Confusion with some of the reading mnemonics

Hi all!
I’ve recently unlocked kanji (finally!), and I’m now learning the meanings and readings for them. However, I’m having trouble memorising some of the readings, as the way they are written in hiragana is sometimes very different from how they’re pronounced. For example, the kanji reading for ‘river’ is taught with a story where cow is the keyword and I assumed it was かう at first, but the actual hiragana is かわ. Another example would be rock from the kanji power: turns out it’s りょく which sounds more like ‘ryoku’. I can memorise these ones easily enough now, but I’m worried I won’t be able to keep up in the future. Has anyone continuously experienced this issue or am I just being overly anxious?
Thanks in advance!


No, I’d say you gave some pretty good examples on how it’s not always possible to create perfect mnemonics. The team is always trying to improve them, but it’s a bit tough.

My practical advice would be that you won’t need mnemonics to memorize all the items. Most of them will be memorized well enough just by following the SRS. In the future, you can always make your own mnemonics or even use Japanese words you already know for them, since readings tend to repeat quite often :wink:


Koichi, he speak crazy moon language :innocent:

In other words, this’ll happen a lot, and it’s especially annoying if you’re not a native English speaker. You can embrace it, ignore it, or make up your own mnemonics, but you can’t escape it. I mostly make up my own when the existing mnemonic just doesn’t make any sense.


Aye, the reading mnemonics only give the vague idea of the reading rather than the precise sound.

Personally, I instead try to think of a word that I already know which includes that particular kanji.



You can ignore the mnemonics if you want to.


I had the same struggles. I had never even used Mnemonics in my life during other studies, so this was completely new for me, and I found that some are a little off from what a native English speaker expects in terms of pronunciation.

One tip I found helpful, but can also be seen as a crutch, so I hesitate to suggest it, is to write the romaji phonetic pronunciation in the reading notes, so when you get it wrong, you can visually see in an English-friendly way the pronunciation.

However, as I am sure you’ve heard, leaning on Romaji is generally advised against because if you know your Kana well enough you should be fine without it, so take this tip with a grain of salt. I think the SRS will do its job properly for you, as I found just 1 level later that I am hardly referring to the Romaji I write for myself at all.

1 Like

Yeah, there is no way to find an English-sounding word that is pronounced anywhere near りょく. That’s just the way it is. With かわ I guess they could use cowabunga, but that will run afoul of the people who don’t want “popular references”.


This has been tripping me up a lot too so I’ve been trying to find my own mnemonics and using them instead. 力 in particular gave me trouble for the same reason, but I just powered through.

Recently I ran across 麦 (むぎ - wheat), which uses mugging as a mnemonic but I think of まぎ when I do that. Instead I used Moogi the Moogle from Final Fantasy harvesting wheat to support life in the winter.

So find something that you can relate to if the mnemonic is too confusing and stick with that.

1 Like

For かわ I used Kava, which is made with water from a river.

1 Like

I use Car Wash for かわ. But the main this is that the mnemonic should trigger your mind to know the right sound through a memory connection, not because it is absolutely perfect, but yeah sometimes if it’s too off, or confused with another sound, then it can be hard to feel confident in that connection. I’d recommend making your own if you come across ones that seem problematic and keeping a list of your personal mnemonics so you can remember to use them again when you come across duplicate readings (which will be literally everything eventually)


I remember reading somewhere that even if the mnemoics don’t give you the exact sounding, the idea behind is to trigger a memory of the reading and from there you should be able to work it out. So even if the story tells of a cow and you think it is read kau (sorry not sure how to add in kana to text), if you read the first ka and it triggers the cow memory you can possibly relate it to kawa. It’s a bit of a stretch but this thinking is helpful in lots of other kanji.

1 Like

i would humbly invite you to join the party (party? maybe?) in Mnemonics Mnania

also just imagine the Japanese conglomerate Kawasaki having a big factory next to the river and dumping all of its pollution in it. Done


This is why I don’t use reading mnemonics, only meaning mnemonics. I remember readings by their reading in a word I already know, or by using one of the radicals.

You will likely notice that many Kanji with the same radical have the same on’yomi. You can leverage that to not bother with mnemonics as well.

But if you are starting Japanese from scratch and you don’t know any words that use the Kanji you’re learning, then I guess just come up with your own mnemonic. I agree that the WK mnemonics are a stretch and a half most of the time…

EDIT: Here’s the script for the phonetic component blah blah [Userscript] Keisei 形声 Semantic-Phonetic Composition


for this mnemonic I think of Mugi from K-ON and her hair color

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.