Drawing Kanji and Vocabulary

So for level 7, at 30 kanjis, I need to Guru 27 Kanjis. Gotcha.

形【かたち】this is the kun’yomi whereas 【けい】is the on’yomi. I think it’s probably safe to assume that the majority of kanji are first introduced in their on’yomi since they’re then used as construction blocks for vocabulary.

The only kanji remaining is ‘house’ or 家. I was under the impression that the on’yomi of a kanji was somehow related to the Chinese pronunciation. か・け, neither remind me of the Chinese pronunciation of that kanji. Thankfully the car か being left at home is easy to remember.


Day 8: Jukugo

I didn’t have much of anything to practice or analyse today. I’d say everything is going smoothly. I got a 100% for a 24 review which is pretty impressive to me anyway.

So anyhow, I decided to google jukugo because it was previously mentioned in my blog and I didn’t know what it was so I looked it up.

I was in for an interesting read:

:memo: 部首【ぶしゅ】are what we know as the radicals.

漢字【かんじ】we all know what this is.

熟語【じゅくご】compound nouns, that’s all there is to it! I was under the impression it was going to be something super complex. Most vocabulary are likely jukugo by that definition (those who aren’t written with a single kanji that is).

JUKUGO (compound words i.e. 種類, 漢字, 哲学, etc.) use the ONyomi.

I’m not sure that’s true. I could’ve sworn there were instances where the kun’yomi was used in compound words. I’d be glad to be wrong though as it’d make my life a whole lot easier. @Jonapedia ?

I don’t know if I could get confirmation about this statement:

HOWEVER … jukugo which have hiragana in 'em (i.e. 手取 , 人当 たり , あざ ) use the KUNyomi, not ONyomi.

Kanji Facts | KANJIDAMAGE.

That’d be a pretty awesome discovery if it turned out to be true. I’m guessing the author is saying that the kanji which precedes the okurigana uses the kun’yomi, not all of the kanjis which are part of the word itself.


True, but if you look at Japanese dictionaries, there are actually two definitions, though we could of course still call both of them ‘compounds’. 熟語 are either compound words (i.e. made up of two or more words) or kanji compounds (i.e. made up of two or more kanji). Since those are pretty similar, it’s fine to remember them as (roughly) the same thing though.

This statement is easy to disprove with a ridiculously common word: 番組(ばんぐみ)=‘TV programme’, which is an on + kun mix. I can give you even more: look up all the possible readings for 生魚. We have せいぎょ, なまざかな and なまうお. The last two are roughly synonyms, and the first means either ‘living fish’ or ‘fresh fish’ (which I guess you could say is roughly the same as ‘raw fish’, which is what the other two mean). Point is though… I’m sorry to break it to the author of that article (who seemed very angry that his Japanese teachers failed to highlight this simple – but sadly incorrect – rule), but we all know that なま and ざかな (=さかな with rendaku) are readings of 生 and 魚 that don’t require any okurigana. Final example: つきぎめ (which should translate as… ‘monthly price/fee agreement’) is usually written as 月極, but it can also be written as 月極め. It’s really a matter of common usage. Yes, pure kanji compounds that contain no okurigana usually use on’yomi, but that isn’t always the case.

Yes, that would be a possibility. Are there compounds that contain okurigana and use a mix of on’yomi and kun’yomi, with the split highlighted by the okurigana? Probably. Can I think of a good example now? Uh… 番組 comes pretty close, but it’s not commonly written with okurigana. [kanji]+し would do the trick, but since it’s a well-established suffix, that feels like cheating… I’m pretty sure examples exist though. I’m just not able to give you one that exists as a standalone word right now.

EDIT: OK, got one. Easy. It’s a term from the world of VTubers and streaming: 凸待ち(とつまち), which is also on + kun. It refers to waiting for viewers/friends to call in for a chat during a stream.


It’s kind of scary that it’s the first google result for Jukugo considering the inaccuracies…
And also:



Oh well. I guess it’s just a very popular site, and perhaps many people have found it helpful.

