Questions about Japanese

Discussion of natural languages, or language in general.
User avatar
Chagen
Avisaru
Avisaru
Posts: 707
Joined: Thu Sep 22, 2011 11:54 pm

Questions about Japanese

Post by Chagen »

This is mainly for Finlay and Clawgrip, since they're the two main Japanese speakers on this board (but if I missed someone please feel free to correct me!). I've got a few questions I've been wondering about. Sorry about so many questions, I just don't know who else to ask! Also I'm on my ADHD meds and therefore can focus a lot better.

Also, if there's a book or online article from a university you can point me to answer these, feel free to do that instead of wasting time answering me. I'd love to have a book on Japanese grammar, especially in how it has evolved from Old Japanese to the at-publication

1: According to dictionaries, 絶対 zettai is a na-adjective, even though it ends in -i (and I've heard zettai ni for "absolutely" in Japanese media). However, I swear I have heard in either anime or visual novels characters say zettaku for "absolutely" as if it were a regular i-adjective (as well as zettakunai, I think). Is this a thing that happens in regular Japanese or did I mis-hear?

2: In various Japanese media I've heard people suffix -ra to demonstratives and pronouns only to make plurals (that are always treated as plural, unlike -tachi. I've heard お前ら omera 僕ら bokura あいつら aitsura こいつら koitsura 君ら kimira 俺ら orera and I think 我ら warera (but never 私ら watashira or あたしら atashira...). Is this common in real-life Japanese or is it more an anime/manga/VN thing? Also I wonder where this suffix came from, as it's a true plural and Japanese has basically none of those.

3: Is 出来る dekiru really a historical combination of 出る deru "come out, leave" and 来る kuru "come"? The kanji seem to imply so, and Jisho doesn't say it's ateji. I can't see how "leave+come" could turn into a suppletive potential for "do". I thought it might have been a combination of de and kuru (I could see "come by/at" possibly becoming "can do", if I squint), but it's not written as で来る, it's dekiru not *dekuru (umlaut does not appear to be a thing in Japanese), and it's a ichidan as opposed to godan.

Does suru even have a normal potential? Would that be *sureru or shirareru?

4: Did some ichidan verbs once have stems ending in more than just -e or -i? Many of kuru's forms all look like an ichidan with a stem ending in -o (e.g taberu:kuru, tabenai:kona, taberareru:korareru?

4b: Where did the distinction between ichidan and godan verbs come from (sorry if this is way too complicated to answer)? Ichidan verbs end only in -eru and -iru: the former looks suspiciously like a godan-derived potential but that doesn't make any sense.

5: Japanese has closed-class verbs, yet has the pair 歌 uta "song" and 歌う "to sing". There's no way to view this other than noun > verb derivation (by suffixing -(w)u); a verb > noun derivation woul give us *utai. I swear I've also seen another godan derived from a foreign loanword, and in the realm of i-adjectives, we have two unusual situations:

-kirai "hateful" is plainly derived from kirau "to hate" but no other verb > i-adjective derivation like this exists AFAIK

-On the internet I've seen people take the loanword エロ ero and turn it into a i-adjective エロい eroi "lewd, erotic". Are there any other causes of derived i-adjectives like this?

6: There are no godan verbs ending in -hu, -pu, -d(z)u, -zu, or -yu (not counting -Cyu or -wu because the former morae are Chinese influence, and the latter verbs became -Vu AKAIK), as well as only one ending in -nu (死ぬ shinu "die"). There are, e.g, plenty of -su verbs, so why no -zu verbs? Likewise, what happened to all the other -nu verbs (unless shinu really was the only one ever)?

-pu verbs would have become -hu (> -fu), and -/h/ in japanese seems prone to dropping so maybe they just merged with -Vu verbs. Still, -zu verbs should just be identical to -su verbs except with -za/-ji/-zu/-ze/-zo. -du has merged with -zu in most dialects, but that would just bolster the amount of -zu verbs. /y/ in Japanese is only distinct before /a o u/, but that would still leave -ya/-i/-yu/-e/-yo...I can't this entire class being bowled over just from that.

-nu conjugation has merged with -mu and -bu in a few places (e.g Old -mite -nite -bite all become -nde, but even then shinu has remained distinct (that a verbal class consisting of a single verb has remained distinct is honestly impressive)

7: I recently had someone on /jp/ translate this rap song for me and some parts of it I found interesting (though I think he messed up in a few places):

-At 3:44 this line occurs:
都の鬼も連日騒ぐ 頭っから浴びる泉質は格別
miyako no oni mo renjitsu sawagu atamakkara abiru senshitsu wa kakubetsu
The capital’s oni live it up day after day, the feeling of quality water over your head can’t be beat...

For some reason the vocalist, Rainyblueytr, says 頭っから atamakkara for "over your head" instead of 頭から atama kara. Is this just a thing to make the rap go more smoothly, because I've never seen this spontaneous gemination of post-positions in any media.
Nūdhrēmnāva naraśva, dṛk śraṣrāsit nūdhrēmanīṣṣ iźdatīyyīm woḥīm madhēyyaṣṣi.
satisfaction-DEF.SG-LOC live.PERFECTIVE-1P.INCL but work-DEF.SG-PRIV satisfaction-DEF.PL.NOM weakeness-DEF.PL-DAT only lead-FUT-3P

Sumelic
Avisaru
Avisaru
Posts: 385
Joined: Sat Mar 28, 2015 7:05 pm

Re: Questions about Japanese

Post by Sumelic »

I don't know much about Japanese at all, but I think I can comment on one of these points.
Chagen wrote:=
6: There are no godan verbs ending in -hu, -pu, -d(z)u, -zu, or -yu (not counting -Cyu or -wu because the former morae are Chinese influence, and the latter verbs became -Vu AKAIK), as well as only one ending in -nu (死ぬ shinu "die"). There are, e.g, plenty of -su verbs, so why no -zu verbs? Likewise, what happened to all the other -nu verbs (unless shinu really was the only one ever)?

-pu verbs would have become -hu (> -fu), and -/h/ in japanese seems prone to dropping so maybe they just merged with -Vu verbs. Still, -zu verbs should just be identical to -su verbs except with -za/-ji/-zu/-ze/-zo. -du has merged with -zu in most dialects, but that would just bolster the amount of -zu verbs. /y/ in Japanese is only distinct before /a o u/, but that would still leave -ya/-i/-yu/-e/-yo...I can't this entire class being bowled over just from that.
If I remember correctly, intervocalic /p/ (later /h/) was actually regularly lost in most contexts, not just prone to being lost. I think the course of development was /p/ > /ɸ/ (p-lenition), /ɸ/ > /β~w/ (intervocalic vocing), and then /w/ ended up being lost before any vowel that wasn't /a/. A relevant passage on Google Books: https://books.google.com/books?id=ix9r6 ... 0h&f=false

User avatar
linguoboy
Sanno
Sanno
Posts: 3681
Joined: Tue Sep 17, 2002 9:00 am
Location: Rogers Park/Evanston

Re: Questions about Japanese

Post by linguoboy »

Chagen wrote:3: Is 出来る dekiru really a historical combination of 出る deru "come out, leave" and 来る kuru "come"? The kanji seem to imply so, and Jisho doesn't say it's ateji. I can't see how "leave+come" could turn into a suppletive potential for "do". I thought it might have been a combination of de and kuru (I could see "come by/at" possibly becoming "can do", if I squint), but it's not written as で来る, it's dekiru not *dekuru (umlaut does not appear to be a thing in Japanese), and it's a ichidan as opposed to godan.
I should have a better answer for this since I just read the relevant passage in Martin last night (trying to answer a question about potential constructions in Japanese). The etymological explanation in Wiktionary sounds so similar I wonder if it isn't lifted from him wholesale (I'll double-check when I'm home) so I'll quote it in full:
Maybe Martin wrote:Over this same period, the meaning gradually shifted from “to come out” to include “to appear, to become manifest”, and from there “to come into existence anew”. This then extended further to “to occur, to happen; to be born; to be created, to be produced”, and thence “to come to fruition; to be completed, to be finished (with positive overtones)”. The underlying sense of latent existence gave rise to the meaning of “to be capable of making or producing something”, and then simply “to be able to do”.
It sounded familiar to me because it recalls parallel developments involving Korean 되다 "become", which, however, is used to express deontic modality rather than potential.
Chagen wrote:5: Japanese has closed-class verbs
Sez who? May favourite recent coinage is ググる "to google", since the final syllable ends up being reanalysed as the verbal termination る on the pattern of earlier トラブる and ダブる.

