Japanese Pitch Accent [split from Q&A]

A forum for discussing linguistics or just languages in general.
User avatar
LinguoFranco
greek
greek
Posts: 613
Joined: 20 Jul 2016 17:49
Location: U.S.

Japanese Pitch Accent [split from Q&A]

Post by LinguoFranco »

Does Japanese's pitch accent have a 'bounded' system?

For example, in Ancient Greek, the accent is right bounded, and (I think) it can fall on the ultimate, penultimate or antepenult.
User avatar
LinguistCat
sinic
sinic
Posts: 324
Joined: 06 May 2017 07:48

Re: Japanese Pitch Accent

Post by LinguistCat »

LinguoFranco wrote: 12 Dec 2020 17:45 Does Japanese's pitch accent have a 'bounded' system?

For example, in Ancient Greek, the accent is right bounded, and (I think) it can fall on the ultimate, penultimate or antepenult.
So within Standard Japanese pitch accent in native non-compound words, there are accented or unaccented words; Within the accented words you have ones that are accented on the first mora (with a drop after), accented middle (only found in words 3+ mora in length, high tones from the second mora until it drops for the final mora), or accented on the last mora. Due to pitch rising slowly in unaccented words, final accented words look unaccented unless followed by most particles, where the tone drops in final accented, but doesn't in unaccented words. Compounds are a bit more complex and depends on what each component's accent pattern. And Chinese loans probably got their pitch patterns at least in part from Chinese, but also greatly simplified to match the "zero or one tone drop per word" rule.

I don't know if that makes it right bounded, but you do need to know both the pattern and the word's length. This is also only talking about the standard accent and not other common accent patterns, which would be an entire other convo and someone already wrote a very big book about it which I have in PDF form somewhere.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Japanese Pitch Accent

Post by Creyeditor »

Judging from what I have read on Japanese pitch accent and from what LinguistCat wrote, it is neither right- nor left-bounded if we look at the abstract accent. I am not so sure about the phonetic pitch movements, since the pitch rise always starts close to the left edge, IINM. Also note that their are unaccented (content) words which is another difference to Ancient Greek.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: Japanese Pitch Accent

Post by Sequor »

I don't know Japanese to tell, but 1) are there any native non-compound words that are four syllables or longer?, and 2) are there any limitations on where the pitch accent can fall if it's in the middle?

If it does not occur on the third syllable in words of four syllables, nor on the fourth syllable in words of five syllables, I'd say that'd be evidence for it being "left-bounded" in some sense, even though we do know words can indeed still take the accent on the last syllable (the 尾高 odaka accent).

(Or vice-versa if the limitation is on the second and third syllables respectively, even though we do know first-syllable accent occurs too).
hīc sunt linguificēs. hēr bēoþ tungemakeras.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Japanese Pitch Accent

Post by clawgrip »

I'm having difficulty coming up with any four-syllable/four-mora words that cannot be further analyzed.

Accent is not general bounded, although it's more likely to be towards right right.

A lot of words with far-left accents are loanwords
レストラン rèsutoran "restaurant"
パンフレット pànfuretto "pamphlet"
インスタント ìnsutanto "instant"

In foreign or Sino-Japanese compounds, the accent of the first element is frequently dropped, leaving only the right element's accent:
インスタントラーメン insutanto-rā̀men "instant ramen"
日本 nihòn "Japan"
日本大使館 nihon-taishìkan "Japanese embassy"
中華人民共和国 Chūkajinmin-Kyōwàkoku "People's Republic of China"
南北線時代 nanbokusen-jìdai "Civil War era"
化学調味料 kagaku-chōmìryō "artificial seasoning"

There is no possible way to analyze Japanese as having a left-bounded accent, because it is the pitch drop, not the pitch rise, that determines the pitch accent for the word; the rise begins on the left edge in order to set up the drop, wherever it occurs. A word will have no pitch rise if the preceding word has no accent, because the high pitch will simply continue over from that word.
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: Japanese Pitch Accent

Post by Sequor »

clawgrip wrote: 18 Dec 2020 01:33I'm having difficulty coming up with any four-syllable/four-mora words that cannot be further analyzed.

