Hacker News Reader

Comment by latentsea | Hacker News Reader

I don't think Kanji makes Japanese easier to read. Korean, ditched it, and it's comparatively easier to read.

Not for beginners. But for fluent speakers, kanji are very necessary to distinguish all the homophones.

E.g. 効果, 硬化, 降下.. are hard to learn, but they're clear. こうか is much easier to learn, but it could mean any of 10-15 unrelated things.

All of which native speakers have no trouble distinguishing between in conversation when there are no Kanji anywhere. It's the same with homophones in any language, usually the context makes it clear because the alternatives don't fit.

The homophones in Japanese and Korean pretty much all come from the vocabulary they share with Chinese which makes up the bulk of the vocabulary for both those languages.

One doesn't use Kanji anymore, and no one seems to struggle to read it?

Japanese on the other hand I have seen even natives struggle to read. Heck even the existence of furigana in novels is an admission of this.

Written kana drops intonation information that's present in speech. Writing with kanji makes up for this, and also allows for more complex sentences that aren't as common in spoken Japanese.

I personally find the most difficult part of reading kana-only text to be detecting word boundaries. It's much easier when kanji is used, and I'm not even a native speaker.

An English analogy isthatyoucouldwritewithoutspacesandbeunderstood but it's more difficult to read and unnatural.

Young gen-z types on Japanese Twitter abbreviate everything, but even they don't drop kanji.

That is what whitespaces are for, which Korean also uses.

Adding whitespace is a pretty simple solution. Heck if you really, really absolutely needed to resolve tonal ambiguity in kana you could add something to Kana to do that. That'd enhance the readability even further since, it's basically impossible for foreigners to learn correct intonation in Japanese unless they explicitly study it and that's on top of memorizing all that Kanji, but it would become explicit. I can recall exactly once in the last 10 years having a conversation where the there was ambiguity between two homonyms and someone asked a clarifying question to resolve it. The vast majority of the time it's just clear from context.

So.. I would say even that ambiguity isn't something people would actually have much a problem with.

> One doesn't use Kanji anymore, and no one seems to struggle to read it?

Chinese/Japanese has a level of written mutual intelligibility. Korean lost it.

> Japanese on the other hand I have seen even natives struggle to read.

It's like a native English speaker encountering new vocabulary. Happens quite often.

> Heck even the existence of furigana in novels is an admission of this.

I'd agree that manga use of furigana helps (perhaps school-aged readers) reading, but furigana in novels are standard tools in the language that authors can use to achieve some effects that is hard to describe to non-speakers.

Sometimes furigana can be used artistically,sure, but that's the exception to the rule and it's by and large a reading aide in the vast majority of cases, and the inclusion of it in novels aimed at adults indicates that without it the author expects a certain percentage of readers may struggle with how to read the Kanji otherwise.

Why does this tool in the language need to exist? The answer cannot be because Kanji make things easier to read, else you wouldn't need tools to help you read Kanji you at times otherwise wouldn't be able to.

If you come across a word you don't know as either a native speaker of English or Korean, you can at least sound it out, which ups the probability you can connect it with a word you've heard before, otherwise since you know how to type it out it's trivial to look it up in a dictionary. If you come across a word you don't know in Japanese as a native speaker and there's no furigana it's a guessing game. The meaning is slightly more obvious to you, so you might be able to guess, but if you can't guess and you care to know and the word is in print then looking it up becomes a bit more of a pain.

Korean didn't completely lose the mutual intelligibility aspect entirely since the underlying pronunciation of the words still remains and can be used to correctly guess the word in a lot of cases. Like 시간 and 時間 as an example, but there's many, many words I've been able to guess in Korean based off knowing Japanese. I was able to score 50% on TOPIK II reading exam after only having studied Korean for 4 months in large part because of this.

