Kanji (漢字)

This entry will go into full detail about everything VCE and Japanese-Language Proficiency Test (JLPT) students must know about kanji. The requirements for kanji knowledge are much greater for JLPT students than they are for VCE students, so I’ll make a note when appropriate as to when VCE students should stop reading if they do not wish to gain additional knowledge about kanji. JLPT students must read the whole thing. 😛

 

The Definition of "Kanji":

 

The first aspect of kanji we should discuss is what does "漢字" actually mean? If, on your Japanese 2nd Language VCE exam, you had a question worth 1 mark asking: "Translate 漢字 into English", how would you translate it (I should note firstly to VCE students that you will never get a question like this on the exam :P)?  The first possible answer that a student might write is "kanji" (one of my classmates did this too in year 11 Japanese). This is not a translation, but a transliteration of 漢字 and as such would not score you any marks.  The next possible response, and perhaps the most common chosen, is "Chinese characters". This is generally correct and if the question was worth 2 marks instead of 1, you would get 1 mark for it. While kanji mostly refers to Chinese characters, there are a number of cases where there are kanji which do not – and never did – exist in China. These kanji are known either as 国字 (こくじ) "national characters" or 和製漢字 (わせいかんじ) "Chinese characters made in Japan". These sorts of kanji will be discussed in further detail later. The literal translation of 漢字 may be chosen by some students, thus their answer would be "Han Dynasty characters". This is also not quite correct for the reasons just discussed. The most correct translation – and one that students would probably never think of – is "Sino-Japanese characters". Throughout this entry, "Chinese characters" will refer to those characters used in Mandarin Chinese without any consideration to their application and use in Japan, while "kanji" will refer to the Sino-Japanese characters.

 

The Chinese character is an ideograph – a symbol which by itself has a meaning and, when placed with other characters, contributes to the overall meaning of the word formed by the multiple characters. A person who knows the individual characters of a phrase but does not know the actual phrase itself could guess the meaning by analysing the individual meaning of each character. Because the vast majority of kanji are Chinese characters adapted for use in Japan, this idea holds true for kanji as well.

 

A Brief History about Kanji:

 

It is not really necessary to know where kanji comes from when you are learning Japanese as a 2nd language. However, as many teachers will tell you, having knowledge about a topic’s history can help in one’s understanding of the topic. Chinese characters were brought to Japan in ancient times (the exact date is unknown but there was a discovery of a gold seal given to Japan by the emperor of the Eastern Han Dynasty in 57AD). Before they were introduced, Japan had no form of writing system and their communication was either done orally or through drawings, etc. As such, Japanese was initially written completely in kanji. Both hiragana and katakana were developed from kanji by an alternative cursive style to the norm and by kanji simplification by monastery students respectively. Prior to WWII, all kanji were either identical to or derived from traditional Chinese characters. During the simplification of Chinese characters in the 1950s, Japan also replaced a number of their kanji with the new simplified Chinese characters – as well as simplifying some kanji independently from China (for example: 気 – discussed in more detail later). There have been numerous attempts to make kanji obsolete in Japan, however all have failed. It has, however, been successful in a number of countries such as North Korea, South Korea, and Vietnam. Today, Japan is the only country in the world to have its own unique writing system intertwined with Chinese characters.

 

Reading an Individual Kanji:

 

To a Japanese 2nd language student whose primary language is English, each kanji will have at least 2 of the 3 following ways of reading a kanji. These ways of reading a kanji are known as "readings". They are:

 

• A translation of a kanji’s core meaning in English. Each kanji could have one or more translations, so the translation you choose to apply will depend on the context.
• An 音読み (おんよみ) "on-reading".
• A 訓読み (くんよみ) "kun-reading".

There is only one kanji which does not have a single reading. It will be discussed in more detail later. Let’s go over the three types of readings in detail:

Translation of a Kanji’s Core Meaning in English:

This reading is not used much when actually reading kanji in books, etc. It’s just there to help the student understand what the kanji itself actually means. Because the languages of English and Japanese have significant differences (apart from the obvious ones), it can be quite difficult to give a single translation to the character. The varying contextual uses of kanji also contribute to this difficulty, which is why kanji can often have a number of definitions in English. For example, the kanji 日 has the following translations in English: "day", "sun", "Sunday". In most cases, because of kanji’s derivation from Chinese characters, the English translation of a kanji will be the same if it was read as a Chinese character. For example: If 花 was read as a kanji, it would mean "flower", and if it was read as a Chinese character, it would also mean "flower". However, there are a number of cases where this is not so. One example is with 私. If that character was read as a kanji it would mean either "private" or "I, myself" depending on the context. If it was read as a Chinese character, it would only mean "private". This is probably due to a development of a kanji’s meaning independent from China. In Japan, these characters – which have different meanings in Japanese to their Chinese meanings – are called 国訓 (こっくん). Because of this change in meaning independent from China, a lot of VCE Japanese 2nd language students who know Mandarin Chinese often make mistakes when translating kanji for the purposes of kanji tests, etc. 