Accent is not general bounded, although it's more likely to be towards right right.

A lot of words with far-left accents are loanwords
レストラン rèsutoran "restaurant"
パンフレット pànfuretto "pamphlet"
インスタント ìnsutanto "instant"

In foreign or Sino-Japanese compounds, the accent of the first element is frequently dropped, leaving only the right element's accent:
インスタントラーメン insutanto-rā̀men "instant ramen"
日本 nihòn "Japan"
日本大使館 nihon-taishìkan "Japanese embassy"
中華人民共和国 Chūkajinmin-Kyōwàkoku "People's Republic of China"
南北線時代 nanbokusen-jìdai "Civil War era"
化学調味料 kagaku-chōmìryō "artificial seasoning"

There is no possible way to analyze Japanese as having a left-bounded accent, because it is the pitch drop, not the pitch rise, that determines the pitch accent for the word; the rise begins on the left edge in order to set up the drop, wherever it occurs. A word will have no pitch rise if the preceding word has no accent, because the high pitch will simply continue over from that word.
Those are great points!!! I imagine a lot of those English loanwords have a first-syllable accent because English itself is generally stress-initial, but I also notice many English noun loanwords are accentless...

While you're here, are you aware of any known instances where (Middle or modern) Chinese tone was adapted in Japanese with a certain pitch accent pattern that matched it in some way?

Also, I'm not familiar with Japanese, but I've seen no evidence so far in whatever I've read that it's good to talk about pitch accent with morae rather than syllables, except that it's the convention to use morae. Do you think I'm right that pitch accent works with syllables instead of morae? Are there any words such as my imaginary "maàto" rather than "màato" (minimal pairs or not)?
hīc sunt linguificēs. hēr bēoþ tungemakeras.
User avatar
LinguistCat
sinic
sinic
Posts: 324
Joined: 06 May 2017 07:48

Re: Japanese Pitch Accent

Post by LinguistCat »

Now that I'm thinking about this, the only 5+ mora words I can think of that are natively Japanese and not (directly) compounds are all older, classical words that are rarely used these days. And if they have a "middle" accent, it's either the penultimate or antepenultimate and it's based on the length of the word, not the word itself. I'd have to go back through my research to find some examples tho.

ETA: With a quick check of this site, there are near minimal pairs for words that have a moraic nasal as the second mora for all accent patterns. I haven't found any for long vowels but I also literally only checked for <あ> without getting farther. I also know there are words that end the same but have the difference of whether the first, second or neither of a long vowel sequence/vowel+moraic nasal has the down step. So that's definitely evidence to me that is IS moraic, at least in the standard dialect.
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: Japanese Pitch Accent

Post by Sequor »

LinguistCat wrote: 18 Dec 2020 08:29ETA: With a quick check of this site, there are near minimal pairs for words that have a moraic nasal as the second mora for all accent patterns. I haven't found any for long vowels but I also literally only checked for <あ> without getting farther. I also know there are words that end the same but have the difference of whether the first, second or neither of a long vowel sequence/vowel+moraic nasal has the down step. So that's definitely evidence to me that is IS moraic, at least in the standard dialect.
Well... this is very strange. I looked at that page you linked to, and I couldn't find any of the things you mention. Could you quote a specific example? All the あ examples there seem to have pitch accents analyzable as syllables.

I think you might be confusing the phonemic and phonetic levels here? I look at 暗算 [aꜛɴdzaɴ] (no drop after indicated) and 安産 [aꜜɴdzaɴ], but these is not a relevant minimal pair, since these can be phonemically analyzed as /aɴ-dzaɴ/ (accent-less) versus /^aɴ-dzaɴ/ (if you allow me to mark pitch accent on a syllable placing a ^ before it). Rising pitch is not phonemic at any rate.

