If English were written like Chinese (1999) | Hacker News Reader

If English were written like Chinese (1999)

zompist.com

228 pointsrogerbraun94 comments

The author really knocked it out of the park with this analogy. This explained things to me that I might have learned through experience but had never been explicitly told before. He even took the time at the end to point out where his own analogy breaks down. Zhongwen.com is also a fantastic site worth a look for students of Chinese. Overall extremely impressive, creative, and interesting.

This article is actually a lot better at explaining the characters system used in both china and Japan, than the articles that have hit the front page today and in the last week or so.

With the added bonus that it's accurate. Referring to these characters as they are originally: Chinese. Some people don't seem to be very happy about that fact.

There's too much politics involved. Try to find any even the slightest mention of proto-Korean language roots (which are Japonic) in Korean.

Who? The word kanji literally means "Chinese characters" lol

In that 口 article yesterday, several commenters seemed angry that the author referred to kanji as "Japanese characters".

So apparently this is some kind of thing, that some people take seriously. New one on me.

"This is your brain on nationalism"

You can imagine being British and learning the American language. Of course people feel cheated if their culture is not credited where it should be.

When someone in Japan, talking about the Japanese language, refers to "Japanese characters", they clearly mean "characters used when writing Japanese". They are not suggesting anything about where the characters originated.

My understanding (as a speaker of neither language) is that Japanese usage has diverged a fair bit from Chinese roots.

The phonetic bits, for example, are based on how Japanese speakers of the time thought the equivalent Chinese sounded, with allowances for Japanese and Chinese having radically different phonics. And then you do that a few more times because the language both evolved in parallel, and you end up in a situation where each kanji can have more than one phonetic representation based on the original Chinese pronunciation. It can also have a phonetic representation based on the Japanese word that means the same thing, which will invariably be something radically different since Chinese and Japanese aren't terribly closely related languages.

So while they might be using the same writing system, the fact that it's a fairly elegant writing system for Chinese doesn't imply that it's elegant in Japanese. Not any more than the Roman alphabet being a very good writing system for Latin implies that English's writing system isn't a bit of a mess.

For not knowing either language, your knowledge is impressive. There are actually Japanese-originated Kanji, some that have no analogue in Chinese.

Japan certainly adapted kanji to fit the Japanese language, but I was not aware of a dispute about the origin of the characters.

Obviously they have diverged; Japan has made efforts to simplify kanji forms, although not to the extent of zh-cn, unless you count hiragana and katakana, which are derived from kanji:

https://upload.wikimedia.org/wikipedia/commons/7/7a/Hiragana...

As a side note, I was taught that Japanese "Chinese" readings have more in common with the Chinese spoken at the time than modern Chinese.

> Japanese "Chinese" readings have more in common with the Chinese spoken at the time than modern Chinese.

Yup. There are even names for the various readings based on which era/dynasty they were borrowed during. E.g. for the readings of 明, "myo" was borrowed first, then "mei" a few centuries later, then "min" most recently.

https://en.wikipedia.org/wiki/Kanji#On.27yomi_.28Sino-Japane...

Kan Ji or Letters of the Han Dynasty

The original article makes no mention of kanji (Han zi). Instead it ends with: "Well okay, I'll come clean. I've actually been secretly teaching Japanese to you. 口 actually really is the Japanese character for "mouth" and 食 really is the symbol for eating."

Not exactly. Those are Chinese characters and Chinese meanings. The article does not teach anything exclusivly Japanese. For anyone who was not familiar, they would go around with the impression that these characters really originate from Japan. You can imagine how the Chinese people or people who study Chinese would feel being shown this cool new way to learn "Japanese".

It's a fucked up writing system that is pretty beautiful as well. Unfortunately it has limited the number of syllables (combinations of sounds) in the language. Why? Because they have to make due with just 2000 phonetic sounds (335 in modern Japanese) while English letters can combine in many more ways and form a much more diverse array of sounds.

As a result Japanese people (probably Chinese too) have a hard time processing some of the English syllables. This creates difficulty in learning English for Japanese and Chinese, and in turn, it leads to isolation. It's like a "cultural moat" separating ideographic languages from phonetic ones. If they tried to give it up, it would be sad, because of all the history and beauty it has, but on the other hand, in these countries children in the 5th grade can't properly read a newspaper because they need to already know 2000 characters and many more combinations of characters by rote. By comparison, a first grader can read a newspaper in English.