Sumelic
Avisaru
Avisaru
Posts: 385
Joined: Sat Mar 28, 2015 7:05 pm

Re: Questions about Japanese

Post by Sumelic »

linguoboy wrote:
Chagen wrote:5: Japanese has closed-class verbs
Sez who? May favourite recent coinage is ググる "to google", since the final syllable ends up being reanalysed as the verbal termination る on the pattern of earlier トラブる and ダブる.
Chagen and some other people have already discussed this topic earlier in the thread Closed and open classes in Natlangs (Especially Japanese)

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Questions about Japanese

Post by clawgrip »

Chagen wrote:1: According to dictionaries, 絶対 zettai is a na-adjective, even though it ends in -i (and I've heard zettai ni for "absolutely" in Japanese media). However, I swear I have heard in either anime or visual novels characters say zettaku for "absolutely" as if it were a regular i-adjective (as well as zettakunai, I think). Is this a thing that happens in regular Japanese or did I mis-hear?
Never heard this. If it were some sort of anime-speak, I would not be surprised, but I don't know a whole lot about anime in general, so all I can tell you is that this is not normal Japanese.
2: In various Japanese media I've heard people suffix -ra to demonstratives and pronouns only to make plurals (that are always treated as plural, unlike -tachi. I've heard お前ら omera 僕ら bokura あいつら aitsura こいつら koitsura 君ら kimira 俺ら orera and I think 我ら warera (but never 私ら watashira or あたしら atashira...). Is this common in real-life Japanese or is it more an anime/manga/VN thing? Also I wonder where this suffix came from, as it's a true plural and Japanese has basically none of those.
Japanese has three plurals, but they are not true plurals, they are associative plurals, which is why they are primarily only used for humans (in the adult world, at least; children (and possibly women more often than men maybe, in informal situations) tend to anthropomorphize). -tachi is a neutral associative plural, while -ra is a slightly informal or derogatory form of it (though it is regularized in karera et al). The polite form is ~方 -gata.

Also I want to point out that お前 is omae; omē is a more slangy and/or confrontational and/or derogatory form of it.
3: Is 出来る dekiru really a historical combination of 出る deru "come out, leave" and 来る kuru "come"? The kanji seem to imply so, and Jisho doesn't say it's ateji. I can't see how "leave+come" could turn into a suppletive potential for "do". I thought it might have been a combination of de and kuru (I could see "come by/at" possibly becoming "can do", if I squint), but it's not written as で来る, it's dekiru not *dekuru (umlaut does not appear to be a thing in Japanese), and it's a ichidan as opposed to godan.
This seems to have been discussed sufficiently above.
Does suru even have a normal potential? Would that be *sureru or shirareru?
No, it does not.
4: Did some ichidan verbs once have stems ending in more than just -e or -i? Many of kuru's forms all look like an ichidan with a stem ending in -o (e.g taberu:kuru, tabenai:kona, taberareru:korareru?
I want to discuss this below, but: I think Classical Japanese ku (the predecessor to modern kuru) is better analyzed as having a stem k-, with a bunch of weird endings. There are some other verbs with only a single consonant ending, such as nu. Weirdest of all is the verb 得る uru, which, in classical Japanese, is just the shimo-nidan verb endings alone, with no verb stem. But this is off topic.
4b: Where did the distinction between ichidan and godan verbs come from (sorry if this is way too complicated to answer)? Ichidan verbs end only in -eru and -iru: the former looks suspiciously like a godan-derived potential but that doesn't make any sense.
Godan derive from Classical yodan, while ichidan derive from all other forms (except for the irregulars, ku, su, ari, shinu, inu). Yodan employed only four vowels for all endings (hence the name), but the combination of the -a ending plus the verb suffix -u, produced a shift from -au to -ō, giving us the fifth vowel.
Ichidan verbs derive from two different sources:
1. original ichidan verbs, which were vowel stems rather than consonant stems. Even in Classical Japanese, they ended only in -i and -e.
2. original nidan verbs. Nidan verbs looked like yodan verbs in the basic dictionary form, but, only used either -i or -e in addition to -u for any endings. For whatever reason (I would like to know why!), at some point they abandoned their conclusive -u form and adopted either the mizenkei or ren'yōkei as their base stem, becoming modern ichidan.

You can tell a historical ichidan from a historical nidan in the modern language by looking at the okurigana: if you see nothing but る, it is very likely a historical ichidan, e.g. 見る, 寝る, but if it has two (and it doesn't include a transitivity marker), it's probably a historical nidan, e.g. 食べる, 落ちる.

The question as to why these only use -i and -e is still unknown to me.
5: Japanese has closed-class verbs, yet has the pair 歌 uta "song" and 歌う "to sing". There's no way to view this other than noun > verb derivation (by suffixing -(w)u); a verb > noun derivation woul give us *utai. I swear I've also seen another godan derived from a foreign loanword, and in the realm of i-adjectives, we have two unusual situations:
It could be a noun derived from a verb though, with the ending dropped for whatever reason. This would be important distinction from your explanation, because it would remove precedent for allowing verbalization.
-kirai "hateful" is plainly derived from kirau "to hate" but no other verb > i-adjective derivation like this exists AFAIK
kirai is not an -i adjective. If you've heard people use it as one, they're not using it in a way that matches with regular usage.
-On the internet I've seen people take the loanword エロ ero and turn it into a i-adjective エロい eroi "lewd, erotic". Are there any other causes of derived i-adjectives like this?
Deriving -i adjectives is not so strange, informally. In addition to modern slang ones like that (I'm sure there are others but I'm tired and want to finish this post), there is an i-adjective-deriving ending, -tai, which you may know in words like 冷たい, 眠たい, 平べったい, etc. This can easily be abbreviated into just -i in lots of slang and dialect words.
6: There are no godan verbs ending in -hu, -pu, -d(z)u, -zu, or -yu (not counting -Cyu or -wu because the former morae are Chinese influence, and the latter verbs became -Vu AKAIK), as well as only one ending in -nu (死ぬ shinu "die"). There are, e.g, plenty of -su verbs, so why no -zu verbs? Likewise, what happened to all the other -nu verbs (unless shinu really was the only one ever)?
I think this was covered, but:
-hu / -pu: /h/ does not occur word-medially within a morpheme except in a handful of cases: haha, ahiru, afureru. p is just a geminated h, and no verbs in Japanese have a geminate as the final consonant of the stem. However, all Vu verbs derive ultimately from h, from earlier p.
-dzu: They existed, but they call got turned into ichidan verbs. One of interest is idzu, which is the old form of 出る deru. For some reason, the initial i- was dropped, so idzu → *ideru → deru. Interestingly, the imperative retains the i in the weird but common form oide, which I'm sure appears frequently enough in anime, and means "come here".
-pu verbs would have become -hu (> -fu), and -/h/ in japanese seems prone to dropping so maybe they just merged with -Vu verbs.
exactly.
Still, -zu verbs should just be identical to -su verbs except with -za/-ji/-zu/-ze/-zo. -du has merged with -zu in most dialects, but that would just bolster the amount of -zu verbs. /y/ in Japanese is only distinct before /a o u/, but that would still leave -ya/-i/-yu/-e/-yo...I can't this entire class being bowled over just from that.
There were -zu verbs, but they were all forms of su (modern suru), and in modern Japanese they are all now -zuru or -jiru. As for -yu verbs, those existed as well, but they were all nidan, so they are all ichidan now, which conceals their origin since the /j/ is dropped, e.g. kikoyu → kikoeru. Still, there weren't so many.
-nu conjugation has merged with -mu and -bu in a few places (e.g Old -mite -nite -bite all become -nde, but even then shinu has remained distinct (that a verbal class consisting of a single verb has remained distinct is honestly impressive)
In classical Japanese there were only two -nu verbs: 死ぬ shinu and 往ぬ/去ぬ inu, which also basically meant die. It's kind of weird, I don't know what's going on.
7: I recently had someone on /jp/ translate this rap song for me and some parts of it I found interesting (though I think he messed up in a few places):