The On-Reading:

The on-reading is the Chinese-derived pronounciation of a kanji. It is not the "Chinese pronounciation" of a kanji as I have sometimes heard it being described. When the Chinese characters were first brought to Japan, the Japanese converted the Chinese pronounciations of those characters to make them sound Japanese. Because some Chinese characters were introduced to Japan during a number of different dynasties, and some of the characters’ Chinese pronounciations changed as the dynasties changed, a number of kanji have multiple on-readings. Each one reflects a converted pronounciation peculiar to a particular dynasty. When writing the on-reading above an individual kanji to aid in pronounciation, it is written in katakana. When writing the on-readings above a kanji compound (i.e. a Japanese word made completely from kanji), it is written in hiragana. In addition, if a single kanji, when read by its on-reading, forms a word, its pronounciation should also be written in hiragana. So, for example, if you were writing the reading for 電, you would write デン. However, if you were to write the reading for 電話, you would write でんわ. In addition, the on-reading for the kanji 天 is written as テン but if the word 天 ("sky") was being referred to, it would be written as てん, not テン. In other words, if you are writing the on-reading down without any consideration as to whether or not the on-reading is actually a word, you would write it in katakana, otherwise it is written in hiragana. Students who are learning kanji from the Kookoo Seikatsu Kanji Workbook will note that a character’s on-reading is written in bold hiragana. This is actually the incorrect method of writing it and is probably used as a matter of consistency so that students do not get confused between on-readings of an individual kanji and the on-readings of a kanji compound.

As anyone who knows both Chinese and Japanese can tell you, the pronounciation of a modern-day Chinese character and a kanji’s on-reading are very different. For example: The on-readings of 木 are モク and ボク, but the pinyin (the Mandarin Chinese word for the Romanised pronounciation of a Mandarin Chinese character) for it is mù. Usually, a Chinese character only has a single pinyin reading while a kanji has more than one on-reading. However, there are a number of cases where a Chinese character can have multiple pinyin readings as well.

The Kun-Reading:

The kun-reading of a kanji is the word in Japanese which has the same meaning as the kanji itself. As stated earlier, the Japanese at one stage did not have a writing system. So when the Chinese characters were introduced, the Japanese associated their own words to the characters as well as converting the Chinese pronounciations to sound Japanese. For example, the character 行 means "to go". The Japanese word for this is いく so the Japanese then associated 行 with いく to make 行く. All verbs, adjectives and nouns follow this association. The word itself has a meaning and the kanji which are associated to the word have the same – or at least very similar – meaning as the word.  By the time literature became prominent in Japan, kanji were used in order for the reader to gain a better understanding of the writer’s views and feelings. Today, it is still used for this purpose as well as a means of significantly reducing the number of characters used (words written completely in hiragana typically have more characters than those written either completely in kanji or a combination of kanji and hiragana). Kun-readings are generally verbs, such as 行く, however they can also be adjectives ( 新しい) and nouns (水). It should be noted though that kanji are never used to write particles. When writing the kun-reading above a kanji to aid in pronounciation, it is always written in hiragana.

It is important to note here that unlike on-readings, some of the kun-readings in some kanji are not completely replaced by the kanji. As you can see with 新しい, the "しい" is not replaced by the kanji 新. The reason for this is generally to be able to identify what sort of conjugation is being used with the verb/adjective (eg: past tense, conditional ば form, etc.). Different dictionaries will have different methods of informing their readers which hiragana of a kun-reading are replaced by the kanji, and which hiragana are not so replaced. The Kookoo Seikatsu Kanji Workbook does this by placing the portion of hiragana that is to be replaced by the kanji in brackets. So in the case of 食べる, the kun-reading is written as (た)べる. In electronic dictionaries, a ・is placed somewhere in the kun-reading. All the hiragana before this ・ are replaced by the kanji. So, considering 長い, the kun-reading in an electronic dictionary would be read as なが・い.

Hiragana Associated with Kanji:

Hiragana is used in conjunction with kanji in two different ways, and it is important that VCE students know these different ways as well as what they are called in Japanese. Firstly, except in the case of single kanji, the pronounciations of kanji written above the character are in hiragana. This hiragana above the kanji is called furigana. Secondly, the hiragana which appears to the right of kanji in verbs and adjectives is called okurigana. So, for example, the verb "to eat" in Japanese is 食べる. 食 is the kanji while べる is the okurigana. Also, in the adjective 新しい ("new"), 新 is the kanji while しい is the okurigana.

Kanji and Japanese Synonyms:

Many students who are just starting to learn kanji may be overwhelmed with the concept of these characters (like was when I first started learning kanji). Students may not understand that the meaning of the character is just as important as the word that it is associated to. So, for example the i-adjective "あつい" ("hot") is written in the following way with kanji: 暑い. There is another i-adjective, also pronounced "あつい", that means "thick". A student who doesn’t understand the concept of the meaning of the kanji character may be inclined to also write "thick" as "暑い" when it in fact should be written as "厚い". In addition, the "い" in いきます is replaced by the kanji 行 to make 行きます when writing the phrase "to go" with kanji. If a student does not know the true meaning of 行 and believes kanji are used only as a means of "shorthand" or similar for a particular pronounciation, they may be inclined to write the verb "います" ("to exist (animate object)") as "行ます", when instead it should be "居ます". While the same hiragana can be used to write synonyms in Japanese, kanji cannot. Each kanji represents a particular idea or meaning. While this idea can sometimes be quite broad, it is never so broad to be able to say that "hot is very similar to thick" when in fact the two adjectives have little correlation with each other. So remember that while you can use the same hiragana to write more than one word with the same pronounciation, in the vast majority of cases – if you intend to write in kanji – you must use different kanji. 

