Radicals vs. phonetic elements vs...

[Woods]

Could be a silly question, but I am trying to figure that out: is it like some elements (i.e. basic characters) are used as radicals only (e.g. 金), others could be either radicals or phonetic elements or parts of an ideogram (馬), and finally I suppose there should also be some basic characters which would never be used as radicals but only standalone or in ideograms?

I'm pretty sure the answer is yes since the list of radicals as far as I remember is limited, while there should be more simple pictograms than the number of radicals; but anyway I wrote the question cause it was on my mind and any advice on how to dustinguish what is what is welcome!

Also is the phonetic element completely arbitrary (i.e. they just care about the sound), or do they try to match it with the meaning as much as they can (i.e. they would choose one particular phonetic element between several similarly-sounding options in order to get a closer meaning and create some sort of ideogram), or are there just a few elements that are generally favoured phonetically (like the horse) and they always use those?

[schnaz]

Hi Wood, I hope that this link proves helpful.
"Colloquial Terms for Chinese Character Components | Pleco Software Forums" http://www.plecoforums.com/threads/coll ... ents.6734/

[linguoboy]

Could be a silly question, but I am trying to figure that out: is it likeare some elements (i.e. basic characters) are used as radicals only (e.g. 金), while others could be either radicals or phonetic elements or parts of an ideogram (馬), and finally I suppose there should also be some basic characters which would never be used as radicals but only standalone or in ideograms?

I'm pretty sure the answer is yes since the list of radicals as far as I remember is limited, while there should be more simple pictograms than the number of radicals; but anyway I wrote the question cause it was on my mind and any advice on how to dustinguish what is what is welcome!

I think it's important to remember how the Chinese writing system developed. Although phonosemantic compounds are the largest class of character now, they weren't originally part of the system. Initially, basic pictograms were simply extended according to the rebus principle to represent other words with similar pronunciations. It was only later that scholars hit upon the idea of combining these with a semantic element in order to hint at their meaning and only much after that that someone attempted to enumerate these semantic elements and use them to classify the existing body of characters. (The first attempt to do this, the 2nd Century CE Shuowen Jiezi, identified 540 radicals. The current system of 214 dates only to 1615.)

As a result, the classification can be extremely arbitrary. You even have instances like 王 where the character itself is simpler than the radical (玉) it's classified under! (Historically, of course, these characters originated independently but had similar forms after Qishihuangdi's standardisation.) There are general principles for identifying the radical in phonosemantic compounds, but these tend to break down when you get to those characters which originated as pictograms or ideographs.

Also is the phonetic element completely arbitrary (i.e. they just care about the sound), or do they try to match it with the meaning as much as they can (i.e. they would choose one particular phonetic element between several similarly-sounding options in order to get a closer meaning and create some sort of ideogram), or are there just a few elements that are generally favoured phonetically (like the horse) and they always use those?

I would say it's the latter. I've never heard of a phonetic being specifically selected based on its meaning, though I wouldn't rule out it happening (particularly given the vast range of character variants which have existed over time).

[Woods]

Could be a silly question, but I am trying to figure that out: is it likeare some elements (i.e. basic characters) are used as radicals only (e.g. 金), while others...

Does that sound so much out of style that you decided to correct it?

someone attempted to enumerate these semantic elements and use them to classify the existing body of characters.

As a result, the classification can be extremely arbitrary.

Yes, and it's also extremely strange that nobody bothers to fix it. At the same time, it's better to keep something that has been in place for 400 years as it is - but maybe some obvious mistakes could be explained and corrected? I don't know indeed, what is easier - to put things in order or to learn the system as it is? Does it become clear to the learner how most characters are organised at some point or is it too large and disorganised of a database that nobody gets the hang of, even for the contemporary characters?


I think the "communists" invented some new characters with their own logic (and new radicals?) - how do they fit in the Kangxi system after that? I'm thinking of 煙 (smoke), of which the basic part was amputated and replaced by 因. By doing so, the proletarians turned it into an ideogram. The 垔 part, according to Wiktionary, was used for its phonetics, and there is no info as to what it meant (by the way, do you know a better place where I can check?). But they kept the radical intact - did they do so with all characters they "simplified"? I was thinking that some radicals had been merged and removed, but it seems like the set remained the same?

烟 is so much easier to remember, especially as long as one does not know what 垔 means; but it feels illogical, because whoever was creating the character for smoke, first wrote something and then associated it with fire, and that thing couldn't have been "reason" - it just doesn't make any sense at all. It would have made sense the other way around, but then 因 should be the radical. Are there any characters where another semantic component was added to an already-selected radical before the Chinese Communist Party came about?

