Learn to Read Chinese with Ease (TED Talk)

古代文字ゲーム kanji game screenshot

古代文字ゲーム kanji game screenshot

There probably were many people in the audience who didn’t know anything about Chinese writing, and so this talk was a brief introduction that went a long way toward demystifying kanji / hanzi for them. However, ShaoLan is not even close to the first person to introduce this concept. Kanji Pict-O-Graphix (1992) did a fantastic job of creating memorable and coherent pictures to match the core meaning of the characters, and even won awards for its graphic design. Even earlier, Heisig published Remembering the Kanji in 1977. Both Rowley and Heisig built on older etymological systems already in place in Japan and China.

The constituent elements are usually called radicals in English (部首; bushu in Japanese) and there’s a decent online introduction at About.com including the traditional patterns used for dividing the characters. (Incidentally, radicals are how you can explain exactly which character you mean when there might be dozens of characters with the same phonetic reading. This is similar to how we disambiguate letter sounds by saying “M as in Mike” in English.) Heisig called these “primitives” instead, but that’s just terminology. There are various methods for dividing and categorizing characters for use in dictionaries, but the most straightforward system for non-native learners is probably the newly-developed SKIP system used in the Kanji Learner’s Dictionary.

Chinese characters originally started as pictograms (representing the thing directly) and were soon adapted into ideograms (representing an associated or more abstract idea). Their meanings have sometimes mutated a bit from the earliest usages, and the forms have changed — sometimes drastically — since then. I have an iPhone game that can give even someone completely unfamiliar with the kanji a taste of how the characters have mutated from ancient Chinese to modern Japanese.

There is some instruction given in Japanese schools about combinations of the early characters. ShaoLan’s eight examples are among the first taught, since they are so easy to grasp. But oddly enough, once the basic concept of radical combinations is introduced children aren’t given explicit instruction in using the radical characters as a tool for guessing at the meaning of unfamiliar characters, or to enhance the retention of new characters. Most instruction heavily features lists and repetitive writing exercises, nothing more sophisticated than brute-force memorization. My wife (who is Japanese) was fascinated when I told her how I studied kanji. She’d never thought of learning characters through mnemonic stories. Instead, everyone just learned writing patterns and practiced the strokes, over and over and over.

In looking for links for this article, I ran across an excerpt on a blog called Nihongocentral that explains the traditional thinking pretty well.

Kanji textbooks for foreigners are based on Japanese books that use the following logic:

The smallest orthographic units in Chinese are strokes. There are six basic strokes, a dot (丶), a horizontal line (一), a vertical line (丨), a diagonal line falling from right to left (丿), a diagonal line falling from left to right ( ), and stroke with a change in direction (乚). These basic strokes have varied formats, such as different dots (e.g. in 氵, 丷, and 灬), different diagonal lines, (e.g. in 爪, 彡), lines with different curves (e.g. in 乙, 阝), and lines with hooks in different positions (e.g. in 乛, 勹).… [Et cetera, ad nauseum.]

Japanese children must learn the Jôyô Kanji, a list of (currently) 2,136 characters, by the end of high school. Roughly half of this number (1,006) are introduced gradually throughout elementary school, but the bulk of them are learned in middle school, with another few hundred and supplemental characters in high school. That means that it takes 9–12 years with the traditional approach to attain normal basic literacy. In contrast, I know non-native Japanese learners who were able to memorize around 1000 characters in just a couple of years using mnemonic approaches. When I was studying regularly, I would learn about 15–20 a day in an hour of dedicated study, though I would of course have to reinforce that study with associated vocabulary and reading practice if I wanted to retain what I’d learned for more than a few days to a week or so.

If Japanese literacy instruction was reformed to use mnemonic storytelling techniques, I think the time spent learning kanji could be considerably reduced. After all, non-native speakers have to learn a large amount of vocabulary and seemingly arbitrary readings along with the core concepts of the characters, while native Japanese kids already know the spoken language. All they need to learn is the idea, not the readings. If a dedicated non-native learner can achieve basic literacy in about 3–5 years, then native speakers should be able to use those techniques to achieve that level of reading ability even faster.

For those who are inclined to learn everything about a subject, I’d advise against “mastering” the writing systems of China or Japan. Kanji/Hanzi are a nearly bottomless black hole of study. I’ve personally seen a set of kanji dictionaries with about 50,000 entries. Yes, that’s a five followed by four zeroes — as in fifty-thousand — not a mistyping of five-thousand. There are an additional 30,000 or more characters that have been found in various ancient Chinese writings, many of which are one-off personal or place names.