How to write in Cuneiform

Well, if you call them kanji, they're Japanese characters. (Japanese characters with a name that literally means "Chinese characters", but still not Chinese characters.) Kanji are very much not syllabic.

But Chinese characters are.

You’re right in a politically sensitive context, but practically speaking Kanji and Chinese characters mostly share the same unicode code points, and you might want to ask yourself whether it’s worth nitpicking over for example whether English uses the Latin alphabet when proper classical Latin does not have J or U…

Chinese characters are syllabic in the sense that most Chinese words are just one syllable. But still the characters have a meaning associated to them, that's why there can be dozens of characters that map to the same syllable.

Or multiple syllables for some characters depending on context, like 了 being le or liao. Chinese characters really aren't syllabic.

What? Le and liao are both syllables.

> Chinese characters are syllabic in the sense that most Chinese words are just one syllable. But still the characters have a meaning associated to them

Well, there is semantic information included in the spelling of Chinese words. Far more than is included in other, non-Japanese writing systems.

But that won't stop the script from being syllabic any more than the same phenomenon in every other written language will stop its script from being syllabic or alphabetic. English script is alphabetic even though way is spelled differently from weigh.

Meaning is not used in determining the pronunciation of a Chinese character. (Except to the extent that the same character may have separate uses, as when 長 is pronounced zhang3 if it means 'grow' and chang2 if it means 'long'.) A character indicates a sound, and it always indicates that sound regardless of the meaning of the word in which it appears. This is as pure as syllabaries get.

Kanji do not share those properties. They are not restricted to single syllables. They frequently stand for several different unrelated words. They do not represent any particular sound. They may be drafted into any word with a vaguely appropriate meaning, even if that word is conventionally spelled with other kanji.

> most Chinese words are just one syllable

This is false; they're mostly two syllables.

No J, U, W, and, prior to ~230BC, also no G; around that time also no Z which had been disfavored; almost no K which was barely used if at all, and Y was a distinctively Greek-only letter in Roman times, so that gives you 20 or 21 letters: ABCDEFHI(K)LMNOPQRSTVX for the Latin alphabet.

A lot of Pompeii graffiti is the alphabet. For example: https://www.ancientgraffiti.org/Graffiti/graffito/AGP-EDR167...

21 letters.

It's post-3rd-century and includes the G; you might have remembered an accurate statistic ("20 or 21 letters") while forgetting to count your alphabet, which only contains 19 or 20.

K is barely used, but you wouldn't be able to argue that it wasn't used at all. It's used in "kal.", which you have to use when you're specifying a date.

> Japan’s kanji alphabet

> if you call them kanji, they're Japanese characters [...] Kanji are very much not syllabic

@2, the reason Japanese speakers use the designation 'Kanji' is the same reason that English use 'Latin', as a reference to their origin. To insist that Kanji are Japanese and not Chinese would be as strange as insisting that English does not use the Latin alphabet: it's not wrong in the sense that Japanese and English have added to their respective scripts, but it's also wrong in that it obscures the fact that the idea, the system, the forms, and the principle values (semantics and sounds) are inherited.

Put another way, if the British were to deny in the future that they're not using the 'Latin' alphabet but the 'British alphabet' (hard cultural Brexit?) and demand the Unicode consortium to split it off to a dedicated block (as they did for Coptic, which was initially considered a mere variant of Greek), we then have to wonder whether the French, German, Polish, Portuguese languages all should get their own dedicated block. That of course is denying the fact that major parts of Europe all use the same Latin alphabet, each with their own quirks and flavors added according to locale.

So when "you call them kanji, they're Japanese characters" but only with respect to usage, meaning, sound, and sometimes form, but not with respect to the overall system or character repertoire, which is shared with Chinese.

Also this entire branch of the thread is dangling from a faulty premise in The Fine Article, viz. "Japan’s kanji alphabet", which screams "writer knows zilch about this". Kanji are not an alphabet. Maybe they mixed it up and wanted to write "Japan's kana alphabet" which is sort-of similar but less wrong. It's acceptable if you think of an 'alphabet' as 'glyph repertoire', but then Cuneiform (or any of the roughly 10~15 orthographies and languages from Sumerian to Ugaritic) would be an 'alphabet' too (only somewhat true for Ugaritic, which is esssentially an abjad).

