Author: LIAO Hanbo 3 Dec. 2022 at the University of Hong Kong; cloudy to overcast
A couple of days ago, I used the word拎[lɪng55] (< *ʰniŋA) (to grab; to carry; to fetch; to take) in a Cantonese short message. It suddenly occurred to me why the Chinese character 令 (lìng/ling6) with the initial consonant /l-/ was used for the phonetic component of this Cantonese word, instead of the one starting with the initial /n-/. The reason I thought of this is that in Baihua (the local name of Cantonese in Guangxi) that I heard when I was a child, the word was ning1, not ling1. To make sure that my question was not an idle one, I contacted my friend Xu Xiaoming, whose mother tongue (one of) is the Cantonese variety in the Daxin County seat of Guangxi, and asked her to confirm that she pronounced the word 拎 with n- rather than l-.
Inherited from Guangzhou Cantonese, Hong Kong Cantonese also merges the original *n- initial into /l-/ in all environment. But for those Cantonese language experts who have the ideal of "correct pronunciation", when they compile their Cantonese dictionaries, they likely use the /n-/ initial to standardize those with the original *n- (which has been merged into /l-/ in everyone’s daily life). For example, the word 寧 ‘peaceful’ would be pronounced as ning4 in Jyutping (a romanisation system for Cantonese), even though they mostly pronounced it as ling4. However, for this 拎 ‘to take/carry’, not only is it phonetic component in the Chinese character, but most Cantonese dictionaries in Guangzhou or Hong Kong also designate it as ling1, with the exception of 周無忌 and 饒秉才’s廣州話標準音字彙 (The Standard Gloss of Guangzhou Cantonese) (p. 61), which treats ning1 as an alternative pronunciation, and its proper pronunciation is still ling1
(click the following icon to search 拎 for its pronunciation in Standard Catnonese)
Taking into account the fact that the Cantonese character 拎, which uses令 as its phonetic component, appeared quite early in Cantonese literature, it can be tentatively concluded that even in the old generation’s accent, where n- and l- are largely contrastive, the word拎was probably one of the first group which merged their original *n- to /l-/. This may be why later Guangzhou dialect experts followed this reality when they standardized their dictionaries. If they had been able to refer to other Cantonese dialects other than the Guangzhou varieties, especially the Guangxi vernacular Cantonese varieties, where n- and l- are still clearly contrastive, they would probably have established ning1 as the “correct pronunciation”.
Talking about this Cantonese 拎 reminds me of some anecdotes from my childhood. I grew up in the county seat of Debao County (德保縣), Guangxi. When I was a child, the only dominant language spoken in the Debao county seat was Yang Zhuang (Qyaang/Káang Thó), and Chinese (Guiliu Mandarin and Standard Mandarin) was only spoken by intellectuals, cadres and teachers during administrative meetings and teaching activities. Baihua (literally ‘vernacular language’, the name of Cantonese in Guangxi) is not a common language in any area of Debao County, but some of the old neighbourhoods in the county seat have Baihua as their family language, as the first two or three generations of them moved in from outside. I have witnessed countless times when a good friend of mine was talking to his grandfather, one in Yang Zhuang and the other in Baihua, and they spoke to each other without any difficulty. Actually, as the former language of commerce, Cantonese has a certain influence in the region. People of a certain age are generally able to speak some Baihua, especially those who are experienced in business and those who often go to the markets to buy and sell. Going back to the topic of 拎 ‘carry’, when I was a child, we used to make jokes about this Cantonese word in particular. Instead of being pronounced as [nɪŋ55] as in Nanning, local Debao people would definitely pronounce it as [nəŋ55], which has a lower and more centralised vowel.
Note: That is, although they share a common Lingnan trait in which the original close front vowel [i] is lowered before the velar endings (-k and -ŋ), the lowering in ordinary Cantonese is smaller, only to the near-close front vowel [ɪ], but in Yang Zhuang and some Cantonese and Pinghua varieties of Guangxi, it is lowered to the central vowel [ə]. This topic is discussed in my PhD thesis titled “The formation of Lingnan linguistic traits : Typological structure and diachronic issues”, and I will not repeat it here.
Unfortunately, this pronunciation of [nəŋ55] in local Debao Baihua accent is the same as the word nhèng (鸋) [nəŋ55] ‘penis’ in Debao Yang Zhuang. So, as children, we would say 拎過來nhèng gò looy [nəŋ55 ko33 lo:j31] (bring it over = penis, come here) with great gusto, causing laughter.
