Division of Humanities

【Abstract】 The question why some Middle Chinese *l- words have fricatives initials ([s],[S]or[¬]) in Northwestern Min has been under hot debate for at least 40 years. After a brief review on the previous archievements, we start our investigation based on the tone behavior of MC *l- and its relationship to other sonorant in in different Min subgroups. The most striking finding is, despite *l- has different initials in Eastern Min and Shaojiang Min, the words in Qu tone share the same two-way tone behavior which is parallel to nasals (such as *m-). The tone behavior of voiceless sonorants among Tai languages informs us that Proto Min has another voiceless series of sonorants besides the normal voiced ones. It is the voiceless lateral that turn out to be fricatives among Northerwestern Min. The more complex tone behavior in Northern Min can also be explain by the fact that voiceless lateral is more likely to last longer than the voicless nasals, which is supported by the typological observation. It is argued that the sound changes from voiceless lateral fricatives to sibilants are perceptual ones.
【Keyword】Northwestern Min, tone, voiceless sonorants, lateral fricative, perceptual sound change

This paper focuses on the phonetic nature and path of evolution of Cantonese short-long stopped tones. Firstly, the phonetic nature of Cantonese stopped tones is explored in light of recent tonal acoustic studies. Secondly, drawing on examples from other Southeast Asian languages, an asymmetry of pitch differences is shown to exist in languages with both short and long stopped tones: the ‘short-high VS long-low’  pattern is much more common than the ‘short-low VS long-high’ one. An evolutionary perspective on this synchronic asymmetry is proposed, in light of (i) Ohala's (1981) listener-based perspective on sound change and hypercorrections, and (ii) facts pointing to an evolutionary path from ‘short VS long’ to ‘high VS low’. Finally, the dating of pitch splitting of Cantonese Yin Ru tone is discussed, as well as the issue to what extent the similarity of the stopped tone between Cantonese and Tai languages is due to substratum, recent influence, or parallel evolution.
[Keywords] long and short vowel, stopped tones, tonal evolution, hyper-correction

Determining the nature of the four Divisions of the Qièyùn is a fundamental problem in the study of the phonetic history of Chinese. Analyses by Pulleyblank and Baxter make it possible to bring out two major changes from Old Chinese to Middle Chinese: a two-way split of the vowel system, and later the lenition of medial -r-. The use of models drawn from Mon-Khmer voice type register languages made it possible to reconstruct the phonetic bases of the four divisions. Div. I groups tense rimes, Div. II groups velarized rimes resulting from medial -r-, Div. III groups lax/breathy rimes. As for Div. IV, it represents rimes with the diphthong ie.

This paper describes Shibei Min, a Northern Min Chinese variety with nine contrasting non-stopped tones that are quite different from other Chinese dialects. Shibei Min has an extremely complex tonal system in which phonation type interacts with pitch contour and length. An auditory description of the nine citation tones is followed by an explanation of the measurement procedure, the acoustic analysis and the results. We demonstrate that duration and phonation type play an important role in distinguishing two low falling tones and two level tones with similar pitch. We argue that so-called ‘voice initials’ are in fact not distinguished by voicing as their phonetic nature differs between word-initial position and word-medial position. The nine tones in Shibei fall into two natural classes and could be represented better in the RL model.
Keywords: Acoustic phonetics, Tonetics, Phonation type, Northern Min, Voiced initials

The reconstruction of the syllable Chi 翅 has long been a  puzzle to students of Chinese historical phonology. The character was used in early Chinese transliterations to transcribe an Indic form with initial *k-, while it was recorded with initial <S>- in Middle Chinese. In this paper, I have proposed a hypothesis that the graph Chi翅 was initially invented for a word for ‘wing’ with initial *k-. When this word was replaced by another word with initial <S>-, the graph remained the same. Therefore, a mismatch between word and character prevents us from recognizing the early form, which is retained as a literary reading of the graph ji 翨. Furthermore, the early form still exists in modern Chinese dialects such as Northern Guangdong Tu Vernacular 粤北土话 and Northern Wu 北部吴语. The modern forms with initial t- are also derived from a further palatalized form of the etymon for ‘wing’ with initial *k-.

