The following is the English version of an article in interview format published in May 1st, 2007 edition of the Japanese Language Educational Journal. The audience for this article is Japanese language teachers, so some of the terms might be confusing for beginners, but I hope this is useful information for those studying Japanese as well.
When non-Japanese study Japanese language, understanding kanji is essential to progress to upper-level proficiency because of the profound relationship between kanji and vocabulary. But for learners from non-kanji backgrounds, it is undeniable that learning kanji is the biggest barrier to becoming proficient in Japanese. Picking up from where the article in the March 1st edition of the Japanese Language Educational Journal left off, we ask Karl Rosvold, who started his Japanese studies from almost zero after arriving in Japan, and was able to progress to passing Level 1 of the Japanese Language Proficiency Test (JLPT), about his experience and beliefs about studying kanji, and ask his advice for studying kanji from the standpoint of someone who has successfully studied kanji as an adult.
Q: What are some of the problems for non-Japanese studying Japanese reading and writing?
Traditional wisdom tells us to teach foreigners kana and kanji the same way they are taught to Japanese children. First start with the kana and then teach the kanji in the order prescribed by the Ministry of Science and Education. But many foreigners who study kanji this way are utterly frustrated with their inability to remember kanji for more than a short period of time, even when they make great effort. As a result, I think there is a sort of myth that the majority of students from non-kanji backgrounds will never really be able to master written Japanese. One could even point to the overwhelming number of Chinese students who pass Level 1 of the JLPT compared to a relatively small number of students from non-kanji countries as “evidence” that this is true.
I think the ultimate goal of the majority of students studying Japanese is the same – conversational ability, listening comprehension, reading comprehension, and the ability to express themselves in writing. But for students from non-kanji backgrounds, remembering kanji is so difficult and different from anything they have studied before that the idea of teaching kanji simply as a transcription of spoken Japanese without teaching about kanji is insufficient.
I think this situation could change significantly if teachers changed the way kanji are taught to non-native speakers. First, it is important to realize that Japanese children studying kanji and adult non-native speakers studying kanji are different in many ways, including their ability of spoken Japanese, their adult knowledge of the world, and their available time to study and practice Japanese. So even though the goal of learning to read and write Japanese is the same for the non-native student and the Japanese child, I think we must admit that it is at least possible that the best study methods for each group may also be different.
Q: You said that when you studied kanji you used a method called component analysis. Please talk about that.
Last time I explained that I used the basic method for studying kanji called component analysis, as introduced in the book Remembering the Kanji I by James Heisig. I was lucky that someone had recommended that book to me because it really helped me make sense of kanji and organize the new information I was learning.
Most Japanese recognize that kanji are made up of components, but I think Japanese mostly consider this fact “interesting” rather than “essential” or even “useful,” so these components are not identified systematically and emphasized when Japanese students learn new kanji in Japanese (kokugo) classes. From my own experience learning kanji, and after thinking about kanji for many years, I believe that if kanji are taught to adult students in a systematic way, this will allow the students to develop frameworks in their minds in which to organize their knowledge of kanji, and thus better understand and retain what they have learned. How can someone effectively remember something that they don’t understand?
So because teachers are a great influence over their students’ study habits and strategies, I hope that if I can convince you, the Japanese language teacher, that this is a valuable approach, you can in turn guide your students to use these methods to help them learn and remember kanji.
Q: Can you give some examples to illustrate what you mean by component analysis?
In component analysis students learn single-element pictograph characters such as 日 (sun) and 月 (moon) the same way Japanese children do -- by seeing how the characters are stylized pictures of the things or ideas they represent. Let us call characters made of more than one such single-element character “compound characters.” Loosely speaking, the elements used to construct compound characters are radicals. But more precisely speaking the term radical refers to the specific part of the kanji by which it is classified in a kanji dictionary. Compound characters contain many identifiable elements that are not included in the traditional list of 214 radicals, so these elements are simply referred to as "parts" or components.
When studying kanji that combine other identifiable elements, however, students first learn the components individually, and then study the new character as a combination of its elements. So 明 (bright) would be studied only after 日 (sun) and 月 (moon) have been learned. In this case, the same is true for Japanese children learning in elementary school, because 日 and 月 are taught in first grade, while 明 is taught in second grade.
