If there is a pattern to Chinese characters

Question

Many people say that written Chinese is a difficult language to learn because the characters don't have any relation to the sound of the word, like in English and other languages, even Japanese. But I've read so far that Chinese characters are composed out of basic building blocks. This led to the list of Chinese radicals.

...

In the list there are 214 radicals, with the last one containing 17 strokes. The first few have 1 stroke, then 2, then 3, ... etc..

So you would assume from that that every single Chinese character could be derived from that. Similar to how every English word can be derived from the set of characters in the alphabet.

But right off the bat, I noticed that the 2-stroke characters aren't even derivable from the 1-stroke characters. I would've thought they built on each other. For example, the 2-stroke 人 is supposed to be derived from the 1-stroke 丿. You can see that it is 丿 plus the mirror image. So I guess technically it can be said to be derived from that. But it's not explicit. Then another 2-stroke character is 冂. But this isn't visibly derivable from any of the 1-stroke characters from what I can see: 一, 丨, 丶, 丿, 乙, 亅. Maybe 1, 2, and 5. 丨一亅, but that would be 3.

Another example is from here.

So two concepts combine into a third concept, and the symbols roughly compose into the bigger symbols. But not exactly. The box in the "elder brother" symbol is slightly skewed, seems to use slightly different strokes, the legs are a little more stretched as well, etc. So it's not as if you compose words from letters, and the letters are just shrunken. I'm not very familiar with Chinese, but it seems that the exact position of the strokes is important.

To me this means that there is no exact system for determining the meaning of a character by constructing it out of smaller components. The best you can do is get a "rough idea" of what the symbol might mean from what your guess is at the components.

But I wanted to learn more about this, and confirm/deny that perspective. Wanted to ask if there is in fact a system for learning some basic components in Chinese (basic strokes and shapes), and then combining those into more complex shapes, such that you can easily learn the gist of any new character you encounter. Because from what I've seen, there isn't such a system, and you just need a dictionary to interpret any new symbol you find.

Answer 1

The Chinese characters have not only a pattern, but many, many patterns. But First to clear up some confusion. Radicals are not usually composed of eachother, but are unanalysable. By analogy, a radical is like a morpheme (it's compositional, it's there for a reason semantic or otherwise), and a stroke is like a phoneme (it's just there so that you've got something to put on the page, or perceive when you're reading, and is void of any semantic significance).

But right off the bat, I noticed that the 2-stroke characters aren't even derivable from the 1-stroke characters. I would've thought they built on each other.

They're not characters, they're radicals (radicals are combined to form characters). Like letters of an alphabet, they're not (necessarily) derived from each other. There is a repertoire of 214 radicals (by some analyses sometimes less or more), and these are put together to form a character.

For example, the 2-stroke 人 is supposed to be derived from the 1-stroke 丿. You can see that it is 丿 plus the mirror image.

The right-hand leg of 人 is not the mirror image of 丿. 人 is also not derived from any of the radicals in your list (the left-hand leg is also not the radical 丿). If you look at http://www.archchinese.com/chinese_character_strokes.html you can see numbers 4 and 5, these are the strokes that make up 人.

Then another 2-stroke character is 冂. But this isn't visibly derivable from any of the 1-stroke characters from what I can see: 一, 丨, 丶, 丿, 乙, 亅. Maybe 1, 2, and 5. 丨一亅, but that would be 3.

If you look at that site again, it's number 3 and 10. which make up 冂.

Radicals, being like morphemes, have not only a surface form, but also it usually has a meaning. Observe:

• 女 is a woman.

• 姐 is a sister.

• 她 means "she".

They all have that same component, 女. It's there to provide a clue about the meaning of the character. Similarly, 说 (talk), 话(speech), 语(language) , 读(read), and many many more, all have the same left-hand side, which is derived from the character 言, meaning speech. This is a pattern, the Phono-semantic compounding

A related issue is that maybe a character contains something with is not a semantic clue. It could be a phonetic clue. Some examples:

• 饿 (hungry)

• 鹅 (goose)

• 莪 (some kind of plant ...?)

• 俄 (Russia)

• 我 (me)

They all contain 我. These are all pronounced in very similar ways. (In many dialects, with an initial /ŋ/ and some sort of back vowel or schwa). This is another pattern, a phonetic series.

