Hacker News Reader

Comment by shard | Hacker News Reader

In related news, the younger generation of kids in Asian countries are getting worse and worse at writing Chinese characters because they can just type the Romanized versions without learning how to write it. Personally I am able to read Chinese at a moderately high level (can read newspapers) but I cannot write most characters off the top of my head. This is what happens when you emphasize reading over physical writing with pen and paper.

View on HN

Is that a bad thing? You can still write in Chinese with an IME, and most uses are on some sort of phone or computer anyways. 27 year old me learned how to text in Chinese but skipped learning the muscle memory to write characters, first I'm left handed so stroke order was bizarrely awkward anyways, but also...it just didn't feel important anymore.

If you live in a Chinese-speaking country then, yes, it's a bad thing.

There are plenty of situations in which you still have to write; at the post office, applying for things, hospitals, places like that. Maybe not every day, I'll grant you that, but you are going to hit barriers when required to write your address or apply for a driving license or similar.

You don't want to be the guy painstakingly copying characters stroke by stroke off his phone while a line grows behind you.

I live in Japan and often regret that my writing is so far behind my reading comprehension. Like, it doesn't need to be AS good, but it would be better if it were close.

I lived in Beijing for 9 years and it didn't seem to be a problem. Yes, I occasionally had to really write something in Chinese (e.g. at the bank when making an account changed), but my hand was basically held while writing it. The post office and hospital never required me to write in Chinese.

My 69 year old father lived in Beijing for a few years recently and I'm pretty sure the company he worked for must have taken care of things like that for them because I can't imagine that man ever attempting to write Chinese.

I guess there's probably a pretty strong market for translators in the big Asian cities with American/European presence.

The company didn’t even have to provide anybody. There was always at least one English speaker at the bank, even though I could get by in Chinese verbally, it’s just one weird rule that requires me to write anything by my own hand at all (other than my signature). A lot of places are like that: they’ll easily take care of the paperwork for you if they know you can’t. Even for foreigners that don’t speak Chinese, they get by, especially in a first tier.

If you have a Caucasian appearance, might not part of it be that the Chinese simply excused you from having to write Chinese more than they would someone who appeared Chinese?

Of course it's a bad thing! Imagine not being able to write your own language.

I think some countries should just bite the bullet and start replacing their archaic writing systems with something more modern

Phonetical alphabet, standardized spelling, etc

Few people nowadays have the time and willingness to learn thousand of characters and their stroke orders. Not to mention the whole "every character has several unrelated pronunciations, but hey, context, right?!" of Japanese.

> archaic writing systems with something more modern

The Latin alphabet is an archaic system that barely worked for Latin and doesn't work at all for English. There's a reason English dictionaries have to include IPA.

> Phonetical alphabet

You say so while typing in an alphabet that doesn't have enough characters to describe all the phonemes of the language you're using, and which is used phonetically in name only.

> standardized spelling, etc

I don't know what gives you the impression that "some countries" don't have standardized spelling.

> Few people nowadays have the time and willingness to learn thousand of characters and their stroke orders.

You're literally talking about the writing system of the word's most widely used language. Obviously a plurality of people do have the patience to do so. In practice it requires roughly the same amount of memorization as English. The only difference is being that most of HN probably learned English early in life and have internalized things like "Peak Vs. Pique Vs. Peek".

Honestly it took me a few seconds to remember what pique even means.

Chinese characters (hanzi) are not an 'archaic writing system'. While hanzi allows for the expression of meaning in addition to pronunciation, which is unique amongst the other writing systems of the world, they are also more phonetic than is commonly perceived.

The majority of Chinese characters (more than 80%) are 'phono-semantic compounds' where 1 part of the character indicates the meaning and the other indicates the pronunciation. And the majority of these compounds follow a surprisingly regular pattern: for instance, the character '召' is pronounced 'zhao' in Mandarin Chinese, and forms the right side of the characters '招' and '昭' - both of which are also pronounced 'zhao'.

