not sure whether this fits into category “kanji” or “reading”.
sometimes it feels like too many onyomi are either “kyou” or “kyuu”. It’s already hard as it is, but when I’m having a bad day and can’t concentrate well, it’s always going like: “you fool, that kyuu and not kyou” and “hahaha - that’s kyou and not kyuu, you idiot!” in the reviews.
so, I’ve taken some time to look for something that could help my brain to keep them apart and i found those tricks, that seem promising to me.
it seems that:
kyou/kyuu
kyou is often used for abstract or conceptional meanings
kyuu is often used for physical or practical meanings.
chou/chuu
chou is often large, extreme or long
chuu often means something like punctual, central or inside
shou/shuu
shou is often used for intangible things and concepts
shuu is often used for words that have something to do with grouping
tou/tsuu
tou is often used for something fixed or stationary
tsuu is often used for something that involves movement or change
kou/kuu
kou is often used for something structural or related to human society and progress
kuu is often used for something raw, natural or empty
my phrasing might not be perfect (english is not my native language and i’ve translated those) and there sure is still a lot of room for improvement.
Honestly, the thing that helps me the most is to learn the vocab. Then I can think to myself “where have I seen this kanji?” and remember a word. Then I know for sure. If it’s something that you haven’t guru’d yet and haven’t gotten vocab in WK for, you can always look at the examples page of the kanji and see if there are any words you recognize.
I feel like you’re seeing patterns that don’t really exist, or that exist mostly due to the fairly small set of kanji you have encountered so far. I don’t think there’s any relationship particularly between the on reading and the underlying meaning of the kanji (except occasionally for cases where a subcomponent of the kanji is being used as both a pronunciation hint and a meaning hint, which does happen sometimes).
Your best bets for on readings, I think, are:
a lot of kanji have a component that is acting as a reading hint (e.g. in 求 球 救 the 求 component is a hint that these all have a きゅう onyomi); as you pick these up you can make better guesses
remember a word that uses this kanji with this onyomi (bear in mind that the only reason WK is teaching kanji readings is to help in learning the vocab readings)
come up with a mnemonic that ties this specific kanji to a reading, for particularly troublesome cases
Remembering the right vowel does seem to be tricky for some reason, though – as a native English speaker I do find I make more mistakes with the vowel sounds than the consonants on words I’m learning and have not quite remembered correctly. I’m not sure why that is.
thanks a lot for your detailed reply. i appreciate that information.
it’s a little bit discouraging though. i usually excel at understanding (in a logical way) and completely fail at learning. i was hoping to find a very logical connection that would help me to trick my brain.
as a native german speaker it’s the same for me and especially with those pairs in onyomi (kyou/kyuu and chou/chuu are the worst for me).
I think it’s OK to theorize, only to debunk some, or a lot, later. (Looking for hidden patterns that seemingly don’t exist.)
Well, it comes from China, at some era in the past, though adapted to Japanese people’s ear. Not sure if it would have as much general trend as onomatopoeia, some surprisingly related vocab, and maybe kunyomi…
A better way could be to strengthen the phonology (vowel moras here) and listening sufficiently, for words, though.
Unfortunately thousands of years of history have rather distorted what logical connections there once were: although “one component to give general area of meaning plus one to hint pronunciation and sometimes also more specific meaning” is the underlying idea of most of the kanji, multiple things conspire to make life hard:
Old Chinese had more syllable sounds than old Japanese, so characters that had distinct pronunciation in Chinese got squashed down, creating lots of unrelated homophones (this is part of why so many kanji have a こう onyomi)
Because borrowings happened in multiple “waves” at different times, the Chinese pronunciation shifted between each wave (this is why some kanji have several different onyomi)
Characters later got simplified, so sometimes the bit that hinted the reading got dropped, or characters that were originally distinct got merged
Meanings drifted, or characters acquired secondary meanings that eventually became the only meaning, so that meaning hints aren’t always actually hinting the meaning
And of course the kanji has no hints at all to the kunyomi…
I also suspect that for a fair portion of the early history of the characters, “make them easy to learn” was almost an anti-goal: being a scribe was a good profession and it wasn’t in anybody’s interests to make it easier to be literate.
Ultimately this means that although kanji history is interesting, learning them boils down to a mix of “whatever mnemonic and other tricks work for you” and “brute force repetition” and “exposure”. It does get gradually easier as you go along IME.
it would already be helpful to me, if it would cover like 2/3 of all daily use kanji. but if it turns out to be completely coincidental and a wrong assumption, then i guess it could easily back-fire on me.
though imo, word choice considered, written language is after all a tool for verbal communication. Sound matters, regardless of symbols used to display. Unless you go somewhere very formal…
Not sure about if there is a meaningful pattern, but thinking about it now, it’s more useful to try to guess a potential word in listening, along with Kanji used in that word. Not exactly just the feeling of the sound.
as a native german speaker it’s the same for me and especially with those pairs in onyomi (kyou/kyuu and chou/chuu are the worst for me).