On the other hand, 近 (near, 2nd grade) is made up of 辶 (road radical) and 斤 (axe, loaf). 辶 is learned in four other 2nd grade kanji: 道 (roadway)、週 (week)、遠 (distant) and 通 (traffic), but the character 斤, as in 「食パンが一斤あります。」 (“There is one loaf of bread.”) is not included in the first 1006 kyoiku kanji (general education kanji taught in the six years of elementary school) at all, so Japanese children would not formally learn this character before learning other characters that use it. Other characters in kyouiku kanji that contain this one along with their grade levels are: 新 (new, 2), 所 (place, 3), 折 (fold, 4) 兵 (soldier, 4), 断 (severance, 5) and 質 (substance, 5).
A student learning kanji according component analysis would learn the identity of a component the first time it is used, and any time another new character is learned that uses the same component, half of the work to remember the character has already been done. Furthermore, if students learn about kanji systematically, including why a particular combination of components has a particular meaning or particular on-yomi pronunciation, or that a particular character is a stylized pictograph, they can use this information to make guesses about new characters they come across and help them remember what they have learned as well.
※ Two well-known books that promote component analysis are Remembering the Kanji mentioned above, and Kanji ABC by Andreas Foerster and Naoko Tamura. Mary Sisk Noguchi also often writes about component analysis in her column "Kanji Clinic" for the Japan Times. You can also read past articles at www.kanjiclinic.com.
Q: What kind of techniques are necessary for studying upper-level kanji, such as the 1000 kanji included in joyo kanji but not in kyouiku kanji? (i.e. the second 1000 kanji)
Some methods instruct students to see a single image in the kanji as an aid to help memory. I think this approach is fine for characters that are truly pictographs. Showing how the characters developed from characters written on tortoise shells to characters etched on metal vessels, to seal style characters, to clerical style to modern block style of writing is interesting. In fact, almost half of the first grade kanji – the ones usually taught first – are indeed single-element pictographs. Six of the seven characters used for the days of the week, for example 月火水木土日 (all of the days except Friday, and all learned in the first grade) are all examples of pictographs.
In contrast, for higher level kanji, there are almost no pictographic characters. I estimate that about 70-80% of the second 1000 kanji are meaning-sound combinations, where the radical gives a clue to the character’s meaning and the remainder gives a clue to the character’s on-yomi pronunciation. To use a single image to remember the whole character is to ignore the real breakdown of the character. Furthermore if the student has to remember a completely new image for each character, such images are not systematic, so an image used to memorize one character cannot usually help to remember a different character. This will not help the student remember a great number of characters.
Many books that teach kanji by component analysis instruct learners to construct an associative story linking the meaning of each component with the meaning of the compound character made up of those components.
If the character itself is a meaning-meaning compound (also called logical aggregates in the Wikipedia article on Chinese Characters) or 会意文字 (kaii moji) in Japanese, the meaning of the compound character is often logical based on the meaning of the components. For example the character 休 (rest) is supposed to represent a person (人) resting up against a tree (木). But the vast majority of kanji learned at advanced levels are meaning-sound compounds, so there is not always a clear connection between the meaning of the components and the meaning of the character they form.
What should be done in cases such as these? Take the character 精 (refine) for example. It is a compound character made up of 米 (rice) and 青 (blue). The component 青 represents the pronunciation sei or shou. But there is no easy explanation why “rice” and “blue” should logically combine to yield the meaning of “refine.”
The reason students study kanji is to learn to read and write Japanese, not to learn about ancient Chinese history or kanji etymologies. If the character’s etymology really is helpful to remember the writing and meaning of the character, then by all means that is a good study method, but if studying the character’s etymology does not help the student remember the character, it would be better to use some other technique to remember the character. So in the example just mentioned, the student can remember that 精 (refine) is made up of 米 (rice) and 青 (blue) by creating an associative story or image. How about the following story? When rice is first harvested each grain is encased in a golden brown hull. The rice then goes through a refining process whereby it is converted to white rice. But excellent quality white rice is really translucent. In fact, if you really look closely at polished rice, you can see that it has a beautiful blue sheen to it. Actually, this story has nothing to do with the real etymology of the character 精, and some imagination might be required to “believe” this story. But true or not, if the student can remember this story, and if the student already knows the components for 米 rice and 青 blue, the student can remember the compound character. If the student is asked to make up the stories by him- or herself, the story might not have any meaning for anyone but the person who created them, and sometimes the stories can be really silly. As a result, sometimes teachers criticize this technique saying that it is not a serious way to study kanji, but I would respond that if such techniques help students remember characters faster or more easily, then it is a good method. Creating an associative image in this way is a kind of mnemonic.
