Learn Japanese with JapanesePod101.com

View topic - Learning pitch accent

Learning pitch accent

Japanese, general discussion on the language

RE: Learning pitch accent

Postby Yudan Taiteki » Thu 01.17.2008 9:04 am

Realize that "unaccented" means the pitch rises on the second mora and doesn't fall after that.
-Chris Kern
User avatar
Yudan Taiteki
 
Posts: 5609
Joined: Wed 11.01.2006 11:32 pm
Native language: English

RE: Learning pitch accent

Postby furrykef » Thu 01.17.2008 3:59 pm

I think the downsteps in pitch are much more important than the upsteps, because downsteps are more sudden and it is this point that is perceived as the accent of a word. In particular, in my understanding, if you have a long series of morae before the next downstep, it's more natural to make the rise in pitch gradual instead of suddenly rising on the second mora and holding that pitch. This is why I say I don't really buy into a high/low representation, and instead prefer to just mark for downsteps.

- Kef
Last edited by furrykef on Thu 01.17.2008 4:01 pm, edited 1 time in total.
User avatar
furrykef
 
Posts: 1572
Joined: Thu 01.10.2008 9:20 pm
Native language: Eggo (ワッフル語の方言)
Gender: Male

RE: Learning pitch accent

Postby skrhgh3b » Thu 01.17.2008 4:28 pm

It is true that the pitch rises from low to high on the first mora as long as the first mora isn't the accented mora.

And there's definitely a rise in pitch from the first to the second mora. If you're making a conscious effort to gradually rise in pitch over the course of however many morae there are before the accented mora, then I think it'll end up sounding unnatural. Even the pronunciation workbook I mentioned earlier warns against this. Besides, pitches actually gradually fall over the course of a sentence (as long as it's a statement and not a question, anyway) just as they do in English. This is why accentuation on the sentence level is easy: it's the same as English.

It could be helpful to actually see the pitch contours with your own eyes. If you can't find any linguistics papers on pitch accent, this webpage has a couple of images of pitch contours of the words hashi.

http://sp.cis.iwate-u.ac.jp/sp/lesson/j/doc/accent.html
Last edited by skrhgh3b on Thu 01.17.2008 4:29 pm, edited 1 time in total.
♪夢も見たくない 幸せなんか要らない
恋もしたくない お金なんか要らない
ぼくに必要な眠りを眠らせておくれ♪
skrhgh3b
 
Posts: 517
Joined: Sun 07.24.2005 3:57 am

RE: Learning pitch accent

Postby furrykef » Thu 01.17.2008 7:43 pm

The problem is that I don't think a pitch contour of three morae gives us much to generalize from. What would the pitch contour of an unusually long "flat" phrase look like, for example?
User avatar
furrykef
 
Posts: 1572
Joined: Thu 01.10.2008 9:20 pm
Native language: Eggo (ワッフル語の方言)
Gender: Male

RE: Learning pitch accent

Postby arbalest71 » Thu 01.17.2008 8:58 pm

skrhgh3b wrote:

It's true, the bad news is the pitch accent of Japanese words have to be learned on a word-by-word basis just as the stress accent of English words have to be. The good news is, unlike English, an estimated 80% of the Japanese lexicon is unaccented.


This doesn't really address my point. I might not have been clear enough. I did a little digging and found this page: http://socrates.berkeley.edu/~hasegawa/Accent/accent.html which explains some of my reservations a bit better.

You're right that the adult ear is not very good at hearing this sort of thing, and it's worth noting that this is actually true of native speakers as well, for slightly different reasons. One of the things that page touches on is that Japanese native speakers will often perceive pitch accent in spoken Japanese where it doesn't exist- in fact they will perceive it in cases where the accent is produced by other means.

