I have no trouble reading but writing kanji has become a problem. I never need to do it and I can’t remember how to write kanji I have no trouble reading.
It’s Japanese people too, to a lesser degree. My own Japanese wife has to pause to remember how to write something every now and then.
This is probably why Japan still adamantly emphasizes writing.
Written resumes/ fax machines ... remain the norm, and while this may seem anachronistic for the rest of the world (pretty much all of which uses either (semi-) phonetic scripts derived from Aramaic or from Brahmi), it makes sense after you come across the Chinese characters.
A similar thing happens with all kinds of iconography, from flags to logos. People can easily recognize many logos, but when asked to draw them they often can't come very close.
Is ワープロ馬鹿 really a term used by native Japanese speakers? As far as I can tell it only really shows up in Japanese->English dictionaries and English forums (see https://www.google.com/search?q="ワープロ馬鹿"+-a+-the).
I enjoyed this read, but I am noticing that people who claim to have aphantasia seem to write about themselves and their experiences an awful lot. I doubt the phenomenon is real.
Some people doubt that sun-sneezing is real, so I can entertain the possibility of being wrong. But sun-sneezing is trivial to demonstrate to doubters, and it doesn’t confer any “I’m special points.” No one would pretend to have it, unlike aphantasia.
This reminds me of one time I mentioned to someone I had aphantasia and their response was “how do you spell!?” Seems wild to me that some people see words in their head to spell them but I guess at least one person does. I do wonder if that means they’d have better kanji recall for writing.
When learning Japanese, I purposely chose to _not_ learn how to write any of it by hand. As the author notes, writing (by hand) is in fact a separate skill from reading. So I decided I would not invest my limited time, motivation, or brain space to writing.
Overall it's been a successful approach, and I recommend it to new learners unless they have a particular interest in being able to write by hand or they feel strongly that writing the characters helps them remember them.
It's only rarely that I have to write anything other than my own name in Japanese. I've practiced my address but writing it in English is fine in 99% of situations. Being able to write properly would save a little embarrassment, but I still believe my language learning time would have a much higher ROI in other areas.
It is a separate skill from reading, but I think it's still useful.
At the very least learn the strokes of common radicals. In my experience things like denshi jishos can be VERY picky about how you input them. It makes word lookup much faster IMO anyways.
I think your advice makes a lot of sense for most learners: prioritize the skills you'll actually use, and don't feel guilty about skipping handwriting unless it personally matters to you
Perhaps if there were fewer radicals this would be less of a problem. Many thousands of characters could probably be generated from a small number of radicals.
32 radicals are enough for ~75% of characters though. My biggest problem with writing is that I default to writing fairly small so if I have to do a character with more than 12-15 strokes it's either way too big or hard to read later.
If native speakers are starting to have character amnesia too, does that suggest in the long run you would expect the writing systems to simplify towards the phonetic syllabaries? Or is the fact that we have computers as a mediating tool going to forestall that and just make things weird?
IMEs have actually caused the opposite a little in Japanese, because you type the phonetic pronounciation and get a list of possible kanji, then with computer use "writing" kanji really means "recognising" and so people will use kanji that they never would have bothered with remembering how to handwrite.
Being able to read but not write Kanji is so common that it's a meme amongst Japanese people -- to the point where it's a game. For example, here you can watch some Japanese television people play a game where they compete to write words in 10 seconds or less:
Completely off topic but that is the shakiest camera work I have ever seen, and I learned to shoot before gimbals were cheap and easily available. For that matter, it's been easy to automatically correct that sort of thin in post for years now. How were they not embarrassed to upload something so unwatchable?
> In other words, what feels like a single, monolithic "literacy" ability is actually two distinct skills, each exercised in different instances and each capable of improving and decaying on its own.
This dissociation has been used to test theories of hemispheric specialization. A good overview is in Neurolinguistic Aspects of the Japanese Writing System by Michel Piradis (1985).
Nice post! Enjoy your blog's overall aesthetic too. Perhaps correlation in sense of style, though, as I also used RTK to learn to write Kanji, loved it, and now, ~15 years after that escapade, am kind of in a similar bucket (can write some characters, but mostly just read). I still think RTK great overall method and would do it again!
Also, shoutout to Fabrice, creator of Kanji Koohii -- that was my first foray into SRS back in ~2007/2008, after which I found Anki (pre-mobile).