-At 3:44 this line occurs:
都の鬼も連日騒ぐ 頭っから浴びる泉質は格別
miyako no oni mo renjitsu sawagu atamakkara abiru senshitsu wa kakubetsu
The capital’s oni live it up day after day, the feeling of quality water over your head can’t be beat...

For some reason the vocalist, Rainyblueytr, says 頭っから atamakkara for "over your head" instead of 頭から atama kara. Is this just a thing to make the rap go more smoothly, because I've never seen this spontaneous gemination of post-positions in any media.
It's artistic license to make the rap flow.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Questions about Japanese

Post by clawgrip »

I looked up the etymology of uta/utau, and there seem to be two theories: the first is that it derives from 打ち合う uchiau (Classical 打ち合ふ uchiafu) "exchange blows", the idea being that the song is done in time to physical activity. The second is that it derives from the verb 訴える uttaeru (Classical 訴ふ uttafu) "complain; plead; appeal; accuse; sue", the idea being that a song is an expression of the emotion of appealing for the understanding of one's feelings.

In either case, it is clear that the noun is a derivative of the verb. I can't think of any other noun derived from a verb that actually has elements removed rather than added or replaced, but given the importance of song in culture, I can imagine some sort of weird sound change or abbreviation happening, and/or it managing to be the lone survivor of an extinct derivational process.

User avatar
Qwynegold
Smeric
Smeric
Posts: 1606
Joined: Thu May 24, 2007 11:34 pm
Location: Stockholm

Re: Questions about Japanese

Post by Qwynegold »

Chagen wrote:-On the internet I've seen people take the loanword エロ ero and turn it into a i-adjective エロい eroi "lewd, erotic". Are there any other causes of derived i-adjectives like this?
Two that I can think of are ナウい (which I think means something like "trendy", from Eng. now) and きもい (slang for 気持ち悪い "gross").
clawgrip wrote:
2: In various Japanese media I've heard people suffix -ra to demonstratives and pronouns only to make plurals (that are always treated as plural, unlike -tachi. I've heard お前ら omera 僕ら bokura あいつら aitsura こいつら koitsura 君ら kimira 俺ら orera and I think 我ら warera (but never 私ら watashira or あたしら atashira...). Is this common in real-life Japanese or is it more an anime/manga/VN thing? Also I wonder where this suffix came from, as it's a true plural and Japanese has basically none of those.
Japanese has three plurals, but they are not true plurals, they are associative plurals, which is why they are primarily only used for humans (in the adult world, at least; children (and possibly women more often than men maybe, in informal situations) tend to anthropomorphize). -tachi is a neutral associative plural, while -ra is a slightly informal or derogatory form of it (though it is regularized in karera et al).
But isn't it also like written language? It appears occasionally in e.g. newspaper articles. I remember reading about a Santa contest with "サンタら".
Image
My most recent quiz:
Eurovision Song Contest 2018

User avatar
finlay
Sumerul
Sumerul
Posts: 3600
Joined: Mon Dec 22, 2003 12:35 pm
Location: Tokyo

Re: Questions about Japanese

Post by finlay »

i will repeat a bit of what's said above
Chagen wrote: 1: According to dictionaries, 絶対 zettai is a na-adjective, even though it ends in -i (and I've heard zettai ni for "absolutely" in Japanese media). However, I swear I have heard in either anime or visual novels characters say zettaku for "absolutely" as if it were a regular i-adjective (as well as zettakunai, I think). Is this a thing that happens in regular Japanese or did I mis-hear?
you can tell because it is a two-kanji compound, and doesn't end in い. never mind that you can write it that way in hiragana. never heard that but it sounds playful, like saying だいじょうない instead of 大丈夫じゃない. obviously wrong of course
2: In various Japanese media I've heard people suffix -ra to demonstratives and pronouns only to make plurals (that are always treated as plural, unlike -tachi. I've heard お前ら omera 僕ら bokura あいつら aitsura こいつら koitsura 君ら kimira 俺ら orera and I think 我ら warera (but never 私ら watashira or あたしら atashira...). Is this common in real-life Japanese or is it more an anime/manga/VN thing? Also I wonder where this suffix came from, as it's a true plural and Japanese has basically none of those.
it's not a true plural, it's a collective plural. the only one i've heard irl is 彼ら
3: Is 出来る dekiru really a historical combination of 出る deru "come out, leave" and 来る kuru "come"? The kanji seem to imply so, and Jisho doesn't say it's ateji. I can't see how "leave+come" could turn into a suppletive potential for "do". I thought it might have been a combination of de and kuru (I could see "come by/at" possibly becoming "can do", if I squint), but it's not written as で来る, it's dekiru not *dekuru (umlaut does not appear to be a thing in Japanese), and it's a ichidan as opposed to godan.

Does suru even have a normal potential? Would that be *sureru or shirareru?
no, it's ateji; no, there's no すれる. but the kanji for 出来る kind of suggest that meaning tangentially
5: Japanese has closed-class verbs, yet has the pair 歌 uta "song" and 歌う "to sing". There's no way to view this other than noun > verb derivation (by suffixing -(w)u); a verb > noun derivation woul give us *utai. I swear I've also seen another godan derived from a foreign loanword, and in the realm of i-adjectives, we have two unusual situations:

-kirai "hateful" is plainly derived from kirau "to hate" but no other verb > i-adjective derivation like this exists AFAIK

-On the internet I've seen people take the loanword エロ ero and turn it into a i-adjective エロい eroi "lewd, erotic". Are there any other causes of derived i-adjectives like this?
they're only sort-of closed class - to join the class they have to fit a certain phonological pattern (hence ググった and so on, even though Google is usually グーグル - a japanese verb doesn't normally have a long vowel there)

i've heard むずい for むずかしい and a few other things like that
7: I recently had someone on /jp/ translate this rap song for me and some parts of it I found interesting (though I think he messed up in a few places):

-At 3:44 this line occurs:
都の鬼も連日騒ぐ 頭っから浴びる泉質は格別
miyako no oni mo renjitsu sawagu atamakkara abiru senshitsu wa kakubetsu
The capital’s oni live it up day after day, the feeling of quality water over your head can’t be beat...