I'm leery of analogies, but this actually has a pretty good analogue in music. I spent quite a while at one point transcribing Charlie Parker solos, and trying to imitate his phrasing. Parker's playing "swings". The conventional wisdom is that this effect is produced by varying note duration- basically by playing 8th notes in a "triplet" feel. The problem is that if you listen closely to Parker's playing it becomes clear that this is not actually what is going on at higher tempos. The note durations are pretty constant, actually, though they don't _sound_ constant. He is producing the illusion of a triplet feel through other means, largely by varying the loudness of the notes. You could probably come up with a notation that captured this, but it would be very complicated, and very hard to turn into music. There are a fair number of people playing bop out there that never really get this and continue to try to achieve that feel by adhering to some simple rules about note duration that turn out to tell only a small part of the story.

I think that the emphasis on tones in isolation is a large part of why English speakers have such a hard time pronouncing Chinese (where pitch plays a much larger role than in Japanese). The problem is that actual speech just doesn't follow the rules. But people come up with their own idea of how Chinese should be pronounced based on some little marks on paper, and some syllables in isolation, before they've had a chance to hear a lot of it, and it is very hard for them to shake this once they've internalised it. Sure, it's impossible to learn to pronounce Japanese perfectly by listening to it- for all practical intents and purposes it is impossible for an adult English speaker to learn to pronounce Japanese perfectly at all. I think that emphasising pitch accent before people have had a chance to get as much by ear as they can is as likely to hurt their accent as to help it. Of course, there will come a point where it will be important in fixing problem spots, but this can only work once you have a good idea of what pitch accent sounds like in the first place, and even then probably only with good feedback.
arbalest71
 
Posts: 142
Joined: Wed 10.11.2006 8:44 pm

RE: Learning pitch accent

Postby furrykef » Thu 01.17.2008 9:03 pm

arbalest71 wrote:
This doesn't really address my point. I might not have been clear enough. I did a little digging and found this page: http://socrates.berkeley.edu/~hasegawa/Accent/accent.html which explains some of my reservations a bit better.


I was wondering when this article was gonna get brought up. But that article doesn't say that learning the pitch accents of individual words is a bad idea. Despite its title, it seems to be complaining not about marking accents on individual words, but on marking accents without providing an understanding of how pitch accent actually works. It's saying that pitch markings alone won't give you good pitch accent, and I agree. That doesn't mean that you don't need to study where the accents belong.

- Kef
Last edited by furrykef on Thu 01.17.2008 9:03 pm, edited 1 time in total.
User avatar
furrykef
 
Posts: 1572
Joined: Thu 01.10.2008 9:20 pm
Native language: Eggo (ワッフル語の方言)
Gender: Male

RE: Learning pitch accent

Postby arbalest71 » Thu 01.17.2008 9:18 pm

furrykef wrote:
arbalest71 wrote:
This doesn't really address my point. I might not have been clear enough. I did a little digging and found this page: http://socrates.berkeley.edu/~hasegawa/Accent/accent.html which explains some of my reservations a bit better.


I was wondering when this article was gonna get brought up. But that article doesn't say that learning the pitch accents of individual words is a bad idea. Despite its title, it seems to be complaining not about marking accents on individual words, but on marking accents without providing an understanding of how pitch accent actually works. It's saying that pitch markings alone won't give you good pitch accent, and I agree. That doesn't mean that you don't need to study where the accents belong.


Well, the title of the article is "Against Marking Accent Locations in Japanese Textbooks", which seems to imply not studying them directly, at least as a beginner, as does the rest of the article.
Last edited by arbalest71 on Thu 01.17.2008 9:19 pm, edited 1 time in total.
arbalest71
 
Posts: 142
Joined: Wed 10.11.2006 8:44 pm

RE: Learning pitch accent

Postby furrykef » Thu 01.17.2008 9:29 pm

Allow me to clarify my position. I think that the accent position in individual words, while inextricably related to phonology, is a separate issue from phonology. That is, the problem of memorizing pitch accents on words is a lexical one rather than a phonological one, not unlike memorizing kanji or memorizing meanings, because it is, after all, a property of individual words. Thus, I think it should be approached accordingly.

There's no substitute for practicing with real native speakers or sound recordings, of course, but the problem of memorization is still going to be there.