What I would like to see is [aꜜɴdzaɴ] versus *[aɴꜜdzaɴ], or "[maꜜato]" versus *"[maaꜜto]". That'd be very convincing. My staring at transcribed pitch accents hasn't been extensive, but I haven't seen anything like *[aɴꜜdzaɴ] or *"[maaꜜto]", so I say I strongly suspect it is applied on syllables...
hīc sunt linguificēs. hēr bēoþ tungemakeras.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Japanese Pitch Accent

Post by Creyeditor »

clawgrip wrote: 18 Dec 2020 01:33 A word will have no pitch rise if the preceding word has no accent, because the high pitch will simply continue over from that word.
Right, I forgot about this part [:)] I heard there is dialectical variation in the precise phonetic implementation of these high "spans" but I can't recall the details.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
LinguistCat
sinic
sinic
Posts: 324
Joined: 06 May 2017 07:48

Re: Japanese Pitch Accent

Post by LinguistCat »

Sequor wrote: 18 Dec 2020 09:30 I think you might be confusing the phonemic and phonetic levels here? I look at 暗算 [aꜛɴdzaɴ] (no drop after indicated) and 安産 [aꜜɴdzaɴ], but these is not a relevant minimal pair, since these can be phonemically analyzed as /aɴ-dzaɴ/ (accent-less) versus /^aɴ-dzaɴ/ (if you allow me to mark pitch accent on a syllable placing a ^ before it). Rising pitch is not phonemic at any rate.

What I would like to see is [aꜜɴdzaɴ] versus *[aɴꜜdzaɴ], or "[maꜜato]" versus *"[maaꜜto]". That'd be very convincing. My staring at transcribed pitch accents hasn't been extensive, but I haven't seen anything like *[aɴꜜdzaɴ] or *"[maaꜜto]", so I say I strongly suspect it is applied on syllables...
Ok I think there's a misunderstanding at the base of this: The "up steps" are actually the start of a continuous rise in pitch due to common processes that occur in pitch accent languages. The only salient feature IS the down step. So phonemically, you only have no accent (flat) or accent (down step), which phonetically looks like a gradual rise in pitch for no accent assuming it isn't the first syllable, or a pitch drop. And since you get a down step in 安産 between the <a> and the <n> and in 暗算 we do not, it is a relevant near minimal pair. If you want to look for better examples (or look to counter this) we'd probably want to look among the verbal adjectives, especially those that end in -shii/しい.

So some examples:
おいしい has no down step and is unaccented.
うれしい has a down step after shi/し

Interestingly, I haven't seen any of these with the down step at the end of the word, so I'd tentatively say that pattern might be disallowed in adjectives.
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: Japanese Pitch Accent

Post by Salmoneus »

LinguistCat wrote: 18 Dec 2020 18:15
Sequor wrote: 18 Dec 2020 09:30 I think you might be confusing the phonemic and phonetic levels here? I look at 暗算 [aꜛɴdzaɴ] (no drop after indicated) and 安産 [aꜜɴdzaɴ], but these is not a relevant minimal pair, since these can be phonemically analyzed as /aɴ-dzaɴ/ (accent-less) versus /^aɴ-dzaɴ/ (if you allow me to mark pitch accent on a syllable placing a ^ before it). Rising pitch is not phonemic at any rate.

What I would like to see is [aꜜɴdzaɴ] versus *[aɴꜜdzaɴ], or "[maꜜato]" versus *"[maaꜜto]". That'd be very convincing. My staring at transcribed pitch accents hasn't been extensive, but I haven't seen anything like *[aɴꜜdzaɴ] or *"[maaꜜto]", so I say I strongly suspect it is applied on syllables...
Ok I think there's a misunderstanding at the base of this: The "up steps" are actually the start of a continuous rise in pitch due to common processes that occur in pitch accent languages. The only salient feature IS the down step. So phonemically, you only have no accent (flat) or accent (down step), which phonetically looks like a gradual rise in pitch for no accent assuming it isn't the first syllable, or a pitch drop. And since you get a down step in 安産 between the <a> and the <n> and in 暗算 we do not, it is a relevant near minimal pair. If you want to look for better examples (or look to counter this) we'd probably want to look among the verbal adjectives, especially those that end in -shii/しい.