---

VCE students do not have to read any further, as the above information is all that they need to know about kanji in VCE. The remaining part of this entry will focus on kanji knowledge required for all levels of the JLPT. VCE students who are unaware of what the JLPT actually is should go to http://en.wikipedia.org/wiki/JLPT for more information. VCE students can sign up for the JLPT at the Japan Seminar House – the official host of the JLPT in Victoria. Their website is: http://www.japansh.jp/jlpt.html. 

 

Reading an Individual Kanji (continued):   

Kanji Having Varying Numbers of On and Kun-Readings:

Throughout your study of kanji, you will notice that not every kanji will have one on-reading and one kun-reading. The reason why a kanji would have more than one on-reading has already been discussed in The on-reading section. Some kanji may have more than one kun-reading because the Japanese spoken language, before kanji were introduced, had numerous ways of representing a particular idea depending on the context. Kanji which only have on-readings had no Japanese equivalent when they were introduced, so the Japanese had to adopt the unknown idea into their own language. There are also some kanji, as stated earlier, which were "made in Japan". These characters are made to be "very similar" to kanji which came from China. They are made to "look like" Chinese characters, they have a core meaning, a kun-reading and some also have on-readings. If a kanji only has a kun-reading, then it is a wasei kanji and it does not exist in China. If a wasei kanji has an on-reading, then it is one of the few kanji which previously did not exist in China but has now been adapted into the Mandarin Chinese language. The only wasei kanji that I know of that has both an on-reading and a kun-reading is 働 (ドウ, はたらく). Please post as a comment any other wasei kanji that you know which is like 働!

The following are some wasei kanji which only have kun-readings:

• 峠 (とうげ) = Mountain pass
• 畑 (はたけ) = Field of crops
• 辻 (つじ)  = Crossroads

The following are some kanji which only have on-readings:

• 漢 (カン) = The Han Dynasty
• 復 (フク) = Return, restoration, repeat
• 茶 (チャ, サ) = Tea

The Kanji Repeat Sign:

The kanji repeat sign (々) is a very odd kanji. It does not have an English translation, it does not have an on-reading and it does not have a kun-reading. As such, it is not possible to actually read this kanji by itself. It is because of this trait that it is not in any standard kanji dictionary. The kanji repeat sign always comes after another kanji and usually gives a plural meaning to the kanji that comes before it. For example: 山 means "mountain" while 山々 (やまやま) means "mountains". As you may be able to guess from the example, reading the kanji repeat sign with another kanji involves repeating the reading used for the first kanji. However this is not strictly true in every case. In many cases with the kanji repeat sign, the kanji repeat sign is read the same as the kanji before it except a ゛ (tenten) or ゜ (maru) is added. So for example, 花々 ("flowers") is read as はなばな, not はなはな. Because each unknown phrase with the kanji repeat sign can potentially have tenten or maru markings, it is better to use a kanji dictionary to look up the phrase rather than trying to guess the pronounciation of it (see below).

The Small ケ (ヶ):

The ヶ is technically not a kanji, but is present in quite a number of kanji compounds. At first glance, a student may guess that the ヶ character is pronounced in the same way that the katakana character ケ is pronounced, which is "ke". However this is incorrect. The pronounciation of ヶ will differ depending on the context in which it is used. The ヶ character is mainly used in phrases which indicate some sort of quantity, or in the names of places. In the case of phrases which indicate quantity, the ヶ is pronounced か. For example: 六ヶ月 (ろっかげつ) "six months" In terms of the names of places, the ヶ is pronounced が. For example: 霞ヶ関 (かすみがせき).

Pinyin:

Most electronic dictionaries these days, in addition to stating the meaning, on-reading, and kun-reading, will also state the pinyin (Mandarin Chinese pronounciation) for the character. For example, if 飲 is looked up in an electronic Japanese dictionary, its pinyin will be listed as both "yǐn" and "yìn". This information is not required for the JLPT nor is it required for Japanese students in general, but is added to this entry as a matter of completeness. However, this feature can help students who know Japanese but are studying Mandarin determine the Mandarin pronounciations for kanji they already know. Of course, due to the fact that one particular character can potentially have a number of different uses in Japan compared to China and vice versa, you should always consult your Mandarin teacher to ensure that you are not using a character in the incorrect context from a Chinese perspective.

