Acocdrnig to a reschearer at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit a porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.
Were you able to read the text above? Why is that? Why can we figure out what all of those words mean if they are all spelled wrong? Isn’t spelling supposed to matter?
The human brain is fine-tuned to recognize patterns. When we learn language, we remember it in context. The words that came before, the words that came after, how the sentence flows; we don’t just remember each word as a standalone, isolated object. When you parse a sentence, you don’t stop and dwell on the spelling and origin of each word, you connect the words on the page to words in your head and move on. In fact, your vocal cords actually move ever so slightly as you read in a process called subvocalization1.
Kanji in modern Japan
Today, Japan is the only non-Chinese speaking country that regularly uses Chinese characters (kanji). And unfortunately for Japanese children and language learners, it’s a necessary part of learning how to communicate within the language. Japanese children spend hundreds if not thousands of hours learning this complex writing system, yet many continue to struggle with it throughout their adult lives.
In this post, I’m going to address some of the major arguments you’ll hear around the kanji abolition debate. I don’t intend for this to be an academic paper, but I will do my best to cite my sources where I can as I visit each argument. This definitely won’t be an exhaustive examination, since this is a complex and deep topic that people have written entire books about. Some fundamental knowledge of the Japanese language will be helpful when reading this post, but it shouldn’t be absolutely necessary.
As for my qualifications, I have been studying Japanese since 2006 and it was one of my two majors in college. I also spent a year in an intensive Japanese language course at Hokkaido University. At the end of this program, I passed level N1 of the Japanese Language Proficiency Test (JLPT). After University, I spent a few years at a Japanese company back in the United States and am now working at a large American company in Tokyo.
I was a bit reluctant to put this non-anime post up on an anime blog, but there was enough interest to justify it. Many anime fans study Japanese to some extent, or at least take an interest in the culture.
When people begin studying Japanese as a second language, the writing system can be a bit overwhelming. With two syllabaries (Hiragana & Katakana) and a standard set of over 2000 Chinese characters to remember, it really can be a daunting hurdle to overcome. You can find countless expatriates from English speaking countries who have lived in Japan for a decade or two and still only read the language at a 2nd grade level.
Often times, kanji is explained as being crude pictographs, similar to the hieroglyphs of ancient Egypt. For example, it’s not hard to follow the logic of the character for tree (木) looking like a tree. And if you have three trees (森), that means forest. Makes sense, right?
However, this explanation spreads misunderstanding about the function kanji actually have in the language. Sure, some kanji totally look like what they represent, but many do not. Furthermore, actually using most of these characters will still rely heavily on the context in which they are used. Problems arise when learners are only taught this overly simple explanation. Simply put, kanji are not hieroglyphs.
Before I started writing this post, I decided to finally get around to reading Marshall Unger’s book “Ideogram”2. I purchased it about 5 years ago but kept forgetting to actually read it. In any case, I’ll be referencing it a good few times for this post as it has a lot of solid examples.
I’m well aware that there are people out there who have graduate degrees in Japanese and are likely more qualified to discuss this than I am, but I will do what I can to walk you through each argument with enough information to make a decision for yourself. If you think I got something wrong or missed something, I look forward to continuing this discussion in the comments.
How we process written language
If we wanted to, we could decide to replace every occurrence of the sequence of letters t h r e e in English with the character 三 and agree to say “three” aloud every time we see it. We could even get fancy and let 三 also stand for the t r i in words like “tripod” and “tricycle” (but not, say, “trip” or “pastries”), using our knowledge of the fact that Greek “tri” and English “three” are etymologically related.
[Unger, 2004, p. 4]
The above example is actually quite similar to how kanji practically functions in Japanese. If we stop and think about the origin of the “tri” in “tricycle”, we would be able to connect it back to the meaning of the “three”, but we don’t normally process it any further than the simple word “tricycle”.