In terms of programming, Chinese is like Perl (even Perl6! - huge, complex, mysterious, beautiful) and English is like Lua (small, elegant, expressive).

"There is a standing joke among sinologists that one of the first signs of senility in a China scholar is the compulsion to come up with a new romanization method"[1]. That's how much they are bothered by the huge initial cost of learning it.

[1] "Why Chinese Is So Damn Hard" - http://www.pinyin.info/readings/texts/moser.html

This is incorrect. You seem to be saying that the writing system is responsible for the fact that English is hard to learn for Chinese speakers because the phonetics are different? Many languages have phonetics very different from English and so are at a "disadvantage" when learning English as a second language -- this is orthogonal to the choice of writing system.

Chinese speakers learn Pinyin from primary school and so know exactly how an alphabetic writing system works. Indeed, many Chinese who are not fully literate in Hanzi know how to read Pinyin.

Furthermore, Japanese children in the fifth grade can already read about 850 characters -- this is not as difficult a feat as I think you imagine.

"As a result Japanese people (probably Chinese too) have a hard time processing some of the English syllables. This creates difficulty in learning English for Japanese and Chinese."

We can also say that American/English have a hard time processing some of the Japanese/Chinese syllables. All the languages I speak a little bit have distinctive sounds that are difficult for most non-native speakers to master unless you put a lot of practices. It escalates to another level to speak a whole sentence, that is, a sequence of sounds. That is why it is easy to tell if a person is a native speaker of a language in most cases.

It is true that Indo-European languages are difficult for Japanese and Chinese to learn. It is due to other reasons, which would be a long answer.

What I am saying is that in English, after learning 26 letters, you can read any combination of them. If you try to see how many possible syllables there are, there would be many thousands. Combinations like "eng" "lish" can't be properly put in Japanese phonetic characters because you can't put "eng" directly, you have to use "en gu ri shu", inserting vowels all over the place. Japanese can't properly learn English because their brains are trained with much fewer possible syllable sounds and they can't make the jump.

But in turn why is it that there are so few syllable sounds in Japanese? It's because each syllable has to have its own ideograph (usually there are multiple ideographs) and you can't easily learn 10000 of them. Being limited to much fewer ideographs you can memorize, you are also limited to fewer sounds you can use. On the other hand, with English, after 26 letters you can read any combination of them, and they form many more possible sounds. So you learn to make all those sounds as a kid. Even if you come from another phonetic language you will probably have a similarly diverse sound vocabulary which will help you transition to English.

So, memory limitations -> fewer characters known than possible English syllables -> a kind of mental straightjacket in producing other sounds that fall outside the allowed ones in their native language -> difficulty in learning English -> cultural isolation. Sad story.

Just look up Japanese people who have learned English from grade 1 to 12, and see how great their pronunciation is. It's so damn hard for them to make the sounds.

I love hacker news. Some personal anecdotes soon summaries into neuroscience theory, and sticks to it

> But in turn why is it that there are so few syllable sounds in Japanese? It's because each syllable has to have its own ideograph (usually there are multiple ideographs) and you can't easily learn 10000 of them.

Please don't make up facts just because they seem plausible to you. A quick internet search reveals that the Japanese language predates any writing system for it:

"Before the 4th century AD, the Japanese had no writing system of their own. During the 5th century they began to import and adapt the Chinese script, along with many other aspects of Chinese culture, probably via Korea. However the Japanese were aware of Chinese writing from about the 1st century AD from the characters that appeared on imported Chinese goods."[1]

Thus Japanese had settled on having few syllable sounds before there was any way to write them down, so their syllables can't have been limited by their script (at least initially).

> Just look up Japanese people who have learned English from grade 1 to 12, and see how great their pronunciation is. It's so damn hard for them to make the sounds.

You should see English speakers try to pronounce Chinese...

[1] http://www.omniglot.com/writing/japanese.htm

I have a couple things to add to this thread, not everything directly related to your comment.

1) Knowing the 26 letters is not remotely enough to be able to read the newspaper in English (earlier commenter made this assertion). Example: "She caught a cough. Such is life and death, dear!" How does knowing the letter "u" help you pronounce half those words? What about "g"? What about "h"? What about even if you put "gh" together? What about "e", "a", and if you shove "ea" together? In this single paragraph I've ballooned the number of letters and letter combinations needed into the 40's, and I've barely just begun. Don't get me started with borrowed words from French and Spanish (c'est la vie!).