Reading Multiple Kanji:

Phrases in Japanese which are formed by multiple kanji rather than using hiragana form the bulk of a Japanese 2nd language student’s vocabulary. These phrases are known as 熟語 (じゅくご). For example phrases like 先生 ("teacher"), 病院 ("hospital"), and 会社社長 ("company president") are all jukugo. Even if you have learnt phrases completely in hiragana, it’s quite likely that there are kanji for those phrases but you just haven’t learnt the kanji yet. It is common for students to come across jukugo in which they know the individual characters, but not the phrase itself. There are two ways a student could proceed to look up the meaning of the phrase: look up the first kanji of the phrase in a kanji dictionary and then look for the appropriate jukugo (note that not every kanji dictionary will have relevant jukugo listed in the kanji’s entry), or determine the pronounciation of the phrase by trial-and-error and looking up each attempt in a standard Japanese dictionary. The latter method will be explored here by detailing how to read unknown jukugo where the individual kanji are known. Although, as you will see, this latter method has a number of flaws and most of the time is unreliable. I do recommend that students first refer to a kanji dictionary rather than trying to guess the pronounciation if they do not know a particular jukugo.

Usually, when reading a single kanji on its own, or reading the kanji of verbs and i-adjectives, the kun-reading should be used. For example, 水 would be read as みず and not すい, 悲しい would be read as かなしい and not ひしい, and 出会う would be read as であう and not しゅつかいう. There are cases, however, where a kanji by itself would be read with its on-reading rather than its kun-reading. Some examples of this are: 天 (てん) "sky" (stated earlier), 元 (げん) "element, trace", 本 (ほん) "book". In these cases, it can be very tricky to work out which reading to use – especially when the on-reading has a meaning distinctly different from the kun-reading. In the case of 本, the on-reading means "book" as previously stated, but the kun-reading – which is もと – means "origin". This uncertainty of which reading to use can be quite frustrating. Usually, the context of the sentence will dictate which reading to choose. For example, consider the sentence: 僕は新しい本を買った. The reading for 本 that one would choose is ほん rather than もと. Since "I bought a new book" makes more sense than "I bought a new origin". Kanji which have okurigana after it are never read by their on-readings. Students rarely have a problem reading single kanji or kanji associated with verbs and adjectives, since if they actually know the kanji then they would also know the meaning of the readings and where to use them.

Reading more than one kanji is often where problems arise. In the majority of cases, a jukugo containing more than one kanji should be read by using each individual kanji’s on-reading. Words like 昼食 (ちゅうしょく) "lunch", 救急車 (きゅうきゅうしゃ) "ambulance", and 社会 (しゃかい) "society" are all read with their on-readings. However there are cases in which the on-reading of the kanji is changed slightly when read in a compound. Take 出 (シュツ, でる, だす, で) for example. The phrase 輸出 (ゆしゅつ) "export" is read entirely with on-readings, which is fine. However, in the case of 出発 (しゅっぱつ) "departure", the つ has changed to っ. "シュッ" is not listed as an on-reading for 出 in any kanji dictionary. Therefore, it would not be possible to find 出発 in a regular dictionary if you guessed the reading was "しゅつぱつ". In addition, there are jukugo in which the on-reading of one or more kanji has been changed significantly and only resembles the original on-reading. An example of this is 学校 "school". The on-reading of 学 is カク or ガク and the on-reading of 校 is コウ. However 学校 is read as "がっこう", not "がくこう". In these cases, the only solution is to look up the first kanji in a kanji dictionary and find the relevant jukugo there. 

There are also quite a number of cases where some of the individual kanji in a jukugo should be read with on-readings and the rest read with kun-readings. In addition, there are some jukugo in which each kanji should be read only by its kun-reading (this is particularly prominent in Japanese names). Take 金色 "gold colour" as an example. The first kanji 金 is read with its on-reading, キン, while the second kanji is read with its kun-reading, いろ to form "きんいろ". Another example is with the phrase 一言 "a word". Both kanji are read with their kun-readings ひと and こと respectively to form ひとこと. Again, the easiest way to determine the reading of the jukugo would be to find the jukugo in the kanji dictionary rather than trying to guess the pronounciation.

When reading the kanji of Japanese names, in most cases only the kun-readings are used. So you can be safe in assuming that by using the kun-readings of the kanji in a Japanese name, you would be pronouncing it correctly. Some examples of Japanese names are: 小松 (こまつ), 山中 (やまなか) and 田中 (たなか). All of the kanji in these names are read by their kun-readings. 

Classifications of Kanji:

Experts estimate that there are around 50,000 kanji in existence (and even more Chinese characters, which are not used in Japan). The average native Japanese person, however, only knows between 2,500 and 3,000 kanji. The government has classified the thousands of commonly used kanji into certain groups which relate to their use in daily life. It is important that 2nd language students become familiar with these classifications in order to understand how Japan has categorised their kanji.

The Education Kanji:

The education kanji (or 教育漢字 (きょういくかんじ)) are the 1006 kanji that all Japanese children must learn during their first six years at school. Most kanji dictionaries actually state what year a particular kanji is learnt by the children during their first six years at school. For example, the kanji dictionary states that 言 is learnt in a student’s 2nd year of school (written in some Japanese kanji dictionaries as: "常: ２年" or in some English kanji dictionaries: "Jōyō-2"). Generally, a strong level 2 JLPT candidate would know the entire education kanji by the time they take the test, or an average level 1 student would know the education kanji by the time the sit for the examination.

