I have got some time on my hands, and the OP's native language (may?) not be English, which would make it harder to understand the dictionary's explanations.
Kanji, many of them borrowed from China (where they are called hanzi), may look complicated to the beginner, who finds no logic or structure in them and may perceive them only as an unsorted collection of different strokes.The task at hand is to classify each of tens of thousands of these symbols, such that they can be down in a dictionary in a certain order - and be found by the reader.
The following first part contains a (not-quite-so) brief overview of some background information necessary to understand the radical methods, the second part contains a tutorial on how to use the New Nelson.
----------------FIRST PART-------------------
Many different methods of doing so have been proposed. The most famous and well-known is the classical radical system, which is closely related to the system used by the
New Nelson. It was created by the authors of the Chinese dictionary
Zihui dictionary (published in 1615). This was later
adopted by Japan with some minor changes.
It is vital we are aware of the fact that this is only one possible classification, introduced for convenient lookup; it is not an inherent property of Chinese characters.
Let me briefly explain this classification scheme. First of all, we need to take a quick look at how new kanji were created in China.
Some kanji are an abstract representation of the things or concept they stand for:
水(meaning "water", it looked like
originally)
魚 ("fish", originally
As you might imagine, it is quite hard to come up with some symbolic represenation for tens of thousands of words, and they would be quite difficult to remember. That is why most kanji are made up of
two parts. Let us take a look at an example:
As we learned above, 魚 means
fish. So, any guesses as to what this could mean? Yes, it means
fish as well. Or rather, a particular kind of fish. It means
carp, a word which sounds like
li in
Chinese (Mandarin). The second part to this kanji is 里, which also sounds like
li in Chinese.
To summarize this, many kanji consists of a
semantic (=indicating meaning) part and a
phonetic (indicating sound/pronunciation) part. The semantic part gives us a rough hint at the kanji's meaning, the phonetic parts how the word it stands for should be pronounced. Sometimes, when there are many kanji with the same sound for the phonetic part, one was chosen which was also related to the meaning.
Please note that the above is an over-simplification, reality is a bit more complex (
kanji != word, sound changes obscure the phonetic, semantic drift the other part &c.), but it should do for now.
Another example:
It means
source of water, spring, sounds like
yuan, and consists of the phonetic 原
yuan and the semantic氵, which is a variant of 水
water.
It is important to note that when combining two kanji into one to form a new kanji, they were sometimes changed or abbreviated (eg for easier writing or visual pleasure).
水 can become 氵 or 冫, 牛 can become 牜, 肉 can look like 月 &c. Luckily, you will not need this for the New Nelson.
These classical radicals are also the first part of the traditional classification of kanji introduced by the Zihui dictionary: As a general rule, each kanji is classified under the meaning-related semantic element, which are called
radicals. (Again, it is a bit more complex than that, but this is the general idea.)
Kanjis with many strokes often consist of several simple (sometimes archaic or obscure) kanji, which we call
elements. A radical is an element that was selected by some person (or dictionary author) for purposes of classification.
So, both 鯉
carp and 鱒
trout would both be classified under 魚
fish. 魚 is itself a radical, so it would be classified under 魚. To identify a complicated kanji such as 朧
hazy (moon), you would first identify its radical as 月
moon.
Now, as you might have noticed, this classification still leaves us with many hundreds or more kanji per radical. Before we can classify them further, we need to take a look at the concept of
stroke ordering. Like radicals, it was introduced for convenience's
(or art's) sake.
Stroke order is a traditional set of rules how kanji should be written with a brush (or pen). It consists of two parts: (1) Identifying how to break up a kanji into
individual strokes; and (2) establishing
an order how they should be drawn (or written).
The second part does not need to concern us, as it is not relevant to the classification of kanji. See the link for more details.
The first part, however, is quite important. After identifying a kanji's radical, we can count the remaining number of strokes, and use this to distinguish kanjis with the same radical from each other. For example, consider 鯉
carp again. It consists of 魚,
which is made up of 11 strokes; and
里 with 7 strokes. Therefore, 鯉 would be classified as 魚-7. By the same logic, 鱒 is classified as 魚-12.
I did not say much about how to identify, or how to count a kanji's stroke. A detailed explanation would be beyond the scope of this post; use the search engine of your choice for further information, and look
here,
here , or
here.