2) Your comment: "You should see English speakers try to pronounce Chinese" - From direct first-hand experience, this is tough for a completely different reason than this comment thread is talking about. You've gone off on a tangent here. The reason Chinese is hard for English speakers is that tonality suggests overall sentence semantics, it does not directly affect any single word, whereas tonality affects every single word in Chinese. I thought we were talking about how syllable pronunciations were expressed by the writing system.

> Thus Japanese had settled on having few syllable sounds _before_ there was any way to write them down, so their syllables _can't_ have been limited by their script (at least initially).

You might be underestimating just how huge an influence Chinese has had on Japanese:

https://en.wikipedia.org/wiki/Sino-Xenic_pronunciations#Ling...

Japanese (as well as Korean and Vietnamese) imported Chinese vocabulary en masse, including a gloss of Chinese translation. That made Japanese much more homophonic since the tones got dropped in the process.

Just compare the kunyomi (native) and onyomi (Chinese-derived) readings for some Kanji and you'll see that the onyomi readings are usually single syllables.

"For example, Sino-Japanese words account for about 35% of the words in entertainment magazines (where borrowings from English are common), over half the words in newspapers and 60% of the words in science magazines."

That said, I don't really agree with GP's contention that using Kanji makes Japanese have a simple sound system. Chinese has a somewhere richer sound system even though it invented the characters (especially if you count tones), and Spanish and Italian have very simple sound systems even though they're written with alphabets.

> What I am saying is that in English, after learning 26 letters, you can read any combination of them.

But those combinations are usually learned via oral tradition and rote rather than actually having anything to do with the letters themselves.

Most notoriously “ou” and “ough”.

We have a solid 14-20 vowels, and 5 glyphs for them.

I suppose Spanish, Korean, and Russian might be easier to pronounce and less rote memorization, but you're certainly right about English.

> What I am saying is that in English, after learning 26 letters, you can read any combination of them. If you try to see how many possible syllables there are, there would be many thousands. Combinations like "eng" "lish" can't be properly put in Japanese phonetic characters because you can't put "eng" directly, you have to use "en gu ri shu", inserting vowels all over the place. Japanese can't properly learn English because their brains are trained with much fewer possible syllable sounds and they can't make the jump.

This claim is significantly misleading.

Every language has its own phonotactics, which includes allowed and forbidden sequences of sounds.

https://en.wikipedia.org/wiki/Phonotactics

English and Japanese both have their own phonotactic structures. While English's phonotactic rules are much more permissive than those of Japanese, and allow more different sounds to be used and to be combined in more ways, and while these differences do affect what native speakers of each language can easily say (or recognize other people saying!), it's a mistake to say that English allows everything while only Japanese has restrictions.

Both languages have restrictions, and both have restrictions that some other languages don't have. For example, Slavic languages allow consonant clusters that English doesn't. Hindi allows distinctions of aspiration in arbitrary locations where English requires particular aspirations in particular positions or at least considers this distinction non-phonemic (and hence extremely hard for English speakers to learn to appreciate even when both sounds in question are part of their phonetic inventory).

If, from noticing the more extensive restrictions on sound structure in Japanese, you get the impression that English speakers can pronounce any sequence of letters from our alphabet, you're missing something significant about how English works.

Edit: as other people have noted in this thread, these differences between these languages are also probably not originally due to differences in writing systems. Often languages that share the same writing system have different phonotactic rules (and different sound inventory, for that matter). It's more likely that many of the important differences predate the writing system, although it's true that the writing system can influence how people learn and think about their language and how it changes over time.

I might have been unfair to the parent commenter here; I just realized that "you can read any combination of them" might mean "you know the intended pronunciation of any English text" (including words you've seen before) rather than "you are able to pronounce any conceivable sequence letters" (including letter sequences that never occur in English words). That doesn't necessarily represent confusion about phonotactics; it might just be a claim that an alphabetic writing system is always/sometimes easier to learn than the full present-day Japanese writing system. I'm not positive this is true of English orthography, but is probably true for some languages written with the Roman alphabet.

