Is it known why a さかや normally has a か, rather than a け like in さけ?
Are there many other -や constructions for stores that change the spelling of the word added to?
(Especially in the ancient times,) there were/are bound morphemes (morphemes that cannot be used in isolation as a word) that end with the vowel a
. The a
at the end of these morphemes cannot appear at a word boundary. These forms are known as 被覆形.
saka- (as in 酒)
ama- (as in 雨)
puna- (as in 船)
ma- (as in 目)
When they are used as the first component of a compound noun, the a
-ending is rescued by being attached to the second component:
sakaya (酒屋), sakagura (酒蔵), sakadaru (酒樽), sakamori (酒盛り), sakazuki (盃)
amaoto (雨音), amagasa (雨傘), amagappa (雨合羽), amayadori (雨宿り)
funatsukiba (船着き場), funanori (船乗り), funazumi (船積み), funabashi (船橋), funayoi (船酔い)
mabuta (目蓋), manako (眼)
These morphemes cannot be used in isolation, but there were ways to modify them so that they can be used by themselves. One such way was to attach the vowel i
(上代特殊仮名遣い乙類イ) after it, which may be either an epenthetic vowel or a derivational morpheme that derives a noun (this part may be controversial). When such vowel attaches, the a+i
sequence became e
due to a phonological rule:
a-i → e
saka-i → sake (酒)
ama-i → ame (雨)
puna-i → fune (船)
ma-i → me (目)
These forms derived in this way are called 露出形.
e
sound should be fairly safe?
Oct 23, 2011 at 8:34
foo-屋
form but not the word foo
in isolation, and you want to figure out the correct form of foo
in isolation? Then, a-e alternation is not the only case of 露出形-被覆形 alternation as you can see in the linked question, but there are probably only a few such cases, and 酒
is one of them.
saka
as you see in saka-ya
is the original, sake
used in isolation is the modified. But as I said, there are only a few cases, so you don't have to worry about changing the form. If you do not change the form where you are supposed to, it will still be understandable, and you will likely notice shortly after.