On the chronology of the changes to Proto-Tai initial clusters *pl-, *ml-, *kl- in Northern Tai*

In: Cahiers de Linguistique Asie Orientale
View More View Less
  • 1 1Université de Toulouse 2 - Le Mirail 4, rue Georges Courteline 31100 Toulouse FRANCE dhuber@univ-tlse2.fr

Reflexes of Proto-Tai initial clusters *pl-, *ml-, *kl- show a rich array of forms across modem Northern Tai, and these forms are worth studying for their phonotactic patterns. These lenitions to pj-, mj-, kj-, or py-, my-, ky-, or eventual palatalizations to tʃ- in Northern Tai are not extensively discussed in the literature: Li (1977) offers some crucial data but with little analysis, Qin (1997) offers further data with a chronology of the changes. It is precisely Qin's chronology that this paper challenges. The paper offers a chronology that is more plausible based on Qin's data and cross-linguistic evidence as well as Northern Tai phonotactics.

  • I I would like to thank the anonymous reviewers for their precious suggestions and comments. Cahiers de Linguistique - Asie Orientale 39(2): 129-155 (2010) 0 CRLAO-EHESS 131 Bd Saint Michel 75005 Paris 0153-3320/2010/039-129

  • 1 A clarification of the dialect labels is needed here. I have followed Li's (1977) classification of Tai dialects based on a set of phonological criteria. He divides them into Southwestern Tai (SWT), Central Tai (CT) and Northern Tai (NT) (see Map 1). In the ethnolinguistic classification presently official in China, two important Northern Tai languages of China include: (1) Zhuang tt of GuAngxi� including its many vernaculars, such as L6ngzh6u .7.t�+I, Ningming 'T'H)j, Bo'ai �J�, Wiíming Jït PI, and (2) BuyUBouyei 1!J1& of Guizhou. Some NT languages spoken outside China include Yay (Dioi) of northern Vietnam and Saek of Thailand. 2 In this table, Siamese and Tianzhou (T'ien-chow) data come from Li (1977), Lips and Shànglín from Qin (1997). To my knowledge, T'ien-chow data, while regularly cited in Li (1977), are not quoted in other reference books on NT varieties. "T'ien-chow" is here taken to stand for "fflj'N", Tianzhou, the present- day Tianzhouzhen ffiflit1ï in Guangxr Province.

  • 3 Qin (1997:300) specifically reports that the Qing Dynasty annals of the Liozh6u Prefecture (1764) indicate the presence of /pi- kl-/ clusters in the NT dialects in question. He cites Zhuàng (NT) monosyllabic words, which were transliterated into Chinese characters in two syllables, indicating that the Zhuàng words still had initial clusters: bila 'fish' and pole 'vegetable' (cf. Siamese plaa and phak, with regular loss of /1/ (ibid p. 299).

  • 4 Special thanks go to the editors for this. 5 Incidentally, Gedney (Hudak 2008) cites palatal reflexes in two CT varieties: Lei Ping has tfop 'bamboo hat' and tfrr 'salt'; Western Nung has tfip 'petal'; and they both have tfau 'hairknot', tfoon 'to close a door/window' and tfaa 'young rice plant'. These invariably come from PT velar clusters. Li (1977:220) also makes it clear that palatal reflexes occur in CT as well, and he regularly cites data from Tay and Tho (both spoken in northern Vietnam) in his entries (1977:220-232). These data show unequivocally that palatalization to /tJ7 is not a distinctively NT feature. I shall not deal further with these CT changes and their implications in this paper.

  • 6 Li (1977:117) also reconstructs PT *tl-. This must have undergone changes long before palatalization could apply since no variety shows palatal reflexes of *tl-.

  • 7 Li (1977:225) reconstructs *kl- for this item. Nevertheless, comparative evidence points to *kr-. See notes 28, 30-32 and data in (4) below as well as Haudricourt (1960:171). 8 It only appears in the B6'Ai expression tfaUBI hamB2 'midnight' where it means `middle' (Li 1977:231). 9 Li (1977:232) proposes PT *yr- for NT, which is consistent with the NT reflexes.

