This is part two of a series on the frustrations of learning the Japanese language. The foregoing assumes you’ve already read part one (though you should be able to get along without it: it’ll just start somewhat abruptly).
Learning Through Reading Japanese
I believe that my voracious appetite for reading is directly responsible for an enhanced understanding of English; and when you want to master a language, it stands to reason that you should expose yourself to as much reading material written in it as possible. This makes it all the more frustrating, that learning Japanese through exposure to its written language is extremely difficult, time-consuming, and tedious.
For instance, suppose you’re confronted with a Japanese sentence such as:
Assuming that you’ve studied your approximate hundred hiragana characters (half of which are just systematic modifications of the basic 46 forms), you’ll have no problems with the へ and きましょうか bits, and since hiragana is a syllabery, you instantly know how they sound, too: even though they don’t represent much in terms of meaning until you know what are the more complicated-looking characters that remain (the kanji).
But what can you do with the kanji (明日図書館, and 行), if you’ve never encountered them before? It’s not like you can look each up in a dictionary by its name, since you don’t know it. Literally all you know at this point is its shape. And that’s what you end up having to look it up by. Many characters can be broken into separate constituent parts that might be common to other characters, or are possibly complete characters in their own rights—for instance, 明 can be broken into 日 and 月, 館 can be broken into 食 and 官. So you count the number of brush/pen strokes it would take to write that character, look it up by that number, then find that component (called a “radical”), and then find the character you’re looking for (possibly by further looking under the total number of strokes to find a character). Once you find the character that matches, flip to the page number that describes that character.
So, we’ve found the page holding the description of the meaning for (say) 行, and its pronunciation. Except there’s not one pronunciation: there are several. Fortunately, in the case of 行, it’s followed by a string of hiragana which, after even a small amount of training, a student will quickly recognize as the final portion of a verb. The list of possible readings for 行 will include various verbs that start with that character, and an indication of what characters would be expected to follow it when performing its function as a verb. Verbs, whether written with initial kanji or not, will virtually always be followed by a string of hiragana which indicate how the verb is being conjugated, and usually indicate which reading the kanji itself has (if one is already familiar with the character, of course). You’ll determine that in this case it should be read simply as the “i” in “ikimashouka?”, which means “shall we go?”.
The 行 had hiragana to help indicate its reading (when filling this role, they are called “okurigana”). No such luck with the others. If I look up 明, I see 14 possible readings, and 10 different verbs/adjectives it could form part of (depending on the okurigana that follows it, and fortunately most of them have the same reading for 明). I can rule out the ones that require okurigana, because there isn’t any in this case. And the remaining readings can be divided into two groups, one group that is used mainly when the character appears as its own separate word, and another that is used mainly when the character appears in combination with other kanji to form a compound word (but there are no guarantees). Since it appears with other characters, it’s reasonable to assume they form some sort of compound, and fortunately the group of readings more commonly used for compounds comprises just 3 of the 14. Ideally, the compound I’m looking for will appear in the list of common compound words that will appear on that page of my kanji dictionary: then I will have identified not only the correct reading for this character, but for the remaining characters in the word, too. Otherwise, I’ll most likely need to look up the next character the same way I did the first, find its list of possible readings, and try looking up various possible reading combinations between the two characters in my Japanese dictionary (the one for looking up words, and not kanji).
Alas! In this particular case, none of the possible readings of 明 are correct here! 明日 is a relatively special case where two characters join together to form a word which is pronounced based on the characters’ combined meaning, rather than their individual readings. Unless it was listed as one of the example compounds (fortunately, this is likely), I would probably have a very, very hard time discovering that the word is “ashita”, which means “tomorrow”. Fortunately, even without knowing this, I probably would have settled on the alternative reading “myōnichi”, which is a legitimate reading of the characters 明日, and also means “tomorrow”, but combines the actual readings of the characters, making it possible to find using the dictionary approach. It wouldn’t be wrong, and I’d be able to continue reading from that point—even understanding the correct meaning—but “myōnichi” is a very formal word, and I’d sound pretty funny to any Japanese person that would hear me use that word in the middle of a relatively casual sentence. There’s even one more possible reading of 明日 (still with the same meaning): “asu”!