> It's because each syllable has to have its own ideograph

False. If you look at the history of writing systems, we have seen at least three instances of logographic systems developing into syllabries. Egyptian hieroglyphs eventually devolved into representing only consonants (vowels in Semitic languages largely represent only inflectional variations, so they're far less necessary). When this language and its descendants passed into Greek, distinct letters were added for vowels yielding an alphabet; when it passed into Indic languages, the consonants were systematically modified (e.g., rotating the glyphs) to make CV letter-form pairs. Mayan broke its logography into CV syllables (much akin to how Japanese derived the kana from kanji).

Mayan and the Indic scripts are great examples of the flexible of syllabic systems. Mayan is not a CV-syllable language, yet they still used CV glyphs. The syllable "bak" could be written, for example, as the glyphs for "ba" and "ka". Indic scripts, representing languages that have phontactics as complex as English, yet they also retain basic CV letters.

Even Japanese plays similar games: each letter doesn't represent a syllable, it represents a mora. The syllable "nan" would be represented as having 2 mora: "na" "n". Similar, the phrase "ですか" is written "de su ka" but is pronounced "de ska" in most dialects.

The Japanese kana orthography isn't really limiting its ability to add more sounds any more than the Latin alphabet limits our ability to represent sounds foreign to Latin speakers 2000 years ago like n of 'sing' (IPA: ŋ) or the th of 'the' (IPA: ð).

What makes English hard to pronounce for Japanese is that our phonotactics are very incompatible. You can see the same effects for English speakers trying to pronounce, say, Czech: "strč prst skrz krk" is pronounced exactly like it looks (well, you have to know that č is the 'ch' of chocolate in Czech), yet it is still difficult for an English person to pronounce. It's not that we can't make those sounds, it's that we can't put them in that order (/ŋgis/ makes for another good example".

The other difficulty Japanese speakers have is that some phonemes aren't recognized as separate--the infamous l/r comes up here. In general, Japanese don't hear a difference between them. Native English speakers can have the same reaction to other cases: both ś and sz in Polish sound like 'sh' to an English speaker, but they represent two distinct phonemes.

One nitpick: ですか is not pronounced "de ska" it's closer to "de s' ka"; They elide the full vowel, but they insert a stop instead of moving on to the consonant (in order to retain the prosody).

> It's because each syllable has to have its own ideograph (usually there are multiple ideographs) and you can't easily learn 10000 of them

As an aside, one side-benefit of a language that doesn't have so many phenomes is that automatic speech creation is much easier.

That's speech creation, not speech recognition.

Using software to speak Japanese fluently, at a given pitch, with a given emotional context is a solved problem. There's even software that can be used to create singing voices.

It only takes about 1000 recorded sounds to do so. In english, it'd take 10,000 multiplied by the number of emotional contexts and speech patterns you'd like to mimic.

>What I am saying is that in English, after learning 26 letters, you can read any combination of them.

No. After learning 26 letters you're no better off than before. You can't read, because you don't know the meaning of the words. Vocabulary is learnt separately from the writing system, unlike in Chinese or Japanese. On the other hand, with 26 kanji, you'd be able to read and write, but not necessarily speak.

Neither system is better - they're different.

> What I am saying is that in English, after learning 26 letters, you can read any combination of them.

Not really. For instance: trough.

I forget the specifics and terminology, but to read English competently, you have to be familiar with at least several hundred different letter-sound mappings.

Just a tiny nitpick: the "ng" in "English" is actually a bit more complicated than just "n" plus "g". The "n" is actually a "ŋ" (the "ing" sound without the hard "g") and the actual "g" is often dropped in colloquial pronunciation and part of the second syllable otherwise. So it's either "eŋ-lish" or "eŋ-glish" but never "eŋg-lish".

And don't get me started on English speakers making fun of the Japanese "r"/"l" sounds when English doesn't even have a proper "r" sound in most dialects -- as a native speaker of a language with an uvular ("back of the throat") "r" sound the alveolar ("front of the mouth") "r" feels more like an "l" to begin with.

I think that if Americans try, it's a lot easier to go to the Japanese phonemes; given that they are effectively a subset of English phonemes (with some variation). That said, it seems that Americans feel like they're making a disrespectful accent when they're getting closer to pronouncing things correctly. They would rather be authentically bad than sound like they're making fun of an accent. Here in Canada we have this problem even when kids are learning Québecois French in school; they think they sound like they're making fun of Francophones when they get close to good pronunciation.

An American English speaker can generally approximate る and similar sounds far more easily than a Japanese speaker will learn to pronounce l or figure out the th in the. That's assuming that the English word they're trying to say doesn't have consonant clusters, which (aside from nasal consonant pairs) simply do not exist in Japanese. It really is not equivalent.