Initially, basic pictograms were simply extended according to the rebus principle to represent other words with similar pronunciations. It was only later that scholars hit upon the idea of combining these with a semantic element in order to hint at their meaning

I'm not sure I get the picture 🤔

So you're saying they were just using a random character with the same sound regardless of meaning?

I've never heard of a phonetic being specifically selected based on its meaning

How come? There are so many characters where the non-radical part has been identified as phonetic and yet the combination has some logic which works as the perfect mnemonic. I don't know if those are stories that people come up with later, but many of them seem too good to be just arbitrary? And why would someone not, given the wide range of available characters with the same pronunciation, choose the closest one in meaning? Were a select few favoured - I'm thinking of 馬, it appears too many illogical places. And by the way, could it have been that some authors favoured certain characters for their phonetics and others - other ones? I guess this happened too far back in time to tell?

according to the rebus principle

It's very surprising that it didn't turn into an alphabet/abjad like Greek and Arabic.

But basically, didn't the Egyptian script work somewhat the same?

[Karavinka]

So you're saying they were just using a random character with the same sound regardless of meaning?

Yes. Don't try to find too much logic there, sometimes they make their damn best to give some logic or semantic connection between the phonemic element with the overall meaning of the word, but most of the times they don't and they don't care.

How come? There are so many characters where the non-radical part has been identified as phonetic and yet the combination has some logic which works as the perfect mnemonic. I don't know if those are stories that people come up with later, but many of them seem too good to be just arbitrary?

As mentioned, they did their damned best. You see, there are usually more than one characters to choose what to use as the phonetic component for any given word. When there are choices, they went with the one that makes most sense both with sound and meaning, but they don't always succeed.

But basically, didn't the Egyptian script work somewhat the same?

Yes, and so did Cuneiform and all known pictographic-ideographic ancient scripts follow the rebus principle. It's important to remember that the characters, the glyphs themselves, don't have any meaning whatsoever. It's the spoken word that is represented by the characters that have meaning. This is why the character can represent the same-sounding word, regardless of what the picture originally represented.

[linguoboy]

Could be a silly question, but I am trying to figure that out: is it likeare some elements (i.e. basic characters) are used as radicals only (e.g. 金), while others...

Does that sound so much out of style that you decided to correct it?

I try not to make corrections that are purely stylistic. The way you had it phrased was unidiomatic.

someone attempted to enumerate these semantic elements and use them to classify the existing body of characters.

As a result, the classification can be extremely arbitrary.

Yes, and it's also extremely strange that nobody bothers to fix it. At the same time, it's better to keep something that has been in place for 400 years as it is - but maybe some obvious mistakes could be explained and corrected? I don't know indeed, what is easier - to put things in order or to learn the system as it is? Does it become clear to the learner how most characters are organised at some point or is it too large and disorganised of a database that nobody gets the hang of, even for the contemporary characters?

Nobody has ever bothered to "fix" the Western alphabet, have they? Sure, the classifications are very arbitrary, but they're also very well established, so it's hard to say that there would be much benefit to changing them around now.

Moreover, modern technology means that there's been a move away from classifying characters by radicals anyway. My first Chinese-English dictionary didn't even have a radical index; instead, characters were sorted by stroke count and first stroke. (For instance, 煙 would be listed under 13 strokes and then under "dot".) All print dictionaries have phonetic indexes nowadays (using Pinyin or another romanisation, sometimes alongside Bomopofo) and there are a wide range of input and lookup methods which don't require any knowledge of the radicals, many of quite long standing. Nowadays you can just point your camera at a character and look it up. So why spend a lot of effort reforming and reteaching a system that's almost obsolete anyway?

I think the "communists" invented some new characters with their own logic (and new radicals?) - how do they fit in the Kangxi system after that? I'm thinking of 煙 (smoke), of which the basic part was amputated and replaced by 因. By doing so, the proletarians turned it into an ideogram.

How do you figure? 煙 is pronounced yān and 因 is pronounced yīn--which is the same pronunciation as 垔 has. It's literally just as useful as a phonetic as the element it replaced.

The 垔 part, according to Wiktionary, was used for its phonetics, and there is no info as to what it meant (by the way, do you know a better place where I can check?).

It says right there in the entry that this is an obsolete term for "to block up". It also links you to the Unihan entry where it is glossed as "to restrain; to dam a stream and change its direction; a mound".

But they kept the radical intact - did they do so with all characters they "simplified"? I was thinking that some radicals had been merged and removed, but it seems like the set remained the same?