Here's another example - the character '包' is pronounced 'bao', and is frequently found as a component in characters pronounced 'bao': '饱' ('full stomach', with the food radical ⻠ on the left) and '抱' ('hug', with the hand radical ⺘). '包' also forms the phonetic component of some characters pronounced as 'pao', which sounds similar to 'bao': '跑' ('run', with the foot radical ⻊), '炮' ('cannon', with the fire radical 火), and '泡' ('bubble', with the water radical ⺡). I had been studying Chinese recently, and understanding the phonetic aspects of the characters has helped me to read them more easily than before.

Having said that, a lot of these phonetic relations date back many years and some of them have been obscured due to language change. But to blindly assume that Chinese characters are 'archaic' is false - as others have commented, you can say the same for English spelling, which also has spellings that date back many years but do not make sense today, like the 'gh' in 'tough', 'dough' and 'caught'. Also, it would be possible to reform Chinese characters such that every character with the same pronunciation would use the same phonetic component, while allowing for additional components to indicate the meaning. And that was what Simplified Chinese did to a certain extent.

Thanks for the detailed response, 'archaic' is not necessarily the best word to describe the problems with it, and hanzi seems to be more straightforward than Kanji

But dealing with a huge alphabet is more complicated than one with (much) fewer characters as the Latin, Arabic and Korean ones.

The English spelling problem is annoying, but most languages don't have such a disparity in writing (which, for the most part, is not that big).

Chinese writing is adapted to write the Chinese language. The only place where this is an issue is languages where there was no standardization (namely every other Chinese languages except Standard mandarin) because even natives don't know which character to use. More interestingly, once someone has knowledge of Chinese character, he can guess the meaning of text in other languages; for instance Korean written in mixed script is incredibly readable to some one with Japanese reading ability.

The Voyager Golden Record has a record of Minnan language which transcription in Chinese character is: 太空朋友，恁好。恁食飽未？有閒著來阮遮坐哦! Anyone in this thread reading some Chinese can guess accurately what the first sentence means. Probably the second too. The third is more difficult but looks like an invitation. Now compare with the romanization: Thài-khong pêng-iú, lín-hó. Lín chia̍h-pá--bē? Ū-êng, to̍h lâi gún chia chē--ô·! Besides pêng-iú looking like Mandarin pengyou is this totally foreign.

They did, but it stopped and then went back as far as I know. E.g. nowadays there are two Chinese writing systems, simplified and traditional. Similar Japanese efforts to ban some rare kanji (e.g. some were only ever used in certain last names) fizzled with the advent of computers that make it easy to type any kanji.

People do not always want things simplified. They also want richness and this may conflict with simplicity. Look at fonts, for example. They gained a lot of complexity recently, there's lots of contextual variations and even two different systems to express it (Apple has its own). The monospace fonts that were invented for typewriters are much simpler than other fonts, but nowadays they are only used in certain specific contexts, like programming, and even there they started to get quite a bit of richness + complexity with PragmataPro being the absolute champion of it.

I'd say the general trend is toward perceived simplicity, but internal richness. Automation, not simplification.

> look at fonts, for example. They gained a lot of complexity recently, there's lots of contextual variations and even two different systems to express it

Yes, but see how many fonts there are for Latin alphabets vs. Kanji/Chinese. (something about dozens of characters vs thousands)

What you describe includes English, however, due to its heritage of being a conglomerate of several different languages. That's why English spellings bees are a thing -- the spelling and pronunciations don't always match up. We'd need to switch to something like Korean, where deviations between spelling and pronunciation can be counted on one, maybe two hands.

> What you describe includes English

Yes, but to a much, much lesser degree

If someone says to you their name is "John" you know how to write it. Foreign names might be weird sometimes (hi Ireland) but usually they're simple.

