I feel like this isn’t something that’s taught a lot in English resources but I find it an interesting thing nontheless.
Instead of there just being ”Onyomi” and ”Kunyomi” there’s actually different types of onyomi that depend on when and where they loaned the words from China.
1: 漢音 (kan-on). Most of them will be these from the tang dynasty, with the name reffering to han chinese people. They tend to sound similar to the next type.
2: 呉音 (go-on) These are the older ones from Wu China. You tend to see them in buddhist language and early loan words.
3: 唐音/唐宋音(Tou-on): Readings from later dynasties of china like Ming and Song, used in the later words. These are quite rare by comparison. They tend to sound completely different from the other readings and stick out like a sore thumb. There’s not many of them.
4: 慣用音.(kanyou-on). These come from people changing or mistaking readings and them getting accepted as standard even though they differ from what it should be compared to the Chinese sound.
It seems like people find all of them to have a bit of a different sound to them.
The dictionary dicitonary.goo.ne.jp tends to show the onyomi type at the bottom entry. Often when people make mistakes reading onyomi words, they often assume it uses the kan-on reading when they’re not supposed to. Very few would guess the tou-on reading without seeing a specific word they know that uses it.
Most of the time the meanings are related to one another, though even then sometimes a specific reading has a specific meaning the other doesn’t. Most of the time Chinese only had 1 reading for a character but sometimes they have 2 with different meanings and that sometimes carries over to Japanese.
\—–
If you can already read Japanese there are ways to predict which reading is what.
For Touon, it’s just that one weirder, less common reading you wouldn’t expect. An exception is when one of the kan/go-on readings is non jouyou and rare, but that’s usually not the case. Luckily, for go-on/kan-on there are some patterns for when a kanji has both types of readings you can use to predict which reading it will be with quite a bit of accuracy.
1: If the pair has a kana row of the voided/voiceless counterparts asin ガ行(g- Starting kana)・ザ行・ダ行・バ行 vs カ行・サ行・タ行・ハ行, then Go-on will be the voiced ”dirty” ones (濁音) and kan-on will be the voiceless ”clean” ones (清音). This is the exact opposite for the next two patterns which have to do with when the two pairs their sounds are different rather than the same.
2A:: N- to Z-: If 1 pair starts with the ナ行 and the other with a ザ行 , then ナ行=Go, ザ行=Kan .
2B::N- to D-: If 1 pair starts with a ナ行 kana and the other with a ダ行 , then ナ 行=Go, ダ行=Kan .
2C: M- to B-: If 1 pair starts with a マ 行 and the other with a バ 行 , then マ 行=Go, バ 行=Kan. Notice in how all three the unclean/voiced sounds (濁音) are the kan-on, while in the first pattern the clean sounds/unvoiced sounds are the kan-on. So kan gets N, N, M and kan gets Z, D, B.
3: -Chi to – Tsu: This one has to do with the ending rather than the start, and only applies to a specific kana, not the whole row. If the last kana in their pairs and with チ and ツ , then チ = 呉 and ツ = 漢
4: it seems like if you have something like myou vs mei and shou vs sei then the ou ones tend to be go-on and the ei ones kan-on but don’t quote me on that haha.
I don’t know what use this has and I’ve known about it for a long time, but I dunno, as I was adding some reading types to my kanji list, I thought it’d be neat to share for fellow nerds like me who happened to not get far enough yet to know or just overlooked it :3