Think about how you read English day-to-day. When you see words like automobile, automatic, or autobiography, do you stop and think about how “auto” means “self” and then piece the word together? In general, you probably read “automobile” as one word. You might even imagine a car in your head as you think about this word. If this is how we process written language in English, why would Japanese be an exception?
In his book “Ideogram”, Unger references a study by Alan Horodeck that shows that how Japanese writers and readers encounter different types of “spelling” errors (table below) [P. 48]. In sample one, they identified writing errors where writers chose an incorrect kanji that had the same pronunciation, but an entirely different meaning 10 times more frequently than any other writing error.
In sample three, they replaced the second kanji with a character that has the same pronunciation, a similar appearance, but a totally different meaning.
In sample four, the second kanji is replaced by a kanji with a very similar appearance, but a totally different reading and meaning.
If the meaning of the kanji was truly the most important aspect, we would expect readers to notice the mistake in both samples at the same rate, as they both have incorrect meanings. However, mistakes like type three where the kanji had the same pronunciation and similar appearance were overlooked by readers five times more often than any other kind of mistake in the study. This lends evidence to the case that kanji is read primarily as a phonetic device, for its sound, and any meaning is secondary.
|Error Type||Kanji||Pronunciation (reading of 2nd kanji italicized)||Word Meaning||“Core meaning” of 2nd kanji|
|2. S-F-M+||迅早||jinsoku||non-word||Early, Fast|
|Error Type||Kanji||Pronunciation (reading of 2nd kanji italicized)||Word Meaning||“Core meaning” of 2nd kanji|
- Error type key: S = sound, F = form, M = meaning
- S+ false kanji can be pronounced like target
- S- false kanji cannot be pronounced like target
- F+ False kanji looks like target
- F- False kanji looks different from target
- M+ False kanji has “core meaning” similar to target’s
- M- false kanji has “core meaning” different from targets
But what about homophones?
This is probably the most common argument I hear In favor of kanji, so I’m also going to spend a large portion of this post debunking it. Homophones are words that have the same pronunciation, but different meanings. If you’ve studied Japanese at all (or just watched enough anime), you’ve probably run into some cases like the word “kami”. It has three commonly used meanings: hair, God, and paper. How can we possibly know which one someone is talking about without using different kanji for each one? Homophones are indeed a logical concern and Japanese really does have a lot of them.
First, let’s take a step back. If homophones present such an immense challenge for the Japanese language, how do people manage to speak to each other without getting confused?
Ask a Japanese person if you can: have you ever been confused which ”kami” someone was talking about? Did you think they were talking about cutting paper instead of hair? Or cutting God? I’d be willing to bet that the average Japanese person doesn’t even give spoken homophones a second thought.
Japanese speakers differentiate between spoken homophones the same way we do in English, through context. Context is what allows language to be language. Without it, speaking would devolve into indistinguishable hisses and hums. Even if the context is as simple as “I’m in the United States. People speak English here”, that still establishes a set of expectations for the participants in a conversation.
Japanese does have a lot of homophones, but so do many other languages, including English. The Oxford English dictionary has identified 464 definitions for the word “Set”, distinguishing it as the English word with the most definitions. The word “run” comes in second place with a mere 396 definitions. But for the sake of keeping this post to a reasonable length, let’s examine a word with not quite as many meanings.
If we look at just the subset of the word “pitch” as a noun, we can easily identify 8 individual meanings according to the Oxford English Dictionary. Have you ever thought twice about which pitch was being referenced when you heard the word? How many of these meanings did you already know?
1. Sporting grounds (soccer, rugby, etc)
2. Degree or strength of a feeling/activity
3. Pitch of sound (think music note)
4. A sales pitch
5. A baseball pitch(throw)
6. The black sticky stuff
7. The movement of a ship up and down in the water or an aircraft in the air
8. The degree to which a roof slopes
Interestingly, Japanese already has indistinguishable homonyms (same pronunciation/spelling, different meaning) with loan words like “keesu”(ケース）or “torakku”（トラック）.”Keesu” is the English word “case”, but is used to mean both a container to put something in and as an instance of a situation. “Torakku” is used to represent both the English words “truck” and “track”; thus actually introducing ambiguity compared to the original English spellings. Since these loan words are written using Katakana, the only way for Japanese readers to distinguish which meaning they convey is to judge based on the surrounding context.