So some examples:
おいしい has no down step and is unaccented.
うれしい has a down step after shi/し

Interestingly, I haven't seen any of these with the down step at the end of the word, so I'd tentatively say that pattern might be disallowed in adjectives.
I'm not sure that that answers Sequor's question?

Your examples show that there's a difference between a bimoraic syllable with an accent, and a bimoraic syllable without an accent. But they don't show that accent position within a syllable is phonemic. There could simply be a phonetic rule "in a bimoraic accented syllable, downstep occurs after the first mora" - but that's a phonetic rule, not a phonemic rule. So Sequor could still be right that accent is phonemically syllable, even though it's phonetically moraic. You'd need to find a word in which the downstep occurs after the second mora of a syllable - and ideall a minimal pair where the accent position within the syllable is the only distinctive feature - to show that he's wrong.


However, I don't entirely agree with him on this (although I don't entirely disagree either). Because there's a third case: in which bimoraic syllables always take accent (if at all) on the first mora, but accent is still phonemically moraic. This - I think - would be a fair description in cases in which accent position were predictable based on mora counts. Which I gather isn't the case in Japanese. But might mora counting be relevant to tone sandhi behaviours, including the realisation of accent in compounds? That is, even though accent may look syllabic on the surface (assuming Sequor's right that there are no deuteromoraic accented syllables), due to a surface phonetic rule neutralising the distinction within syllables, it's possible that phonemic moraic accent may still be revealed in other ways.

To give a fictional example of what I mean, let's imagine a rule that says "where two accents are adjacent, the second is deleted", but then "where two accents remain, the first is deleted". Then let's imagine the words /^ku/, /^santo/, and /ta^nto/. We would then have the compounds:
/^kusanto/
/kuta^nto/
I think this would be good cause to say that accent was phonemically moraic, even if a surface neutralising rule meant that the actual pitch realisation in /^santo/ and /ta^nto/ was the same (i.e. /ta^nto/ is realised [^tanto])

However, I've no idea whether this is actually how Japanese works.
User avatar
LinguistCat
sinic
sinic
Posts: 324
Joined: 06 May 2017 07:48

Re: Japanese Pitch Accent

Post by LinguistCat »

Salmoneus wrote: 18 Dec 2020 19:58 I'm not sure that that answers Sequor's question?

Your examples show that there's a difference between a bimoraic syllable with an accent, and a bimoraic syllable without an accent. But they don't show that accent position within a syllable is phonemic. There could simply be a phonetic rule "in a bimoraic accented syllable, downstep occurs after the first mora" - but that's a phonetic rule, not a phonemic rule. So Sequor could still be right that accent is phonemically syllable, even though it's phonetically moraic. You'd need to find a word in which the downstep occurs after the second mora of a syllable - and ideall a minimal pair where the accent position within the syllable is the only distinctive feature - to show that he's wrong.
...
Ah, that does clear it up a little. Depending on if we're only counting "long vowels" or any segment that could be argued to be "a syllable" the answer might be different.

I have found some minimal pairs or near minimal pairs for some vowel sequences. I should probably go through my resources though and see what patterns there are and what limitations there are for each accent pattern.

酔い yoi' (drunkenness, from yo'u "to get drunk")
良い yo'i (good, alternate pronunciation of i'i meaning the same)
扇ぐ ao'gu (to fan), 仰ぐ (to look up at)
治す nao'su (to cure or heal)
尚 na'o (furthermore, in addition)
陥るochii'ru (to fall into)
食い入る kuii'ru (to eat into. NB compound of ku'u "to eat, vulgar" and iru "to enter")
(I have found no native Japanese words with -i'i- aside from the noted above)
催す moyoo'su (to hold an event, give a dinner party)
通す to'osu (to stick through, to force through)

I will comment that I have not found a syllabic nasal with an accent after it, though I haven't searched comprehensively; If they do exist, they're marginal. And I'm not even sure what it would mean for someone to claim there's a difference between an accent before a sokuon vs after it (aside from maybe /s:/), though I have seen some resources differentiate, for example, ma'tte vs asat'te. There are a long of vowel sequences that are only differentiated by the placement of the down step. That could be a difference of syllabification or it could be that pitch assignment is partially moraic and partially syllabic.
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: Japanese Pitch Accent