Essentially, it boils down to learning the stroke order and count of the basic elements composing more complex kanji. Note, for example, that 口 consists of *3* strokes - first the left | vertical stroke, then the upper-right edge, and finally the hoizontal __ to the bottom. 人 consists of 2 strokes, the two slanting strokes / and \ . Once you know that, you it should not be hard to conclude that 囚 must consist of 5 strokes [囚(5) = 口(3) + 人(2)]
We are now ready to tackle our original problem, how to look up kanjis in the
New Nelson dictionary.
Nelson's original dictionary
The Original Modern Reader's Japanese-English Character Dictionary, inspired by the traditional system, uses a lookup system that is
http://www.tuttlepublishing.com/books-by-country/the-original-modern-readers-japanese-english-character-dictionary-hardcover-with-jacket said:
Scientifically arranged by a logical extension of the traditional radical system so as to make the finding of a given character almost fool–proof, saving hours of time.
The
New Nelson reverts back to the traditional radical system, but introduces the UniveralRadicalIndex. It allows you to find a kanji given any element it contains, not only the one element called radical that has been selected by the author using some more-or-less intuitive rules. In the traditional system, the radical is (sometimes) the meaning-related semantic element. But say you are looking up 鯉 because you do not know what it means, how could you determine its radical? Nelson solves this by allowing you to look up this kanji as either 魚-7 or 里-11.
Other books and kanji classification schemes that should be mentioned:
-
Hadamitzky&Spahn: Similar to the traditional system, but uses only 79 radicals and a few simple graphical rules to determine a kanji's radical.
-
Halpern He uses a new graphical approach of indexing kanji, called the SKIP (="System of Kanji Indexing by Patterns"). I had no problems finding kanji with this method and definitely reccomend you should give it a try.
-
jisho kanji by radicals: More or less the same as the
New Nelson, but electronically. To find 鯉, open the page in your browser, and select 魚 and 里 from the list.
- Handwriting recognition: Available online here
Jisho.org , also used by many
Denshi Jisho and also available for most major OS's and mobile devices. Free and/or open source engines may require precise knowledge of a kanji's stroke order, commercial engines are (in my experience) more forgiving.
No, it doesn't have to look pretty. The software is only interested in the strokes you are drawing, not the final image (this is not OCR!).
-
Several other methods, such as the
4-corner method, kanji on/kun reading(s),
De Roo Code
---------------SECOND PART-------------------
Now, let's actually take a look at looking up : ) The kanji we want to look up is 鯉.
First of all, let us use the normal radical method. So open your dictionary...
We guess 鯉 is some kind of fish and should therefore be listed under the 魚 radical (there are many fish-name kanji with 魚 the left hand part). 魚 consists of 11 strokes...
From the list of radicals with 11 strokes, we locate 魚
..and note its number 195. The main section of the dictionary shows the current radical number to the top left or right corner, in ascending order. We locate the 魚 section...
Kanji are separated by the number of remaining strokes, in ascending order. 鯉 has got 魚(11) + 里(7) = 18 strokes, 11 of which are for the radical 魚, so that leaves 7 remaining strokes. We go to the section for 7 remaining strokes...
Taking a quick look through the few kanji under this heading, we quickly locate 鯉...
...mission accomplished.
Alright, now let's say we did not know that 魚 was the radical. We can still find this kanji with the URI method. Basically, this method works the same way as what we did just now, except that we use another element as the "radical."
We note that 鯉 consists of both 魚 and 里. We can use either of these two parts to look up the kanji. Let's choose 里.
Taking a look at the first page again, we find that 里 has got the number 166
Next, we go to the URI section of the dictionary and locate the section for
radical 166. 里 has got 7 strokes, so that leaves us with 18-7=11 strokes. We find the sub section for 11 remaining strokes, and quickly find the kanji...
And that's it. As long as you can count strokes and identify at least one element of the kanji, you're free to go. You will need some practice, but finding kanji with this method should not be too hard. It would help to have some knowledge of Japanese and kanji in general, however.
Finally, consider a kanji like 殺 (I won't spoil its meaning, you can look it up yourself). It conists of 几(2) + 木(4) + 又(2) + メ(2) = 10 strokes. The first three elements are radicals. Using the URI, you can look this up under either of these:
几-8 (radical 16, page 1422)
木-6 (radical 75, page 1513)
又-8 (radical 29, page 1442)
These will all tell you that 殺 has got the index number 2994, which is found on page 622.
