things directly relevant to the word itself, and getting it firmly entrenched in memory.