Post by Sequor »

Looks like Salmoneus' wording clarified what I intended.
LinguistCat wrote: 18 Dec 2020 22:51陥るochii'ru (to fall into)
食い入る kuii'ru (to eat into. NB compound of ku'u "to eat, vulgar" and iru "to enter")
(I have found no native Japanese words with -i'i- aside from the noted above)
By "the noted above", do you mean うれしい [ɯꜛɺes\iꜜi] ureshìi? The page you linked shows many with -iꜜi-, like いいもの [iꜜimono], so I take it you made a typo and meant to write "no...words with -ii'- aside from" ochiìru and kuiìru.
催す moyoo'su (to hold an event, give a dinner party)
通す to'osu (to stick through, to force through)
Now that is what I call a wonderful suggestive pair.

I can't help but groundless-ly, very speculatively wonder whether a typo was involved in this dataset here, so that this is actually [moꜛjoꜜosɯ] moyòosu, and similarly ochìiru and kuìiru, but this is not a comment about analysis anymore (rather just the data collection). I am now definitely agnostic of my previous belief from yesterday at least. To say nothing of Salmoneus' good objection that a distinction could show up in morphophonology, in the behaviour of words when entering compounds or being inflected (probably undocumented in accent dictionaries, except implicitly when you look at separate compound entries).
And I'm not even sure what it would mean for someone to claim there's a difference between an accent before a sokuon vs after it (aside from maybe /s:/), though I have seen some resources differentiate, for example, ma'tte vs asat'te.
Man... hrm.
hīc sunt linguificēs. hēr bēoþ tungemakeras.
User avatar
LinguistCat
sinic
sinic
Posts: 324
Joined: 06 May 2017 07:48

Re: Japanese Pitch Accent

Post by LinguistCat »

Sequor wrote: 18 Dec 2020 23:27 By "the noted above", do you mean うれしい [ɯꜛɺes\iꜜi] ureshìi? The page you linked shows many with -iꜜi-, like いいもの [iꜜimono], so I take it you made a typo and meant to write "no...words with -ii'- aside from" ochiìru and kuiìru.
Ah sorry that's my own mistake. I actually meant 良い pronounced /iꜜi/ which いいもの is derived from. I also had in mind trying to find words with ii where it was not a -shii adjective since they all end with /iꜜi/.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Japanese Pitch Accent

Post by clawgrip »

Moyoosu is not a word I tend to use or have much familiarity with, but another website lists them the same: moyoòsu vs. tòosu.

I can also say definitively that naòsu and nào have different accents, as indicated above, and tòosu matches nào, on the first mora.

Also, kind of a tangent here, but morae are definitely a thing in terms of word length as well. The Tokyo subway station 国会議事堂前 Kokkai-gijidō-mae is a good example: each set of two morae is pronounced in roughly the same length of time: koQ-kai-giji-dō-mae
GrandPiano
mayan
mayan
Posts: 2080
Joined: 11 Jan 2015 23:22
Location: USA

Re: Japanese Pitch Accent

Post by GrandPiano »

Sequor wrote: 15 Dec 2020 19:361) are there any native non-compound words that are four syllables or longer?
ホトトギス hototógisu "lesser cuckoo" is one.
LinguistCat wrote: 18 Dec 2020 22:51I have found some minimal pairs or near minimal pairs for some vowel sequences. I should probably go through my resources though and see what patterns there are and what limitations there are for each accent pattern.
This actually depends on the vowel sequence. Some vowel sequences allow the accent to occur on the second vowel and some don't (unless the two vowels belong to different morphemes). Specifically, it seems like the vowel sequences ai, ae, and oi (maybe also ui?) only permit the accent on the first mora when they occur within a single morpheme, and other vowel sequences allow the accent on either mora. Long vowels also generally only allow the accent on the first mora.