For some reason the vocalist, Rainyblueytr, says 頭っから atamakkara for "over your head" instead of 頭から atama kara. Is this just a thing to make the rap go more smoothly, because I've never seen this spontaneous gemination of post-positions in any media.
i think this is to do with the rap, but i'm fairly sure there's places where gemination can happen spontaneously in japanese, like やはり>やっぱり so i don't see why から would be any different

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Questions about Japanese

Post by clawgrip »

finlay wrote:
2: In various Japanese media I've heard people suffix -ra to demonstratives and pronouns only to make plurals (that are always treated as plural, unlike -tachi. I've heard お前ら omera 僕ら bokura あいつら aitsura こいつら koitsura 君ら kimira 俺ら orera and I think 我ら warera (but never 私ら watashira or あたしら atashira...). Is this common in real-life Japanese or is it more an anime/manga/VN thing? Also I wonder where this suffix came from, as it's a true plural and Japanese has basically none of those.
it's not a true plural, it's a collective plural. the only one i've heard irl is 彼ら
I've heard several of these, such as aitsura and so forth. But you are right.

re. サンタら, it is possible for the suffixes mentioned to be used as standard plurals, but this is a derivation from the associative plural, not the other way around.
3: Is 出来る dekiru really a historical combination of 出る deru "come out, leave" and 来る kuru "come"? The kanji seem to imply so, and Jisho doesn't say it's ateji. I can't see how "leave+come" could turn into a suppletive potential for "do". I thought it might have been a combination of de and kuru (I could see "come by/at" possibly becoming "can do", if I squint), but it's not written as で来る, it's dekiru not *dekuru (umlaut does not appear to be a thing in Japanese), and it's a ichidan as opposed to godan.

Does suru even have a normal potential? Would that be *sureru or shirareru?
no, it's ateji; no, there's no すれる. but the kanji for 出来る kind of suggest that meaning tangentially
Actually it is not ateji, it is exactly as was described above. Chagen makes the mistake of translating 出る as "leave", when its basic meaning is "go/come out; emerge; exit", so 出来る literally means "come out; emerge" (cf. 出てくる). It is admittedly odd-looking, though. The original Classical Japanese verb was 出で来 ideku (出づ idzu + 来 ku), and this of course has an associated noun form, 出で来 ideki, modern 出来 deki (because as I mentioned, the i was dropped from this root for some reason). The meaning of this word is along the lines of workmanship, the way something was made, the result of something, etc. e.g. 出来が悪い deki ga warui "it was poorly made (lit. 'the production is bad')"

This noun was then reverbalized as a kami-ichidan verb (showing that things like ググる, メモる, 事故る have been happening for longer than one might think), resulting in the modern ichidan verb 出来る dekiru. The verb 出来る is frequently used to mean "be completed" and the etymology rather clearly follows from this in the way that the quote in linguoboy's post describes it.
-On the internet I've seen people take the loanword エロ ero and turn it into a i-adjective エロい eroi "lewd, erotic". Are there any other causes of derived i-adjectives like this?
they're only sort-of closed class - to join the class they have to fit a certain phonological pattern (hence ググった and so on, even though Google is usually グーグル - a japanese verb doesn't normally have a long vowel there)

i've heard むずい for むずかしい and a few other things like that
Things like むずい, きもい, うざい, やばい all originate as abbreviations of previously existing い adjectives. エロい, ナウい, グロい are a bit different, since it is a loanword turned into an い adjective, which breaks the pattern since loanwords generally become な adjectives. This is a relatively new development in Japanese, so there are no established, non-slangy words of this type.
i think this is to do with the rap, but i'm fairly sure there's places where gemination can happen spontaneously in japanese, like やはり>やっぱり so i don't see why から would be any different
I feel the need to point out that やはり>やっぱり is not really spontaneous; rather, the word yahari is said to come from the same root as 柔らかい yawarakai and it existed as a mimetic やはやは yahayaha. Yahari is very similar in form to the mimetic CVC:Vri pattern seen in words such as yukkuri, tappuri, kotteri, gisshiri, kossori, batchiri, etc., and there is of course precedent for reduplicated mimetic words to appear also as this pattern, cf. kosokoso and kossori. In truth, the fact that yahari is not geminated is weird to me, as is the fact that it never became yawari.

Otherwise, gemination arises either as a compensation for the loss of something else (e.g. hiotokohyottoko: preserves mora count), ease of pronunciation (e.g. ma + akamakka: length shifts from vowel to consonant, ten + ōtennō: nnō easier to pronounce than n'ō) or as an indicator of emotion/intensity, (e.g. sugoisuggoi). The quote from the rap would fall into the third of these categories.

User avatar
Chagen
Avisaru
Avisaru
Posts: 707
Joined: Thu Sep 22, 2011 11:54 pm

Re: Questions about Japanese

Post by Chagen »

Thank you everyone! Also, Clawgrip, as you could probably tell, I am not very good at Japanese ^^; Stuff like assuming that kirai is an i-adjective is pretty embarrassing! Also, while watching anime yesterday, I heard zettai nai for "absolutely not" so it is a na-adjective even in anime. Though I swear I've heard zettaku. Could have been something else, but I swear I've heard it somewhere.

I have another question about that song, by the way, mainly about the translation I received:
(I've made the URL's point to the times in the song I mention, but I've also mentioned the times in case that doesn't work)

1: At 0:49 this line occurs:
鬼が出るか蛇が出るか 頭上掲げるは唯一無二の魂 地獄の話
oni ga deru ka ja ga deru ka zujou kakageru wa yuiitsumuni no tamashi jigoku no hanashi

The translator listed this line as "What’s next? Oni? Snakes? Hanging high in the air, a one-of-a-kind soul, the story of hell!". Jisho/Rikaichan list 鬼が出るか蛇が出るか as an idiom meaning "You never know what might happen". Is this true? I've never heard this idiom, though I am well aware that I have a very skewed perception of Japanese, given that 99% of my exposure to it is anime and video games.

(You guys are probably rolling your eyes at my weeaboo self lol XP)
Sumelic wrote: If I remember correctly, intervocalic /p/ (later /h/) was actually regularly lost in most contexts, not just prone to being lost. I think the course of development was /p/ > /ɸ/ (p-lenition), /ɸ/ > /β~w/ (intervocalic vocing), and then /w/ ended up being lost before any vowel that wasn't /a/. A relevant passage on Google Books: https://books.google.com/books?id=ix9r6 ... 0h&f=false
Yeah I've noticed this. I have seen /h/ (and [ɸɯ]) almost exclusively word-initially, outside of a few rare words such as ご飯 gohan, 溢れる afureru as Clawgrip mentioned (interestingly, a light novel I'm reading is called Afureta...same verb?), and in the song I keep mentioning, 五線譜 gosenfu (which even struck me as a very unusual word when I first heard it).

2: The ending of the first bridge into the chorus (1:20) is this:
さぁ踊れや騒げ 飲めや歌えやお手を拝借
saa odore ya sawage (nomeya utaeya o-te wo haishaku)
Dance and live it up~ (Drink! Sing! Take my hand!)

The translator viewed nomeya utaeya as simple imperatives, but Jisho says that that's another fixed phrase meaning "revelry". What I'm wondering is if the ya in both words is the one that means "and", or if it's an old form of the imperative and the modern "e-grade + ∅" is just a truncation of that.

I personally find it a little odd that the same grade is used for both the potential/conditional and the imperative (i.e nomu "to drink" > nomeru "can drink", nomeba "if (he) drinks", nome! "drink!"). E-grade is obviously an irrealis and imperatives don't strike me as particularly irrealis.


