I mean, are there any more strange shapes in the characters, as (smallest) components, that are not listed among the 214 Kangxi radicals?
Yes, a lot. Radicals are not the building blocks of characters, really, they are a set of components used to sort characters in traditional dictionaries (those that arenāt sorted by pronunciation, which is normal today). If you want to refer to parts of characters that recur in other characters, just call them character components!
So, the 214 radicals you refer to is a rather arbitrary subset of components. There are many hundreds of common components, but itās not possible to have a definitive list as it depends on how you define your terms.
Of course, you could reduce all characters to smaller and smaller components and then into strokes, and that way you could say that everything is covered by the radicals, but this truly meaningless and misses the point entirely. It requires you to break down units that should not be broken down.
For example, take a character like ä». The radical here is äŗ»(äŗŗ), which leaves ä¹. Both äŗ» and ä¹ are functional components (äŗ» provides meaning, ä¹ is a sound component, even if thatās not obvious in Mandarin). ä¹ is not on the list of 214 radicals, so thatās an example of a common component that is not in itself a radical. If you look it up, youāll see that the radical of ä¹ is ä¹, but now youāre breaking down something that really shouldnāt be broken down, as thatās certainly not a functional component. Also, what do you get from ä¹ after removing ä¹? What function does it have?
Thereās a very large number of such examples where you could break it down, but really shouldnāt. Remember, with traditional dictionaries, they needed a way to sort characters, so all characters needed to have a radical to sort them by, but this does not mean that the radical has anything to do with the actual composition, structure or function of the character.
If youāre using Skritter, check out our character course! It goes through all the basics of characters in sixteen video episodes with associated characters (about 160 in total).
3 Likes
Thank you for the careful explanation! I am very close to understanding this. But please finish your ārhetoricalā question with respect to the example! 
(even if I should not do it). What do I get after removing ä¹? You mean, there is no such component at all? Which means that ä¹ is an empty component and the remaining part is something that is not even an empty component? Or if I had a good imagination, I would say it consists of, e.g. å and äøæ, which now can be considered as āproperā empty components?
Thank you, I have already gone through all the 16 videos several times, and each time I understand something new that escaped my attention before. And still, I have doubts about certain things!
Finally, all the things above mean to me that: although there is some logic in the construction of Chinese characters, the logic is limited, and many of the characters (or components of it) just have to be learned in the way they are without looking for some didactic decomposition of them. Am I close?
One more thing: You have mentioned at some place, that the radicals in the characters do not even stick to the semantic component, although it is typical. Sometimes, (in rare cases), it is the phonetic component, and in your example above with ä¹, it is neither of them? (The āMakerā just found some part of that doodling that resembles to one of the radicals already in an existing list and just defined that as the radical of that character (irrespective of its role in the character)ā¦)
I think the easiest way to think about it is to consider how characters developed. When new characters were created, people had something in mind. Sometimes, they wrote something that resembled their idea of a real-world object (pictographs), but beyond basic characters like that, they combined characters already in existence. They did so with some thought, so each component was included for a reason.
What Iām saying is that it makes sense to break characters down into those functional components (everybody agrees on that, I think), but it doesnāt make sense to break down components if they arenāt themselves also compounds. ä¹ is not a compound, and neither is äŗŗ, so breaking these characters down into smaller parts (strokes) doesnāt tell us anything about their origin or function. I would argue that itās not helpful for remembering them either. Itās a bit like trying to make sense of the word āhelloā by breaking it down into āı n c - l l c āā. Can you do it? Yes! Is it meaningful? No.
Then there are characters like é±¼ and č½, which actually arenāt compounds even though they look like compounds if you donāt know anything about their origins (youāll be familiar with these two examples from the course). In these cases, I think itās helpful to think of them as compounds because they are so complex and the components do recur in other characters and are rather common in themselves.
Sometimes, (in rare cases), it is the phonetic component, and in your example above with ä¹, it is neither of them?
ä¹ is not a compound, so talking about functional components is not meaningful. The radical is an arbitrarily chosen stroke thatās used to index the character, thatās all.
ā
So, generally speaking, never break things down to individual strokes, unless that stroke really means something or indicates something (the top line in 天, the bottom line in ę¬). Do break compounds down (rather obvious). What you want to do with characters that look like compounds but actually arenāt is up to you, but I tend to use the empty components of characters like č½ to make them easier to remember, while keeping in mind that the character has nothing to do with āmoonā, ādagger/spoonā and āprivateā.
1 Like
As Olle has said, radicals are not the building blocks of characters, really, they are a set of components used to sort characters in dictionaries. What I want to add is that they are designed to help one to find out a character in the dictionary when one knows neither the reading nor the meaning of the character. Just as you have deduced, one searches a character by matching some part of that doodling that resembles to one of the radicals already in an existing list. The radicals do not always relate to the semantic nor phonetic components, if they have to, it would hinder the search because neither of these information is known beforehand.
Throughout history many radicals systems have been designed. The earliest one from čŖŖęč§£å has 540 radicals. The current one used in Mainland China for the simplified characters has 201 radicals. So one character can belong to different radicals under different systems.
It is crystal clear now! Thank you Olle and pts! I will try to use some traditional dictionary just for practice (I do not know the reading or meaning of most of the characters )
1 Like
And to add to this, which further adds to the confusion, the list of radicals (any list actually) is a list of components that can be the radical in other characters, it doesnāt mean that they always are. I think this is where much of the confusion comes from. Beginners look at a list of radicals, see ę°µ(ę°“) and ęØ, and think that ę· is composed of three radicals. Iāve yet to come up with a neat and intuitive way of describing how all this works that is suitable for beginners and isnāt too long.
Yes, you have also mentioned that in one of your article on the character course forum page. I think you have explained it there quite clearly! Thanks! It helped me a lot!
1 Like