The box in the "elder brother" symbol is slightly skewed, seems to use slightly different strokes, the legs are a little more stretched as well, etc

Earlier I introduced the phoneme/morpheme analogy, and I did that to explain this phenomenon. Just like the same English morpheme may be spelled "in-" or "im-", depending of its phonemic context, that box 口 (not really a mouth) might get munched up a bit. It is just reacting to its environment, just like that English prefix did.

To me this means that there is no exact system for determining the meaning of a character by constructing it out of smaller components. The best you can do is get a "rough idea" of what the symbol might mean from what your guess is at the components.

That's right, just like how you cannot know exactly how to write a word from just knowing how to pronounce it, you always don't know how to pronounce or interpret a word from its spelling. Not really any different from English in that sense.

your examples 己, 已, 巳

They are not related, they just look a bit like each other. Like the English words "rite", "rota", and "Rita" they actually have nothing to do with each other. Same deal with 土 and 士, or "interior" and "inferior". Looks like a pattern but is not one. The exactness does matter yes, but just like most human communication, there's enough redundancy that many actual mistakes are just subconsciously corrected by the receiving end without noticing.

Wanted to ask if there is in fact a system for learning some basic components in Chinese (basic strokes and shapes), and then combining those into more complex shapes, such that you can easily learn the gist of any new character you encounter.

There is no system, but many patterns. So people who know Chinese better than me, if they encounter a character they do not know, but they do know the word that character is depicting, then they will often by hunches and educated guesses read it correctly.

Because from what I've seen, there isn't such a system, and you just need a dictionary to interpret any new symbol you find.

To read and write Chinese needs a steep learning curve, for the reasons you mentioned. But if you know Chinese well, then it's not really a problem in practise.

Answer 2

To understand how Chinese characters work, we need knowledge of the following concepts:

• Morphemes have properties of both meaning and sound
• Chinese morphemes are overwhelmingly composed of a single syllable (sound). Each Chinese character represents a morpheme, and thus simultaneously represents a syllable and meaning
• The purpose of Chinese characters is like the purpose of any other writing system; that is, to remind the reader of the morphemes (or words) used in the text that is being read
• If the Chinese character is decomposable, the purpose of the components is to remind the reader of what morpheme is being referred to. As with @melissa_boiko's comment, the sole purpose of components are to hint at the sound, meaning, or both the sound and meaning of the character (morpheme)
• If the character is not decomposable, then it is either some kind of pictogram or ideogram, hinting purely at the meaning of the morpheme, or a rebus loan, hinting purely at the sound of the morpheme
• About compositional logic of characters:
  • Characters are not made of strokes, just like the letter E does not contain the letter F
  • Characters are not made from radicals, just like English words are not functionally made from letters, and morphemes are not made from a sequence of IPA symbols;
  • Morphemes are not to be decomposed; character decomposition only results in components which hint at the morpheme (by sound and/or meaning), not "components of the morpheme" (which is an undefined concept)
  • The logic of Chinese character structure is only fully given by the character's usage around the time of its inception. The older the character, the more likely it suffers structural shape corruption and stroke stylisation. Thus, analysing characters in their modern regular script form is a conceptual mistake
    • This is exactly like trying to decompose English words. The older the English word, the less likely a naive decomposition from its sequence of letters will result in a correct etymology; the newer the English word (e.g. scientific vocabulary coined using Latin), the easier it is to see its root morphemes.

---

Character evolution tables are often used to illustrate how characters have changed over the years. Here are two examples:

Character evolution table for the character「若」(see http://xiaoxue.iis.sinica.edu.tw/yanbian?kaiOrder=1335). The morpheme that this character originally referred to is /*nak/ (submission/obdience). /*nak/ is cognate to /*na/ (如, to comply/agree with) and /*nˤak/ (諾, to agree).

The academically accepted description of this character is one of a kneeling person with unkempt hair and hands raised in submission. This character is a pictogram and is not actually decomposable at all, but a sequence of shape changes across time and geographical space, unrelated to linguistic function, has made the character appear in the modern form as if it was composed from「艹」(/*[tsʰ]ˤuʔ/, grass) and「右」(/*[ɢ]ʷəʔ/, right hand/direction).

These components are not related to the morpheme /*nak/ in either sound or meaning, and hence are a result of character shape corruption.