(Also, one cool thing I want to show is a remix of Yukemuri Tamashii Onsen the same group did later, called Yukemuri Raga Onsen "Steamy Ragga Hot Springs", where they remix it into a Reggae song and member Hazu puts on his best Jamaican accent. I find it really cool how /hu/ is actually pronounced [hɯ] and not [ɸɯ] to make the accent even more authentic; /ti/ and /si/ are still pronounced [tɕi ɕi] though)

3: AFAIK, Japanese originally did not allow any hiatus and had a very strict CV syllable structure (I don't know if initially open onsets were allowed). It wasn't until Chinese loanwords came en masse that CyV and the moraic /n/ appeared (speaking of that, why the hell is it a Uvular nasal /ɴ/? Nothing else in Japanese or Chinese even approaches an Uvular POA). However there are some words which have these yet MUST be native Japanese words (usually because they're verbs), such as the well-known 頑張る ganbaru "do well, persevere" (if I had a nickel for everytime I heard ganbare or ganbatte in anime...honestly, I've rarely heard 頑張る in any inflection but those two) or しゃべる shaberu "chat". I wonder what these words originally were. At first I guessed that ganbaru might have been ga(n/m)ibaru or summat, and underwent the same kind of simplifcation as, say, yomite > yonde, but I don't think that NibV > NbV is a thing like mite > nde was.

Gemination also was apparently not a thing before Chinese came in, which I find odd as Chinese doesn't even have phonemic gemination.

4: Are there any other natlangs which have analogues to onbin (i.e certain phonological processes that only occur during inflection, and are morphologically conditioned). You could say it's just morphophonology, but onbin only occur during verbal inflection and nowhere else. It really feels like an unnaturalistic feature in a noob conlang (Japanese feels like a weird conlang to me most of the time tbh).

For instance, /k g/ both drop before /i/ in verbal inflection, with /g/ also voicing a following voiceless consonant (e.g *yakita *oyogita" > yaita oyoida "grilled, swam". However in verbs this only occurs in the past tense. -ku and -gu verbs still have normal i-grades such as the -masu form (yakimasu oyogimasu). My guess for this is that the polite form is clearly i-grade in the other verbs (noborimasu, machimasu, tsukaimasu, shinimasu(?), whereas the past tense is just a complete mess (nobotta, matta, tsukatta, shinda...) not connected to any particular grade, so maybe analogy has run through the forms and fixed stuff.

Then there's the i-adjectives which have *ki > i regularly, explaining why their other forums suddenly gain a -k- (atsui > atsukunai)

Sorry for all these really long questions, I took my ADHD meds today so I can focus really well and I'm really interested in this language due to being such a big weeaboo XP. Also sorry if any of these questions are dumb as hell.
Nūdhrēmnāva naraśva, dṛk śraṣrāsit nūdhrēmanīṣṣ iźdatīyyīm woḥīm madhēyyaṣṣi.
satisfaction-DEF.SG-LOC live.PERFECTIVE-1P.INCL but work-DEF.SG-PRIV satisfaction-DEF.PL.NOM weakeness-DEF.PL-DAT only lead-FUT-3P

Sumelic
Avisaru
Avisaru
Posts: 385
Joined: Sat Mar 28, 2015 7:05 pm

Re: Questions about Japanese

Post by Sumelic »

Chagen wrote:
3: AFAIK, Japanese originally did not allow any hiatus and had a very strict CV syllable structure (I don't know if initially open onsets were allowed). It wasn't until Chinese loanwords came en masse that CyV and the moraic /n/ appeared (speaking of that, why the hell is it a Uvular nasal /ɴ/? Nothing else in Japanese or Chinese even approaches an Uvular POA). However there are some words which have these yet MUST be native Japanese words (usually because they're verbs), such as the well-known 頑張る ganbaru "do well, persevere" (if I had a nickel for everytime I heard ganbare or ganbatte in anime...honestly, I've rarely heard 頑張る in any inflection but those two) or しゃべる shaberu "chat". I wonder what these words originally were. At first I guessed that ganbaru might have been ga(n/m)ibaru or summat, and underwent the same kind of simplifcation as, say, yomite > yonde, but I don't think that NibV > NbV is a thing like mite > nde was.

Gemination also was apparently not a thing before Chinese came in, which I find odd as Chinese doesn't even have phonemic gemination.
Regarding [ɴ]: Again, I'm only interested in Japanese, not knowledgeable about it, but I haven't seen much evidence in what I've read for classifying this as phonemically uvualar, i.e. /ɴ/. It seems like most phonemic analyses treat it as placeless (often using the symbol N, like Q for geminates and R for long vowels, although the analysis of the latter two seems like it might be more controversial and might just be shorthand). How frequent is the [ɴ] allophone compared to the non-uvular allophones? Doesn't it mainly occur utterance-finally, or phrase-finally? (I don't know the scope of Japanese allophony rules.) I know before plosives, affricates or nasals it assimilates in place, and before fricatives it generally is converted to a nasalized semivowel, and I think it can also become a semivowel between vowels. The uvular position seems like it would just be a phonetic realization of placelessness, like how some Spanish accents realize word-final nasals as /ŋ/.

I think your guess of syncope leading to /NC/ clusters in native-strata words sounds plausible; Clawgrip or someone else with knowledge will have to explain if it is right. Some of them might just be the results of one-off irregular sound changes, none of which was really a "thing" in general.

I've read that in some contexts in Japanese where voiceless stops are subject to gemination, voiced stops are converted to clusters of /N/ + voiced stop. Even if gemination was not a thing before Chinese came in, it seems various gemination processes have acted on native Japanese words due to the factors Clawgrip mentioned.

M Mira
Lebom
Lebom
Posts: 92
Joined: Thu Mar 05, 2015 8:21 pm
Location: Taipeium, Respublica Sinarum

Re: Questions about Japanese

Post by M Mira »

Chagen wrote: 1: At 0:49 this line occurs:
鬼が出るか蛇が出るか 頭上掲げるは唯一無二の魂 地獄の話
oni ga deru ka ja ga deru ka zujou kakageru wa yuiitsumuni no tamashi jigoku no hanashi

The translator listed this line as "What’s next? Oni? Snakes? Hanging high in the air, a one-of-a-kind soul, the story of hell!". Jisho/Rikaichan list 鬼が出るか蛇が出るか as an idiom meaning "You never know what might happen". Is this true? I've never heard this idiom, though I am well aware that I have a very skewed perception of Japanese, given that 99% of my exposure to it is anime and video games.

(You guys are probably rolling your eyes at my weeaboo self lol XP)
It is an idiom:
http://ejje.weblio.jp/content/鬼が出るか蛇が出るか
Sumelic wrote: If I remember correctly, intervocalic /p/ (later /h/) was actually regularly lost in most contexts, not just prone to being lost. I think the course of development was /p/ > /ɸ/ (p-lenition), /ɸ/ > /β~w/ (intervocalic vocing), and then /w/ ended up being lost before any vowel that wasn't /a/. A relevant passage on Google Books: https://books.google.com/books?id=ix9r6 ... 0h&f=false
Yeah I've noticed this. I have seen /h/ (and [ɸɯ]) almost exclusively word-initially, outside of a few rare words such as ご飯 gohan, 溢れる afureru as Clawgrip mentioned (interestingly, a light novel I'm reading is called Afureta...same verb?), and in the song I keep mentioning, 五線譜 gosenfu (which even struck me as a very unusual word when I first heard it).
Word-medial /h/ is permitted/preserved in Sinitic or Sino-xenic words (on'yomi kanji words), probably because kanji are treated as discreet objects, and thus the onset of every kanji is technically word-initial.

BTW, intra-kanji /h/ (from Middle Chinese entering tone coda -/p/) were indeed subject to the said sound change.