- Kef
User avatar
furrykef
 
Posts: 1572
Joined: Thu 01.10.2008 9:20 pm
Native language: Eggo (ワッフル語の方言)
Gender: Male

RE: Learning pitch accent

Postby Nibble » Thu 01.17.2008 10:08 pm

I think that the emphasis on tones in isolation is a large part of why English speakers have such a hard time pronouncing Chinese (where pitch plays a much larger role than in Japanese). The problem is that actual speech just doesn't follow the rules. But people come up with their own idea of how Chinese should be pronounced based on some little marks on paper, and some syllables in isolation, before they've had a chance to hear a lot of it, and it is very hard for them to shake this once they've internalised it.


Can you provide some sort of example of this? I would argue that the tones most definitely do follow regular rules (with modification for tone of voice, etc), and that many English speakers have a hard time pronouncing Chinese because they simply can't pronounce the tones.

Having personally known several students of Chinese who felt that studying tones in books was meaningless and it was much better to learn it "by ear," I must say that I have strong reservations about your assertion that this is how people should learn.
Last edited by Nibble on Thu 01.17.2008 10:09 pm, edited 1 time in total.
User avatar
Nibble
 
Posts: 269
Joined: Thu 07.28.2005 6:39 pm

RE: Learning pitch accent

Postby arbalest71 » Fri 01.18.2008 1:14 am

Nibble wrote:

Can you provide some sort of example of this? I would argue that the tones most definitely do follow regular rules (with modification for tone of voice, etc), and that many English speakers have a hard time pronouncing Chinese because they simply can't pronounce the tones.


To a native speaker, and to a somewhat lesser degree to someone who has learned to speak well, there are simple, regular rules that actually have predictive power. But this is because they have a large body of implicit knowledge to work from- given that knowledge, the simple rules work (assuming that you consider the sandhi rules simple- they're not that complex in the abstract, but consciously applying them in real time is not all that simple- they pretty much have to be internalised). It's hard to pick out an example because I'm not talking about exceptional situations, but about the normal situation- pretty much every sentence spoken in a normal fashion is an example (and it doesn't help that it has been 10 years since I last studied Chinese).

I think that what you're saying may beg the question a bit, in that it amounts, as far as I can see, to saying that your notational system describes the spoken language accurately, at the resolution that your notational system can capture. But what I am saying is that that resolution isn't adequate, that there's a lot of information that is really crucial, but too fine-grained to be captured by that notation. It's just very hard to see that if you happen to have that information available implicitly.

As an extreme example, say you take someone and show them the iconic graph of the tones that is in every Mandarin textbook, and explain sandhi to them, and then ask them to start learning Chinese from the printed page, without ever letting them hear Chinese spoken. Now, let's also stipulate that you spend some time teaching them how to pronounce things in neutral tone, to try to isolate the tone aspect as much as possible... there's a universe of possibilities that they could come up with, all consistent with the information you've given them. Let's say you then have them learn Chinese this way for a few years. By that time they will have invented their own version of the Chinese tonal system, and internalised it. It might be very difficult to ever break them of the habits they've developed.

Nibble wrote:Having personally known several students of Chinese who felt that studying tones in books was meaningless and it was much better to learn it "by ear," I must say that I have strong reservations about your assertion that this is how people should learn.


Ah, well I didn't make that assertion. Nor did I say something as simple as what you are attributing to those students. One way or another, you are going to have to be explicitly aware of tones if you want to learn to speak Chinese even passably (I'm not sure the same is true of Japanese and pitch accent). I'm not entirely sure that memorizing them based on diacritical marks in a textbook is the only, or even the best way to learn them, but I can't imagine omitting them from textbooks.

Part of the problem is that to describe a learning method as learning "by ear" is pretty ambiguous. If you mean that you're just going to magically pick up the tones by listening to a stream of Chinese go by... well, I think you are going to be in for a rude shock. OTOH, I think that a method in which you were required to try to learn the tones of vocabulary, _used in context_, before having them given to you might be very useful, and it would really help even with the predictable changes, i.e. sandhi.