> on the other hand, in these countries children in the 5th grade can't properly read a newspaper because they need to already know 2000 characters

A child exiting reception grade in China, i.e. entering Grade 1, is expected to know 1000-1500 characters. They certainly don't know all combinations of characters these words form, but the learning starts young.

It's more useful if you say the child's age, since I'm not sure which country "grade 1" refers to. If it's the USA, I'm not sure how old that is either.

From 6-8 years old. Parents in China can 'speed up' or 'slow down' their child's education by putting them in school early or late.

It's a remarkable fact, I think, that spoken modern standard Chinese (Mandarin) has only about 400 distinct syllables ignoring tones. With the 4 tones plus neutral pronunciation, there are about 1300, maybe 1500 (including rare ones) distinct syllables. They're basically all of the form

  consonant (optional) + vowel/diphthong + n/ng (optionally)

(typically called "initial + final" in this context).

(Note that older Chinese "dialects" such as Cantonese or Teochew retain many more distinct syllables. For example, the infamous "shi shi shi shi" "poem", consisting only of "shi" sounds in Mandarin (neglecting tones), contains 11 distinct syllables (neglecting tones) in Teochew.)

English or German, on the other hand, have somewhere around 8000 distinct syllables. (Think of "strict", "fractal", "Angstschweiß", "Hampsthwaite", "strengths", etc.)

By the way, you may not have encountered "Hampsthwaite" before, but you probably know how to pronounce it. This doesn't work in Chinese.

Honestly, though, of all languages written with an alphabet (Latin or no), English is the worst when it comes to mapping from writing to pronunciation. Sure, it might better than Chinese, but flaunting it in public looks rather absurd even when comparing to, say, French.

And that's not even comparing to languages with a phonemic orthography (where the mapping from pronunciation to spelling is essentially injective) like Finnish or Turkish (which is also extremely regular in terms of grammmar, I'm told). English is terrible in this regard.

Now try pronouncing Lancashire, Gloucestershire, Worcester, Derby, Edinburgh, Argyll… Granted, the latter two are Gaelic in origin, but it's not like English spelling — especially of place names — is terribly phonetic.

English orthograhpy is in fact widely known for having [less systematicity](http://ncf.idallen.com/english.html) than many of its Indo-European neighbors.

>"in these countries children in the 5th grade can't properly read a newspaper because they need to already know 2000 characters and many more combinations of characters by rote."

This is completely false. I spent seven years working in Taiwanese schools, and saw a few precocious seven year olds reading newspapers and virtually any 5th grader able to. It's very similar to the situation in English speaking countries, though I would say that first graders growing up with characters are probably a bit behind their English-speaking counterparts and a bit ahead by the time they're in fifth grade. This only applies to developed places like Japan, Taiwan and Hong Kong. Obviously a poor child from rural China will have a different set of challenges and be behind.

The increasing use of icons and emoji suggests that English will become like Chinese!

(Ever try to look up an icon in a dictionary? This puts paid to the idea that icons are decipherable by people who don't know the language. Copyrighting the icons makes even that infinitely worse, as it prevents standardization.)

A running joke in Gibson's "Idoru" is that the lead character, through all her adventures, carries an icon dictionary.

An icon? no. There are however, online hanzi dictionaries where the user draws a character, here's one:

http://ce.linedict.com/dict.html#/cnen/home

See the paintbrush next to the magnifying glass in the search field.

Also not for icons but for math symbols: http://detexify.kirelabs.org/classify.html

Really useful when you want to use "that squiggly thingy" but don't know the TeX command.

I usually jump straight to http://shapecatcher.com/ for arbitrary Unicode.

I'm not sure where you've seen this "increasing use of icons and emoji". When I encounter people (this is pretty much limited to teenagers) use emoji, they have very little communicative value.

In the modern era the use of pictographs has become Chinese's Achilles' heel: the hanzi are not sortable. The very things that define the Chinese are what makes it stupidly difficult to get computers to grok the language.

By which way you mean Chinese is not sortable?

They usually sort by number of strokes.

There are many ways to sort. Typically sorted first by radical, then by number of strokes.