Not all characters were simplified in the same way. Often, a radical or phonetic is replaced with a simplified form--sometimes both (e.g. 鍾 → 钟). But sometimes a simpler but more obscure character replaces a common one, e.g. 干 for 乾 (a substitution that was common even before the mass simplification). In these cases, obviously the character is reclassified according to the components in the simplified form.

烟 is so much easier to remember, especially as long as one does not know what 垔 means; but it feels illogical, because whoever was creating the character for smoke, first wrote something and then associated it with fire, and that thing couldn't have been "reason" - it just doesn't make any sense at all. It would have made sense the other way around, but then 因 should be the radical.

How do you know what they were thinking?

因 itself didn't originally mean "reason" anyway. It's a representation of a man on a mat and was the original form of the character now written 裀 with the clothes radical; its use for "cause, reason" is an instance of a phonetic loan character.

Are there any characters where another semantic component was added to an already-selected radical before the Chinese Communist Party came about?

I don't understand what you're asking. The radical is the semantic component of a phono-semantic compound character.

Initially, basic pictograms were simply extended according to the rebus principle to represent other words with similar pronunciations. It was only later that scholars hit upon the idea of combining these with a semantic element in order to hint at their meaning

I'm not sure I get the picture 🤔

So you're saying they were just using a random character with the same sound regardless of meaning?

That is exactly what I'm saying. Any basic introduction to the history of Chinese characters will explain this. Here's the relevant section of the Wikipedia article, for instance: https://en.wikipedia.org/wiki/Chinese_characters#Rebus

according to the rebus principle

It's very surprising that it didn't turn into an alphabet/abjad like Greek and Arabic.

But basically, didn't the Egyptian script work somewhat the same?

Pretty much.

[Woods]

Moreover, modern technology means that there's been a move away from classifying characters by radicals anyway.

A move? Can you give examples?

Wiktionary, based on hypetext, is great for connecting characters to their compounds or elements they're made of, but it's awful for putting the word in context - is isn't telling you anything about when and where the character was used and whether it still is, and it doesn't tell you enough about whether it's an important one.

Nobody has ever bothered to "fix" the Western alphabet, have they?

Because they're made of twenty-something characters that can be learnt in an hour, not of 50 000 pictogrammic/phonetic/picto-phonetic/ideogrammic/a-combination-of-any-of-the-above lifelong-learning characters?

So why spend a lot of effort reforming and reteaching a system that's almost obsolete anyway?

To give a full grasp of what characters are available in a learnable way.

And why would it be "almost obsolete"?

I think the "communists" invented some new characters with their own logic (and new radicals?) - how do they fit in the Kangxi system after that? I'm thinking of 煙 (smoke), of which the basic part was amputated and replaced by 因. By doing so, the proletarians turned it into an ideogram.

How do you figure? 煙 is pronounced yān and 因 is pronounced yīn--which is the same pronunciation as 垔 has. It's literally just as useful as a phonetic as the element it replaced.

A-ha, well that makes more sense then - and if 因 was used phonetically for other characters when the Chinese script evolved, it would also feel authentic.


Do you have any info as to when the majority of presently-used characters were created? Or the different stages / time periods when they were added?

The 垔 part, according to Wiktionary, was used for its phonetics, and there is no info as to what it meant (by the way, do you know a better place where I can check?).

It says right there in the entry that this is an obsolete term for "to block up". It also links you to the Unihan entry where it is glossed as "to restrain; to dam a stream and change its direction; a mound".

I think I didn't phrase that right - I meant what the picture meant, i.e. what is drawn?

But sometimes a simpler but more obscure character replaces a common one, e.g. 干 for 乾 (a substitution that was common even before the mass simplification).

Yeah, this is where it gets messy! So how do you fit a new character into a 400-year old classification and still claim it's 400 years old?

Are there any characters where another semantic component was added to an already-selected radical before the Chinese Communist Party came about?

I don't understand what you're asking. The radical is the semantic component of a phono-semantic compound character.

I mean instances where ideograms (i.e. characters with two semantic elements) were made by first selecting the radical and then the non-radical part.

Initially, basic pictograms were simply extended according to the rebus principle to represent other words with similar pronunciations. It was only later that scholars hit upon the idea of combining these with a semantic element in order to hint at their meaning

I'm not sure I get the picture 🤔

So you're saying they were just using a random character with the same sound regardless of meaning?

That is exactly what I'm saying.

There are so many characters where the non-radical part has been identified as phonetic and yet the combination has some logic which works as the perfect mnemonic. I don't know if those are stories that people come up with later, but many of them seem too good to be just arbitrary? And why would someone not, given the wide range of available characters with the same pronunciation, choose the closest one in meaning?

🤔

[linguoboy]

Moreover, modern technology means that there's been a move away from classifying characters by radicals anyway.