The fact that ai and ae only allow the accent on the first mora is most obvious when you look at verbs. Japanese verbs have two accented patterns: one where the dictionary form is unaccented (e.g. 止まる tomaru "to stop") and one where the dictionary form is accented on the second-to-last syllable (e.g. 休む yasúmu "to rest"). For the second pattern, the accent would fall on one of two vowels in a vowel sequence, we see different behavior depending on the vowel sequence:

倒す taósu "to throw down"
but
入る háiru "to enter"
帰る káeru "to return"
通る tóoru "to go past"

I don't have a verb example for oi, but there's the noun 恋 kói "love", which you would expect to have the accent on the last mora since it's historically a nominalization of the verb 恋う kóu "to love" (nominalized forms of accented verbs with 3 or fewer morae are normally accented on the last mora, e.g. 話す hanásu "to speak", 話 hanashí "speech").

For this reason, I wonder if it would make sense to analyze /ai/, /ae/, and /oi/ as diphthongs and other sequences as two monophthongs in hiatus; in other words, to analyze taósu as three syllables but háiru as two syllables (I think /ui/ may be another sequence that behaves as a diphthong, but I can't think of an example offhand).

I'm not really sure what to make of moyoósu. Maybe tóoru is /toː.ɾu/ but moyoósu is /mo.jo.o.su/?

The other cases where the accent falls on the second mora in one of these sequences are not exceptions because the two vowels belong to different morphemes. 陥る ochiíru is a compound of 落ちる ochíru "to fall" and 入る iru "to go/come in". 酔い yoí is a nominalization of the verb yóu, as was noted.*

One might bring up apparent counterexamples like 考える kangaéru "to think, to consider", 老いる oíru "to grow old", and 強いる shiíru "to compel". Based on these verbs, it seems like, at least for the purpose of accent assignment, the -eru and -iru of ichidan verbs are treated as morphemes (and not -ru as you might expect based on how these verbs inflect). For the sequence ae specifically, though, this seems to be in the process of changing, as kangáeru is now a common pronunciation of 考える that coexists with kangaéru.

*This seems to contradict the kói example from earlier. My guess is that since the verb kóu is no longer in common use, its nominalized form was reanalyzed as a monomorphemic noun and its accent shifted accordingly. The verb yóu, on the other hand, is still commonly used, so yoí is still thought of as bimorphemic.
Last edited by GrandPiano on 06 Jan 2021 19:18, edited 2 times in total.
User avatar
Sequor
sinic
sinic
Posts: 438
Joined: 30 Jun 2012 06:13

Re: Japanese Pitch Accent

Post by Sequor »

GrandPiano wrote: 06 Jan 2021 07:03For this reason, I wonder if it would make sense to analyze /ai/, /ae/, and /oi/ as diphthongs and other sequences as two monophthongs in hiatus; in other words, to analyze taósu as three syllables but háiru as two syllables (I think /ui/ may be another sequence that behaves as a diphthong, but I can't think of an example offhand).

I'm not really sure what to make of moyoósu. Maybe tóoru is /toː.ɾu/ but moyoósu is /mo.jo.o.su/?
That would be surely disturbing, yet amazing, if true.
hīc sunt linguificēs. hēr bēoþ tungemakeras.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Japanese Pitch Accent

Post by clawgrip »

GrandPiano wrote: 06 Jan 2021 07:03One might bring up apparent counterexamples like 考える kangaéru "to think, to consider", 老いる oíru "to grow old", and 強いる shiíru "to compel". Based on these verbs, it seems like, at least for the purpose of accent assignment, the -eru and -iru of ichidan verbs are treated as morphemes (and not -ru as you might expect based on how these verbs inflect).
Respectfully, I find your reasoning doubtful. You claim that "ai and ae only allow the accent on the first mora," and invalidate verbs that violate this premise by claiming that "the -eru and -iru of ichidan verbs are treated as morphemes." You then use this as grounds to "analyze /ai/, /ae/, and /oi/ as diphthongs and other sequences as two monophthongs in hiatus."