In Japan you have to ask everyone how to write their name because there's no standard way of spelling. And every character has (completely) different pronunciations depending on context https://www.thejapanesepage.com/tag/kanji-pronunciation/

I was more referring to the phonetic alphabet and standardized spelling part of your post, and not to the Japanese part. That's why I proposed Korean as the language to satisfy your requirements, as English does not, as exemplified by the name John, which is pronounced [jon].

I actually count English spelling as a strength and not a weakness, otherwise we'd have the situation of multiple semantically distinct and unrelated words being spelled the same[0] while at the same time having to maintain distinct orthographies for every minor dialect[1].

[0] https://ko.m.wikipedia.org/wiki/%EC%88%98%EB%8F%84_(%EB%8F%9...

[1] https://en.m.wikipedia.org/wiki/North%E2%80%93South_differen...

English is also unable to avoid the issue of having multiple semantically distinct and unrelated words being spelled the same, for example "stock" or "light" or "scale". I don't know enough about languages to say which ones have avoided this issue. Regarding your second point about not having to maintain distinct orthographies, could you explain in more detail how English avoids that?

> English is also unable to avoid the issue of having multiple semantically distinct and unrelated words being spelled the same, for example "stock" or "light" or "scale".

Fair point.

> Regarding your second point about not having to maintain distinct orthographies, could you explain in more detail how English avoids that?

I'm not an expert on phonics but from what I've seen from attempts at writing any language phonetically regular is that in order to achieve phonetic regularity for everyone you either have to force everyone to pronounce words identically (impossible for a language as widely used as English) or you have to maintain entirely seperate spelling systems for each dialect. This number goes up the more phonetically regular your writing system is. See this example in the wikipedia entry on the Shavian script, a phonetically regular alphabet for English[0]:

> Spelling in Androcles follows the phonemic distinctions of British Received Pronunciation except for explicitly indicating vocalic "r" with the above ligatures. Most dialectical variations of English pronunciation can be regularly produced from this spelling, but those who do not make certain distinctions, particularly in the vowels, find it difficult to produce the canonical spellings spontaneously. For instance, most North American dialects merge 𐑭 /ɑː/ and 𐑪 /ɒ/ (the father–bother merger). Canadian English, as well as many American dialects (particularly in the west and near the Canada–US border), also merge these phonemes with 𐑷 /ɔː/, which is known as the cot–caught merger. In addition, some American dialects merge 𐑧 /ɛ/ and 𐑦 /ɪ/ before nasal stops (the pin–pen merger).

With exceptions, the mapping of letters to language in English mostly seems to occur at the level of morphemes and not phonemes. Look at this wiktionary entry for "schedule" for example[0]. All dialects spell it the same but there are multiple distinct pronounciations listed in the IPA section that would all lead to different spellings in a completely phonetically regular system.

[0] https://en.m.wikipedia.org/wiki/Shavian_alphabet

[1] https://en.m.wiktionary.org/wiki/schedule

>(UK) IPA(key): /ˈʃɛ.djuːl/, /ˈʃɛ.dʒuːl/, /ˈskɛ.djuːl/, /ˈskɛ.dʒuːl/

>(US) IPA(key): /ˈskɛ.dʒʊl/, /ˈskɛ.dʒəl/, /ˈskɛ.dʒu.əl/, /ˈskɛ.dʒul/

>(India) IPA(key): /ʃɛ.djuːl/

>(Canada) IPA(key): /ˈskɛ.dʒu.əl/, /ˈskɛ.dʒuːl/, /ˈʃɛ.dʒu.əl/, /ˈʃɛ.dʒuːl/

Actually kanji only don't make sense in the context of only reading them. First learn how to speak and listen. Then as soon as you write a bunch of kanji you will realize how consistent the stroke order is and that there are a lot of repeating elements (loads and loads of boxes...). There are still some annoying exceptions but those are the minority. Learning the meaning of each kanji will make it very easy to connect the kanji with the intended word.

Reply on news.ycombinator.com