The Jōyō Kanji:

The jōyō kanji are the 1,945 kanji which the government designated in 1981 to be the only set of kanji which could appear in Japanese publications (excluding novels, biographies, etc.) without aids for pronounciation. Most magazines, newspapers and the like adhere to these regulations and only use hiragana for other not-included words or use furigana above the kanji not in the jōyō list. The jōyō kanji includes all of the current education kanji plus an extra 939 kanji which are learnt during years 7 to 9 of school. Japanese dictionaries omit what year kanji are learnt in if they are learnt in year 7 or above, while in some English kanji dictionaries "Jōyō" is simply stated with the year omitted. All level 1 JLPT candidates would have learnt the entire set of jōyō kanji as well as some others by the time they sit for the exam.

Kanji for Names:

The kanji for names (人名用漢字 (じんめいようかんじ)) are the 2,928 kanji which include both the jōyō kanji as well as 983 other kanji which are used especially in Japanese people’s names. In Japan, the government permits parents of newborn children to use both jōyō kanji and the extra 983 kanji to create a name for their child. Sometimes the phrase "jinmeiyō kanji" refers to the entire 2,928 set of characters, or to just the 983 kanji used especially for names. Level 1 JLPT students would probably only know a few of the extra 938 kanji.

Simplified and Traditional Chinese Characters in Kanji:

Since kanji originally came from China, all of Japan’s kanji (except for the wasei kanji) are either traditional or simplified Chinese characters. It is not necessary for the JLPT to know which kanji is wasei, traditional, or simplified (since you’re not likely to know unless you know Mandarin or you have a Chinese friend who can tell you), but as a matter of completeness information about it is included here. In ancient times, the kanji which were brought to Japan were all traditional, and today most kanji are still traditional Chinese characters. However, during the early post-WWII era where China simplified their characters, Japan followed suit and simplified a significant number of characters according to how China simplified theirs. Included in this portion of simplified kanji are the kanji that were simplified differently to China’s methods. One such example is the kanji 気. In traditional Chinese this character is 氣, and in simplified Chinese the character is simplified to 气. Neither Chinese character sets have the character 気. In Japan, the traditional kanji 氣 was simplified to 気 rather than 气, and so 気 does not exist in China. Although today no official figure is given as to what percentage of kanji are traditional and what percentage are simplified, my Chinese friend and I observed a large number of kanji and have come to the conclusion that around 70% of kanji are traditional (this includes the wasei kanji) and the remaining 30% are simplified (includes the kanji simplified separately from China). This figure, of course, is just our estimate and should not be taken to be any sort of "official figure" in essays, reports, etc. about kanji that you may write. 

Some kanji which are also traditional Chinese characters include:

• 開 (ガイ, あく, あける, ひらく, ひらける) "to open"
• 買 (バイ, メ, かう) "to buy"
• 為 (イ, ため, なす) "do, sake"

Some kanji which are also simplified Chinese characters include:

• 五 (ゴ, いつつ) "five"
• 虫 (チュウ, むし) "insect"
• 国 (コク, くに) "country"

Kanji Lookup Methods:

There are quite a number of ways one can look up kanji in a kanji dictionary. These different ways are called "lookup methods" and JLPT students must know – at the very least – what each lookup method means and be able to use one method efficiently and effectively. JLPT students would not be expected to know how to use all the methods to look up kanji. This section will go into detail to the extent that it explains what each method "actually is", but will not explain how to use those methods (save for the lookup method "kanji by parts"). The efficiency and effectiveness of each method will also be evaluated. For detailed explanations as to how to use each method, your paper kanji dictionary should have appendices containing such information or alternatively you can search on the web for tutorials.

The On/Kun-Reading Lookup Method:

This is the easiest of all lookup methods. In terms of paper kanji dictionaries, it involves looking in an appendix list of every single on/kun-reading found in the dictionary for the on/kun-reading that matches the kanji you are looking for and then going to the relevant page reference to find the entry for the kanji. For electronic dictionaries, the on/kun-reading is typed into a search box and a list of kanji which has the reading as a on-reading or as a kun-reading is shown. The user then selects the kanji they are searching for. Most paper dictionaries combine their on/kun-reading appendix so that the entire appendix has both on-readings and kun-readings, rather than having each type of reading in a separate appendix. Electronic dictionaries allow you to enter more than one on/kun-reading in the search box for a single search, so it enables the user to type in all of the on and kun-readings for the particular kanji they are searching for and hence reduce search time. In both paper and electronic kanji dictionaries, if only one reading is used for searching, it is much better to use the kun-reading since most kun-readings are unique and are not duplicated with different kanji. On-readings, on the other hand can potentially be the same for hundreds of different kanji.

This method is extremely efficient and effective if you know already know what the kanji you are looking for looks like and also know either one on or kun-reading. But is obviously useless if you don’t know what the kanji looks like. This method can also be difficult if someone told you the reading for a character and you look it up based on what you heard, since you may have misheard them, or you cannot tell the difference between じょう and じょ, etc. when spoken.