> Why does this tool in the language need to exist? The answer cannot be because Kanji make things easier to read, else you wouldn't need tools to help you read Kanji

This just isn't true. Even most native JP speakers agree that kanji are oppressively hard to learn and remember, so if it were feasible to get by with kana alone, then at least some native speakers would do it in some contexts. But outside of language learning it's virtually never done, and there's a reason for that.

Also, I think you're overlooking that Chinese and Korean have a lot more vowels/tones to work with than Japanese. There are a lot of Chinese-derived compound words that are homophones in Japanese but not elsewhere.

So how are homophones in Japanese understood when expressed verbally?

How much trouble do you actually have with them?

"Science" and "chemistry" are homophones in Japanese: We have special disambiguation reading for "chemistry", bake gaku, used only when misunderstanding is suspected.

There are numerous other examples. Those are all unnatural sounding, mostly industry/field specific, and not replacing the main homophone readings.

1: https://dailyportalz.jp/application/files/thumbnails/width64...

> It's the same with homophones in any language,

It really isn't. In conversation fluent JP speakers tend to avoid compounds that would be ambiguous, or add distinguishers like "学校の校歌". Honestly, try converting a paragraph of text from a newspaper to all kana, and having a native speaker read it.

Try have them be educated in a kana only system and then have them try read it. They'd probably do just fine. You'd expect anything you've spend a decade doing to be easier than the thing you've spent much less time doing.

かんちょうがかんちょうでかんちょうにかんちょうされた。 Is probably a sentence that definitely requires Kanji to understand the precise meaning given how many homophones かんちょう has, but it's a toy example.

> Try have them be educated in a kana only system and then have them try read it.

I can't, because there are no native speakers who learned that way, as I'm sure you know :D

But there are many learners like that, and my experience in Japan is that anyone who doesn't learn kanji has a very low ceiling on their vocabulary, even if they use the language daily for decades.

Because the reality is that it's hard to memorize 1K kanji, but if you do it then it's relatively easy to learn 10K+ compound words. Without kanji, to reach fluency somebody would have to memorize 5-10 completely unrelated meanings for "kouka", then 5 more for kakou, and so on for every combination of common single-kanji readings.

I mean - if you're in Japan, you surely know people who try to get by without kanji. Do you know any who've reached fluency? Like who could use and understand 5-6 different "kouka"s without any idea of the kanji they use? If your premise here is true then people like that should be the norm, since learning would be so much easier for them compared to those wasting their time on kanji?

So... how do Koreans and Korean learners do it? Not to mention other languages that used to use Kanji but dropped it?

You have to imagine the entire education system and everyone in it got changed to Kana only and Kanji was subsequently removed from modern literature from that point onward. That's the thought experiment. It was even tried successfully in places...

I don't know anything about Korean. It's a different language, and I'd imagine that some of the things I'm saying here don't apply to it, but that's a guess.

For thought experiments, you're assuming your consequent - obviously a society raised without kanji would get by without kanji, though I think the language would have to change somewhat. But for non-hypothetical people that exist now, I think the things you've been saying about kanji not making it easier to read Japanese just aren't true.

Japanese learning materials for Japanese 5 and younger are hiragana only, and I remember finding it annoying.

As someone who isn't even fluent in Japanese, I find it easier to read text with Kanji (in contexts where I am relatively fluent) than without.

Japanese without Kanji is like English (or any Latin alphabet language) without punctuation or spaces or capitalization. And also if English had a ton more homophones. You basically need to word-split and disambiguate as part of the reading processing; it's painful.

This is because this is how you've trained your brain to process Japanese.

Korean includes white space for this reason, and since everyone gets used to reading it with no Hanja, they don't find it awkward.

So, it's perfectly possible to ditch Kanji and add spaces and have a way more accessible language.