I have to admit that there's apparently no very commonly used word that comes to mind to complete the phrase "Japan's kana ___" that is not wrong ('alphabet'), or specific ('syllabary'), but general and could also be used to fill the gaps in "the Sumerian cuneiform ___", "Chinese ??? ___", "Egyptian demotic ___", "Egyptian hieroglyphics ___", except for "glyph repertoire". "Inventory" is maybe less arcane but also very generic. "Glyph" in the sense of "symbol used for writing" is maybe too academic for a lot of people. "Characters" is another popular choice. "Chinese characters", "Japanese Kana characters", ... but "Latin characters"? Maybe. "Character set" has become a comparatively widely used term already, so yeah.

> Kanji are very much not syllabic. But Chinese characters are.

The even more interesting aspect that (strangely enough) people like John DeFrancis apparently glossed over when discussing the nature of the CJK writing systems is that Japanese is quite a bit closer to 'ideographic' writing. For one thing, the phonetic clues that DeFrancis rightfully stresses so much are often much less useful in Japanese, and on the other hand, characters often have both a 'borrowed' and a 'native' reading while retaining the basic meaning; for example, 犬 can be read 'ken' (a reading borrowed from Chinese) or 'inu' (an indigenous reading), both signifying 'dog'. These characters work like Hindu-Arabic numerals in that the writing indicates am 'idea', a sense (a numerical value) that can be read out in multiple different sounds depending on language, or even within one language (ex. 3, 3rd, 30; 2, 2nd, 20).

> count your alphabet, which only contains 19 or 20

off-by-one errors are to be expected...

K is such a fringe case. I checked back and you're right, it was consistently used for that specific case 'kal.' and for some names. Interestingly though I don't think any (major) romance language retained that spelling afterwards. It's a bit like the British 'ö' in 'Coöperative'. Given that I remember one single time I saw that spelling in the wild while visiting the UK, should I now go and write angry letters to all editors that "English really has 27 letters!!"? How rare is a letter allowed to be used before we declare it "not part of the alphabet"?

> It's a bit like the British 'ö' in 'Coöperative'. Given that I remember one single time I saw that spelling in the wild while visiting the UK, should I now go and write angry letters to all editors that "English really has 27 letters!!"?

Well, no. The 'ö' in 'coöperative' is not a single grapheme; it's two graphemes. The same mark, which we might call U+0308 COMBINING DIAERESIS, is used with the same significance on any arbitrary letter (OK, any vowel), indicating that the letter is not part of a digraph and should be interpreted alone. This is how the distinction is drawn in writing between a coöp [where hippies work; each 'o' is a separate vowel] and a coop [where chickens live; the 'oo' is a single vowel].

The example I generally use to illustrate the conceptual difference is that, in Mandarin pinyin, é and è represent identical vowels, whereas in French é and è represent two different vowels. There's just one letter "e" in the pinyin example (along with two tone markings), but there are two letters in the French example.

(The French themselves would disagree - they don't include letters like è in their official alphabet - but they are wrong.)

Though English, French, etc sharing the same unicode codepoints is not as far as I'm aware problematic, whereas Chinese and Japanese characters sharing codepoints does cause wrong behaviour (wrong shape of character used in text because renderer has to guess whether to use a Chinese or Japanese font) and has to be worked around by marking up text as "this is Japanese"...

You've gotten so nitpicky that you've apparently started mixing your own terms: graphemes, letters; and you're doing so incorrectly to boot (Muphry's Law, perhaps).

There are characters (U+308 in this codepage), letters ("E" and "e"), accent marks (which require an accompanying letter), and graphemes ("e", but also "ee").

Instead of iteratively correcting each other by refining and redefining words in each response, how about we communicate in good faith on the actual topic: cuneiform.