This paper provides the first acoustic description of the tonal system of Shibei Min, a Northern Min Chinese variety. Shibei Min has an extremely complex tonal system with nine non-stopped tones.  Theses nine tones fall into two natural classes and could be represented better in the RL model. In previous studies, some tonal contrast was described as voicing contrast of initial consonants.  Based on acoustic results, we argue that the main phonetic difference is not voicing, but syllable-level phonation-type contrast in word-initial position, and consonant closure duration in word-medial position.

Děng 等 ‘rank’, ‘division’ is a central concept in Chinese historical phonology. However, the interpretation of rank and the reconstruction of division remain major points of controversy. The former is part of the broader debate on the place of Chinese rime tables in Chinese linguistic history, and the latter is part of the ongoing enterprise of reconstructing Middle Chinese and Old Chinese phonology.

Whether Middle Chinese (here forth MC) *j- has more than one origin has been a highly controversial question in the field of Chinese historical phonology. Recently, Baxter-Sagart's new Old Chinese reconstruction (Baxter & Sagart 2014, here forth B&S) has brought this question into attention again by proposing two distinct origins for MC *j, one from Old Chinese lateral **l- (alternates with MC *d-) and another from Old Chinese uvular **ɢ- (alternates with MC *k-) respectively.
In this paper, I aim to approach this question based on writing evidences from excavated texts. After examining several minimal pairs between Old Chines **l- and **ɢ- proposed by B&S, I observe that evidences from Old Chinese writing do not support the separation of these two categories. Rather, direct alternations between MC*d- and MC *k- are also found for certain words, usually also involves MC *j-. Putting these evidences together, the writing evidences from excavated texts suggest that there may be only one single origin of MC *j-, and they form a large category together with some words of MC *d-, *k-. However, the issue of reconstruction needs further studies.

This paper has confirmed five falling contours in the Jianyang variety of Northern Min, {62, 52, 54, 43, 32}, which are typologically identified as High-Falling in Rg U, High-Falling in Rg M, Slight-Falling-H, Slight-Falling-L, and Pure Low Tone. The finding of this extreme case of falling contours has enriched the universal tonal inventory by adding two falling tonotypes: High-Falling in Rg U and Slight-Falling-L.

[Abstract] In a survey of 207 tonal languages, none has been reported to have more than two falling tones (Maddieson 1978). In contrary, to have three or four falling tones is quite prevalent for Min dialects of Chinese, and many of them have more falling tones than level tones. The tonal pattern of Min dialects challenges some predictions of maximum number of falling tones. For example, three falling-tone labels are listed in Chao’s (1930) “tone letters”, but only two falling tones are allowed in many phonological theories (e.g. Wang 1967, Yip 1980). 
  This talk focuses on the most “extreme” case of Jianyang Min, which is reported to possess five falling tones (Norman 1969, 1995, Li 1991, Akitani 2008). Yip (1980: 206) argues that the three slight falling tones in previous description should be reinterpreted as underlying level tones effected by downdrift. Therefore, it does not violate her prediction on the maximum number of falling tones.
  The discussion here is divided into three parts. Firstly, acoustic and perceptual data are presented to demonstrate that the slight falling tones are indeed “falling”, at least in Jianyang Min speakers’ perception. Then, further perceptual experiment is conducted to examine how native speakers are capable of distinguishing such large number of falling tones. It is found that the slope of pitch contours, rather than pitch height, is the main perceptual cue for differentiating the falling tones. Finally, it is suggested that future work on tonal typology should take the diversity of perceptual cues into account. From this perspective, the slope-dominant tonal languages, such as Jianyang Min, are fundamentally different from the height-dominant tonal languages, such as Miao dialects with a large number of level tones (Kuang 2013).