That doesn't solve the homophone issue though.

usually a parenthesis with corresponding character is used to be explicit or to avoid confusion but its strictly for chinese loan words like:

ex) 시간 (時間)

compared to furigana (note that its not even possible to display the phonetic hiragana): 第二巻

younger generation are no longer learning Chinese so more English/European loan words are directly used which ironically fixes the issue.

it is impossible to converse in Korean without using English/European loan words

ex) 아이러니 (irony)

This allows new ideas/concepts to quickly disseminate in the collective Korean psyche. Constantly new words are being invented, slangs used by primary/middle school are unknowable.

Abbreviation/concat combos to create totally new words:

ex) 씹상타치 (way **ing above average) also written as ㅆㅅㅌㅊ (wfaa) literally means

If you look at old Japanese video games before the hardware could do Kanji, they used spaces to separate words. But when the game was capable of Kanji, the spaces went away.

Without Kanji, it severely degrades readability. One has to reconstruct the word from syllables, which introduces another layer of cognitive load.

In Korean, it works similarly as well though, most people nowadays are quite used to not incorporating Hanja in sentences over multiple decades, to the point where it would be impractical to mingle Hanja in Korean.

Yes, so basically the arguments around lack of Kanji leading to worse readability are actually hitting upon the fact that readability suffers short-term not because Kanji enhances readability,but because they're simply not used to processing the language only through kana, and that were they to acclimatize to that, it becomes readable again and in fact easier to read than before.

I do wonder whether kana might be slightly easier to read if ん were written as part of the preceding syllable like it is in hangul.

Kana would be slightly easier to read if we spent as much time reading in Kana as we have in Kanji.

Hangul has some funny rules around patchim that need to be memorized. Kana does a great job avoiding this, so on balance kana is probably just fine compared to Hangul.

I don't think so. Kana just don't have enough entropy compares to Kanji. A kanji can be compose with up to twenty strokes with high variety of stroke patterns. Those excessive complexity make it identifiable even in extreme situations. (Blurry or tainted or whatever situation). In some case, a kanji with half of its size masked still be decoded without any ambiguity. But this will never work with an voice based language.

> Without Kanji, it severely degrades readability. One has to reconstruct the word from syllables, which introduces another layer of cognitive load.

Which is what every language with an alphabetic writing does, and it works just fine. It is not "another layer of cognitive load", it is just a different layer, one that can be said to be much lighter or other languages would not have switched centuries/millenia ago.

The real problem of Japanese is the massive amount of homophones coming from Chinese. It is already a problem in Chinese, but even worse in Japanese due to the smaller phonetic repertoire.

> Which is what every language with an alphabetic writing does, and it works just fine. It is not "another layer of cognitive load",

Disagree. Once you get accustomed to reading kanji, and did not learn to visually parse (except very briefly as a young grade school child) nearly all of the words that you see regularly as logograms first, and groups of sounds second, the experience would be akin to reading English while afflicted with a strangely selective amnesia hole for entire words. Such that reading a word like 'shoe' would not instantly evoke an association with a piece of footwear but would have to be (admittedly very rapidly) sounded out letter for letter each time, instead of scanning the entire unit as a whole.

That's what reading a word normally represented by a familiar kanji character but "expanded" into hiragana feels like, and slightly more pronounced if it's, for some hipster reason, written in katakana.

> Disagree. Once you get accustomed to reading kanji

And how many years does it take to? What about words that you have never seen before? What about ambiguous or uncommon readings that require furigana even for fully educated adults?

As I said in the sibling comment, I live in Japan and every Japanese person I have met complaints about the massive effort it took them to learn how to read and write.

> every Japanese person I have met complaints about the massive effort it took them to learn how to read and write

This stupid phenomenon is due to the fact that Japanese Gov decided to teach only arbitrarily 1000 kanji to school kids and this number decrease every 10 years.

https://ja.wikipedia.org/wiki/%E5%B8%B8%E7%94%A8%E6%BC%A2%E5...

You can see how stupid it is.

> People complain that kanji being used in the prefecture are not include in the list (no included so no obligation to teach in school) e.g.阪鹿奈岡熊梨阜埼茨栃媛, so the gov finally add those to the next revision.