五線譜 is indeed unusual, as /nhV/ should normally become /npV/. Or is /ɸ/ exempt from the change?
3: AFAIK, Japanese originally did not allow any hiatus and had a very strict CV syllable structure (I don't know if initially open onsets were allowed). It wasn't until Chinese loanwords came en masse that CyV and the moraic /n/ appeared (speaking of that, why the hell is it a Uvular nasal /ɴ/? Nothing else in Japanese or Chinese even approaches an Uvular POA). However there are some words which have these yet MUST be native Japanese words (usually because they're verbs), such as the well-known 頑張る ganbaru "do well, persevere" (if I had a nickel for everytime I heard ganbare or ganbatte in anime...honestly, I've rarely heard 頑張る in any inflection but those two) or しゃべる shaberu "chat". I wonder what these words originally were. At first I guessed that ganbaru might have been ga(n/m)ibaru or summat, and underwent the same kind of simplifcation as, say, yomite > yonde, but I don't think that NibV > NbV is a thing like mite > nde was.

Gemination also was apparently not a thing before Chinese came in, which I find odd as Chinese doesn't even have phonemic gemination.
頑張る either came from 眼張る (also ganbaru, the "gan" here means "eye" and is in on'yomi) or 我に張る (ga ni haru)

Dunno about しゃべる, but perhaps it came from the Edo dialect/accent? It has ヒ->シ, and by extension, ヒヤ->シャ

Gemination probably predates Chinese influence, as the Chinese entering tone (actually closed syllables ending in a stop) became geminated word-medially but underwent vowel epithesis word-finally. It looks more like shoehorning into preexisting patterns than creating something anew.
4: Are there any other natlangs which have analogues to onbin (i.e certain phonological processes that only occur during inflection, and are morphologically conditioned). You could say it's just morphophonology, but onbin only occur during verbal inflection and nowhere else. It really feels like an unnaturalistic feature in a noob conlang (Japanese feels like a weird conlang to me most of the time tbh).

For instance, /k g/ both drop before /i/ in verbal inflection, with /g/ also voicing a following voiceless consonant (e.g *yakita *oyogita" > yaita oyoida "grilled, swam". However in verbs this only occurs in the past tense. -ku and -gu verbs still have normal i-grades such as the -masu form (yakimasu oyogimasu). My guess for this is that the polite form is clearly i-grade in the other verbs (noborimasu, machimasu, tsukaimasu, shinimasu(?), whereas the past tense is just a complete mess (nobotta, matta, tsukatta, shinda...) not connected to any particular grade, so maybe analogy has run through the forms and fixed stuff.

Then there's the i-adjectives which have *ki > i regularly, explaining why their other forums suddenly gain a -k- (atsui > atsukunai)
https://ja.wikipedia.org/wiki/音便
(Japanese)
It seems that the process(es) took place elsewhere as well, but not obvious as the original forms disappeared entirely, unlike in verbs where other inflections preserves the original forms.

User avatar
finlay
Sumerul
Sumerul
Posts: 3600
Joined: Mon Dec 22, 2003 12:35 pm
Location: Tokyo

Re: Questions about Japanese

Post by finlay »

Chagen wrote:頑張る ganbaru "do well, persevere" (if I had a nickel for everytime I heard ganbare or ganbatte in anime...honestly, I've rarely heard 頑張る in any inflection but those two)
haha this is definitely where anime is skewed i think - you definitely hear those two a lot but you'd hear it more in other inflections if you were living here, like よく頑張りました means "well done"

and man don't worry that we think you're a weeaboo. we don't really care. embrace your weeabooness ;)

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Questions about Japanese

Post by clawgrip »

Chagen wrote:assuming that kirai is an i-adjective is pretty embarrassing!
It's an easy mistake to make, since it not only sounds like an i-adjective, it also looks like one (嫌い). It actually derives from the verb 嫌う kirau, "to dislike", which as far as I know only ever appears in the passive (嫌われる kirawareru "to be disliked"). As an aside, 好き suki is the same, deriving from 好く suku (→ 好かれる sukareru).
I swear I've heard zettaku.
Most likely you heard 贅沢 zeitaku, which literally means luxury or luxurious, but is much more conversational and is used much more frequently than the English counterpart, such as when people indulge or splurge, etc.
Sumelic wrote: If I remember correctly, intervocalic /p/ (later /h/) was actually regularly lost in most contexts, not just prone to being lost. I think the course of development was /p/ > /ɸ/ (p-lenition), /ɸ/ > /β~w/ (intervocalic vocing), and then /w/ ended up being lost before any vowel that wasn't /a/. A relevant passage on Google Books: https://books.google.com/books?id=ix9r6 ... 0h&f=false
Yeah I've noticed this. I have seen /h/ (and [ɸɯ]) almost exclusively word-initially, outside of a few rare words such as ご飯 gohan, 溢れる afureru as Clawgrip mentioned (interestingly, a light novel I'm reading is called Afureta...same verb?), and in the song I keep mentioning, 五線譜 gosenfu (which even struck me as a very unusual word when I first heard it).
I'm going to give more information than is necessary:
Old Japanese had the consonants /p w j/ (among others). /p/ could appear before all vowels, while /w/ appeared before all but /u/ and /j/ appeared before all but /i/. By the time of Early Modern Japanese, /e/ merged into /je/, while /we/ remained distinct. Meanwhile, /p/ became /ɸ/ everywhere. Then /ɸ/ became /w/ word-medially*, and /h/ before /a i e o/, remaining as /ɸ/ only before /u/. Following this, /we/ merged into /je/ and /wi/ merged into /i/, leaving /w/ only before /a/. Eventually (probably less than 150 years ago), /je/ became /e/, leaving /j/ only before /a u o/.

*Word-medially is a tricky term here. In almost all roots, pre-modern Japanese /p/ or /ɸ/ undergoes the change to /w/ (→Ø), which a very small number of exceptions, notably あひる ahiru and 溢れる afureru as you mentioned above. There is also 母 haha, but I think reduplicants gets to slip by this rule. There's also yahari, which I mentioned above.

When we're dealing with compounds, the rules become less clear. In the vast majority of compound words where root-initial /h/ appears word-medially, it remains /h/ (or is voiced to /b/, or geminated to /p/). However, in certain names, the rule does occur, e.g. 藤原 Fujiwara (from fuji + hara). Some are variable, e.g. 八幡 can be Yawata or Yahata.
Let's also not forget the topic particle は wa, which underwent this change, as well as the directional particle へ e (/pe/ → /ɸe/ → /we/ → /e/).

Regarding 五線譜:
Rendaku (conditional voicing of the first consonant of a non-initial morpheme) is subject to limitations imposed by branching constraints: namely, rendaku cannot normally happen when a morpheme is added to a preexisting compound. The word gosenfu is actually gosen + fu. Since gosen is already a compound of go + sen, rendaku is disallowed from occurring on fu. When you get into names of people or places, though, weird things can happen (Akihabara?)
2: The ending of the first bridge into the chorus (1:20) is this:
さぁ踊れや騒げ 飲めや歌えやお手を拝借
saa odore ya sawage (nomeya utaeya o-te wo haishaku)
Dance and live it up~ (Drink! Sing! Take my hand!)