/quote
This is probably because English (and maybe German?) typically has fixed consonants but variable vowels. If you said “kyou-cumber” instead of “cucumber” you’d be understood, although people might think you were saying it in a different accent or funny voice. So our brains look at the consonants for meaning, the vowels just fill the spaces between.
Three years in and kyou/kyuu still costs me review points on bad days, so the abstract vs physical framing is genuinely the most useful mnemonic angle I’ve seen for that pair.
If there’s a specific item where I repeatedly get slightly confused on the reading (rather than not remembering it at all), like saying shou when it should be shuu, I practice saying the word over and over until it sticks. Especially with rendaku. I used to say chukoku so many times for 中国 instead of chugoku, I must’ve failed that word at least 8 times. Now when I see 中国 all I think of is me saying chuGOKU over and over and over. It’s just a matter of making the reading sound more natural to you.
So far, this helps me a lot. but I also take other user’s warnings serious. It might be limited to beginners vocabulary/kanji.
Anyway, even if it’s only helping me for a short while and with a limited set of items. It’s still worth it for me, as long as I keep in mind that it’s just a temporary trick to remember and not a rock-solid rule.
I think you’re spot on with that. German language also depends far more on consonants than on vowels and ‘o’ and ‘u’ are especially exchangeable. As you said, it sounds funny, but people will still understand.
It’s even a typical comedic way to mimic the infamous speech pattern of a certain historical asshole in written text (writing ‘o’ for any occurrence of ‘u’) to make it sound stupid in your head.
I highly appreciate all of your advices and explanations. Still, I sometimes think that it’s possibly easy to forget that (depending on the challenges we’re facing when we’re very young) our brains tend to wire themselves in very different ways.
Some people excel at learning, other people excel at comprehending. I think, most enviable are those who possess a balanced measure of both abilities. Sadly, I almost completely fail at learning vocabulary. Of course repetition works for me too somehow, but not nearly as well as it seems to work for other people. Please don’t think that I’m looking for some kind of magical solution, that would work with no learning at all. Of course I do the repetitive learning thing all the time too.
So, even though it might seem like a waste of effort from your point of view: For me, even the most vague logical pattern would be highly appreciated. And even if it would apply only half of the time.
If any of you have any tricks other than just repetition, I’d be totally interested. No matter how absurd they might seem and even if those tricks could possibly seem useless to other people.
I still mix up certain onyomi pairs more than I should, so i get the appeal of any trick that helps them stick even temporarily. WK mnemonics in general work this way anyway, they’re scaffolding you use until the reading is fast enough that you don’t need the cue anymore, and then it just fades and the reading is there on its own.
trying to identifying patterns by yourself is risky since they might be misapplied - and borne out of your more limited exposure to the language.
what i do is either confirm patterns (or learn about them) from learning material or from AI - or more commonly, enjoy reading about the etymology of the kanji. even if it doesn’t directly work the same way a mnemonic does, it helps make the kanji become more “familiar” and not just an arbitrary set of strokes (e.g. understanding the etymology of 待, 持, and 特 helps me distinguish between them as a supportive method in addition to the mnemonic).
one important pattern for example is the fact that many (not sure how many %) kanji have a “meaning radical” and a “phonetic radical”. this means for example that many kanji featuring the 正 radical have an onyomi of “sei” or “shou”. this is something even native japanese people rely on a lot btw - with kanji they don’t know or are more rare.
but like others have said, learning kanji in the context of words is by far the best way. i’ll give an example. there is a great japanese film called: “love exposure” where a girl joins some cult and then her friend (slash step-brother) tries to get her to snap out of it by kidnapping her (in a brainwashed state) from them. at some point they physically struggle and then she shouts at him: “はなせよ! はなせよ!” i.e. let go! let go! as he tries to prevent her from running back to the cult). this scene has been burned into my mind, so now when i see 放す in wanikani i have no problem to immediately recall the reading even without any mnemonic. (same goes for 三つ、顔、愛, 希望、鏡 etc. all words that are deeply ingrained now in my brain - from that memorable scene)
so the best thing you can do (imho) is to collect as many of these “memorable examples” for each word - that would give you the widest coverage of kanji as possible.
ofc this is easier said than done - there are some leech kanji/vocab that i simply cannot nail down or find good examples for - but on the other hand that’s also a good sign: if you rarely run into a term in the wild, it also means it’s probably less useful or important for you to know it.
one last comment - what helped me to try and artificially create these memorable scenes myself - is to try and visualize the mnemonics (not necessarily WK ones, sometimes my own making) using AI models like nano banana. this means im not just remembering a “story/narrative” but remembering a specific visual scene.