Q: When students are doing reading practice, do they ever have difficulties reading characters or looking up words in a dictionary?
In addition to learning and remembering kanji in the first place, looking up words one doesn’t know can be very time consuming for students. Japanese adults usually use a kokugo dictionary, which orders words based on their pronunciation, rather than a kanwa dictionary which has entries for each kanji along with its compounds. This is because for the most part Japanese adults already know the correct pronunciation of the word by looking the kanji. For foreign students of Japanese, this is usually not the case, so they need to use character dictionaries (kanwa jisho).
For example, 大 (big) has many readings, such as dai and oo-kii, but also the word 大人 (adult) has the special reading otona, but then the compound 大人気 is read daininki when it means “very popular” or otonage when used to mean “unmanly” or “childish” in the word otonagenai. So even if the student knows the character 大, unless the student has already studied each of these compounds, he or she cannot look them up in a kokugo dictionary where words are ordered by pronunciation. So the student’s ability to rapidly look up compound words is a critical factor to making continued progress when the student has to do a lot of reading comprehension.
For each kanji there is exactly one “official” radical, but for beginning students, it is often difficult to determine which component of the character that radical is. For example, in the character 聞 (listen), the student will probably wonter whether to look up the character using 門 (gate) or 耳 (ear). If the element the student chooses is incorrect, after once looking up the character and not finding it, he or she must again count strokes and start again.
A student once told me “When I do reading comprehension homework, I spend all of my time looking up characters in a dictionary and I have no time to actually study Japanese!!” Japanese teachers are familiar with this problem and don’t want their students spending all their time looking up characters, so they often add furigana (kana written on top to show the pronunciation of kanji) and create vocabulary lists for their students. But because students cannot rapidly look up characters on their own, they cannot freely choose texts to read and study on their own. Therefore I think it is essential to have an easier way for students to look up kanji and compound words.
Recently many electronic dictionaries allow users to find characters using any of its components, not just the official radical, so if someone knows the components very well, using such an approach on a computer can help a student find characters very quickly, especially if more than one component is used to find the character. Unfortunately for the non-native Japanese language learner, electronic dictionaries require users to know the name of the component in Japanese.
Q: Please talk about the software called 漢壁 (kanpeki) you have been developing to help non-native students study kanji.
If I were to compare reading and writing Japanese correctly to a sport, I would say it is as difficult as juggling while riding a unicycle. If I wanted to learn something as tricky and difficult as that, I would practice each technique separately. Only after I could confidently juggle while standing on the ground and ride a unicycle without juggling would try to put them together. What does that have to do with learning kanji? I think that it makes sense to study the writing of kanji, the reading of kanji and how to use compound words all separately and then put them together later. To study all of those aspects of kanji at one time is overwhelming.
It is my sincerest wish that non-native speakers can become truly proficient in Japanese including the ability to read and write. To help make this a reality, I would like to see students and Japanese teachers to use the best study and teaching methods possible. For that reason I am in the process of creating software to help students study kanji through component analysis, and to help them rapidly look up kanji. The name of this software is 漢壁, or kanpeki which is made up of the first character in the word kanji and for which the second character is read kabe (wall) in kun-yomi. The name comes from my belief that kanji are the greatest barrier (wall) to Japanese proficiency. It is my hope that this software can help students and their teachers overcome that barrier to Japanese proficiency. Please try this software and look-up utility on the website that I am starting in conjunction with the development of this software. On that website I also hope to create a forum where students and teachers of Japanese can exchange opinions and ideas about learning and teaching kanji. If you have any questions or comments, please feel free to contact me through e-mail or through the website.
Send e-mail to: kanji.kabe@gmail.com
Website: www.kanjikabe.com