[yejanll]

The Greeks and the Romans got it right; a small set of characters that can be combined to form any word. Complexity from the composition of simpler elements, not inherent. Computer interaction via keyboard makes the superior design all the more obvious. Those guys were ahead of their time. Ave imperator, morituri te salutant.

[larkost]

I would argue that a system like the Japanese Hiragana or Katakana are a better system: they are each 56 letters that directly correspond to syllables. So "ha" is a single letter, as is "he", as is "be". With this nearly everything you say is directly translated both directions, and there are fewer complications (there are always a few, for example in Japanese one of those 56 letters is "n".. so no vowel, and in many dialects you say "s" for the "su" character if it is on the end of a word, and there are a few oddities around letters involving "y").

English, being the composite/mongrel language that it is has really complicated patterns for how you put letters together. For example the "i before e except after c as in neighbor and weigh" sort of thing (which does not cover all of the exceptions of course). This sort of thing has lead to the existence of spelling competitions in the English-speaking world (spelling bees). My Hungarian wife was surprised that such a thing existed. In Hungarian it is much closer to see-what-you say, with only a few exceptions (not that the rules are kind on English-speaking Hungarian learners like myself).

[shiroiushi]

>I would argue that a system like the Japanese Hiragana or Katakana are a better system: they are each 56 letters that directly correspond to syllables.

No, they're really not. First, they have 46 characters (each), not 56, though there are another 36 combination characters like ちゃ. Regardless, the problem here is that number comes from the total number of allowed sounds in the entire language. Japanese has an extremely small number of total possible sounds in the language compared to most other languages, particular western ones, and almost all syllables are of the form consonant+vowel: there's basically no way to write, for instance, a word ending in a hard "t" sound, so when Japanese adopts such words, it adds a vowel ending like "tu", and does this for every syllable with a hard consonant without a following vowel. Because of this, loanwords can be really hard to recognize even if you're a speaker of the language that word was adopted from (usually English these days), because the sounds don't map over very well.

And because there's so few total possible syllables, there's a huge number of homophones. The main reason kanji is still around is because it resolves ambiguity and makes it much, much easier to read Japanese text: trying to read text that's all in hiragana (or katakana) is cumbersome, even if spaces are added (Japanese text doesn't normally have spaces).

[int_19h]

Syllabaries are great if the language has relatively simple phonotactics. Once you start having things like triple consonant clusters or a dozen different vowels, it breaks down pretty quickly.

The other thing that makes alphabets more popular in the long run is that they spread easier because they're easier to adapt to different languages compared to syllabaries (indeed, it's not uncommon for a syllabary to become semi-alphabetic as part of such adoption).

[w0de0]

I believe you mean the Ugarites, Phoenicians, & other northwest Semitic peoples. They developed from cuneiform syllabaries the abjad which the Greeks subsequently adopted (being their language's second written form, preceded by Linear B before an interlude of illiteracy). The Greeks added only non-diacritic vowels.

[lucidguppy]

There are benefits to both systems. Western readers have a hard time reading older writing - while eastern readers have no problem with very old texts.

On the other hand - western scholars can understand what the spoken word sounded like - but eastern readers have a much harder time what ancient words sounded like.

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

Western writing systems "decay" faster. Look at french writing - the spellings are phonetic for the time they were first put to paper - but they sound nothing like the current pronunciations.

[bsder]

> Western readers have a hard time reading older writing - while eastern readers have no problem with very old texts.

That's simply not true.

Ancient Chinese calligraphy and language is so different that you have entire PhD fields about it.

By contrast, as someone who has studied basic Latin in high school, I can read stuff from the walls of Pompeii without issue. I can directly read Latin texts from 700AD or so with the standard difficulty of reading handwriting.

See: http://www.edr-edr.it/edr_programmi/view_img.php?lang=en&id_...

Now, perhaps if I were Chinese, I could read ancient graffiti on the Great Wall, but nobody seems to have ever mentioned that.

[shiroiushi]

This is completely backwards and comically wrong. Japanese text looks very little like it did even 150 years ago; the characters have changed completely, and the language was highly standardized during the Meiji era. Try taking a native Japanese speaker to a museum with old Japanese texts or handwriting and see if they can read any of it; they generally can't understand much of it at all.

Modern Korean people can't even read stuff older than a century or so because the language changed from using Chinese characters to the home-grown Hangul character set, and that was only completed a few decades ago.

By contrast, English speakers can read Shakespeare just fine mostly, with a little difficulty understanding some words that are no longer used.