Well… we have repetition of the Is, I suppose! Seriously though, apparently ‘petition’ can be pronounced both as ‘pi-TI-shun’ and ‘puh-TI-shun’, so perhaps that’s what happened. I personally prefer the second pronunciation, but Oxford lists the first in its definition. However… the recording on Lexico definitely sounds more like the second, so perhaps even Oxford isn’t sure. Hahaha.

1 Like

It’ll teach me to call out someone’s spelling!

Anyhow, there’s something I find odd about WK and jukugo words.

If I take 自由【じゆう】, freedom. When I look up the tab about the reading I get the following explanation which I regularly get for other words as well:


This is a jukugo word, which usually means on’yomi readings from the kanji. If you know the readings of your kanji you’ll know how to read this as well.

This would imply that this word composed of two kanjis uses a combination of the on’yomi of both individual kanji.

However 自 has two on’yomi: し・じ whereas 由 has three: ゆう・ゆ・ゆい. In other words, even if in this case the on’yomi is used for the reading of both kanjis… it doesn’t really help since there’s isn’t just one on’yomi for both kanjis. There are various possible combinations and yet there’s just one correct reading, じゆう.

After checking it seems like all of the compound words are not provided with mnemonics but rather the sentence I’ve quoted above.

1 Like

I find that when there’s multiple onyomi readings, often there’s only really one that ‘feels’ right to me. This isn’t always the case though (たい andだい for 大 always get me). Most of the time though if I need to try and recall them (in the case that I don’t just recognise it instantly) I just kind of go through the readings I know and see if it feels right


Do you know if each reading for 大 occur in different circumstances or are they completely arbitrary?

1 Like

As far as I can tell, completely arbitrary, but if someone knows a magic rule for differentiating I’m all ears

1 Like

Hahaha. No, no, you were right: that word is definitely spelt wrongly. Different people might pronounce it slightly differently, but the word certainly ought to be ‘repetition’. English spelling doesn’t make a lot of sense, unfortunately, but we all have to live with it. Hahaha.

Yeah, I’ve seen some other people say that this isn’t enough, and I understand that. However, perhaps WK is referring to whatever on’yomi readings have been taught by the SRS? I’m not sure.

I mean, sometimes there are other possibilities that exist, but in essence, I second this: it’s not exactly about ‘feeling’ right, perhaps, but generally speaking, there are only one or two readings for the entire compound that mean exactly what you want it to mean. You can definitely try to come up with more specific mnemonics to help you remember the reading of the compound, but the other route is just to… keep practising until you know that for a given compound, one reading exists (or sometimes two), and the other is completely wrong.

Maybe the Japanese internet has some answers, but I don’t feel like bothering right now. I personally think it’s like 国: I believe it’s こく most of the time, but sometimes it’s just ごく for some reason. No real explanation. I can’t think of any specific differentiating factors between the two on’yomi of 大 anyway.

1 Like

こく becoming ごく is called rendaku! There are some patterns to this, but of course, there are also plenty of exceptions! My understanding is that it generally happens primarily to kun’yomi readings (so 国 is an unusual case), and typically it’s the second kanji that does it, but there are circumstances that stop it from happening, plus occasional irregular readings that break the rules, haha!

I haven’t actually done any formal investigating into this yet myself, but I’ve picked up a lot of the patterns just from using the rendaku information script and reading the rendaku section for each new vocab word that I learn. A lot of the stuff in the script went over my head at first, but the longer I’ve had it installed, the more I’ve begun to get a sense for the patterns. Of course, you can pick up on a lot of this eventually without using a script, but the script is nice because it directly spells out what is happening, so it helps you conceptualize it better.


Regarding 国, it’d be interesting to have stats on its reading.

米国「べいく」Why is it こく for USA but 中国「ちゅうく」for China considering they are both countries?

外国「がいく」こく for Foreign country…

When it comes to heaven, 天国「てんごく」ごく for heaven…

I doubt it. It’s too early for me to tell but I’d wager that code handles all compound words the same way by displaying the same sentence. Perhaps providing mnemonics for each compound word would be too massive a task or maybe it’ll come later on.