  • 10 This sign means, here and in all the following tables, that the item is unattested in the language according to Li (1977). zu A question mark means, here and in all the following tables, that Li does not give a reflex and I have not found a reflex for the language in the other main sources I am using. This does not necessarily mean, however, that the reflex is unattested in the language.

  • 12 Li (1977:222) notes that there is -w- in most Central Tai. Gedney (Hudak 2008:122) has Western Nung kai and Ningming kjai, but confirms that other CT have kw-. This change can be due to a constraint *(k)jVj- in CT. " There is a peculiar phonological variation, also noted by Li (1977:223), in the final consonant, but this is not visible in the table: Saek tl-kltmk, Yay tfipltfap, Wumfng klip in Gedney (Hudak 2008:76), while the other data have -t. '4 Gedney (Hudak 2008:122) cites this item with close /o/, i. e. tfop.

  • 'S Li (1977:227) also cites Lingyun �i: tf8l). This variety is spoken in Western Guangxijust like Tianzhou and Xilin. Li only cites a handful of relevant examples from this language. I have reproduced all of them in this paper. 16 Li (1977:227) cites data for Shan and White Thai (both SWT) as well as Nung (CT), and claims that no reflex is attested in NT at all. 1 have included the item both for the sake of completeness and because the SWT and CT data do not show palatalization. 17 Li (1977:228) also cites Lingyun tfap 'to chase'. '8 Li (1977:228) also cites Lingyun tfoogB 1. 19 Etymology is uncertain (Li 1977:228). 2° This form in B6'Ai, and indeed in NT in general, means 'just right'.

  • 21 Cognates of Siamese klmn 'to swallow' are unattested in the varieties Li cites. Gedney (Hudak 2008:159) provides Saek tlrucrmltrcuum and Wuming klwag. zz The correspondences for 'salt', Al: Siamese kll4l; L6ngzhou k*u; Bo'ai tfuu; Yay kua; Tianzhou kua; Xtlin kua. In Yay, Tianzhou and Xilin the cluster had simplified to k- before palatalization could occur (Yay and Tianzhou normally have palatalization), due to the following labial vowel. The lack of a palatal reflex in these varieties is therefore not an exception to palatalization. In B6'ii, the cluster had palatalized as expected for velar clusters in this variety. Central Tai Lei Ping has tfrr (Hudak 2008:77). z3 The development of the labial clusters is unknown in these varieties: did they preserve the clusters or did the clusters develop into something else?

  • 24 Li does not specify whether or not this applies in Xilin, Tho and Lingyun. 25 Further research must establish whether or not the lexical bias, 9 items with a labial cluster versus 30 with a velar cluster, has any bearing on the palatalization. Z6 Some dialects are grouped together in the table because they have identical reflexes for all these items. Numbers in the table indicate Qin's notation of tones, which are irrelevant for the discussion.

  • 27 The data here show the regular developments. Li (1977:93) notes that words in *ml- have a rather complicated history across the dialects and its reconstruction is rather tentative. In the items in (4), Siamese deleted the /m/ in (naam )laai 'saliva', but split the cluster in !!!fkig-(� 'papaya'. Li (1977:173-175) does not mention eventual palatal !J1!- reflexes of PT *ml- clusters, even though WOming and Xilfn both have initial palatal nasal /p/. Bo'iti has /n/- for PT *ml-. Furthermore, recall that some words like 'head', and 'egg' derive from Cr- clusters as confirmed by Haudricourt (1960:171). See (2b) for items in Li having the same origin. See notes 30-32 below. 28 An earlier table in Qin's article has klai3 instead of klai I. It is immaterial for the present discussion. zs Gedney (Hudak 2008:122) gives the Wuming cognate. It shows that the cluster comes from a PT Cr- rather than Cl-cluster. This item was not identified as such in Li. See footnotes 7, 27 and data in (2b). 3° Gedney (Hudak 2008:80) gives the Wuming cognate. Bo'iti and Yay (as well as CT varieties Lei Ping and Western Nung) have palatal reflexes here: tfau See footnote 27. 31 Gedney (Hudak 2008:78) cites kron for Siamese and ffan for Western Nung. Li (1977:226) cites, beside the Siamese form kron, Ltngyun and Xilin palatal form tfon. He notes, however, that all Tai dialects except Siamese kron seem to go back to *PT Cl- (ibid.). Siamese is apparently deviant here.

