For the record, it’s the casual word for paws! Like cat toebeans.
The pronunciation is what seems a bit weird to me.
I feel like it should be にっきゅう or にくぎゅう, since it seems to follow all of the rendaku rules. Why is it pronounced that way?
My best guess is that it might be confused with 日給 (daily wages) or 肉牛 (beef cattle) respectively, but I still don’t know.
Because it’s language and all the rules on planet earth are ascriptive (I hope I am using the word correctly). Basically, language exists and grammar rules are ascribed to it. It actually makes no sense because it just is the way it is. There are general patterns that we can observe and put into linguistic and grammar books but in the end, language just works, because it works for people.
I wouldn’t fret over these things. Just accept that language is weird and there will never be a proper explanation for everything.
At least that is my take on it, maybe the next guy/gal will have an explanation.
It’s possible that the word only started getting used after the rule of merging the small tsu and the ka line (which I really don’t think is called rendaku. In linguistics, this kind of consonant duplication is called ‘gemination’).
Chinese kango words don’t rendaku. There is too much ambiguity, at least when compared to wago words. This high probability for misunderstanding is most likely why kango compound words do not rendaku , for the most part. These kanji compound words, made up of two or more on’yomi readings are called jukugo (熟語), and only a small percentage of them rendaku.
With the benefit of hindsight let’s look at the article again:
WARNING: While we are calling these “rules,” they’re more like guidelines. Rendaku can be difficult to understand because it’s riddled with exceptions and can be unpredictable. These “rules” exist to help you get a better idea of how and when to predict rendaku, but they’re in no way absolute or unbreakable. Language is tricky like that.
Japanese origin words do rendaku.
Chinese origin word do not rendaku.
I think the article is a great resource. I’ve read it more than once and listened to the related podcast. My ability to predict rendaku, like your own, I imagine, has improved quite a lot. But there’s a lot of rules to master and the whole wago/kango aspect alone can be troublesome because it’s an order of magnitude more difficult compared to the points about voiced and unvoiced.
Yeah I think there’s a reading note on one of the vocab items that mentions the shortening tends to occur when the final syllable/mora/kana/etc is く or つ.