The translator viewed nomeya utaeya as simple imperatives, but Jisho says that that's another fixed phrase meaning "revelry". What I'm wondering is if the ya in both words is the one that means "and", or if it's an old form of the imperative and the modern "e-grade + ∅" is just a truncation of that.
I don't honestly know this expression, but it's probably the imperative particle や ya (which I am honestly not familiar with, but which does exist and is used after the imperative form, although if someone used it at me I feel its meaning would be quite obvious).
I personally find it a little odd that the same grade is used for both the potential/conditional and the imperative (i.e nomu "to drink" > nomeru "can drink", nomeba "if (he) drinks", nome! "drink!"). E-grade is obviously an irrealis and imperatives don't strike me as particularly irrealis.
Oddly, the conditional in Classical Japanese uses the irrealis for conditionals; the realis form used in modern Japanese indicates reasons in Classical Japanese: i.e. 飲まば nomaba "if (I) drink" vs. 飲めば "because I drink". The imperative was a completely separate stem, which is identical to the realis stem for yodan verbs but different in other conjugations. cf.:

Code: Select all

yodan:    nom.a   nom.i   nom.u     nom.u   nom.e     nom.e
k.nidan:  tab.e   tab.e   tab.uru   tab.u   tab.ure   tab.eyo
n-irr:   shin.a  shin.i  shin.uru  shin.u  shin.ure  shin.e
However there are some words which have these yet MUST be native Japanese words (usually because they're verbs), such as the well-known 頑張る ganbaru "do well, persevere"
In the first of M Mira's descriptions, 眼張る, the 眼 gan is Sinitic in origin. In the second, it's just a contraction. In any event, this word came about in the Edo period, so rules governing Asuka-period vocabulary are not relevant.
(if I had a nickel for everytime I heard ganbare or ganbatte in anime...honestly, I've rarely heard 頑張る in any inflection but those two)
Pretty sure either today or yesterday I heard something along the lines of さあ、今日も1一日頑張りましょう".
しゃべる shaberu "chat"
This is a bizarre word and I have never found an etymology for it. However, there are the words 背負う shou and 仰る ossharu, which also have an unusual /ɕ/. 背負う is rather transparently a contraction of seou (the kanji make this clear, plus it can still be pronounced seou), while ossharu is a contraction of ohosearu (actually owosearu if you remember what I said above; Classical Japanese spelling indicates etymological differences). So perhaps shaberu also involves a contraction of /ɕia/ or /sea/. Best I can do. Sorry!
/k g/ both drop before /i/ in verbal inflection, with /g/ also voicing a following voiceless consonant (e.g *yakita *oyogita" > yaita oyoida "grilled, swam". However in verbs this only occurs in the past tense. -ku and -gu verbs still have normal i-grades such as the -masu form (yakimasu oyogimasu). My guess for this is that the polite form is clearly i-grade in the other verbs (noborimasu, machimasu, tsukaimasu, shinimasu(?), whereas the past tense is just a complete mess (nobotta, matta, tsukatta, shinda...) not connected to any particular grade, so maybe analogy has run through the forms and fixed stuff.

Then there's the i-adjectives which have *ki > i regularly, explaining why their other forums suddenly gain a -k- (atsui > atsukunai)
In very polite speech (and in Kansai-ben), /k/ is dropped from the ku ending as well, e.g. よろしゅうございますか yoroshū gozaimasu ka. This may sound weird, but I ask you to recall お早うございます ohayō gozaimasu (hayakuhayauhayō). Yoku wakaranai in Tokyo is yō wakarahen in Kansai, for example.

As for the past tense forms, it's a systematic dropping of the vowel and consequent mutation of the initial consonant in the cluster to fit Japanese phonotactics, or spreading of voicing, etc. So:
iwitari → *iwta → itta
hanashitari → hanashita
mochitari → motta
shinitari → shinda
yobitari → *yobta → yonda
yomitari → *yomta → yonda
kiritari → *kirta → kitta

plus the onbin ones:
kakitari → kaita
nugitari → nuida

It would not surprise me if for the ones with da, there was an intermediate stage where voicing spread, like:
shinitari → shinida → shinda
nugitari → nugida → nuida
yomitari → yomida → *yomda → yonda

User avatar
Chagen
Avisaru
Avisaru
Posts: 707
Joined: Thu Sep 22, 2011 11:54 pm

Re: Questions about Japanese

Post by Chagen »

Thank you for all the answers. However, in my onbin question, I wasn't asking about the actual processes. I wanted to know if there were any other natlangs which have a set of "sound changes" that only apply during verbal inflection (or maybe any kind of inflection--Japanese just has inflection in verbs only). ki > Øi is not a regular sound change in Japanese AFAIK--nouns, and verbal/adjective roots retain /ki/--it's only in suffixes (and root-suffix boundaries) that this particular change occurs.

The part about *h become *w word-medially then finally dropping before everything but /a/ does explain why the topic and direction particles are spelled は and へ yet pronounced /wa/ and /e/. However, this makes me wonder something. This only happens word-medially--that is, initial *h remains. Therefore, isn't this conclusive proof that Japanese particles are morphologically a part of the word they agree with? If they weren't, then those two particles would have remained /ha he/. Since they became /wa (w)e/, they were considered word-final. This then makes me wonder why these couldn't be analyzed as case markers. Nothing can intervene between a particle and its argument, and stuff like これは > こりゃ (kore wa > korya) is more likely if the particles are suffixes as opposed to free-standing words, if that makes sense (similar simplifications have mostly happened in verbal suffixes: wakaranai/tsumaranai > wakannai/tsumannai).
Nūdhrēmnāva naraśva, dṛk śraṣrāsit nūdhrēmanīṣṣ iźdatīyyīm woḥīm madhēyyaṣṣi.
satisfaction-DEF.SG-LOC live.PERFECTIVE-1P.INCL but work-DEF.SG-PRIV satisfaction-DEF.PL.NOM weakeness-DEF.PL-DAT only lead-FUT-3P

Richard W
Avisaru
Avisaru
Posts: 363
Joined: Sat Oct 16, 2010 8:28 pm

Re: Questions about Japanese

Post by Richard W »

Chagen wrote:Thank you for all the answers. However, in my onbin question, I wasn't asking about the actual processes. I wanted to know if there were any other natlangs which have a set of "sound changes" that only apply during verbal inflection (or maybe any kind of inflection--Japanese just has inflection in verbs only).
The voicing assimilation of the -s and -ed inflections in English seems to be an example.

vokzhen
Avisaru
Avisaru
Posts: 352
Joined: Sat Aug 09, 2014 3:43 pm
Location: Iowa

Re: Questions about Japanese

Post by vokzhen »

Ingush voices all final fricatives in suffixes, but not in roots. I think I've run across a language that did it to all suffix fricatives, but I can't remember for sure and may be just misremembering Ingush.

User avatar
finlay
Sumerul
Sumerul
Posts: 3600
Joined: Mon Dec 22, 2003 12:35 pm
Location: Tokyo

Re: Questions about Japanese

Post by finlay »

Chagen wrote:Thank you for all the answers. However, in my onbin question, I wasn't asking about the actual processes. I wanted to know if there were any other natlangs which have a set of "sound changes" that only apply during verbal inflection (or maybe any kind of inflection--Japanese just has inflection in verbs only). ki > Øi is not a regular sound change in Japanese AFAIK--nouns, and verbal/adjective roots retain /ki/--it's only in suffixes (and root-suffix boundaries) that this particular change occurs.

