@roan – I have often found it odd that teachers would have a bias against teaching pitch accent; I always assumed they took such a stance because of their lack of understanding of the system.
“…as students may be inclined to believe that they mishear when Japanese people do not follow this pattern”??? What is that supposed to mean? You won’t teach something in case people don’t get it perfectly? Nonsense.
Now, I’ve never heard you, so your teacher may have thought that prioritizing in another area would be more beneficial, but it seems very odd that a pronunciation teacher would not want to teach pitch.
As for incorrect devoicing interfering with pitch, I don’t see what he means. The only way I can imagine any influence is if you’re dealing with an English speaker who imparts English stress to Japanese words and ends up putting stress on a vowel that should be devoiced. But then again, if you did bother to teach pitch, the speaker wouldn’t be using English stress…
First of all, I haven’t said anything about my accent. I haven’t particularly got a problem with devoiced vowels, or pitch for that matter. I decided to take this course because I know, as an adult learner, that I sound foreign. In any case, it really is fascinating.
Also, the teacher did not say that pitch is not important, but that it is not the most fundamental aspect of the Japanese accent.
Using the examples he gave, こころ has a LHL pitch pattern if the word is in isolation. However, put it in a sentence, e.g. 心が広い, the pitch drops, and never goes above the first こ, right to the end of the sentence. If the second こ has a higher pitch then it sounds unnatural. A learner, seeing that こころ is LHL may think that they are mishearing a native speaker and hypercorrect their pronunciation to match the dictionary’s pitch marks.
In order to maintain a natural intonation, vowels are devoiced in standard Japanese. Early on you learn that the final [u] in です is not pronounced, but rather than it being a terminal [s] it is actually an [s]+devoiced [u]. This is because the mouth should still form an [u], but leave it unvoiced. The same goes for the example Steve gave, kinosh’ta. The [i] is devoiced, not omitted completely.
The phenomenon gets more complicated with vowels that are only devoiced to maintain the correct intonation.
(teacher’s example)
雨が降る The [u] in ふ of 降る is voiced here
雨が降っている The [u] in ふ is not voiced.
If you voice the ふ in ふっている you will not have a natural Japanese accent.
/i/ and /u/ are devoiced when they occur between 2 voiceless consonants. It’s not really connected to intonation per se. In “furu”, the f is voiceless but the r is not. In “futteiru”, both the f and t are voiceless. They also devoice at the end of words if they are after a voiceless consonant, like in desu. It’s a linguistic phenomenon called assimilation. Sounds take on the characteristics on the preceding or following sound.
Now the exact amount of devoicing will depend on the speaker, the situation, and the surrounding consonants. I think that the i in shiku is more devoiced than the i in kiku, but maybe that’s just me. I also think there’s less devoicing in Kansai than in Kanto, but again maybe that’s just me.
Your example of 心が広い, if correct, really surprizes me. As far as I know, a noun’s pitch doesn’t change unless it’s in a compound, and I would have expected LHLLLHL for that phrase. I’ll need to double check that.
The pitch accent system is very complex. I think it’s useful to understand what it is and how it works, but I don’t think you can “teach” someone how to speak with standard pitch.
Kanto has a lot of devoicing, and Kasai does resist it more, apparently.
The voicing and devoicing of vowels is related to the surround consonants, but as you say Botrun, it is driven by a number of factors.
Alex, in order to see the true pitch you have to look at the sentence as a whole. It still maintains a LHL pattern, but the fundamental frequency falls. Therefore, the H can have a lower frequency than the first L, but is still higher than a similarly placed L. When paired with a 述語, the tendency is to fall until the next clause, which is marked by a sudden increase in pitch.
Where words are spoken in isolation, the base frequency remains more or less constant.
All languages that have pitch have some kind of descending plateaus, if you will, whereby H is increasingly lower until some kind of reset occurs. If that’s what you meant, then yes, the H in 広い is lower than the one in 心, but pitch doesn’t change on 心。
I see, you were refering to pitch frequency of H syllables lowering within a phrase; I had first understood that pitch accent changes within a noun, which it doesn’t.