There's more than one ordering of the radicals, and choices to be made when characters have the same radical and the same number of strokes

Paper dictionaries are very thumbable though, the current radical is usually highlighted in a way that makes it easy to flick through and find what you need

> I'm not sure where you've seen this "increasing use of icons

Consider all the icons used for "print" as if the letters P R I N T are unclear. Even for non-english speakers, it's trivial to look up the word in a dictionary, and impossible to look up the icon. Ditto for replacing "ON" with |. It's just madness.

> and emoji.

Perhaps their recent appearance in Unicode, and the several screens of them that appeared on my iphone texting app.

This is already encouraged in the latest iMessage, which highlights emojifiable words and makes it easy to emojify them: http://i.imgur.com/r26ABsw.jpg

The letter 'M' in the Roman alphabet is derived from the symbol for running water in phonecian, and the letter 'A' is a rotated cow's head.

Phonecian is partly derived from Egyptian hieroglyphics. I love this stuff.

https://en.m.wikipedia.org/wiki/Phoenician_alphabet

This is a fantastic guide / insight for how Chinese characters work, and does a great deal to dispel the myth that Chinese characters are virtually random - that each individual character requires independent memorization.

As someone who has studied both Chinese and Japanese, this article read very fluidly. Curious how other readers have found this?

As someone with practically zero understanding of the Chinese writing system beyond the basic notion of what an ideogram is, it was very helpful in aiding understanding (assuming it is accurate).

People might be interested to know that Egyptian Hieroglyphics, despite looking even more like pictograms are actually an alphabet.

Another cool fact, in hieroglyphics there is more than one way to write a word, because, unlike most alphabetic systems, some characters are multi syllabic and can represent two or more syllables.

Wasn't it rather a syllabary?

I've never heard it described that way.

And like most semitic languages, hieroglyphics mostly record the consonantal skeleton of a word.

There are additional symbols (determinatives) to disambiguate homonyms.

[work] + [fight] + [sun] is poor Huffmanization of the common English suffix -tion. That's what, 14 strokes when English just takes 7?

Well, maybe that was the point, though -- that the priorities of the language may change over time, and eventually you're dealing with 2000 year old Cockney rhyming slang baked into your written language. But that's just the opinion of one guy sitting on his Vannevar.

That is a part of the analogy which definitely applies -- Chinese script (even Simplified) often completely ignores Huffman coding-like principles. Otherwise we wouldn't have words like 繼續 "continue" (41 strokes).

For comparison, the letters in "continue" have approximately 15 strokes total.

続く is 14 strokes, so Japanese beats English!

I could go on ... and "go on" is only 6 strokes.

You could just use the ellipsis, and use only three strokes. Or even simpler, an em-dash at one stroke. :)

Or two strokes and a dot in cursive. A principle that also applies in Chinese and Japanese

继续 21 strokes jixu 8 strokes

继续 is 21 strokes when printed, but less when foreigners in China like me write them when we handwrite them quickly.

继 is 5 strokes handwritten (纟,䒑,丨,八,乚 each written in one stroke) and 续 is 4 (纟,十over乛,氵,人 each as one stroke), so that's 9 altogether.

Mainland Chinese who learn handwriting at school have a way of writing each character with only one or two strokes.

If you're only counting pen-down-pen up as a single stroke regardless of smooth movement, then any cursive word in any latin-based alphabet is a single stroke, unless it has an 'i/j' or some sort of diacritic in it (some people cross 't' in their cursive, some people do it separately).

People like you looking for less strokes in China mix both English and 中文 words in a passage, which is silly in my opinion.

>sitting on his Vannevar.

Took me a minute and some Googling. Did you just make that up? It's in perfect spirit.

Yeah, that's my original rhyming slang. Thought of it about a year ago. But as is usual for jokes, sometimes you have to wait for years to find the perfect time and audience to use it.

I've been waiting for years for somebody to do something like this with emoji. All we'd need is a way to add radicals and a central dictionary.

In a way, eyes and mouths are like the radicals for emoji.

I'm not sure how you'd sort emoji though, since there's no stroke order.

He wrote this before the rise of emoji. We may see something like this as a cell phone writing form. Teenagers would use it to be incomprehensible to olds.

One thing that is glossed over slightly in these discussions is the distinction between collections of sound-meaning compound characters where the rebus is the same but the words aren't cognate, and where the words are cognate.

I've read in some places that the scribes attempted to use the same rebus for cognate words but I everything I could find online was a re-hashing of wikipedia (or wherever the wikipedia article is sourced from) which states "However, the phonetic component is not always as meaningless as this example would suggest. Rebuses were sometimes chosen that were compatible semantically as well as phonetically."