I think that if a dependent morpheme has phonetically merged with the verb stem such that the two adjacent vowels are now to be considered a single diphthong rather than a sequence of two monophthongs in hiatus, then the original grounds, that "-eru and -iru of ichidan verbs are treated as morphemes ... for the purpose of accent assignment," would no longer seem to be valid. Namely, I would argue that if two morphemes have merged into a single phonological unit, then the now-phonetically irrelevant morpheme boundary can no longer be used to determine anything about the phonological composition of that unit.

Then there's also the matter of words like マイカー maìkā, 喘ぐ aègu, and ツイン tsuìn, which also run against this theory, as the two vowels are within the same morpheme.

When considering the diversity of Japanese dialects, which vary widely in their pitch patterns and conjugations, it seems a little silly to insist that single morphemes that have remained in the same place for at least 1,200 years are preventing the alteration of an accent in some way.

I think that instead of coming up with far-fetched explanations, we are much safer in explaining exactly what we see: that there are no true diphthongs in Japanese, but that in in any given sequence of two vowels, the accent, if there is one, is more likely to occur on the first than the second.
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: Japanese Pitch Accent

Post by Salmoneus »

clawgrip wrote: 26 Jan 2021 13:06 I think that if a dependent morpheme has phonetically merged with the verb stem such that the two adjacent vowels are now to be considered a single diphthong rather than a sequence of two monophthongs in hiatus, then the original grounds, that "-eru and -iru of ichidan verbs are treated as morphemes ... for the purpose of accent assignment," would no longer seem to be valid. Namely, I would argue that if two morphemes have merged into a single phonological unit, then the now-phonetically irrelevant morpheme boundary can no longer be used to determine anything about the phonological composition of that unit.
I'm not sure I follow your reasoning here. You seem to be arguing in a circle: since this morpheme boundary is phonetically irrelevant, it cannot tell us anything. But of course, Piano is saying that it CAN tell us something, and hence isn't phonetically irrelevant at all. You can't use irrelevance as disproof of relevance - since irrelevance itself is only demonstrable from the lack of apparent relevance.

You seem to be supporting the assumption of irrelevance by arguing that these verb sequences "are to be considered" diphthongs. But Piano is saying that they are not to be considered diphthongs, because of these purported effects on accent placement. You seem to want to define "single diphthong" and "single phonological unit" in some sort of, as it were, purely phonetic way - which you can do if you want, I suppose, but it vitiates your argument: the purely phonetic status of a sequence cannot determine its phonological significance. There's no reason, surely, why a single phonetic unit could not have a phonologically-relevant morpheme boundary within it. Consider, for instance, English [ S ] [edit: uuuurghhh!], where it arises from the junction of /t/ (or /s/) and /j/: it would be very easy for an alternate-English to have a rule that creates, say, umlaut where the following syllable contains /j/, and to have this rule triggered by /tj/ and /sj/ sequences phonetically realised as unitary [ S ] - or, indeed, for there to be some rule that remains sensitive to whether that [ S ] underlying contains an /s/ or a /t/.
When considering the diversity of Japanese dialects, which vary widely in their pitch patterns and conjugations, it seems a little silly to insist that single morphemes that have remained in the same place for at least 1,200 years are preventing the alteration of an accent in some way.
Why would that be silly? Surely that would be quite sensible?
I think that instead of coming up with far-fetched explanations, we are much safer in explaining exactly what we see: that there are no true diphthongs in Japanese, but that in in any given sequence of two vowels, the accent, if there is one, is more likely to occur on the first than the second.
You haven't really explained what would be 'far-fetched' about GrandPiano's analysis, which seems quite direct and simple. Your own theory, meanwhile, fails to explain the purported facts that their theory explains, and hence must be considered inferior. Of course, you also suggest that their facts are wrong, in which case their theory may also be wrong, but that's a different issue!
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Japanese Pitch Accent [split from Q&A]

Post by Creyeditor »

I thought the arguments rests on the counterexamples with monomorphemic dipthongs accented on the unexpected vowels.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
Post Reply