Sylablle stress: Neglected critical issue?

[Laska]

First let me thank Roddy and everybody for such an amazing resource.

I posted this in the resources forum because I'm looking for resources that address my issue, but please move it to speaking if it's more appropriate there. Also, my apologies in advance if this is a duplicated discussion.

Recently I've become interested in relative syllable stress (length/intensity--whatever you want to call it) in Mandarin two-character words.

For example, it seems to me that in words composed of two 3rd tones (理解 li3jie3, 老虎 lao3hu3, 勉强 mian3qiang3), the first syallble is relatively stressed; whereas, in, for example, words composed of a 3rd tone + 4th tone (half 3rd + 4th), the second syllable is relatively stressed (肯定 ken3ding4, 榜样 bang3yang4, 满意 man3yi4). Most words, however, seem to have more or less equal stress on both sylablles, except for words in which the second syllable is qingsheng; these have stress on the first sylablle.

Frankly none of the many learning materials I use adequately address this issue. It is important because if I pronounce, for example, lijie with what seem to me correct tones but the accent on the second syllable, I won't be understood. I believe that a lot of the frustration that Mandarin learners encounter with not being understood has to do with tones, tone sandhi and syllable stress (intensity, length, whatever). These three issues, especially the last one, receive, however, inadequate treatment in Mandarin learning materials, in my opinion.

Having done some cursory web searches, I was surprised to see that whether Mandarin is syllable-timed or stress-timed is somewhat controversial. Is this correct?

Are there any linguists out there who could shed some light on the issue of regular stress patterns, such as the ones I mentioned above? I would be very grateful. Also, I think it is of great importance to Mandarin learners.

[carlo]

I think you're right in saying that there is such a thing as stress in Mandarin, and that's usually not taught very well if at all. However you should be careful about what you call 'stress'. 3+4 character combinations seem to have a regular stress on the second syllable, but that's probably only because in your native language (I'm guessing) stressed syllables are pronounced with a falling contour. Stress in Mandarin, on the other hand, is related to syllable duration and amplitude (and possibly pitch register), not the shape of the pitch contours.

[Laska]

Thanks Carlo. I had considered the possibility that my native language (English) was refracting my perceptions of "stress" in Mandarin. Do stressed syllables have a falling contour in English? Yes, I suppose they do... That's very interesting. Come to think of it, I had heard somewhere that Danish is the only Indo-European language with rising stress contours. Is that true? Can someone confirm? Very interesting!

How about my perception that 3+3 character combinations have the stress on the first syllable? In my ear, 2+3 and 3+3 characters have exactly the same tone contours and stress, as one would predict from the tone sandhi. So de2yu3 (德语) sounds similar to li3jie3 (理解). To me they sound like "DEyu," "LIjie," with the first syllable louder than the second.

Also, the second syllable of 3+2 characters sounds stressed, for example, lv3xing2 (旅行), ke3neng2 (可能).

Am I equating extremes of pitch with stress? Or is there really a loudness or length effect here... For example, perhaps it takes longer to pronounce second tone in "neng" than the half-third tone in "ke."

[carlo]

Look at it this way: higher pitch means that your vocal cords vibrate faster. So in a two-syllable word 2+3, you may have a tendency to perceive the point of highest pitch in the contour as 'stressed'. However, this is only an approximation related to your perception of stress as an English speaker.

Try to get a native speaker of Chinese to say something like 西安加工波音七三七飞机机身. These are all first tones, but if the speaker gives prominence to any given syllable (eg, it's 737s and not 747s, so he will stress 三) you'll notice that this syllable is at a higher frequency than the following ones. This is one example of 重音 in Mandarin. A famous linguist (Chao Yuanren) described this as an expansion (/contraction) of the pitch range, as if the 4 tones were regular shapes drawn on a rubber band that can be 'stretched' in a number of ways..

[mrtoga]

I understood that when two third-tone syllables together form one word (like li3 jie3) the first syllable changes to second-tone. Is this not causing the "stress" that you refer to? Put another way, third-tone is the only tone that does not either start or finish high (apart from neutral tone), and is a little longer than the others, so when used with other tones the other tones seem to adopt the stress.

Does this make sense to anyone?

[Laska]

Mrtoga--Yes! That does make sense to me. Especially because third tones followed by other tones become half-third tones (slight falling tones). I can sum up what I'm hearing in two-syllable words as follows:

• When a half-third tone is followed by a falling or rising tone, the second syllable sounds stressed
  
• In 2+3 combinations (which includes, of course, 3+3), the first syllable sounds stressed.
  
• Syllables with falling tones (4th tone) always sound relatively stressed.
  

I think Carlo is right in pointing out that I am perceiving this effect through the prism of my native language, English.

Carlo--thanks for sharing the rubber band analogy and the sentence with all first tones. I will try that out next time I meet with my tutor. Fascinating!

I have been doing searches for Chao Yuanren. I see on Wikipedia that his grammar of Chinese has been republished by the Commercial Press in China. I also have started exploring the 北大中文论坛, which you recommended in a previous thread on tones. Do you have a link to the post you mentioned by Shao on syllable stress?