I'm not sure how important this really is or how many characters that share a common rebus are cognate. But to my aesthetic senses I much prefer characters to contain etymological information than just the pronunciation when the character was first written.

Is this meant to be serious (surely not) or rather like Mark Twain's take on spelling reform?

It's really meant to explain what Chinese characters actually are and how they work, by an elaborate analogy.

As regards Twain: the take you are thinking of was probably not written by him, but wrongly attributed to him after his death.

He did, however, write an article[0] about spelling reform later in life, in which he sincerely advocated the use of a simplified "longhand, written with the shorthand alphabet unreduced." That is, he proposed the use of Isaac Pitman's phonographic alphabet—the basis of what was then the most popular shorthand in the English language—without the brief forms, phrasings, and abbreviations that allow stenographers to write (by omission) at the sound of speech, but slow down the reading back of what they've written; in Twain's preferred shorthand, every sound would be on the page.

[0] - https://books.google.com/books?id=KoBYAAAAYAAJ&dq=what%20is%...

(Unfortunately, the Project Gutenberg transcription of this essay does not include the plates of Twain's shorthand. This is why I've linked the Google Books scan.)

I didn't realize it wasn't actually his.

Thanks for the link, unfortunately Google doesn't seem to want to show me the content. I wonder if they are implementing some kind of region coding (I'm in Norway).

I always wonder whether people advocating phonetic spelling have ever encountered someone who speaks a different dialect. In Twain's case it is pretty certain that he did as the Huckleberry Finn books include dialect dialogue. How did he think such people would use a phonetic script or worse a phonological one? Did he expect everyone to suddenly start speaking the American analogue of what in the UK we call Received Pronunciation (RP)? If not then surely another person's script would be even harder to read than it is already.

> I always wonder whether people advocating phonetic spelling have ever encountered someone who speaks a different dialect.

For just this reason, English spelling reform was probably doomed from the start. Any new standard seems just as arbitrary as the old one to a speaker of a non-standard dialect. The chosen set of vowels—whatever it might be—would ring particularly untrue to the ears of millions of English speakers, who might use any 12 to 20 vowels of a pool of a few dozen in their own speech.

My own example: I grew up in a part of the U.S. that makes no distinction whatsoever between the vowels in 'thought,' 'lot,' and 'father'—yet these are three separate sounds in Received Pronunciation, and might be divided into two elsewhere. Not only can I not make three different sounds for these vowels—I do not know how they would differ!—I cannot tell the difference between them when listening to someone who can. (Perhaps the 'a' of 'father' I could note in contrast to the other two with some conscious effort.) If I were to use a phonological spelling system that split this vowel group into three, I would have to memorize by rote—again!—the spelling of many words. What was supposed to be an easy system that did away with rote memorization, turns out to be more of the same. Perhaps this second learning curve is not so steep—no 'ough'-es—but I've already gotten past the first one! Why bother again?

If, on the other hand, a phonemic alphabet were provided—something like a more elegant IPA—making our spelling as idiosyncratic as our speech, we would just have a new set of problems. First, second language learners would find no refuge from the dizzying amount of dialects and accents while they were still learning fundamentals. This would be a sore spot for any language, but especially for a common lingua franca like English is today. And then, the language of the law, the academy, and business might recede still further from students unprivileged in class, birthplace, and schooling, if they first learned to write in their own dialect, not that of the ruling classes.

There is an advantage to an orthography having some remove from the spoken language, as it provides a common ground for speakers of different dialects to communicate. Chinese ideograms do particularly well here—to draw us back toward the article—as entire articles might be written which could be understood equally well by two speakers of mutually unintelligible dialects, dialects so far removed from one another that they could be called separate languages.

I don't think I'm letting you in on anything new here, but as it is a topic I've put some thought into, I took the opportunity to turn into a windbag.

Sorry about the Twain link. Here's Gutenberg[0], if you're interested in the text all the same.

[0] http://www.gutenberg.org/files/70/70-h/70-h.htm#link2H_4_001...

FYI we're close to this with the use of acronyms and abbreviations, not just emoticons, given English's phonetic basis, etc.

https://xkcd.com/1709/

What a breath of fresh air is a 1999-era web page. Fast loading, simple, readable.

Reply on news.ycombinator.com