The Stroke Lookup Method:

This lookup method involves counting the total number of strokes a kanji has, looking for the relevant number of strokes section in the appendix of your kanji dictionary, and then searching for the right kanji. In electronic dictionaries, the user can enter the number of strokes the kanji has and then a list of all the kanji in the database which have that number of strokes will come up and the user can select the one they are looking for. The counter for "stroke" is the kanji 画. In both electronic and paper kanji dictionaries, a kanji with six strokes would be listed as 6画 (ろくかく).

This method is good if you are fairly competent at stroke-counting and you either have the character in front of you to look over or you have a very clear image of the character in your mind. However, this method can potentially be very slow if you know nothing about the character’s components or radicals (see below) – which speed up the stroke counting process. In addition, the fact that there may be 300+ kanji that have six strokes can make the actual search process incredibly slow and draining – even worse if you’ve mistakenly counted the number of strokes. Stroke-counting can become incredibly tedious and difficult when the character becomes rather complex and hence mistakes are more likely to be made. Stroke-counting is a problem for many beginner-intermediate Japanese 2nd language students who do not know Mandarin, or any dialect of Chinese. A "stroke" in both Chinese characters and kanji is generally defined as the amount of a character one can draw without taking their pen (traditionally "brush") off the page. There will be no detailed explanation of stroke counting here – such explanation would be another blog entry in itself. However most kanji dictionaries have some explanations as to how to count strokes in their appendices. In particular, I recommend beginners of stroke-counting read "Appendix 2 – How to Count Strokes 画数の数え方" In Jack Halpern, The Kodansha Kanji Learner’s Dictionary (New Edition, 2001).

The SKIP Lookup Method:

The System of Kanji Indexing by Patterns (SKIP) is a revolutionary lookup method for kanji and is not used in Chinese dictionaries. It makes the process of looking up kanji by stroke-count a lot quicker. Students that primarily use the stroke-count lookup method and don’t know much about the components of a particular kanji or about radicals should use this method instead. Again, explaining how to look up kanji using this method would take a whole blog entry so I will not go into detail about it here. Basically, the SKIP method is a three-step process in determining the "SKIP number" of a kanji, which can then be used to quickly find the kanji in the dictionary. The SKIP method states that all kanji can be divided into four distinctly different patterns: 1. "left-right" pattern (eg: 相), 2. "up-down" pattern (eg: 恋), 3. "enclosure" pattern (eg: 国), 4. "solid" pattern (eg: 下). Determining which group the kanji that you wish to look up is in is the first step to discovering what the SKIP number for your kanji is. Most tutorials on how to determine the SKIP number are quite confusing, however the website http://dict.regex.info/cgi-bin/j-e/skip?SASE=/cgi-bin/j-e/kanjidoc has quite an understandable – albeit simplistic – explanation of the method which should enable you to have little trouble finding kanji using the SKIP method. Wikipedia’s article on SKIP lookup (http://en.wikipedia.org/wiki/Kodansha_Kanji_Learner%27s_Dictionary) may also be useful to you.

There are still a number of problems with this lookup method, however I firmly believe that the SKIP method is the most useful of the methods available. The fact that students must still count strokes may pose a problem for some, however the problem of counting the number of strokes in complicated characters is effectively obsolete because the SKIP method splits the kanji into parts, which makes it easier to count them. Also, determining the SKIP number can also be a hassle if the pattern is not immediately obvious – especially if it is your first time using a kanji dictionary and you don’t know much about kanji. The SKIP lookup method is not available in any electronic dictionary and is only available in a very small number of paper dictionaries (it is a fairly new lookup method (made especially for gaikokujins like ourselves) and it will take time for the dictionary writers to adjust to it). The Kodansha Kanji Learner’s Dictionary was published in such a way that the SKIP lookup method was the primary method of looking up kanji in that dictionary, although its explanation on SKIP is perhaps too detailed and verbose for the average kanji learner to comprehend effectively. Hopefully, more kanji dictionaries written for non-Japanese will have SKIP facilities in the future.

The Radical Lookup Method:

The radical lookup method is one of the fundamental ways of searching for a kanji in Japanese dictionaries and looking up a Chinese character in Chinese dictionaries; it has existed longer than any other lookup method in both China and Japan. A "radical" (Japanese: 部首 (ぶしゅ) literally: "section header") is small part of a kanji which forms either the root of its meaning, or contributes to the overall meaning of the character. For example the radical 氵(called: さんずい, meaning: "water") forms part of the kanji 洋 (ヨウ) "ocean". Each radical has a name and a meaning, but they are not necessarily kanji; some are kanji while others are just "parts of kanji" and are not kanji themselves. For example, the kanji 日 functions also as a radical in the kanji 星 (セイ, ショウ, ほし) "star". However, the radical ⺮ (called: たけかんむり, meaning: "bamboo"), which appears in many kanji such as 第 (ダイ) "ordinal number prefix", is not a kanji. Some kanji can be made completely of radicals, however usually they contain just one or two radicals and the remaining part of the kanji is not classified as a radical. Some radicals are kanji but their appearance has been changed. For example, the kanji 心 (シン, こころ) "heart" looks like 忄sometimes when it is used as a radical. When 心 looks like 忄in a kanji, the radical form is no longer considered to be a kanji. The たけかんむり radical above is another example of this.

