Teaching unexpected readings of phonetic components

[chrix]

So, I'd like to open up a debate on this other thread I started.

Do you think it would be useful for Chinese learners to teach them about that. I was thinking about some "general expectations", more common and less common alternations, exceptions and so forth.

Any ideas on that?

What level would appropriate, or would you start right from the beginning, and how would that relate to the teaching of radicals.

Also, how distracting do you think would it be to actually teach it from a perspective of historical linguistics? Or would a strictly synchronic approach be better?

Oh, and please keep the examples coming on the other thread

[jbradfor]

As a frequent poster to "that other thread", I assume you thought you would hear from me

Yes, I think this should be taught, in three ways.

The first way is very indirect. From the beginning of learning, when deciding which words should be taught in a textbook, I think the "difficulty" of the word to be learned should be taken into account. [if it isn't already.] While I would expect "difficulty" to be somewhat vague, as I would expect it to vary from person to person, I would also expect that on average some words/characters take longer to learn than others. And having a character with the same/similar pronunciation and just a difference in radical is to me often the easiest to learn. Conversely, a character with a different radical but very different pronunciation I find the most confusing.

The second way is more direct. When a new character is introduced, I think it would be useful to point out similar characters that have already been taught. This would obviously be later (maybe starting second year?), as the student needs to know enough characters for a comparison. I would find this useful as it would make is easier for me to learn to distinguish them, and it would remind me of the older character. And in this, characters with same phonetic / different radical but an "unexpected reading" should be particularly highlighted.

The third way is that I would like a tool that would take all the characters I know, and combine them by phonetic part. I'm doing that manually, but a tool would be nice.

Beyond that, I'm not sure how useful this would be as a pedagogical tool. I'm not at all familiar with the history, so I'm not sure how much that would add. But I would think that would be taught starting at the third year at the earliest.

[chrix]

Thanks for your response, it sounds very reasonable to me. I do think one should start teaching radicals only after students have at least a couple of hundred characters under their belt, and that would be a good time to introduce them to some regularities in phonetic components as well.

(Another reason for this would be that some of the most irregular characters are just some of the most frequent ones.)

As far as phonetic components go, I've been looking for a components database in the public domain. Another idea I've had is to use the Unihan database, as that has information on radical and stroke count (though I forget whether it has that info for both simplified and traditional). But you still would need to do some hand-coding, there wouldn't be a way around that...

[Lu]

The second way is more direct. When a new character is introduced, I think it would be useful to point out similar characters that have already been taught. This would obviously be later (maybe starting second year?), as the student needs to know enough characters for a comparison. I would find this useful as it would make is easier for me to learn to distinguish them, and it would remind me of the older character. And in this, characters with same phonetic / different radical but an "unexpected reading" should be particularly highlighted.

I would actually start to do this much earlier. Perhaps not formally, but when going through the 生字, explain what they mean, but also how they are put together: what the radical is (if it is fairly obvious, not the weird ones right away), what the phonetic component is, and where the students may have seen it before. And if a reading is unexpected, tell them so. Don't test students on all of this, but ease them into it, so they know the system. Once they learn more words and start looking up words, they will have the tools for it.

[chrix]

It depends on what you mean by "much earlier", but at the very beginning it doesn't make too much sense, as students would have to grapple with the very concept of hanzi. Also, the very first characters do tend to be quite idiosyncratic from a systemic point of view, and wouldn't be suitable examples.

Speaking of teachers, I've heard people who took university-level Chinese courses tell me that they would have liked to know about the Six Principles when they studied characters (about which I had just told them), so it's certainly an important factor. But as far as I understand, most textbooks, even those with a character supplement, don't do a good job of explaining the structure of characters. I hope I'm wrong and it was just my textbook

[Lu]

It depends on what you mean by "much earlier", but at the very beginning it doesn't make too much sense, as students would have to grapple with the very concept of hanzi. Also, the very first characters do tend to be quite idiosyncratic from a systemic point of view, and wouldn't be suitable examples.

I think such things can actually help in grappling the concept, as it shows that it's not just a random bunch of strokes, but that there's a lot of logic in it. Barely 30 characters in, as I recall, we learned 他 and 也 in one lesson, and all noticed the similarity. And the dissimilarity in pronounciation. At which point the teacher can make a short remark about how usually similar phonetic components means similar pronounciation.

[jbradfor]

@chrix

I've heard people who took university-level Chinese courses tell me that they would have liked to know about the Six Principles when they studied characters

What are the Six Principles? I assume you do not mean Six principles of Chinese painting?

[chrix]

I meant the 六書, usually translated as "Six Writings", but I think "Principles" is better.

http://en.wikipedia.org/wiki/Chinese_character_classification

These Six Principles were created thousands of years after the characters by the great Xu Shen and his contemporaries, so they're not always correct as far as the earliest characters are concerned, but by and large they do work out.

[jbradfor]

It has a name!

Yeah, I basically figured that out myself, not in that much detail, but the basic idea. But had I been taught that earlier, it would have saved me some time.

So maybe this thread should be expanded to discuss how much "character theory" should be taught, and at what level?

[chrix]

Sure, I mean the entire thread is about how to teach the most important character class (what was it 60%-80%?), so it would all be connected.

Also, does anyone know of textbooks or any other learner-oriented books that actually actively teach this kind of thing? Like a 六書 Heisig maybe

[chrix]

EDIT: Talking about the character 盜.