For someone who is in China (Shanghai), do you have any recommendations on getting access to more scholarly or comprehensive information on Chinese phonology, prosody and tones? Are there any good online resources? Books published on the mainland in Chinese or English? Particularly good books published in English that would be worth ordering on Amazon or as an e-book? Your suggestions are really valued!

[carlo]

It's Carlo, without an s. Try googling for 橡皮带, 语调。I think the original quote appears in 汉语口语语法, but I'm not sure. I know of no book in English, but I recommended a Chinese book in a previous thread, and you can learn a lot from reading the 北大 forum (look for the sticky threads in the 语音 folder).

[Laska]

Thanks! And sorry about your name.

[roddy]

I recommended a Chinese book in a previous thread

This one?. If so, I second the recommendation - lots of 'damnit, that's right. Why didn't my textbooks tell me this' moments.

[Jive Turkey]

Having done some cursory web searches, I was surprised to see that whether Mandarin is syllable-timed or stress-timed is somewhat controversial. Is this correct?

Actually, the whole notion of syllable or stressed timed languages is now controversial. Plenty of teachers are still coming out of teacher training programs taking it as given that English is a stressed timed language. However, the idea of stress/syllable timing has not held up when langauge has been analyzed with speech lab equipment. If you look in TESOL Quarterly or the other big applied linguistics journals, most of the articles on timing published in the past decade have questioned rather than supported the division of languages into stress/syllable timed. Those who still do work on timing now talk about it in relative terms, saying perhaps that English is more stress timed than Putonghua, but Putonghua is more stress timed than Cantonese. And of course as has been pointed out by others in this thread, the concept of "stress" varies from language to language.

[anonymoose]

I think low pitched sounds tend to sound quieter, so for any word that consists of a combination of a 3rd tone syllable plus a syllable of any of the other three tones, there will be an apparent stress on the syllable which isn't the 3rd tone.

[Laska]

Thanks for you replies. Very interesting note on stress-time versus syllable-timed. And Anonymoose, what you are saying makes sense to me.

I got my hands on the book that Roddy and Carlo suggested, 汉语语言教程 published by 北京语言大学出版社. Its discussions of the 3rd tone and of tonal coarticulation were really enlightening!

I also picked up Zhao Yunren's grammar. I'm really glad to be turned on to this.

In addition, I managed to find a copy of Tone Sandhi: Patterns across Chinese Dialects by Matthew Y. Chen, which I had seen online but never thought I would find in a bookstore in China. It's been published locally by Foreign Language Teaching and Research Press, 7-5600-2387-8/G 1107. So far I've only read its discussion of tone sandhi and tonal coarticulation in Beijing Mandarin. It provided a great complement to 汉语语言教程 by discussing rules and systematicity in coarticulation patterns and looking at actual frequency data.

One of the biggest surprises for me was to see how low the 3rd tone goes. According to the data Chen is using, in two-syllable words, it drops way below any other tone and way below its “prototypical” pronunciation in an isolated syllable. Also, in T2T3 combinations, the T2 is actually pushed higher: it becomes MH+. The T3 starts higher, then goes all the way to the bottom of the frequency range. So together the contour looks like MH+M+L. That is to say, the 2nd tone in T2T3 combinations rises to a perceptibly higher frequency than other 2nd tones, which may partly account for the “stress” I am perceiving.

汉语语言教程 says your vocal chords should be in “the comparatively relaxed to most relaxed state," when you pronounce the 3rd tone. Looking at how low the frequency of the 3rd tone dips, I feel as though it is actually pronounced at the bottom my normal voice range, so that it almost rattles a little bit. To put it another way, a trained singer with a large vocal range would have to focus on not "singing” the tone to make sound right.

Does this description sound right?

[cjbaker]

I have noticed that many English speakers acquiring Mandarin appear to initially interpret the tones as English stress. Indeed, the pitch contours of many Mandarin words do appear in some contexts in English. Here are a couple of anecdotal observations of myself and friends learning Mandarin that make me believe English learners often employ their stress framework to approximate Mandarin tones:

Words like 六 liù can be approximated by two English syllables, 'lee'+'oh', the first stressed and the second unstressed. However, English speakers will reverse the pitches of stressed/unstressed syllables in many questions like "你吃了六个饺子了吗?", changing 六 to liú.

In words like 美国 mei3guo2, the first syllable mei3 can be acquired as a stressed syllable in a "surprised"-type English sentence, like "What!! you're going to A*mer*ica?". However, this will end up being realized as mei4guo in many situations if English prosody takes over, like "我不是法国人，我是美国人!".

As for the issue of syllable stress in L1 Mandarin (that of native speakers), I believe it is limited to words with 轻声 (neutral tone), in which it is fixed on the first syllable - correct me if I'm wrong. Perception of other stresses in Mandarin by English-speaking learners may actually be consciousness of the mechanism you're using to produce an L2 approximation of contour tones. This would be analogous to Mandarin learners of English realizing that they hear the word "dog" as "do-ge" and "cat" as "ka-te", because this is how most learners approximate the L2 phonology using their L1.

I have also noticed that non-English, non-Chinese L1 speakers (some French friends I know - French has no syllable stress or tone) tend to have rather different tone errors.