Remembering the Kanji, Volume I
(1977; 4th ed. 2004)
James W. Heisig
Heisig presents a simple but anti-traditional method for remembering the meanings and writing of kanji. He takes the simple idea that many students already try —taking the meaning of radicals and telling a story that gives the meaning of the kanji— but refines it into a proper method. First, rather than radicals, he refers to "primitives" which can be either radicals or other kanji, which allows building up of meanings in a better way by giving 'syllables' in addition to the 'letters'. Second, the simplest part of the method is reordering the presentation so that simpler kanji are presented before and alongside kanji which use them as primitives (e.g. 日、月、冒、朋、明、唱、晶、、). This is a consequence of the method being to teach all the 常用漢字 together, as opposed to the traditional method which teaches them in a way that you can test your knowledge incrementally. Third he focuses, not on the gist of a primitive, but instead chooses a specific "keyword" to help distinguish primitives with similar meanings (e.g. 如 vs 肖). This symbolism gives hooks for the stories to hang off of without them becoming a vague blur that's unhelpful for constructing the kanji from its keyword. Fourth, the focus is on studying from keywords to kanji, instead of the other way. Focusing on writing/recall rather than reading/recognition strengthens our associations since the latter comes for free from the former, but not the other way around.
Learning kanji is often named the hardest part of learning Japanese, and many voice skepticism at Heisig's claim that the method can teach the most common 2000 kanji in a few months. A quick web search will bring up many diatribes both for and against Heisig; if nothing else, it's certainly polarizing. Personally, I'm a big fan of it. I have never liked memorization, and the traditional approach of just writing the 常用漢字 over and over simply doesn't work (hence the notoriety of learning kanji). One of the things I like most about the book is the ways in which it formalizes the simple technique of telling stories. In the modern era we have lost our oral traditions for memory and many have lost their creative ability to tell stories. Heisig gives techniques to rekindle the oral and creative traditions of memory. It's alien to the modern era, which is no doubt why so many dislike it, but it feels like home to me.
Another piece of the controversy is that Heisig does not teach the readings of the kanji in this volume. At first I too was very skeptical. The separatist approach of teaching speaking vs reading in JSL is my principal complaint against JSL. After giving Heisig a try, however, I think that his approach is valid. The big thing to remember is that he is teaching kanji, not words. To the new learner of Japanese this distinction may seem baroque, but it is mirrored in the vast differences between kanji dictionaries vs word dictionaries (cf. etymological dictionaries vs 'real' dictionaries). Because of the nature of Japanese writing, it's not very helpful to learn the readings of kanji without learning specific words and compounds since a kanji's pronunciation is based in large part on the word it's in (e.g. 今日 is read 「きょう」 which is not built from the readings of 今「コン・いま」 and 日「ニチ・び、か」). Heisig's second volume deals with the (音読み) readings of kanji, for those interested in pursuing them.
The only complaint I'll level against the book is that the author's background in Christianity and Freudianism can be quite overt in some of the stories. This lends a distasteful missionary flavor in parts, and I find the Christian stories disruptive to my memory (both because of my distaste for them, and because they are not as familiar to me as they are to him). Those without anti-Christian tendencies probably won't notice nor care. In counterpoint to this singular complaint, the primary goal of Heisig is to teach the technique, not the stories. The major part of the book provides only the kanji and keywords for them, leaving the reader to formulate the story to bind them together. For as infrequent as they are, in the portion where stories are provided it can be helpful to have a few which are disruptive (so long as they are illustrative) since this can drive the reader into coming up with their own stories earlier. Ultimately, as Heisig says, none of the stories will be helpful unless they click for the learner.
I'm a big fan of the book and the technique, and I highly recommend it. For those who want to get polarized before shelling out for the whole thing, the introduction and first (of three) parts (125 pages) is available for free online. For those interested in the rhetorical technique of mnemotechnics or general issues of pedagogy, here is an interesting paper about Heisig's approach to L2 learning.