Hey there folks,

I've noticed that lately, I cannot rely completely on the "On" readings as a trustworthy reading method for when kanji are combined to make up a word. Let me explain: 頃 for example has the On reading of "ころ" yes? So this is fine and dandy until you combine it with 昼頃 and now you've got "ひるごる." This "ごる" comes from a morphing rule that I am not grasping, to be honest. Another example: 算　（さん） until you combine it with 割り算 and now it's (ざん.)

I'm all for learning by example, but I'd like to know if there's a hard rule set out there for the metamorphosis of the Japanese syllabary due to certain kana coming after one another. If not that's cool too.

Basically, the rules that exist are too complicated and have too many exceptions to be useful. Just learn when it happens (it's not like it's hard, it's just voicing the consonant--and many people will understand even if you mess it up).

Some such rules are:
(With Chinese loan words) the consonant is voiced when it comes after a nasal sound (ん）
患者：かん＋しゃ　ー＞　かんじゃ

With native Japanese words (not On readings), the consonant is often changed to a voiced consonant if the first word is modifying the second. If it's not, if it's just expressing two different concepts, it does not usually become voiced.
行き所：いき　＋　ところ ー＞　いきどころ　Destination
vs
行き来：いきき　coming and going

上げ底：あげ　＋　そこ　－＞　あけぞこ　false bottom
vs
上げ下げ：　あげさげ raising and lowering

There are a few other similar rules, which are not too useful in practice.

Just to point out that ころ is a kun reading (not an on reading), and that ごろ is in fact listed in edict.

@fakewookie

Thanks for clearing that up.

@Tzadeck

I think I understand. They aren't completely different from the On reading and are just voiced differently. That much I understand, but it seems like there are little nuances hard to ignore like いつ usually changes to いっ when another kanji comes after it. On the other hand these patterns always have exceptions like you say, so it's not good to rely on them.

Sileh Wrote:@Tzadeck
I think I understand. They aren't completely different from the On reading and are just voiced differently. That much I understand, but it seems like there are little nuances hard to ignore like いつ usually changes to いっ when another kanji comes after it. On the other hand these patterns always have exceptions like you say, so it's not good to rely on them.

Yeah. A lot of it is just about what is easier to pronounce, and you kind of get a feel for it. For example, as for つ turning into っ, you often see it before the consonant sounds of k t p h and s. It's just because it's hard to pronounce a つ followed by those sounds.

Tzadeck Wrote:Yeah. A lot of it is just about what is easier to pronounce, and you kind of get a feel for it. For example, as for つ turning into っ, you often see it before the consonant sounds of k t p h and s. It's just because it's hard to pronounce a つ followed by those sounds.

I'm not sure how much it has to do with being hard or easy to pronounce -- つ plus those sounds occur quite often in the language outside of compound words (待つか?, 雨がぽつぽつ降る, あいつほどじゃない, etc.)

yudantaiteki Wrote:

Tzadeck Wrote:Yeah. A lot of it is just about what is easier to pronounce, and you kind of get a feel for it. For example, as for つ turning into っ, you often see it before the consonant sounds of k t p h and s. It's just because it's hard to pronounce a つ followed by those sounds.

I'm not sure how much it has to do with being hard or easy to pronounce -- つ plus those sounds occur quite often in the language outside of compound words (待つか?, 雨がぽつぽつ降る, あいつほどじゃない, etc.)

I meant when it occurs in Chinese loanwords, and not equally before all those sounds (Those sounds are obviously all the sounds that cause the vowel to be devoiced, as it is in all your examples. But not every devoiced example is hard to pronounce, for reasons I don't really understand analytically.)