Kisha no Kisha ga Kisha de Kisha shita
“Your company’s journalist returned to the company by (steam) train.”
This is a famous sample sentence used by developers at Canon back when word processors were groundbreaking technology. This may seem like a troublesome sentence to parse without kanji, but thankfully, the sentence doesn’t actually make much sense if you mix up the meanings. Functionally, this isn’t much different than the English sentence “A canner can can anything that he can, but a canner can’t can a can, can he” . Hard to look at, but not actually that prone to misunderstanding.
While I’m sure non-kanji homophones would allow us to make any number of ambiguous sentences in Japanese, it’s rather trivial to do the same in English as well. A look at the samples below should serve as a reminder how we ultimately require context to make sense of language.
“She was driven from her home by her husband”
Did she get a ride from her husband or did he kick her out of her house?
“I saw a man on a hill with a telescope.”
There’s a man on a hill, and I’m watching him with my telescope.
There’s a man on a hill, who I’m seeing, and he has a telescope.
There’s a man, and he’s on a hill that also has a telescope on it.
I’m on a hill, and I saw a man using a telescope.
One writer, James B. Hobbs, was able to identify 3,625 homophone groups in English . Japanese may have more meanings assigned to individual words, but a simple examination of English makes it clear that there is no shortage of homophones to deal with. Considering the ease with which homophones are understood in spoken language and the existence of so many homophones in English and other alphabet-using languages, I think it’s fair to say that the role of kanji in distinguishing homophones has been greatly exaggerated.
Kanji can actually introduce confusion
Some kanji can actually require context for you to decipher the correct reading and meaning. The kanji 大人気 has two common readings: “daininki” and “otonage”. The meanings are “very popular” and “maturity”. Without the surrounding context, the reader has no clue which pronunciation to use. This is no different than English words such as “bass”, “lead”, or “bow”, but in the case of Japanese, these would be avoided using an existing pronunciation-based writing system like hiragana.
The blind man can be better educated than his more fortunate brethren who are endowed with good sight; for the former, by acquiring the forty-seven letters of the I-ro-ha syllabary, through the Braille system, can read history, geography or anything written in that system; whereas he who has eyesight cannot read the daily papers unless he has mastered at least 2000 characters.
-Nitobe Inazo (Writer and diplomat. The man featured on the 5,000 yen note)
Another way that kanji can be unnecessarily confusing is through the use of different kanji for very similar meanings. Let’s look at the kanji for ”hot”. Well, that depends, is it hot to the touch or hot as in climate? Which one do you use to talk about emotions? When speaking, this isn’t an issue as they would all be the same word: “atsui”. But in writing you have to learn and differentiate between these two different characters.
|暑い||Hot climate/air||計る||Measure time or temperature/judge||変わる||Change (into)|
|熱い||Hot to touch/emotions||測る||Measure length/land||代わる||Take place of/instead|
|量る||Measure amount (weight/rainfall)||替わる||replace/succeed|
We have modern kanji abolition case studies
Luckily, we don’t have to sit here and wildly speculate what kanji abolition might look like. Within the past century, we’ve seen two languages almost entirely abandon Chinese characters in their writing systems. Both South Korea and Vietnam underwent language reforms and switched to new writing systems that do not rely on Chinese characters. Due to it’s unique political climate, let’s skip over North Korea for this discussion.
The most recent and dramatic example of this can be seen in South Korea after World War II. Before implementing their post-war language reforms, it was common that you would see a mix of both Hangul3 and Chinese characters (Hanja) in Korean writing. After the war, South Korea would earnestly begin phasing out Chinese characters in the 1970’s and they would become almost negligible by the year 2000. It took about five centuries for Hangul to become the dominant writing system in South Korea, but Chinese characters have now been relegated to minor traditional uses for things such as historical names.