I’d certainly be interesting to learn more about the nature of on’yomi or Japanese readings in general. 何日‣なんにち but not なんじつ. There has to be reasons why certain readings were adopted over some others in different vocabulary words. This case isn’t a case of rendaku however, it’s a completely different vowel, not just the softening of a consonant.

I also found another example which disproves that on’yomi are used in compound nouns: ankle ‣ 足首「あしくび」. あし is the kun reading, not the on reading.手首 also uses the kun yomi of 手 so maybe this one is contextual?


I’m aware that the phenomenon is called ‘rendaku’, yes, written 連濁. I did look through the Tofugu article about the patterns, and that helped me get a sense for when it happens, but the article itself immediately raised a ton of exceptions, which led me to decide not to attempt to learn the rules by heart. As even the Tofugu article says, much of the information provided is nothing but ‘theories’, and the rules – if they do exist – are still being ‘researched’.

My point was this: can you explain why 戦国, for example, is せんごく? The phenomenon has a name, certainly, but do we know the reason? As far as I can tell, the answer is no. Similarly, why is ‘minister’ 大臣(だいじん), whereas a ‘grand meeting’ (that is, a conference, a tournament, a rally etc.) is 大会(たいかい)? Maybe research has answers for us, but at the very least, I’m pretty sure that the list of ideas in the Tofugu article cannot explain this. That’s why I prefer to have a default on’yomi in my head, and to learn the exceptions one at a time, all the while keeping in mind the cases in which rendaku is less likely so I have a higher chance of getting things right, even though I know words of Chinese origin are less influenced by typical rendaku rules.

Not exactly, though it’s true that there’s a pattern for words ending in くび(首): they all seem to refer to the narrowest part of a particular body part. 足首, 手首 and 乳首 all follow this pattern.

Definitely, and the reasons would be interesting to find out about. However, for this particular case, I think it’s just that にち is usually the reading that’s used as a counter. I haven’t seen じつ outside of compounds, and it’s usually not related to numbers of days.


I didn’t look close enough at your post and thought you were Zizka, my apologies! I know that you don’t actually use WK. But hopefully the rendaku script is useful to someone else!

Regarding rendaku, the research in the article suggests that 戦国(せんごく) possibly redakus so that it does not get confused with other words that are read せんこく (Yomichan brings up 宣告 and 先刻 that both have that reading, whereas 戦国 is the only word that comes up with せんごく). Both of the kanji in 戦国 are also pretty commonly used for a whole bunch of different things, I think? Because the word needs both of the kanji together to convey the meaning (戦 or 国 alone does not convey the same concept), that apparently also makes it more likely to rendaku. Of course, like you said, these are all just theories, and no one really knows for sure, but both of those explanations make sense as possibilities.

I’m not remotely an expert, but it does seem like the rendaku phenomenon is separate from the sound changing in the first kanji? Or at least, rendaku have their own patterns and tendencies that operate separately from other variations in the language (hence the ability for people to find patterns at all, even if they aren’t universal).

From my experience with it so far, it’s not really the kind of thing you can really memorize a set of rules for up front, just gradually get a feeling for over time, but it does help to think about each individual word as you encounter it (which is why I love the rendaku script, because it saves me the work of doing it myself). The majority of the words in WK follow the general rules in the Tofugu article (according to the forum thread for the script, only about 170 WK words are exceptions. WK has 6410 vocab words, so roughly 97% of them follow those general rules). When I do encounter a word that breaks the rules, it stands out to me, and because it stands out, I’m more likely to remember its reading.

I recently learned 足し算(たしざん) (addition) and 引き算(ひきざん) (subtraction), and the rendaku script helpfully informed me that those two (alongside multiplication, division and calculating with an abacus) are the five types of calculation in WK that rendaku. Clearly, even if I don’t understand it, and even if it isn’t super widely applicable, there is some sort of pattern going on there with words referring to calculation, and it’s something I should keep an eye out for in the future! (The rendaku script also makes note of the countries).