When the first Chinese and kanji dictionaries were created, characters were indexed by their radicals in order of increasing stroke-count of the radical. The user would look for the radical part of the character in the radical index, go to the relevant section containing all the characters that are listed under that radical and then find the character they were looking for. This is still the practice today for radical lookup in modern Chinese and kanji dictionaries. Since Chinese dictionaries have fewer lookup methods than kanji dictionaries, greater emphasis is placed on Chinese 2nd language students to learn the radicals than Japanese 2nd language students. Native Chinese people from previous generations, when using computers, prefer to type characters by using radicals to form the character rather than pinyin (which is mostly used by non-native Chinese students and the younger generations of native Chinese). The number of radicals that are currently used in both Japan and China is 214. There are a bit more radicals than this, but they are not often used in either country. Radical lookup is still the preferred method among native Japanese, however it is becoming increasingly obsolete with Japanese 2nd language students since there are other easier lookup methods to choose from.

Each kanji is indexed in a dictionary by one radical only. So in our above example of 洋, you would find it by first locating the radical 氵. If a kanji has more than one radical, such as 明 (contains the radicals 日 and 月), it will only be indexed by one of those. Knowing which one it is indexed by will not be explored in detail here, but again you should refer to the internet, textbooks, or appendices of kanji dictionaries for tutorials. However, in many cases, the indexed radical appears to the left of the character. In electronic dictionaries, the user can type in the number of strokes of the radical that is contained in the kanji they are searching for. A list of all the radicals with that amount of strokes appears, and the user selects the correct radical. Then a list of all the kanji appears that contains that radical and the user chooses the kanji they were looking for. When a kanji is searched in a dictionary, the "primary radical" (that is, the one that is used to search for the kanji in the dictionary), is also listed followed by the kanji "部". So if the kanji 明 searched in a kanji dictionary, the section describing what the primary radical is would state: "部首: 日部" or something similar.

This is perhaps the least effective of all the lookup methods to a Japanese 2nd language student. The reason being is that you have to learn the names, strokes and meanings of 214 radicals on top of the huge amount of kanji you already have to learn, and if there are other ways to lookup kanji than having to memorise extra symbols I’m sure students will go for the other option. If you do know the radicals, however, then it can make looking for kanji that you know the appearance of – or even just a fragment of its appearance – a very efficient process. In my opinion, methods like SKIP and kanji by parts are just as efficient and they do not require one to memorise extra symbols.

The Kanji by Parts Lookup Method:

The kanji by parts lookup method is available in electronic dictionaries only; no paper dictionaries to my knowledge have this lookup method integrated in their indexing system. This method is very similar to the radical lookup method in that it involves the student searching the individual "components" of a kanji in order to find the kanji they are searching for. However, its key difference to the radical method is that there is no extra memorisation of symbols required; the method uses kanji which form "parts" of the kanji that you are seeking to find the correct kanji. For example: the kanji 語 – if radicals are ignored – is actually formed by placing the kanji 言 and 吾 (ゴ, われ, わが, あ) "I, myself" together. Alternatively, you could break up 語 a different way and say it is made up of 言, 口 and 五. Both methods of breaking up 語 are correct and using both methods would eventually lead you to finding the kanji 語. 

The kanji by part method is often not listed as an "official" lookup method, and it is difficult to find tutorials on. User manuals of electronic dictionaries are likely to explain the method, however most electronic dictionary manuals are published in Japanese. So unless you are in level 2 or level 1 JLPT, you are likely to have a lot of trouble reading it. For these reasons, I decided to explain properly how to search for kanji by parts in this entry. To search for a particular kanji by using the kanji by parts method, you first look at the kanji that you’re trying to find in the dictionary and see if you can break it up into one or more "smaller" kanji, if you like. You should only look for "smaller" kanji that you know the on and kun-readings to. After you’ve identified these "smaller" kanji, type up either their on-readings or kun-readings in the "parts" (or similar, depending on which electronic dictionary you’re using. I always use the Canon Wordtank electronic dictionaries) search box of your electronic kanji dictionary (you may have to separate each reading with a "&" sign or similar). The dictionary then searches for a kanji which contain "smaller" kanji that will have the readings you typed up as either on-readings or kun-readings, and then it is a matter of selecting the one you want. Again, it is better to choose the kun-reading over the on-reading if possible since the kun-readings are generally unique between kanji. The fact that some kanji also act as radicals is unimportant, you can choose either to search for the kanji using radicals, or using kanji by parts. For example: 明 is made up of the radicals 日 and 月. But these two radicals are also kanji. Therefore looking up 明 by either using 日 and 月 as radicals or as "smaller" kanji will still lead you to 明. When you encounter situations where there are "smaller" kanji and radicals in one kanji, the kanji by parts method is always quicker because you do not have to bother counting the number of strokes in the radical and searching for the correct radical. Of course, since kanji by parts can only be used in electronic dictionaries, there’s nothing stopping you from using both the radical lookup method and the kanji by part lookup method to search for the one kanji.