---

Character evolution table for the character「喪」(/*s-mˤaŋ/, to mourn).「喪」was originally composed of「桑」(/*[s]ˤaŋ/, mulberry tree) and multiple「口」(/*kʰˤ(r)oʔ/, mouth).

This example highlights the functional analogy between Chinese characters and Egyptian Hieroglyphs, and typifies the majority of Chinese characters' function.「桑」acts as a phonetic rebus, that is, it is only used for its phonetic value, and「口」acts as a determinative, that is, only used for its semantic value (here, indicating crying and wailing for the deceased at a funeral > mourning).

Modern Chinese characters are heavily constrained by "accepted strokes and shapes fit into an imaginary square", so it is very difficult, if not impossible, to see where「桑」is located in「喪」. For reference, the character evolution table for「桑」:

---

To correct some aspects of the question and the other answer (especially its emphasis on radicals):

1. Chinese character strokes are an artefact of script regularisation, that is, the gradual transformation/stylisation of writing from oracle bone inscription shapes to regular script (for examples, see the character evolution tables above). Like most (all?) other writing systems in the world, regular script strokes are not, and do not represent, a functional unit of any Chinese language. Before brush calligraphy was invented, strokes weren't even defined, and the concept of a stroke is not applicable to inscribed characters on hard materials like bones and bronzes.

2. Chinese characters are not composed of radicals. Radicals are dictionary section headers, and function exactly like the first letter of an English word in the context of a dictionary. Please also note that the concept of radicals first appeared over 1,000 years after the first oracle bone characters were attested - radicals are not morphemes!

   In choosing the system of radicals in a dictionary, the dictionary compiler only needs to be concerned with whether the radical appears as a graphical component sufficiently frequently in characters. Characters grouped under seemingly nonsense radicals like the vertical line 丨 could not be reasonably grouped under another radical, and「丨」does not contribute function to any of the characters listed under it apart from being a stroke in those characters.

   Chinese character radicals are thus arbitrarily decided by the compiler of the dictionary. The 214 radicals referred to in the other answer is strictly from the Kangxi Dictionary and its derivatives. Shuowen Jiezi, for example, has a system of 540 radicals.

   

   The above is an example of the functional analogy between radicals and English first letters.

   • 「重」(Baxter-Sagart OC: /*N-toŋʔ/, heavy) and「野」(/*lAʔ/, town outskirts > wild, wilderness) are both organised under the radical「里」(/*(mə.)rəʔ/, village), because the bottom of「重」and the left hand side of「野」looks as if「里」is present.
     • In「野」(town outskirts),「里」(village, or rather its constitutive components 田 field and 土 earth) acts as a semantic determinative;
     • 「重」does not functionally contain「里」at all.「重」(/*N-toŋʔ-s/; heavy) was originally a combination of semantic「人・亻」(person) and simultaneously semantic and phonetic「東」(/*tˤoŋ/; picture of a bag), depicting a person carrying a heavy bag, indicating the meaning heavy.「土」(earth) was added later and the three components「人・亻」,「東」, and「土」were fused together very early onwards.
   • Similarly, the English words are all categorised under the heading "A", but while words like about and above are actually composed of the morpheme a-, words like aeroplane are not. These words are all grouped under "A" because "A" seems to appear in these words, regardless of whether "A" is actually a morpheme or whether it appeared "accidentally" through centuries of orthographic shift.
   • For the other two examples, the radical of「若」is「艹」(non-functional), and the radical of「喪」is「口」(semantic determinative). Radicals frequently but unreliably give sound or meaning hints to the morpheme of the character.

---

So, is there a pattern to Chinese characters? Yes indeed!

• As @melissa_boiko mentioned, many of these morpheme and syllable hints appear repeatedly as part of other characters. However, they can only be picked up through repeated exposure.
• Chinese is a natural language which co-evolved with its script for over 3,000 years, and thus do not expect to be able to derive the morpheme that a character represents, or the character representation of a morpheme, from first principles - as @curiousdannii mentions, don't fall for the etymological fallacy.
• Chinese paleography and linguistics are growing fields. Many of the explanations that you see in even authoritative resources will be incorrect or outdated. In this light, as a learner, it may not be worth pursuing the correctness of a character explanation if it's too difficult to find.

---

References:

• Baxter, William H.; Sagart, Laurent (2014) Old Chinese: A New Reconstruction
• Schuessler, Axel (2007). ABC Etymological Dictionary of Old Chinese
• 季旭昇《說文新證》