Very interesting, definitely the first step towards writing a character for me is picturing it in my head. If it's particularly challenging I might, still in my head, project it onto the page and that seems to give my brain the spatial data to begin translating it to real world movements with the pen.
But even still I also can barely write maybe 5% of the kanji I can read. As well words are often made of multiple kanji, but if you showed me the kanji separately I don't always recognise them as part of a word I do know. Recalling a kanji into my minds eye doesn't seem to be part of the skillset of reading, maybe just a by-product of long term repeat exposure.
That reads as a foreigner suggesting to abandon many thousands years long history and culture just because the characters are too difficult for him.
It doesn't even make sense because today we have computers and ML. If you want to write something, you just type the spelling, and the program automatically converts it into proper characters. And if you don't understand the character, AI can translate it for you. I am sure this can be integrated into electronic glasses.
Japan has had pro-romanization societies since the 1920's, and even during the last attempt at large-scale script reform after the war, it wasn't just the Americans pushing it: Many Japanese were enthusiastic about moving to a phonetic script because they perceived it as more efficient and modern. Likewise, not every American administrator was in favour of reforming away the kanji, far from it.
J. Marshall Unger's Literacy and Script Reform in Occupation Japan is a good treatise on the subject.
I also gave you two examples of cultures who already did it: Koreans did it autonomously, the Vietnamese did it during occupation. Both are successful today.
But yes "preserve history" at the expensive of daily life. I'm sure you speak Latin too, plenty of history lost there.
This is a common misconception until you study the language seriously. Japanese nominally has only 50 sounds, known as mora. In practice there are more because people smush them together, but this tends to be for very common words or dialectical variations by region. There are also spoken pitch markers, but they don't have any written indicia.
Because there is such a small number of basic sounds, homophony (words with the same pronunciation but different meanings) is absolutely rampant. This is not too big of a problem when speaking because the context is usually obvious, but writing anything abstract using phonetic script alone will quickly become problematic. It's also tedious to read long texts in phonetic script in the same way that it's tedious to read lots of upper case text in the Latin alphabet.
Another language feature is that Japanese compound words are often formed by just taking 1 or 2 more from several longer words. This would be like shortening 'American-Russian diplomacy' to 'amru diplomacy'. Very clear when written with specific characters, confusing if it is just phonetic. Then, word pronunciation changes within compounds, eg 川 kawa means river, but 山川 yamagawa is mountain river, and these changes are ubiquitous, probably 95% of words go through this phonetic alteration when folded into another word.
It gets more complex again because there are so many particles, words that indicate grammatical inflection of some kind and can drastically change the meaning of a sentence if you use the wrong one. The most common ones are just a single mora and most others are just two. But there are lots of particles. 10-15 essential ones, ~30 that are used all the time, ~65 you need to know for fluency, and more that 150 you might come across in written form.
I could go on and on. The basic reason for all this is that Japanese is fundamentally different from other languages because it developed on islands, and because the weather patterns around those islands made sailing in and out more difficult than islands in calmer latitudes separated by greater distances. They adopted Chinese characters because they didn't have a written language at the time they encountered them, then adapted those characters to develop the phonetic scripts. There's a huge amount of technical debt that would be impossible to unravel.
Do it like Korea
Korean has more different sounds, eg 14 consonant and 10 vowel sounds, vs ~10 consonant sounds and 5 vowel sounds in Japanese. You could also argue there was a higher level of literacy over a longer period, but that requires a lot of history and geography explanation to support - the point being that it would have been much harder to impose by imperial fiat unless it had been done much earlier.
This is very common in Chinese now. The older generation, many of whom didn’t learn pinyin, just use voice input to send messages; the younger generations just use pinyin input and similarly can’t handwrite beyond the simplest characters.
The phenomenon of forgetting how to write is called 提笔忘字 (tíbǐwàngzì - to pick up the pen and forget the character). It was previously covered here on HN: https://news.ycombinator.com/item?id=41959256
Writing out Chinese characters definitely helped me learn to read them as well (not surprising, since engaging multiple senses helps one retain information better), even though in practice I never had opportunity to write them out because I always used either a phone or computer.
Unlike the author, I found most of the mnemonics as much trouble to learn as the characters themselves, and soon stopped using that approach. It just didn't work for me.