However, the most common etymon for “penis” for the Tai languages (including Zhuang) is a cognate developed from Proto-Tai *ɣwajA, such as way (暐) [ʋɐj31] in Yang Zhuang, and ควย [kʰuəj33] in Thai. In addition to this etymon, the word nhèng ‘penis’ is more frequently used in the Yang Zhuang, and has even developed into an impolite negation (the use of genitalia as a negative maker as an areal trait of Lingnan is also mentioned in my PhD thesis). This word can in fact be traced back to proto-Tai *ʰnɯŋB. Nevertheless, as far as I know, only the Yang Zhuang varieties of Debao, Jingxi and Napo and some small dialects of the nearby Zuojiang Zhuang retain the meaning of ‘penis’ for this etymon, while the other Tai dialects that retain this etymon, such as all the Southwestern Tai (SWT) languages and many of the languages of Central Tai (CT), such as Zuozhou and Longzhou in Southern Guangxi and Tay in Vietnam, the word developed from Proto-Tai*ʰnɯŋB are used only as the numeral ‘one’, such as หนึ่ง [nɨŋ21] in Thai. This semantic change is easy to understand, and it is not surprising that the meaning ‘one’ was developed for ‘penis’ as all men or male animals are with a single penis.
As to why proto-Tai had two etymological roots, *ɣwajA and *ʰnɯŋB, to express the meaning of ‘penis’, my research suggests that there was a semantic distinction. The former was the broader sense of ‘penis’, while the latter originally referred to the ‘penis’ of underage children or young animals, and later, although the latter also developed into the broader sense of penis in Yang Zhuang, there is an emotional difference between the use of way and nhèng. While way is considered to be more vulgar or impolite in tone, nhèng has a lovely sense of speech, which may also be the emotional basis for the latter’s development of the meaning of ‘one’ in other dialects.
Those who know about the Kra-Tai languages will probably know that most of the languages of this family have shifted their inherent numerals to Chinese loanwords, with the exception of languages from the Kra and Hlai branches, which retain some numerals from proto-Austro-Tai. In the modern Kam-Tai languages, cardinal numbers from ‘two’ to ‘ten’ are all loans from ancient Chinese and are therefore quite uniform, but the number “one” has different roots. Apart from the etymon *ʰnɯŋB, which is commonly used to express ‘one’ in SWT and some CT languages, the rest of the CT dialects (including Yang Zhuang), and almost all of the Northern Tai (NT, or Yai/Qywáy) dialects, use a different etymon *ˀdiəwA to express the numeral ‘one’, as in the case of ndew /ˀde:wA1G/ in Yang Zhuang and ndeu [ˀde:w24] in Standard Zhuang (of NT). In SWT and some CT dialects where the etymon from *ʰnɯŋB is used to express the number ‘one’, the etymon *ˀdiəwA has developed to mean ‘single; alone’, as in Thai เดียว [diə̯w33]. For example, in the case that A asks “how many people have come” and B replies ‘only one’, B can simply say คนเดียว [kʰon33 diə̯w33] (literally ‘person single’), which conveys the mood of ‘only one person (has come)’. At this time, Tai dialects that developed *ˀdiəwA to be as ‘one’, as in the case of Yang Zhuang will use the Ancient Chinese loanword dook /to:kDL2/ (< 獨 ‘alone’) to fill in the position on which *ˀdiəwA takes up when expressing the semantic meaning of ‘only one person’ as in Thai คนเดียว, such as Yang Zhuang geon dook /kɔnA2 to:kDL2/ ‘(only) one person’.
One of the more unusual dialects is the Zuozhou dialect (named after its origin in the town of Zuozhou 左州in Chongzuo 崇左, Guangxi) of Southern Zhuang (or CT), particularly those Zuozhou varieties spoken in the area around Jingxi and Napo, which I have studied a lot in the past. It uses ndew to express the cardinal numeral ‘one’ to count, and dhook (alone) to express the tone of ‘only one’, both of which are clearly influenced by the surrounding dominant language, Yang Zhuang.
Note: The common denominator between Zuozhou Zhuang, which is an CT language, and Thai and Lao, which are of SWT, is that the original voiced stops in Proto-Tai have developed into voiceless aspirated stops, but most of the CT languages and some of the SWT languages, such as Northern Thai, Shan and Tai Lue, developed them into voiceless unaspirated stops.
However, nhèng as ‘one’ is used when expressing tu kày nhèng ‘a chicken’ or ghen nhèng ‘a person’. The use of the etymon *ʰnɯŋB to express the numeral ‘one’, as in Zuozhou Zhuang, Longzhou Zhuang and the Tay dialects of CT, is a sign of the fact that the separation of the meanings of ‘penis’ and ‘one’ in *ʰnɯŋB took place in Southern Tai (the direct ancestor of both CT and SWT), before the divergence of SWT and CT. This semantic diversity of this item is preserved in the CT dialects spoken in today’s Sino-Vietnamese borderlands, while the SWT dialects in the hinterland of the Indo-China Peninsula are all descendants of the Southern Tai groups that developed *ʰnɯŋB with the meaning of ‘one’.
Kommentare