[4bpp]

Japanese is not really the right example, because unlike Chinese the writing is really more phonetic than not (and when it isn't, as in kanbun texts, it's basically an idiosyncratic translation to a completely different language). I imagine that the reasonably educated Chinese should be able to read something like the https://en.m.wikipedia.org/wiki/Xi%27an_Stele from the 8th century just fine, given the regular typeface. English, from the same time period, has Beowulf, which is incomprehensible to any lay reader.

Once calligraphy/handwriting is involved, the situation on the Western side is not much better either. Modern Anglosphere children probably would struggle with 19th-century cursive like https://www.pinterest.com/pin/375276581427478862/ ; in Germanic countries, the handwriting system underwent deeper changes, so nobody apart from selected nerds and antiquarians can read Kurrent as in Goethe's letters - https://commons.m.wikimedia.org/wiki/File:Goethe_Brief_(nich... - or even the newer Sütterlin. Contra what some posters here claim, Roman cursive (https://en.m.wikipedia.org/wiki/Roman_cursive) is right out. I don't think this should be conflated with the question of whether the writing system is understood by future readers - as an imperfect computer analogy, an ASCII text document is in some meaningful sense more futureproof than an Autodesk Animator .FLI, even if the former is on a five-inch floppy and the latter is on a USB thumb drive.

(As for the effects of the Japan's Chinese character simplification, I think they are a bit overstated. I accidentally bought a 旧字体 copy of Mishima's Haru no Yuki at a book sale once, and at least as an L2 speaker I didn't find it particularly more painful to read than I find unmodernised Shakespeare as an L2 speaker of English.)

[shiroiushi]

The OP never said "Chinese", he said "eastern". Japanese and Korean are both eastern.

[4bpp]

That's a fair objection, but given the context I assume he was not actually having anything other than Chinese in mind. Otherwise one might as well bring up Vietnamese, which has a similar history and lexical makeup as JP/KR and is spoken about as far east, but is now written phonetically in Latin-based script with hardly any readers of the old Chinese-based writing system remaining.

[shiroiushi]

Yes, Vietnamese is another great objection, as it's similar to Korean that way. Given that China, Korea, Japan, and Vietnam are the 4 major "eastern" countries then, and only 1 of those might be true for his claim, then why did he say "eastern" instead of just "Chinese"? That's like claiming the American continents (North and South) use English because it's used in the United States. People in Quebec and almost every place south of Mexico will be quite offended, rightfully so.

[lucidguppy]

Yeah - that was my mistake.

[xanderlewis]

Where did you get that idea from? I can read most books in Japanese, but probably not a word of most reasonably ‘old’ texts. I think most native speakers would struggle as much as English speakers do with, say, medieval-era English.

[fsiefken]

Yes, and when you strip out the vowels and squeeze the individual syllables together - the syllable almost becomes a chinese character or a llm token. Paleo-Hebrew (10th century BCE) was a century older then ancient Greek (9th century BCE). Like Phoenician (from which ancient Greek derived) it shared close roots with Proto-Sinaitic.

The Proto-Sinaitic alphabetic script is the oldest (1800–1500 BCE) and evolved from Egyptian hieroglyphic symbols. It contained simplified characters representing consonants, The Phoenician alphabet came later, around 1050 BCE, evolving from Proto-Sinaitic. It became a widely used script with 22 consonantal characters and was highly influential, serving as a foundation for both the Paleo-Hebrew and Greek alphabet. The Etruskan alphabet was adapted from the Greek alphabet in the 8th centry BCE and the Roman alfabet was adapted from the Etruskan alphabet in the 7th century.

Alphabets with 20–30 letters seem to be close to a neurolinguistic optimum for balancing simplicity with expressiveness. The Armenian script was designed by monk and linguist Mesrop Mashtots in the 5th century CE to enable the translation of the Bible into Armenian. With 39 letter it represents Armenian phonetics. The Khmer alphabet with 74 characters evolved from the ancient Pallava script, which was developed in Southern India around the 4th century CE. By the 7th century CE, the Khmer people had adapted the Pallava script, creating an early form of the Khmer script. This script was initially used to write Sanskrit and Pali, the languages of Hindu and Buddhist texts.

[int_19h]

One can have a larger inventory of letters if the glyphs themselves are consistently patterned in some way. E.g. if "c" is /ts/ and "č" is /tʃ/, then one can expect "š" to be /ʃ/, and "ž" to be /ʒ/ (which is indeed the case in many Slavic alphabets).