> This is also why I believe that language is a bottleneck for thought. Most of what you remember is nothing like an approximate copy of the things you experienced in real life—even in the specific case of text, memory is not even remotely like a paraphrase of previously read words. Many of our thoughts happen in a highly abstracted and distilled form, interacting and connecting with each other as a network that simply cannot be faithfully converted into a sequence of words, however long.
Perhaps the most interesting quote in an interesting article.
I spent years in Taiwan studying traditional Chinese and even at the height of my proficiency there were plenty of rarer logographs that I'd frequently stumble over - only able to draw "blurry approximations" of them depending on my familiarity.
Coming from a phonetic language with only 26 letters, it was such a surreal feeling being able to effortlessly read a character but be unable to reproduce it.
> On the surface, this atypical trait seems to explain quite well why I can draw a blank when asked to write the kanji for "plant" (植) from memory. I don't see the character in my mind, so it makes sense that I can't reproduce it on paper.
While the author's aphantasia may have posed some recall issues - it wouldn't explain why they had ever been able to reproduce 植. Kanji has the concept of radicals AND stroke order. One could make the case that perhaps the author's motor cortex is simply storing the equivalent of LOGO programming language instructions for reproducing the logograph.
Take away your mind's ability to find and chunk (木, 十, 具) by showing them "radical"-less characters and I'm sure it would be even more difficult.
Writing characters without knowing the radicals is kind of like spelling without really clearly understanding the alphabet. The stroke order is easy to remember, the radicals make the full characters easier to remember.
What drove me crazy when I started to learn Japanese was that a Kanji characters can have so many different pronunciations. The most egregious example is 煙草, whose pronunciation is the really just tabacco(tabako, or タバコ). I knew the etymologies and the historical context on why Japanese evolved like that, but it's nonetheless hard for me to remember all the pronunciations, at least initially.
Also, I find that knowing Kanji is essential in appreciating part of Japanese culture. Take their addresses, for example: Kinukawa is really meaningless, but 鬼怒川 is such a amazing name. Similarly, Akihabara means little if all we know is the pronunciation, yet its Kanji 秋叶原 is such a beautiful and poetic name that invokes complex emotions.
That's why I focus on learning words rather than individual kanji characters. Kanji characters have so many different readings, and the Japanese language is full of exceptions that it's not worth the time unless you're passionate about it, eventually you will pick up the readings of most common ones and you will be able to guess new words automatically, writing them no thank you, I barely write in my own language without a keyboard.
82 comments
[ 2.6 ms ] story [ 80.6 ms ] threadIt’s Japanese people too, to a lesser degree. My own Japanese wife has to pause to remember how to write something every now and then.
Written resumes/ fax machines ... remain the norm, and while this may seem anachronistic for the rest of the world (pretty much all of which uses either (semi-) phonetic scripts derived from Aramaic or from Brahmi), it makes sense after you come across the Chinese characters.
https://magazine.adler.co.uk/promotional-idea/we-asked-100-p...
edit: s/word/term
Some people doubt that sun-sneezing is real, so I can entertain the possibility of being wrong. But sun-sneezing is trivial to demonstrate to doubters, and it doesn’t confer any “I’m special points.” No one would pretend to have it, unlike aphantasia.
Overall it's been a successful approach, and I recommend it to new learners unless they have a particular interest in being able to write by hand or they feel strongly that writing the characters helps them remember them.
It's only rarely that I have to write anything other than my own name in Japanese. I've practiced my address but writing it in English is fine in 99% of situations. Being able to write properly would save a little embarrassment, but I still believe my language learning time would have a much higher ROI in other areas.
At the very least learn the strokes of common radicals. In my experience things like denshi jishos can be VERY picky about how you input them. It makes word lookup much faster IMO anyways.
https://www.youtube.com/watch?v=pqQQqLno9hw
I (and every other learner) have the same problem. It's not special, and has nothing to do with aphantasia.
This dissociation has been used to test theories of hemispheric specialization. A good overview is in Neurolinguistic Aspects of the Japanese Writing System by Michel Piradis (1985).
Also, shoutout to Fabrice, creator of Kanji Koohii -- that was my first foray into SRS back in ~2007/2008, after which I found Anki (pre-mobile).
But even still I also can barely write maybe 5% of the kanji I can read. As well words are often made of multiple kanji, but if you showed me the kanji separately I don't always recognise them as part of a word I do know. Recalling a kanji into my minds eye doesn't seem to be part of the skillset of reading, maybe just a by-product of long term repeat exposure.