But ultimately, yeah, it does seem to be the kind of thing you just have to deal with one word at a time. Though if native speakers are able to develop a sense for it somehow (as the article claims), then probably the more exposure and the more vocabulary we learn, the more we’ll develop a sense for it, too.


I hope so too. Thanks for sharing it nonetheless. It does seem like a good resource.

Mhm, I did look through the article once more, and I guess that’s a possible reason. My question would be though… Japanese has so many homophones anyway, so what’s one more? Of course, I’m not unhappy with the fact that there’s rendaku here by any means, but I’m sure there are other words for which our lives might be made less confusing with rendaku.

Yes and no. 戦 is about ‘battle’, and 国 just means ‘country’ or ‘nation’, so I’d think that combining the two very easily yields ‘countries at war’ or ‘warring states’.

However, yes, the article did raise the idea that the two kanji functioning as ‘one word’ is a factor in rendaku, and it’s true that 戦国 refers to a single specific concept in Japanese history (the Warring States Period, of which China also had one), so it does make some sense to suggest that it’s some sort of ‘single word’, just as 天国 refers to ‘heaven’, which one cannot directly derive from a combination of the two kanjis’ meanings.

Same here. When something goes against my expectations, I’m more likely to remember it because it’s surprising.

Interesting. I guess they might be a form of common usage then, or perhaps they are seen as single concepts.

I’m hardly ‘done’ with learning Japanese, but I think I have an idea of what they mean. Some words just feel like they ‘should’ contain rendaku. I don’t think the rules are hard and fast, and most likely even native speakers make mistakes with unfamiliar words at times, but certain things feel too ‘disconnected’ without rendaku e.g. なまざかな, which I mentioned earlier. なまさかな felt really strange to me.

Oh yeah,

Yes, by definition, rendaku is about how the second word’s pronunciation changes.


Just to be clear, the script tells you whether a word rendaku(s) or not, right?

1 Like

Yes! And it also tells you why it does, if it follows one of the basic rules. If it’s an exception, it will point out that it is one, and if there are similar words in WK that the exception also applies to, it will list them. The only words it doesn’t have information on are words that have irregular readings for the kanji, which you have to just memorize individually.


Day 9: 時代

Yeah, not too original a title hmm? I just ran of ideas. I’m at the end of the level now so it’s just your typical review.

Regarding the script, I’m dubious. I don’t want to rely on scripts to remember things but on the other hand I can see how it would be informative to have them covered and explained as they show up.

時代【じだい】period. I think I got 90%+ in my last review so it’s getting harder to discuss mistakes although I find that if I talk about (invest mentally) mistakes, I am a lot less likely to repeat them.

The thing with 時代 is that in its composition (‘time’+‘substitute’) doesn’t remind me of the word ‘period’.

So it seems like it can be used like a suffix to indicate a certain time period, such as in this case, the time period of college.

Or something more time specific like here. I googled it and it refers to a Japanese historical period (794 to 1185).


Unfortunately you’ll need to remember whether a reading rendakus or not regardless of whether or not you use a script to point out why it’s happening, haha! You’re of course on your own without a handy script for vocab words outside of WK, but if you look at the script when you learn new words here (or manually look at the Tofugu article over and over again, though that’s substantially more work), you’ll gradually start to pick up on the patterns yourself, and it’ll make it easier for you to predict whether an unknown word rendakus, or in the case of the exceptions, it’ll at least help you remember that the word is an exception. Basically all it is a tool to get you thinking about these things when you learn a new word. I didn’t even realize how much I’d learned from it until I got a couple months into my studies and realized that I was suddenly able to recognize what was happening when I encountered new words outside of WK.


Eras occur successively and substitute for/replace each other. I think that’s the link. I personally use 代 more commonly in the sense of ‘generation’ or ‘time period’ than in the sense of ‘substitute’ even in Chinese, because when it has that sense, it’s usually accompanied by something else that indicates which meaning is being used. I’m not sure what the trends are in Japanese though.