Another interesting fact to note is that a kanji which contains "smaller" kanji will sometimes have the same on-reading as the "smaller" kanji. Taking 語 as our example again, the on-reading of this kanji is ゴ. If we look at the kanji 吾, which is a part of 語, its on-reading is also ゴ. Additionally, if we chose to break down 吾 into parts, we would also note that the kanji 五 has the on-reading ゴ as well. Therefore, in the majority of circumstances, if you wished to look up kanji by parts in a paper dictionary you could do so by determining the on-reading of one of the "parts" and then – by assuming its on-reading would be the same as the one you are searching for – use the on/kun-reading lookup method to find the right kanji. I must stress however that this will not always work, so do not complain to me if it doesn’t work out when you try it at home. 😛 One example where using the method just discussed will not work is with the kanji 劣 (レツ, おとる) "inferiority". This kanji is made up of the kanji 力 ( リョク, リキ, ちから) "power" and 少 (ショウ, すくない, すこし) "a little". Neither of the two on-readings of the "smaller" kanji are found in the kanji 劣. It should also be noted that this "pattern" with on-readings in the "parts" of kanji and normal kanji does not work with kun-readings. The reason being is the same as I stated earlier: the kun-reading for each kanji – in the vast majority of cases – is unique.

It should also be noted that in many cases the meanings of the "smaller" kanji all contribute to the meaning of the "larger" kanji. Once again, using 語 (meaning "language") as our example, the individual meanings of 言 and 吾 are "say" and "I" respectively. No doubt there is a connection between "I", "say" and "language". In addition, observing the parts of 劣, we can see that there is also a connection between "a little", "power" and "inferiority"; a person is "inferior" if they have "less power" compared to someone else, etc. – at least that is how inferiority was perceived during the time Chinese characters were brought to Japan. We must remember that while the meanings of words may change with the times, the meanings of the kanji do not. 

The kanji by parts method is by far the most efficient and effective method of searching kanji available to a user of an electronic dictionary. No stroke counting is required, and no memorisation of extra symbols is needed. Its major downfall is that if you cannot identify any "smaller" kanji in the kanji you are looking for, then the method is useless. Also, if you select a common kanji like 口 to be you "part" to search with, there could potentially be hundreds of kanji which have 口 as a part and hence search time may increase significantly.

The Certified Japanese Kanji Proficiency Test:

There is a test similar to the JLPT that is especially directed towards a student’s knowledge of kanji. It is known as "The Certified Kanji Proficiency Test" (日本漢字能力検定試験 – "にほんかんじのうりょくけんていしけん"), or commonly abbreviated to the "kanji kentei". The kanji kentei tests everything about kanji – including on/kun-readings, stroke order, radicals, etc. It is actually designed for native Japanese, however there are an increasing number of Japanese 2nd language students taking the proficiency test. More information about the test can be found at: http://en.wikipedia.org/wiki/Kanji_Kentei.

Credits:

I certainly could not have managed to compose this huge entry entirely by myself. So in this little section I’ll state what references I used and which people gave me some of the information that I’ve included here.

References:

• Miyako Iwami, Shigeko Miyazaki, Masako Nagai, Kimiko Yamamato, Japanese for Busy People II (Revised Edition, 1994) – "The Kanji in Books II and III"
• Keiko Aitchison, Kookoo Seikatsu Kanji Workbook (2nd Edition, 2000)
• Jack Halpern, The Kodansha Kanji Learner’s Dictionary (New Edition, 2001)
• Wikipedia, Kanji, http://en.wikipedia.org/wiki/Kanji at 7 April 2007 – "History", "Local developments", "Orthographic reform and lists of kanji", "Related symbols"
• Wikipedia, Radical (Chinese character), http://en.wikipedia.org/wiki/Radical_%28Chinese_character%29 at 7 April 2007 – "Shape and position of section headers in characters"

Special Thanks To:

• Kentarou-Sensei, Lyn-Sensei, Miho-Sensei and Yuki-Sensei for their help on the various uses of kanji in Japanese writing and in particular to Lyn-Sensei for discussing some differences in use between Chinese characters and kanji and where Japanese 2nd language students who know Mandarin may make mistakes with their kanji
• Ruolan-Laoshi and Lisa-Laoshi for discussing how Chinese characters are used in China
• The Chinese students at RDSC, in particular Jennifer-chan, John-kun and Kiko-chan for showing me some differences between simplified and traditional Chinese characters and Chinese character lookup methods and especially John-kun for helping me determine what percentage of kanji are simplified and traditional Chinese characters

 

If anyone would like further explanations on how to lookup kanji using a particular lookup method, please post your request as a comment and I’ll make it my next entry. If you have any other comments or questions, please post them too! My "comment" section of this blog is looking pretty empty at the moment, so please help to fill it up by posting. Thanks!