  • 32 HdngxiAn has plook7. 33 Lùzhài has PY33J{7. 3a Another table in Qin's article has mla: il instead of mla: i2. The latter seems correct. It is immaterial for the present discussion. 'S The Shinglin word has ap-initial reflex. It is not known whether this is part of a regular change in this dialect or is specific to this item only. 36 I suspect this sound is a high central unrounded lal or a high back unrounded /ui/.

  • 37 Recall that in northern, as opposed to southern, NT all laryngeal contrasts had been lost in initial clusters.

  • 3$ PT *ml- behaves like PT *pl- in NT, but for the palatalization. In fact, reflexes of *ml- are unknown in palatalizing NT varieties. Data are surely needed here. 39 Qin (1997:302) cites 3 words from this dialect: ku 'salt', kai 'far', kaatj 'middle', all with initial /k/ but none with /p/. I have included this dialect because the context in Qin implies that in this dialect the labial and velar reflexes behave alike, which is a welcome addition to the typology here. This dialect has a northern and southern variant in Edmondson (1994:185). Apparently the northern variant figures here.

  • CLEMENTS G. N. and HUME E. V. (1995). The internal organization of speech sounds. In : GOLDSMITH John A. (ed.). The Handbook of Phonological Theory. Oxford : Blackwell Publishing. Pp. 245-306.

  • EDMONDSON Jerold A. (1994). Change and variation in Zhuang. In : ADAMS Karen L. and HUDAK Thomas John (eds.). Papers from the Second Annual Meeting of the Southeast Asian Linguistics Society. Arizona State University, Program for Southeast Asian Studies. Pp. 147-185.

  • HARRIS John and Geoff LINDSEY (1995). The elements of phonological representation. In : DURAND Jacques and KATAMBA Francis (eds.). Frontiers of Phonology. Atoms, Structure, Derivations. Harlow : Longman. Pp. 34-79.

  • HAUDRICOURT André-G (1960). Note sur les dialectes de la région de Moncay. Bulletin de l'Ecole Française d'Extrême-Orient (Paris), vol. 50, no. 1, pp. 161-177.

  • HUBER Daniel (2006). On the representation of coronals and velars across theories. In : GYURIS Beáta (ed.). Electronic Proceedings of the First Central European Student Conference in Linguistics, 24-26 May 2006, Budapest. http://www.nytud.hu/cescl /proceedings.html.

  • HUBER Daniel (2007). Velars and Processes: Their Treatment in Phonological Theory. Unpublished PhD Dissertation. Budapest : ELTE.

  • HUBER Daniel (2010). On khw-f alternations in Bangkok Thai and other Tai languages. Mon-Khmer Studies Journal (Mahidol University at Salaya, Thailand), vol. 39, pp. 155-165.

  • HUDAK Thomas John (2008). William J. Gedney's Comparative Tai Source Book. Honolulu : University of Hawai'i Press. (Oceanic Linguistics Special Publication ; 34).

  • LI Fang-Kuei (1977). A Handbook of Comparative Tai. Hawaii : The University Press of Hawaii. (Oceanic Linguistics Special Publication; 15).

  • QIN Xiaohang (1997). Evolution of the initial consonant clusters "pl", "kl", "ml" in the Hongshuihe vernacular of Zhuang. Mon-Khmer Studies Journal (Mahidol University at Salaya, Thailand), vol. 27, pp. 299–302.

  • ZHĀNG Jūnrú et al. (1999). Zhuàngyŭ Fāngyán Yánjiū ??? ??? [A Study of Zhuàng Dialects]. Chéngdū ??? : Sìchuān Mínzú Chūbănshè ???.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 41 13 0
Full Text Views 57 0 0
PDF Downloads 2 0 0