My kanji dictionary says, the upper part is not 次 orginally, but a character with a 三點水, which is the original form of 涎 xián "saliva". So even though it has nothing to do with 次 it is apparently still a "true" 會意 character (the lower part being 皿 "plate", the character describes a situation of food theft: someone sees food and starts to lust after it so much that they start salivating onto the plate)

Edited February 21, 2010 at 07:15 PM by chrix

[Guest realmayo]

OK. Plenty of this stuff is a bit beyond me. When you say a "true" character, do you mean that it was, basically, created for a particular word (meanigng) long ago, and that process didn't involve selecting any components for their phonetic element?

[chrix]

by "true" I mean that the component was not selected for its phonetic value. Many 形聲 characters are in fact 形聲會意 characters, that is the phonetic component also has some semantic relation (and the category of 形聲會意 does not appear in the Six Principles aka 六書)...

(Moving this discussion from the other thread)

[chrix]

OK, one problem there is with the concept of 會意 and 形聲 is that originally, there were no "radicals".

Characters were used write words with similar meaning and words with similar pronunciation. So a character like 月 yuè would also be used to express the concept "bright" (míng), and also "to hack off your foot as punishment" (yuè). At some point the concept of radicals was introduced to differentiate between all of these, so now you have 明 and 刖 and so forth.

When Xu Shen wrote down his Six Principles in 說文解字, the radicals were already firmly entrenched, so that's how he came up with 會意 and 形聲. Only modern research in our time has established that his analysis did not reflect the original way characters worked.

Now, what do you do with this? I have a feeling that including all of this in teaching characters would just confuse people too much? On the other hand, it explains why there are 會意形聲 characters quite well, and it would counter the tendency that too many people take the Six Principles as absolute truth.

So two more problems with this:

- we only have records for a small number of characters from the "radical-less" stage

- a lot of characters have been created after Xu Shen's time, so in a way his principles hold good a large number of characters (the question just would be for which characters )

[Guest realmayo]

Regards the initial post above: after deciding to study characters properly, alone, I did come across descriptions of how the things were formed, but I didn't really understand these descriptions for ages, either they were too complicated, or, given my then-limited studying of characters, too abstract, too much a history lesson.

Perhaps if I was being taught by a teacher, who could teach well, I would have grasped the detailed concepts early on. But even so, without several hundred characters learned, I'm not sure I'd have made much use of the concepts -- beyond the radical + phonetic combination.

In fact, it would have probably made it more difficult to learn them, as I stressed out about trying to work out which category a given character fell into.

More generally, in my eyes an ideal course would separate reading/writing from everything else. I bought, too late for me, an excellent book called Fundamentals of Chinese Characters Amazon link, which started with the basic but also common characters, eg 言、木、目、母。 These all need to be learned some time, and they derive from pictures (I forget the technical term) -- so why not learn them first?

Of course, if you're learning speaking then you need stuff like 我、是、什么的，which are more complex ... maybe it would be best to learn how to recognise those, but for writing, follow a different road, as suggested by the book I mention.

EDIT: if I remember right, the book does a great job of introducing those categories, but very firmly, clearly, and with examples using the simple characters already learned.

[chrix]

More generally, in my eyes an ideal course would separate reading/writing from everything else. I bought, too late for me, an excellent book called Fundamentals of Chinese Characters Amazon link, which started with the basic but also common characters, eg 言、木、目、母。 These all need to be learned some time, and they derive from pictures (I forget the technical term) -- so why not learn them first?

Of course, if you're learning speaking then you need stuff like 我、是、什么的，which are more complex ... maybe it would be best to learn how to recognise those, but for writing, follow a different road, as suggested by the book I mention.

It's good to hear that there are some books around that seem to teach the Six Principles in an accessible way... I think you'd need both of what you said, the absolute beginner just needs to wholly memorise the characters, and not characters like 木/水/火, but those they will encounter in the lesson text, and especially those with grammatical meaning 我/你/的.

Also, there's a lot of irregularities in basic and frequently used characters, so these methods wouldn't help much anyways. For a certain number of basic characters you just need to bite the bullet and learn them by heart.

The term for characters derived from pictures is pictographic (*), or

象形. With a similar category called 指事 ("ideographic" (*)) they only account for a tiny portion of Chinese characters, but most of them are very important characters, either because they're high-frequency characters or because they occur as components in many other characters. So a good book should present two lists of characters to be learned by heart, a) the basics for the grammar and to start you off with the textbook, and B) 象形 and 指事 characters.

(*) the English translations for both 象形 and 指事 controversial, but let's not get into this just yet..

[Hofmann]

It depends on the kinds of students one teaches. For high schoolers, I think it wouldn't be too helpful to teach them about character classifications until they have a few hundred characters in their vocabulary to serve as examples, and even then I wouldn't go into much detail. For graduate students, I would give all the 六書 stuff to them straight up, using historical linguistics whenever it presents information more clearly, cutting no corners.

I have met too many Japanese students who didn't know that hiragana have such (安以宇衣於) a relationship with Kanji.

[Glenn]

Just out of curiosity, how far along were those Japanese students? I can't remember when I was introduced to the kanji that the kana came from in terms of my level. I mostly learned on my own, though, with some classes thrown in for good measure, but I certainly didn't follow any sort of strict, traditional path. It may not have been until I looked at Heisig's second book that I saw them (around my third year or so) or it may have been as early as when I was using Hadamitzky and Spahn's Kanji and Kana (first year).

[Hofmann]

Just out of curiosity, how far along were those Japanese students?

One has been studying for 8 years.

[Lu]

Then again, I suppose some figured it out on their own. I don't even have any systematic knowledge of Japanese, and I figured the relationship out when I looked at kana (and already knew some Chinese).