The part about *h become *w word-medially then finally dropping before everything but /a/ does explain why the topic and direction particles are spelled は and へ yet pronounced /wa/ and /e/. However, this makes me wonder something. This only happens word-medially--that is, initial *h remains. Therefore, isn't this conclusive proof that Japanese particles are morphologically a part of the word they agree with? If they weren't, then those two particles would have remained /ha he/. Since they became /wa (w)e/, they were considered word-final. This then makes me wonder why these couldn't be analyzed as case markers. Nothing can intervene between a particle and its argument, and stuff like これは > こりゃ (kore wa > korya) is more likely if the particles are suffixes as opposed to free-standing words, if that makes sense (similar simplifications have mostly happened in verbal suffixes: wakaranai/tsumaranai > wakannai/tsumannai).
Basically the reason they're not case suffixes is because in coordinated noun phrases (e.g. 犬と猫が…) the case applies to the whole thing. A true case would produce something more like *inu-ga to neko-ga - both would have to be in nominative case. But actually the particle ga applies to the whole noun phrase. And as you say they're not true postpositions because they're phonologically part of the previous word. So we call them clitics, or use particles because it's a useful catchall term.

There's also the fact that particles like wa and mo replace other particles. That can't happen with cases either

User avatar
LinguistCat
Avisaru
Avisaru
Posts: 250
Joined: Thu Apr 13, 2006 7:24 pm
Location: Off on the side

Re: Questions about Japanese

Post by LinguistCat »

I understand if this needs to become its own thread, and I don't want to hijack, but I thought I might get more attention here than on it's own. Does anyone have a good list of words/roots from Early Middle Japanese?

I've found some info about sound changes over time from then to Modern Japanese, and there are some things that are difficult to reconstruct if you don't already know that the spelling has been changed to reflect the pronunciation. From about 2 hours of Googling and reading, I saw it mentioned that what is now ん was likely originally closer to [mu] or [m], っ and つ were written as the latter and probably was pronounced as [tsu] in all cases (at least for a time and at least in some dialects).

But both of these are relatively easy to account for compared to a lot of mergers that have happened, especially with h-w-y-Vowels. I'm guessing some things can be sussed out just by looking at modern conjugations for some things, but it would be nice to be able to check for something and then try to back derive only if I have to, to derive an AltUniverse relative.
The stars are an ocean. Your breasts, are also an ocean.

Travis B.
Sumerul
Sumerul
Posts: 3570
Joined: Mon Jun 20, 2005 12:47 pm
Location: Milwaukee, US

Re: Questions about Japanese

Post by Travis B. »

Where modern Japanese has long consonants - marked with a little tsu - you are implying that they had [tsu] consistently? Really?
Dibotahamdn duthma jallni agaynni ra hgitn lakrhmi.
Amuhawr jalla vowa vta hlakrhi hdm duthmi xaja.
Irdro. Irdro. Irdro. Irdro. Irdro. Irdro. Irdro.

Sumelic
Avisaru
Avisaru
Posts: 385
Joined: Sat Mar 28, 2015 7:05 pm

Re: Questions about Japanese

Post by Sumelic »

Travis B. wrote:Where modern Japanese has long consonants - marked with a little tsu - you are implying that they had [tsu] consistently? Really?
Obviously modern Japanese small tsu doesn't always correspond to historical [tsu]. For example, no one would suppose that is the case for geminates formed due to compensatory lengthening with simultaneous long vowel shortening that clawgrip mentioned, or assimilation of /k/ to a following consonant. But presumably these were spelled differently before these processes took effect.

I believe what vampyre_smiles is saying is that he's heard that at some point in the pre-reform Japanese writing system, the "tsu" character always represented /tu/ ([tsu]). I don't know if this is true or not. Many of the long consonants of modern Japanese do come from assimilation of historical /t/ to a following consonant after the loss of an intervening vowel.

User avatar
LinguistCat
Avisaru
Avisaru
Posts: 250
Joined: Thu Apr 13, 2006 7:24 pm
Location: Off on the side

Re: Questions about Japanese

Post by LinguistCat »

Sumelic wrote:
Travis B. wrote:Where modern Japanese has long consonants - marked with a little tsu - you are implying that they had [tsu] consistently? Really?
...

I believe what vampyre_smiles is saying is that he's heard that at some point in the pre-reform Japanese writing system, the "tsu" character always represented /tu/ ([tsu]). I don't know if this is true or not. Many of the long consonants of modern Japanese do come from assimilation of historical /t/ to a following consonant after the loss of an intervening vowel.
Yes, that's basically what I found in my search. Granted, other sound changes later would obscure that a bit. But this is why I'd prefer a word list of some sort rather than a list of sound changes to reverse engineer things. Then again I also only researched for about 3 hours after being up all night. I may have over generalized something that I read or just straight up misunderstood.
The stars are an ocean. Your breasts, are also an ocean.

User avatar
finlay
Sumerul
Sumerul
Posts: 3600
Joined: Mon Dec 22, 2003 12:35 pm
Location: Tokyo

Re: Questions about Japanese

Post by finlay »

in hakodate the Dock area is written down as どつく in hiragana with a big tsu, but pronounced ドック as you'd expect from an english loanword. it bothered the shit out of me. but this might be what you mean? it's obviously a relic of the old spelling system or something

User avatar
LinguistCat
Avisaru
Avisaru
Posts: 250
Joined: Thu Apr 13, 2006 7:24 pm
Location: Off on the side

Re: Questions about Japanese

Post by LinguistCat »

finlay wrote:in hakodate the Dock area is written down as どつく in hiragana with a big tsu, but pronounced ドック as you'd expect from an english loanword. it bothered the shit out of me. but this might be what you mean? it's obviously a relic of the old spelling system or something
Something like that. If nothing else, big tsu was the only one used at one point and the size of little tsu was variable for a while and in different places, according to what I read. And there's obviously a few hold overs with big tsu from what you've said. I suppose I was assuming the original sound change was tC>C: in certain circumstances, which later went from being written つ to っ, and later sound changes that caused other long vowels just ended up lumped into that. Is that (basically) correct?
The stars are an ocean. Your breasts, are also an ocean.

User avatar
clawgrip
Smeric
Smeric
Posts: 1723
Joined: Wed Feb 29, 2012 8:21 am
Location: Tokyo

Re: Questions about Japanese

Post by clawgrip »

vampyre_smiles wrote:I've found some info about sound changes over time from then to Modern Japanese, and there are some things that are difficult to reconstruct if you don't already know that the spelling has been changed to reflect the pronunciation. From about 2 hours of Googling and reading, I saw it mentioned that what is now ん was likely originally closer to [mu] or [m], っ and つ were written as the latter and probably was pronounced as [tsu] in all cases (at least for a time and at least in some dialects).
Old Japanese, as you may know, had a strict CV syllable structure, but this changed with the influx of Chinese vocabulary. As I understand it, at this point, coda /m/ and /n/ were distinguished, and /p t k/ as well, to some extent or another. Coda /p/ and /k/ quickly got an epenthetic /u/ added after them, but /t/ did not. Final /m/ and /n/ also eventually merged.

Coda consonants typically were not marked at all in writing, but sometimes as you say, む (which was not really distinguished from ん) was used to mark final /n/, and then also final /t/. Eventually, as writing became more explicit, む/ん was used for coda /n/ (as well as /mu/), while つ was used for final /t/, through analogy with く ku and ふ pu/fu, though I think this was more used in foreign loanwords than Japanese and Sino-Japanese vocabulary, so totte would be more often written as 取て than 取つて (small letters like っ were not really used yet). I believe む and ん were officially distinguished in 1900, but it wasn't until the post-WW2 spelling reforms that っ became the standard way of marking non-nasal geminates.
But both of these are relatively easy to account for compared to a lot of mergers that have happened, especially with h-w-y-Vowels. I'm guessing some things can be sussed out just by looking at modern conjugations for some things, but it would be nice to be able to check for something and then try to back derive only if I have to, to derive an AltUniverse relative.
I have already explained the h-w-y vowels in this post, though you'll need to look at a classical Japanese dictionary rather than a modern one to figure out the etymology.

Post Reply