Perhaps you’ve noticed at this point that I’ve now said 明日 is a single word. Of course, it’s a word that appears in a sea of kanji: 明日図書館. How the hell are you supposed to figure out where one word ends and another begins? The answer: brute force. If, when trying to look up what word might be formed from 明 and 日, you see that there’s a word 明日, well then the next kanji might start a new word, so you react accordingly. (The remaining characters turn out to form the word for “library”; the meaning of the full sentence was: “Shall we go to the library tomorrow?”.)
Of course, it’ll suck if it turns out that there’s another word that continues with that kanji… the other day, someone on a Japanese-language IRC channel wrote “今日本語を勉強しています”, which means “I’m studying Japanese now”. But there’s a bit of ambiguity in the start of the sentence: 今 is the word for “now”, and 日本語 is the word for “Japanese language”. But 今日 is a word meaning “today”, and 本 by itself means “book”: “今日本を読みました” would mean “I read a book today.” Notice that the first three characters (kanji) of both sentences are the same, even though they have no words in common (not counting the particle “word” を). You don’t know whether the first couple of words are “today, book” or “now, Japanese” (obviously, the word order is completely different from that of English), until you reach the character 語. If you’d never seen any of these characters before, and were looking each of them up individually, it might take you quite a while to figure out what was written if you first started down the track of 今日 being the first word, and trying to figure out what word might be formed by 本語, only eventually discovering your mistake after an hour or so of banging your head against the table! A well-placed comma will go a long way in such a case, but by now I think you’ve got a very good picture as to why learning Japanese through reading it, has the potential to shorten your lifespan.
As you saw with 明日, “ashita/asu/myōnichi”, a given string of kanji can sometimes have multiple possible readings. This is not particularly uncommon. However, sometimes a string of kanji can not only have multiple readings, but those readings can actually have different meanings! Take 見物, which can be read either as “kenbutsu/to go sightseeing”, or as “mimono/thing worth seeing”. Only the surrounding context can make it clear which is meant. (In practice, 見物 is a rarer reading, and a writer might choose to write 物 in kana instead, to alleviate any ambiguity.)
When I sit down to make an attempt at reading a Japanese book, I’ll often end up with several other books nearby to help me do so. There’s the actual book I’m trying to read, the kanji dictionary for looking up unfamiliar characters, and the vocabulary dictionary. I’ll also often have a secondary kanji book that is less useful for discovering what a character is, but more helpful for providing advice and information on how to remember the character. And, I might even have another book or two to look up points of grammar that I’m having trouble grasping.
As I read a passage, I look up my characters, possibly look them up again for tips on what the individual components might mean, look up unfamiliar words, and if necessary look up an unusual verb or adjective use. I can easily spend twenty minutes, maybe even up to an hour, on a single sentence. These days I try to rely more on electronic dictionaries, which save both time and disk space, but the process remains largely tedious. As I learn more characters, vocabulary and grammar, each additional passage takes me less time, but it can still be an ordeal.
But, after mastering reading and gorging myself on Japanese literature, will I finally have gained an intimate familiarity with the language that I can take and apply directly in using the spoken language in conversation? Well… yes and no. Obviously, Japanese is Japanese, and a lot of phrases I get from books can go straight into conversations too. But, the usual style of written Japanese is still significantly different in tone, level of formality, etc, from spoken Japanese. Nobody speaks the way a typical Japanese book reads, except maybe radio announcers and other people who aren’t actually speaking to a particular person. If you spoke sentences you’d gotten from a book, in many cases you’d sound like you were quoting (which, of course, you would be). Fortunately, reading (modern) novels means that you may pick up some actual conversations in the form of dialog, and manga (Japanese graphic novels) are a popular way of learning the colloquial language—but note that both of these can often depict very casual conversation styles that would be considered quite rude if you employed them in conversation outside of family and close friends.
Wrapping things up
I was going to follow this section up with some ideas about learning Japanese vocabulary and kanji efficiently; but after working on that for a while, I’ve decided that the ideas are still too immature to put here yet. Meanwhile, I think I’ll write an article that talks about my favorite Japanese-learning resources instead, and where they still manage to fall frustratingly short, and what I think the perfect Japanese course might look like. Probably, some of the ideas I had for efficient learning will seep into that article.