Another argument I’ve heard thrown around is that kanji is needed for differentiating the names of people. However, Korea somehow manages to do this even though just the three surnames of Kim, Lee and Park combined account for nearly half of the last names in use in Korea.4 As for homophones, it would seem that it has its fair share of them, yet I am under the impression that they get by just fine.5
Up through World War I, Vietnam also used Chinese characters for writing. They mainly used a Chinese characters based logographic system called Chữ Nôm6 for writing the Vietnamese language, but many formal documents were still written using Literary Chinese7. Starting in the 1500’s, Portuguese and French missionaries began using Latin script to record Vietnamese, as the existing writing systems were too much of a barrier for them to overcome. This Latin-based writing system would go on to become the favored method of writing Vietnamese as they began to push for Independence after World War I.
In both countries, this transition has come at some cost. By switching away from their historical scripts, it does introduce a hurdle for anyone wanting to read historical texts. This would indeed be a downside if Japanese were to transition away from using kanji, but it does have its limitations. Native English speakers struggle to parse texts as recent as Shakespeare (1500’s) while deciphering something like Beowulf (8~10th century) is nigh impossible. This is much the same for Japanese speakers as well. Without specialized study, the average Japanese reader cannot make much sense of classic works like The Tale of Genji (1021) or The Tale of the Bamboo Cutter (900’s).
Surely, there are many people who value the cultural history and aesthetic values and would mourn the loss of kanji for this reason alone. Additionally, due to its function combining multiple syllables into a single character, replacing kanji with a syllabary or alphabet alternative would likely increase the length of written Japanese. You can see this in action if you take a look at Japanese Twitter compared to English Twitter. Japanese users are able to pack much more information into a single tweet compared to us English speakers.
If nothing else, Korea and Vietnam have shown that abolishing Chinese characters is possible from a purely technical perspective. There were most likely a number of complications and consequences from these movements, but both languages have continues to exist and be functional in their new(and improved?) form to this day.
So what can be done?
In the end, I’m not very hopeful for a revolution in the Japanese language. Hell, the United States passed the “Metric Conversion Act” in 1975, and we all know how successful that has been. Converting something as limited in scope as a measurement system would be a minor task compared to overhauling an ancient writing system.
Here are a few topics i did not get around to covering in this post:
- Historical value of maintaining a similar writing system
- Cultural and aesthetic value of kanji
- The burden kanji places on school children
- The barrier kanji creates for learners and how that could impact foreign trade
- Possible typing speed efficiency gains by eliminating kanji conversion
- Benefits of sharing characters/meanings with Chinese
- China’s own reform of the characters – “Simplified”
While I think we’ll continue to see kanji in use for decades, if not centuries to come, the way Japanese speakers process and think about language has certainly changed over the last few decades. As word processors, computers, and western culture have become more prominent in Japan, knowing the Latin alphabet has become a fact of life. In fact, most Japanese people (especially younger generations) type on computers in using the standard QWERTY alphabet layout to type out the hiragana to convert to kanji. Similar to how we’ve come to rely on spellcheck in English, we are seeing more and more Japanese natives forgetting how to write kanji as they have come to rely on the computer converting the characters for them.
Darwinism provides us with a useful lens with which to examine language evolution. Following the principles of Darwinism, survival does not indicate anything inherently good or bad. It simply means that any evolutionary changes were useful enough or benign enough to allow the genetic code to be passed along. We grow up speaking the languages we do because they were passed on to us, not because we chose it for its superiority as a method of communication.
- Kanji Abolition Theory(Japanese only): https://ja.wikipedia.org/wiki/%E6%BC%A2%E5%AD%97%E5%BB%83%E6%AD%A2%E8%AB%96