A move? Can you give examples?

I already did: You have many options for looking up characters which require no knowledge of traditional Kangxi radicals at all.

Wiktionary, based on hypetext, is great for connecting characters to their compounds or elements they're made of, but it's awful for putting the word in context - is isn't telling you anything about when and where the character was used and whether it still is, and it doesn't tell you enough about whether it's an important one.

Unihan includes frequency data and information on what data sets the character is included in (e.g. Jōyō kanji, which is a list of basic kanji used by the Japanese Ministry of Education). It's linked to from the Wiktionary entries.

Nobody has ever bothered to "fix" the Western alphabet, have they?

Because they're made of twenty-something characters that can be learnt in an hour, not of 50 000 pictogrammic/phonetic/picto-phonetic/ideogrammic/a-combination-of-any-of-the-above lifelong-learning characters?

You can learn a complete alphabet in "an hour" but not 214 radicals?

So why spend a lot of effort reforming and reteaching a system that's almost obsolete anyway?

To give a full grasp of what characters are available in a learnable way.

What do you mean "available"?

And why would it be "almost obsolete"?

For the third time: Because you don't need to know the radical to look up a character.

That's all the system is there for: To make it easy to find characters in lists. But now we have computers.

I think the "communists" invented some new characters with their own logic (and new radicals?) - how do they fit in the Kangxi system after that? I'm thinking of 煙 (smoke), of which the basic part was amputated and replaced by 因. By doing so, the proletarians turned it into an ideogram.

How do you figure? 煙 is pronounced yān and 因 is pronounced yīn--which is the same pronunciation as 垔 has. It's literally just as useful as a phonetic as the element it replaced.

A-ha, well that makes more sense then - and if 因 was used phonetically for other characters when the Chinese script evolved, it would also feel authentic.

It was. Again, you can find this information in the Wiktionary entry. (Look in the section called "Derived characters".)

Do you have any info as to when the majority of presently-used characters were created? Or the different stages / time periods when they were added?

I've never seen a breakdown like that. I imagine some Chinese or Japanese character dictionaries might contain that kind of information.

The 垔 part, according to Wiktionary, was used for its phonetics, and there is no info as to what it meant (by the way, do you know a better place where I can check?).

It says right there in the entry that this is an obsolete term for "to block up". It also links you to the Unihan entry where it is glossed as "to restrain; to dam a stream and change its direction; a mound".

I think I didn't phrase that right - I meant what the picture meant, i.e. what is drawn?

The Shang dynasty form looks to me like a picture of something blocking a watercourse.

But sometimes a simpler but more obscure character replaces a common one, e.g. 干 for 乾 (a substitution that was common even before the mass simplification).

Yeah, this is where it gets messy! So how do you fit a new character into a 400-year old classification and still claim it's 400 years old?

I don't understand what you're asking.

Are there any characters where another semantic component was added to an already-selected radical before the Chinese Communist Party came about?

I don't understand what you're asking. The radical is the semantic component of a phono-semantic compound character.

I mean instances where ideograms (i.e. characters with two semantic elements) were made by first selecting the radical and then the non-radical part.

That's not how ideograms were made. Most if not all were invented before there was any sort of classification by radical.

Initially, basic pictograms were simply extended according to the rebus principle to represent other words with similar pronunciations. It was only later that scholars hit upon the idea of combining these with a semantic element in order to hint at their meaning

I'm not sure I get the picture 🤔

So you're saying they were just using a random character with the same sound regardless of meaning?

That is exactly what I'm saying.

There are so many characters where the non-radical part has been identified as phonetic and yet the combination has some logic which works as the perfect mnemonic. I don't know if those are stories that people come up with later, but many of them seem too good to be just arbitrary? And why would someone not, given the wide range of available characters with the same pronunciation, choose the closest one in meaning?

🤔

If this is what happened, there's no real way to reconstruct it. These scribes didn't leave notes behind.

BTW, can you give me some examples of some of these "perfect mnemonics"?

[Karavinka]

http://zhongwen.com/bushou.htm

The linked website has Chinese characters broken down by its components and covers about 4k characters, which should be more than enough to keep you occupied until you get past the intermediate and reach advanced level. Also, most of them contain example compound words and everything is hyperlinked.

As an alternative, try Heisig. I kind of despise it but a lot of people swear by it. It's a book for Kanji but knowing Kanji is almost knowing traditional set, and there's a separate volume for simplified.

By the way, you seem to be very interested in spelling reforms and making them more "logical" but it just ain't going to happen so just take the languages for what it is. Seriously, the only way to learn the Chinese characters is to stop trying to make sense of them in abstract and keep learning words, and yes words, not individual characters one at a time.