This page contains examples of the citation tones of a young female speaker from Sanhezhen, in S.E. Tiantai county, Zhejiang province. This variety belongs to the Taizhou subgroup of the Wu dialects of East Central China. I have presented them so that you can look at their acoustics and listen to them at the same time. You can also listen to their mean values synthesised as glottal pulses. The tones are interesting primarily because two of them (Yinshang/IIa, Yangshang/IIb) have extrinsic laryngealisation. This is usually realised as a glottal-stop or creaky voice in mid-Rhyme. Chao Yuen Ren noted this in 1928 for the neighbouring Taizhou dialect of Huangyan. In addition to the expected eight tonal shapes from Middle Chinese, there is also an extra convex tone, the origin of which may be morphotonemic (involving an extra underlying tone).
ELICITATION
The informant was recorded by Prof. Zhu Xiaonong in 1997, when she was 22 years old. He used three separate lists of of chinese characters to elicit the data, read out by the informant at different times in the recording session (if the informant was unable to read the character, she said 跳过 "skipped"). The results from two of the three lists are presented here. One list, with about 230 characters, was designed to elicit likely contrasting segments. The characters intended to elicit information about the dialect's Onsets can be viewed by clicking: Onsets (Shengmu 声母); characters for Rhymes are here: Rhymes (Yunmu 韵母). Because this list was designed to elicit segmental differences, the tones of its constituent characters occurred randomly. I have called tones obtained from items in this list Shengyun Citation tones. These data are presented first.
Prof. Zhu compiled a second list to elicit the tones in sequential groups. This list consisted of about 85 characters arranged by their Middle Chinese tones. I have called tones from this list Grouped Citation tones. The informant usually read out the name of the Middle Chinese tone ("yinping" "Ia" etc.) before reading the list of characters with that tone. The acoustics for the grouped citation tones are presented second. For some reason, tokens of the last three tones (IIIb, IVa, IVb) were not recorded.
In the figure legends below, the names of the characters are given in Pinyin, not their Sanhezhen phonolgical representation. Note that there are clearly cases where the informant's reading was not a cognate of the MSC form implied by the Pinyin. The relative values of the tonal F0 are plotted in the figures below the table, where you can compare and listen to the mean Shengyun and Grouped data.
SHENGYUN CITATION TONES
click on individual figures to expand for inspection; click again to contract
|
Yinping/Ia: tone 
|
Yinshang/IIa: tone 
|
Yinqu/IIIa: tone 
|
Yinru/IVa: tone 
|
|
Yangping/Ib: tone 
|
Yangshang/IIb: tone 
|
Yangqu/IIIb: tone 
|
Yangru/IVb: tone 
|
|
tone 
|
GROUPED CITATION TONES
click on individual figures to expand for inspection; click again to contract
|
Yinping/Ia: tone 
|
Yinshang/IIa: tone 
|
Yinqu/IIIa: tone 
|
Yinru/IVa: tone no tokens recorded |
|
Yangping/Ib: tone 
|
Yangshang/IIb: tone 
|
Yangqu/IIIb: tone no tokens recorded |
Yangru/IVb: tone no tokens recorded |
MEAN VALUES
The figures below show the mean values of the Sanhezhen speaker's citation tonal F0 as a function of mean duration (csec.). Their mean pitches can be listened to by clicking on their legends ("Ia" etc.). The drop and recovery in F0 signalling the laryngealisation in tones IIa (black) and IIb (green) can be clearly seen. You can also see, and hear, that tones Ia (red) and Ib (blue) have similar F0 trajectories and pitch to the laryngealised tones IIa and IIb, but without the drop.
You can also see that the F0 is slightly lower for the Shengyun tones than for the Grouped tones. This can also be heard: for example by clicking first on the Shengyun "IIIa" and then in the Grouped "IIIa" you can easily hear the pitch of the Grouped IIIa tone is slightly higher. All Grouped tones in fact have a slightly higher audible pitch. This may have to do with the fact that they were recorded earlier in the session than the Shengyun tones. Or the differences may relate to some unconscious psychological effect from the way they were elicited. In either case they are a nice demonstration of contemporaneous within-speaker variation in tone.
