I think they have that mixed up with hiragana and katakana. Kanji are Chinese characters.
https://www.merriam-webster.com/dictionary/alphabet
> al·pha·bet
> a set of letters or other characters with which one or more languages are written especially if arranged in a customary order
And then it shows an Alphabet Table beginning with Hebrew and Arabic.
But Chinese characters are.
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.
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"?
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.)
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.
> 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).
How to Write in Cuneiform, the Oldest Known Writing System in the World
You can argue that, "well, obviously!" but correctness and exactness are what makes science, history, journalism, etc good, and allowing incorrectness like this is a step backwards.
I read a history book when I was a teenager (can't remember which one, unfortunately), and the author wrote a preface that said something along the lines of "Everything in this book is based on the published information I could discover during my research period of April to September 1999. I have chosen to write in absolutes--stating many things as certain and clear--but in reality there is still much we do not know about this time period. No history author should say their writing is fact and any good historian will make it clear that their work is composed of assumptions layered on assumptions. Please read these works with this in mind."
If you don't have a preface like that, you should add "known" to your title/sentence! I will argue with someone all day over this! I will die on this hill!
Writing has only been invented independently a few times in history, so it seems reasonable that cuneiform could be the first.
Why do we not see more writing in the archeological record? Maybe agrarian societies both motivate writing and are required to provide the free time to invent it? Or perhaps it was written on media that's subject to decay? If some society developed writing on tree bark 100,000 years ago, none of that is going to survive and we'd never know.
Of course, if you’re a fallibilist you believe that it’s always possible that you’re making a mistake. It seems unnecessary to always add “unless I am mistaken,” because that hedge always applies.
How to Write in Cuneiform, the Oldest Known (by the author) Writing System on Earth, the third planet from the Sun in the Milky Way galaxy, as of 2025 as long as you're a human without a major disability that would prevent you from using these techniques or are at least a being with similar hands and arms also able to obtain the necessary materials and can read and comprehend modern English if you aren't too busy doing other things and expect to live long enough to complete the task
You often get nitpickers going after some small technically correct detail which may be true but no reasonable person in the intended audience would ever actually need to be told. No one reading the original title would assume that the author had omniscient knowledge of the whole human history of writing beyond present archaeological fact and this doesn't need to be pointed out.
Petroglyphs are not a form of writing, and the Kush tablet along with a few others are considered to be precursors of the proto-writing – at best.
So I reached for my trusted Ouija board to ask whether writing predates Sumer. It spelled, with unsettling clarity: «Y E S . B U R I E D . D E E P». Then it paused. «N O T Y E T M E A N T T O B E R E A D». Mysterious? Yes. Confirming? Not quite.
Has anyone considered that possibility?
This seems unlikely.
https://en.wikipedia.org/wiki/Baghdad_Battery
The Baghdad Battery is the name given to an artifact consisting of a ceramic pot, a tube of copper, and a rod of iron fixed together with bitumen. It was discovered in present-day Khujut Rabu, Iraq in 1936,[1][2] close to the ancient city of Ctesiphon, the capital of the Parthian (150 BC – 223 AD) and Sasanian (224–650 AD) empires, and it is believed to date from either of these periods.[3]
Its origin and purpose remain unclear. Wilhelm König, at the time director of the laboratory of the National Museum of Iraq, suggested that the object functioned as a galvanic cell, possibly used for electroplating, or some kind of electrotherapy. There is no electroplated object known from this period, and the claims are universally rejected by archaeologists. An alternative explanation is that it functioned as a container for magic spells for protection, defense or curses.[2]
Ten similar clay vessels had been found earlier. Four were found in 1930 in Seleucia dating to the Sassanid period. Three were sealed with bitumen and contained a bronze cylinder, again sealed, with a pressed-in papyrus wrapper containing decomposed fiber rolls. They had been held in place with up to four bronze and iron rods sunk into the ground, and their cult meaning and use are inferred. Six other clay vessels were found nearby in Ctesiphon. Some had bronze wrappers with badly decomposed cellulose fibers. Others had iron nails or lead plates.[2]
There was no "Analytical Engine" in ancient Rome. I don't even know what this could be referring to.
As your link mentions, the "Baghdad Battery" was debunked long ago, it's just a jar for storing scrolls.
The modern symbology of circuit diagrams was invented in the 20th century.
None of those cuneiform structures even work as circuit diagrams.
It's coincidence. The Dendera lightbulb isn't a lightbulb either.
I was really looking forward to seeing how that got translated into something other than English.
But in the fine print, they say this instead:
> We take the letters from your message and transliterate by syllable, as nearly as we can, into [Old Persian] cuneiform.
The choice of Old Persian cuneiform is lazy, much like those Egyptian cartouche name necklaces that use a hieroglyphic alphabet rather than even attempting to write out a name as ancient Egyptians would.
all you need is essentially a basic website and billing + $0.30 of clay + basic kiln or oven + broken reed or bamboo stick.
chances are the orderer won't even know if you just made it up.
I highly recommend watching his lectures and British museum feature on YouTube if you’re interested in the ancient world.
Nice article: https://www.telegraphindia.com/7-days/the-ark-that-finkel-bu...
Excellent lecture: https://www.youtube.com/watch?v=s_fkpZSnz2I
I studied Greek and Hebrew in college, Latin in high school. In each the very first night's homework was to memorize the characters and their pronunciation.
Multiple ANE cultures used cuneiform (Ugaritic, Akkadian, Sumerian, Hittite, and so on). The time to master each depends on your native language, the target language, and exposure to similar languages. The writing system is not the hard part.
All of the examples you mentioned are derivatives of the Phoenician alphabet, which have around 20 to 30 characters each. Even with case sensitiveness and diacritics, I think they still add up to under a hundred characters.
Cuneiform character sets are in the order of magnitude of the several hundreds or even thousands, depending on the language[1], so I imagine that the experience is closer to learning to read Chinese or Japanese and less like Hebrew and Greek.
That being said, I've never tried to learn neither cuneiform or hanzi, so I'm just guessing based on the number of characters.
[1] https://en.wikipedia.org/wiki/Cuneiform#Sign_inventories
Cuneiform was also used over such a vast period of time that significant evolution took place. e.g. As numbers and mathematics evolved, there were sometimes different symbols for the same numbers depending on what was being counted. Scribes often had to learn several sets of numerals and when it was appropriate to use each of them.
The modern reader needs to learn, not only languages, but contexts and also be aware of how the script evolved over time.
Cuneiform generally evolved to become simpler and less contextual as time went by, but there remained a lot of characters to learn by the common era. The Phoenician alphabet was a huge step forward precisely because it was much simpler and easier to learn. Shaving years off of the learning process turns literacy into a common skill that many can obtain, rather than a select few whose families can afford to send them to a school for many years.
Now all we need is LLM support for translation. Is there enough cuneiform content available for training?
Some discussion on HN: https://news.ycombinator.com/item?id=44280168
thaumasiotes•14h ago
Writing systems using cuneiform include Sumerian, Akkadian, Hittite, Hurrian, Ugaritic, Old Persian...
The article states that "Properly written out, these syllables join up into a flowing calligraphy that your average, educated Babylonian would be able to read at a glance", so presumably they're thinking of Akkadian. Why not say so?
(Does it make a difference? Consider that our first attempt to read the name of Gilgamesh came out as "Izdubar".)
wl•11h ago