While at the same time Chinese people are learning 3X more. No one ever complain about the difficulty of Chinese character after all.

Sounds like an education problem. Traditional Chinese, which is the defacto language used in Taiwan doesn't even have any comparable phonetic alphabet (beyond a phonetic pronunciation alphabet in the form of bopomofo) such as hiragana and katakana.

It's effectively "all kanji" as it were, and yet Taiwan has one of the highest literacy rates in the world, and I never met any Taiwanese when I lived there (for years) that complained that Chinese was too difficult.

In Chinese tho with extremely rare exceptions all the characters have only one reading.

Japanese has onyomi and kunyomi. The onyomi also come from different periods in Chinese so there's multiple onyomi for most Kanji.

Then you get two Kanji words that come in all varieties. Most are onyomi + onyomi, but you get some that are onyomi + kunyomi or kunyomi + kunyomi or kunyomi + onyomi.

There's also not really any solid rules to it, and when there are, there are plenty of exceptions.

It's a real nightmare of a system. A fun one though.

And then you have nanori, the non-standard readings used for people and places names that are impossible to read without furigana or already knowing the name. One that really surprised me was a village called 愛子 (a common female name read as "Aiko") near Sendai but in this case read as "Ayashi".

> The real problem of Japanese is the massive amount of homophones coming from Chinese

So you're saying that verbal communication doesn't work in Japan and everyone just texts each other? "I'm sorry Mr Honda Kawasaki but due to how Japanese language works it's impossible for me to tell whether you want to buy three oranges or cook prostate cancer, please send me a letter" "Okay I will fight the sky colander"

Most countries at some point had to simplify their languages in order to promote literacy. Korea didn't ditch hanja just for shits and giggles, it did so in order to make it easier for schools. Japan never really had to face this problem at a scale that required complete removal of kanji because by the time people got such ideas Japan was already quite literate, so kanji stuck around. Plus, Japan is an extremely conservative society, they only ever change anything once all other options have been exhausted.

Same reason why English spelling is so ridiculous. It's not that English is such a unique language that it absolutely requires a spelling system that doesn't make sense and effectively forces everyone to memorize each word's spelling aside from it's pronounciation (wow just like kanji), it's just that English spelling has never been a problem to a degree that required a systematic solution, so now we're stuck with what we have. If we suddenly decided to make a giant reform of English spelling to have it reflect actual pronounciation, the resistance would be equally giant.

Don't take me wrong, I fully agree with you.

I actually live in Japan, I am trying to learn the language, and it is royal PITA. Heck, every single Japanese person I have asked has complained about their language being so ridiculously difficult. They wish their language was easier, but as you say it is also such a conservative society that it will never change.

In comparison, my mother language is Spanish, a language with a highly phonemic spelling. My girlfriend is trying to learn it, and she always commends how once you learn a few basic rules, you can read anything.

Korean is easier to pronounce, but Japanese is easier to read. Try reading a Japanese children's book that doesn't use kanji. It's excruciating.

That hasn't been my experience of both languages.

Most HN users who are coders should be able to read and write Korean within a few hours...

Good luck with any other language. Japanese is tougher since you have to memorize thousands of what are basically hieroglyphics and learn two separate alphabets (one for native Japanese, another for foreign).

With Korean you can express many sounds phonetically (with some abstractions for non-korean sounds)

I don’t speak or read Korean but I am studying Japanese.

I think GP was trying to say that kanji helps:

たまねぎ玉ねぎ

いつつ五つ

In both of these examples the words are the same. I’m still early enough in my studies that I don’t know the rules of when someone might choose to write one way or the other, but I’ve seen examples of ads that “spell it out” with hiragana. (Which is harder for me to read, which is what GP was trying to convey imo)

I've been fluent in Japanese for over a decade and am about 6 months into studying Korean.