The example I used above (not letting people hear any actual spoken Chinese) may seem far-fetched, but I think a lot of programs are like that to a lesser degree. They often start by teaching students the tones _in isolation_ at the same time that they are taught a few phrases, like "ni hao" and "wo hen hao" (both featured prominently in the first two lessons of the first text I used). Of course, they haven't touched on sandhi at that point, so you wind up with students pronouncing every syllable in an exaggerated parody of third tone. Many are still doing so a year later. This is the natural outcome of relying too much on diacritical marks. The problem could be avoided if you started them off listening and repeating those phrases, without allowing them the opportunity to make up their own version of Chinese.

The tone marks in vocabulary lists are a pointer, but they are often mistaken for what is being pointed at. In the Chinese program I attended, they were very interested in whether or not you could see the finger, and not that interested in whether or not you could see the moon.

Anyway, part of the problem here is something I also observed in the newbie... kanji thread. To properly qualify what I am saying would make for a much longer post than this, so I am having, as YT did in that thread, to make flatter statements than I would like to.
arbalest71
 
Posts: 142
Joined: Wed 10.11.2006 8:44 pm

RE: Learning pitch accent

Postby Yudan Taiteki » Fri 01.18.2008 10:10 am

You can do both, though. The Chinese classes I'm taking right now have audio files (an interactive DVD, actually) as the main study aid, with a companion textbook that has pinyin and marked tones. In class you are not allowed to have any notes or books open and the class is all in Chinese, so your tones are getting corrected a lot.
-Chris Kern
User avatar
Yudan Taiteki
 
Posts: 5609
Joined: Wed 11.01.2006 11:32 pm
Native language: English

RE: Learning pitch accent

Postby Shirasagi » Fri 01.18.2008 10:33 am

I do not really speak Chinese anymore, aside from some greetings and useful phrases -- remnants of high school Chinese learned over 15 years ago. But I know tones and pronunciation, so I can at least read pinyin. I have the feeling I sounded (and perhaps still sound) like the folks arbalest71 mentioned -- very exaggerated, very strict interpretation of the tones. But I had something of an epiphany while watching Wong Kar-wai's 2046, and being entirely enthralled by Zhang Ziyi sexily whispering. Suddenly it struck me. She's whispering a tonal language.
Last edited by Shirasagi on Fri 01.18.2008 10:33 am, edited 1 time in total.
Josh Reyer
------------
頓ニ纜ヲ斬テ大荒ニ入レ。
長岡桃嶺房成
Shirasagi
 
Posts: 443
Joined: Wed 02.14.2007 10:50 am

RE: Learning pitch accent

Postby AJBryant » Fri 01.18.2008 3:54 pm

Shirasagi wrote:
She's whispering a tonal language.



Wow.

Yeah.




Tony
User avatar
AJBryant
Site Admin
 
Posts: 5313
Joined: Sun 10.09.2005 11:29 am
Location: Indiana
Native language: English
Gender: Male

RE: Learning pitch accent

Postby Yudan Taiteki » Fri 01.18.2008 5:21 pm

When Chinese is sung the tones disappear; I don't know about whispering, though.
-Chris Kern
User avatar
Yudan Taiteki
 
Posts: 5609
Joined: Wed 11.01.2006 11:32 pm
Native language: English

RE: Learning pitch accent

Postby furrykef » Fri 01.18.2008 6:05 pm

I just tried it and tone can still be conveyed somewhat in whispering, though not very well. The second and third tone in particular seem to blur together, and the first and fourth would probably blur in rapid speech. Still better than no tonal information at all.

Cantonese still uses tones in singing, from what I understand. Mandarin usually doesn't, though. It's kind of funny, considering that fitting tones to music would probably be much easier to pull off in Mandarin, since there would be more words in each tone to choose from.

- Kef
Last edited by furrykef on Fri 01.18.2008 6:06 pm, edited 1 time in total.
User avatar
furrykef
 
Posts: 1572
Joined: Thu 01.10.2008 9:20 pm
Native language: Eggo (ワッフル語の方言)
Gender: Male

PreviousNext

Return to Japanese General Discussion

Who is online

Users browsing this forum: No registered users and 9 guests