So time to sunset the system, surely? I don't know why so many countries are so obstinately hanging onto something so difficult.
Do it like Korea if you don't want to go the Vietnamese way.
It doesn't even make sense because today we have computers and ML. If you want to write something, you just type the spelling, and the program automatically converts it into proper characters. And if you don't understand the character, AI can translate it for you. I am sure this can be integrated into electronic glasses.
J. Marshall Unger's Literacy and Script Reform in Occupation Japan is a good treatise on the subject.
EDIT: Also, this was published in 1877 if I did my date conversions right: https://archive.wul.waseda.ac.jp/kosho/bunko17/bunko17_d0035...
I also gave you two examples of cultures who already did it: Koreans did it autonomously, the Vietnamese did it during occupation. Both are successful today.
But yes "preserve history" at the expensive of daily life. I'm sure you speak Latin too, plenty of history lost there.
Because there is such a small number of basic sounds, homophony (words with the same pronunciation but different meanings) is absolutely rampant. This is not too big of a problem when speaking because the context is usually obvious, but writing anything abstract using phonetic script alone will quickly become problematic. It's also tedious to read long texts in phonetic script in the same way that it's tedious to read lots of upper case text in the Latin alphabet.
Another language feature is that Japanese compound words are often formed by just taking 1 or 2 more from several longer words. This would be like shortening 'American-Russian diplomacy' to 'amru diplomacy'. Very clear when written with specific characters, confusing if it is just phonetic. Then, word pronunciation changes within compounds, eg 川 kawa means river, but 山川 yamagawa is mountain river, and these changes are ubiquitous, probably 95% of words go through this phonetic alteration when folded into another word.
It gets more complex again because there are so many particles, words that indicate grammatical inflection of some kind and can drastically change the meaning of a sentence if you use the wrong one. The most common ones are just a single mora and most others are just two. But there are lots of particles. 10-15 essential ones, ~30 that are used all the time, ~65 you need to know for fluency, and more that 150 you might come across in written form.
I could go on and on. The basic reason for all this is that Japanese is fundamentally different from other languages because it developed on islands, and because the weather patterns around those islands made sailing in and out more difficult than islands in calmer latitudes separated by greater distances. They adopted Chinese characters because they didn't have a written language at the time they encountered them, then adapted those characters to develop the phonetic scripts. There's a huge amount of technical debt that would be impossible to unravel.
Do it like Korea
Korean has more different sounds, eg 14 consonant and 10 vowel sounds, vs ~10 consonant sounds and 5 vowel sounds in Japanese. You could also argue there was a higher level of literacy over a longer period, but that requires a lot of history and geography explanation to support - the point being that it would have been much harder to impose by imperial fiat unless it had been done much earlier.
The phenomenon of forgetting how to write is called 提笔忘字 (tíbǐwàngzì - to pick up the pen and forget the character). It was previously covered here on HN: https://news.ycombinator.com/item?id=41959256
In my experience people often use input methods that aren't pinyin. I tend not to be able to type on other people's phones.
Unlike the author, I found most of the mnemonics as much trouble to learn as the characters themselves, and soon stopped using that approach. It just didn't work for me.
Perhaps the most interesting quote in an interesting article.
Leaves me speechless or something
Coming from a phonetic language with only 26 letters, it was such a surreal feeling being able to effortlessly read a character but be unable to reproduce it.
> On the surface, this atypical trait seems to explain quite well why I can draw a blank when asked to write the kanji for "plant" (植) from memory. I don't see the character in my mind, so it makes sense that I can't reproduce it on paper.
While the author's aphantasia may have posed some recall issues - it wouldn't explain why they had ever been able to reproduce 植. Kanji has the concept of radicals AND stroke order. One could make the case that perhaps the author's motor cortex is simply storing the equivalent of LOGO programming language instructions for reproducing the logograph.
Take away your mind's ability to find and chunk (木, 十, 具) by showing them "radical"-less characters and I'm sure it would be even more difficult.
Also, I find that knowing Kanji is essential in appreciating part of Japanese culture. Take their addresses, for example: Kinukawa is really meaningless, but 鬼怒川 is such a amazing name. Similarly, Akihabara means little if all we know is the pronunciation, yet its Kanji 秋叶原 is such a beautiful and poetic name that invokes complex emotions.
"I used to do drugs waaaay baaaack - there!" And he points to the side of the stage.