I understand the issues of Kanji vs no Kanji well. Korean successful ditched it, isn't painful to read, is far more accessible to read for beginners, and doesn't suffer from an extreme long-tail of ambiguous difficult readings like Japanese does.

With Japanese no matter how many vocab you learn, you hit new words like 仲人, think you know how to read it correctly, can never quite be sure, look it up every time as a consequence and are surprised often enough at the reading that you never really settle into a sense of confidently being able to read new words correctly. It sucks.

In contrast I was able to score 50% on the reading section of TOPIK II after only 4 months of study.

So, on balance I'd say reading Korean is way easier because they ditched Kanji.

Thanks for clarifying. I don't disagree that Korean does seem easier, but I still think you are slightly confusing what the GP is trying to say.

I've never seen "仲人" before but I'm pretty sure it's some kind of person.

https://jisho.org/search/%E4%BB%B2%E4%BA%BA

仲 go-between (which I never saw before but I knew the radicals as "person middle" but didn't know what they were combined, but this one made great logical sense)

人 person

So while I can't "read" it (in Japanese) I can know what it means pretty confidently as kanji very regularly mean the same thing in compound words.

If I saw "なこうど" I'd have no idea because those hiragana don't mean anything to me until I learn the meaning.

Am I making sense? Like the first time I saw 花火 I knew "flower fire" and was able to guess firework.

same with 大人 being adult.

I'm not saying you are wrong that Korean is easier -- I'm saying, learning kanji can make it easier to understand a lot of meaning with never actually being able to "read" the words. and the reading is absolutely hard because of kunyomi and onyomi etc etc.

Being able to guess the meaning of new words was neat earlier on in the Japanese journey, but in the end the problem of "gah but the how hell do you actually read this?" was a greater detriment than that was a benefit.

In contrast if I saw なこうど I could at least be perfectly confident I was reading it correctly even if I didn't know what it meant. Sometimes I may be able to guess from the context at least partially what it means, but if not, then I could simply opt to move on having collected an instance of seeing the word. I might then later hear it elsewhere, or perhaps see it again and if I encounter it enough times I can get curious and look it up.

I could do the same thing with Kanji except I'd have to look it up anyway to be confident I was reading it correctly. Else I just don't know what the word is, so its harder to mentally file it anywhere in my brain. I found this lead to a very long-tail of pain when reading Japanese that didn't abate even when I got up to around 17k vocab in Anki after which I just said bugger it.

So, on balance I prefer the set of problems that no Kanji poses over the set of problems that Kanji poses.

I vastly prefer the ability to potentially infer the rough meaning of an unrecognized word, then the ability to pronounce it.

As an ESL CELTA certified teacher for years, their rubric also seems to back this up in order of relative importance: it's meaning, then form, then finally pronunciation.

I don't know if a rubric for English is as applicable to Japanese.

You're just trading one set of problems for another. Those aren't even problems English has.

> I've been fluent in Japanese for over a decade and am about 6 months into studying Korean.

Me too! Nice to talk with someone with a similar background :)

> you never really settle into a sense of confidently being able to read new words correctly. It sucks.

Native speakers don't magically gain the ability to read correctly every new word. So it is fine to hit the dictionary every now and then!

In the case of 仲人, I can guessimate 仲 (naka) and 人 (hito), but recall that 素人 and 玄人 are pronouced <long vowel>~uto. I would try to pronounce it as nakouto (the correct spelling is nakoudo). So people do gain a heuristic for reading.

Also Kanji provides a mnemonic device after learning the meaning of the word. (One who makes/improves 仲).

>So, on balance I'd say reading Korean is way easier because they ditched Kanji.

Yes, unless you're Chinese or Japanese speaker ;)

I think kanji can make more clear the meaning if it can be understood, but not for pronouncing. Kana is other way around. (In case it is difficult, it is also possible to add furigana.)

I think it is beneficial to have both in Japanese writing.

Reply on news.ycombinator.com