Qian Gu · Richard VanNess Simmons Common Phonology of the Chinese Dialects Common Phonology of the Chinese Dialects Qian Gu (cid:129) Richard VanNess Simmons Common Phonology of the Chinese Dialects 123 QianGu Richard VanNessSimmons Schoolof LiberalArts AsianLanguages andCultures NanjingUniversity RutgersUniversity Nanjing, China NewBrunswick, NJ, USA ISBN978-981-15-3101-9 ISBN978-981-15-3102-6 (eBook) https://doi.org/10.1007/978-981-15-3102-6 JointlypublishedwithNanjingUniversityPress TheprinteditionisnotforsaleintheMainlandofChina.CustomersfromtheMainlandofChinapleaseordertheprint bookfromNanjingUniversityPress. ©NanjingUniversityPress2020 Thisworkissubjecttocopyright.AllrightsarereservedbythePublishers,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,reuseofillustrations,recitation,broadcasting,reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,computersoftware,orbysimilarordissimilarmethodologynowknownorhereafterdeveloped. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublicationdoesnot imply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevantprotectivelawsand regulationsandthereforefreeforgeneraluse. Thepublishers,theauthors,andtheeditorsaresafetoassumethattheadviceandinformationinthisbookarebelieved tobetrueandaccurateatthedateofpublication.Neitherthepublishersnortheauthorsortheeditorsgiveawarranty, expressorimplied,withrespecttothematerialcontainedhereinorforanyerrorsoromissionsthatmayhavebeen made. The publishers remain neutral with regard to jurisdictional claims in published maps and institutional affiliations. ThisSpringerimprintispublishedbytheregisteredcompanySpringerNatureSingaporePteLtd. Theregisteredcompanyaddressis:152BeachRoad,#21-01/04GatewayEast,Singapore189721,Singapore Table of Contents Introduction 1 Ⅰ. The theory and method behind the Common Phonology 1 Ⅱ. The nature and composition of the Common Phonology of the Chinese Dialects 4 1. The initials of the Common Phonology 5 2. The finals of the Common Phonology 6 3. The tones of the Common Phonology 7 Guidelines for the Data Tables 9 Main Text and Data 11 1. o、uo 11 2. a、ea 22 3. ua 30 4. e、ie、ye 33 5. u、y 37 6. ai、oi、eai、uai、uoi 71 7. ei、uei 86 8. i、ui 102 9. ou 141 10. au、eau、iau 152 11. əu、iəu 172 12. om、op、am、ap 193 13. eam、eap、iem、iep 204 14. əm、əp、im、ip 217 15. on、ot、uon、uot 225 16. (e)an、(e)at、uan、uat 240 17. ien、iet、yan、yet 257 18. ən、ət、un、ut 281 19. in、it、yn、yt 301 20. aŋ、ok、eaŋ、eok 315 21. uaŋ、uok、iaŋ、iok 339 v vi Table of Contents 22. eŋ、ek、ueŋ、uak、uek 356 23. ieŋ、iek、yak、yeŋ 368 24. əŋ、ək、uəŋ、uək 389 25. iŋ、ik、yik 397 26. uŋ、uk、oŋ、ouk 403 27. yŋ、yk、ioŋ、iouk 425 References 435 Alphabetical Index 437 Postface 465 Introduction Ⅰ. The theory and method behind the Common Phonology For a long time scholars have considered the common phonology of Chinese to be embodied in the Qièyùn 切韵 system. The Qièyùn is the culmination of rime book development in the Six Dynasties and basically reflects the essential features of the phonology of Chinese in the seventh century. Its influence in subsequent generations was deep and wide. In the 20th century the Qièyùn became the most important reference for the research in the history of Chinese and in the investigation of the Chinese dialects. But the Qièyùn is ultimately a product of its time. In it we see the results of phonological categorization that was undertaken in the medieval period. The evolution and historical stratum of the Chinese dialects are not discernible in the Qièyùn. For this reason we have undertaken to outline a common phonology of the Chinese dialects with data from living, representative dialects and using a comparative methodology to derive the “lowest common multiple”. The resulting outline of comparatively derived phonological categories is representative of the common phonology of the modern Chinese dialects that can be taken to be a “modern Qièyùn” for the reference and use of scholars. Using a common phonology as a reference in the research on the Chinese dialects allows us to avoid some of the shortcomings of slavishly relying solely on the Qièyùn: (1) The Qièyùn phonology is derived through a process of reconciling “the correct and incorrect features of north and south and the common and contrastive elements of the ancient and modern” (nán-běi shì-fēi, gǔ-jīn tōng-sè 南北是非, 古今通塞) in which “all distinctions are included and mergers are avoided” (cóng fēn bù cóng hé 从分不从合) . It thus reflects a combinatory-synthesis of the regional languages of its day but differs rather greatly from the realities of the modern dialects. Our common system is drawn entirely on the basis of dialect data, from which we have derived the common phonological categories through careful analysis. It thus corresponds closely to the realities of the modern dialects. For example, the syllables (morphemes) duō 多 {te-1}‘many’, cuō 搓{tshe-1}‘rub with hands’, zuǒ 左 {tseQ-1} ‘left’, and duǒ 朵 {tweQ-1}‘a blossom’, tuǒ 妥 {thweQ-1}‘appropriate’, in the Qièyùn belong to the traditional guǒ-shè 果摄 classification, respectively in kāikǒu 开口 and hékǒu 合口 categories.a But in the actual dialects themselves there is no kāikǒu—hékǒu distinction with these syllables. a We represent the traditional Qièyùn forms inside curly brackets ‘{xyz}’ using David Prager Branner’s Neutral Transcription System. For a description and details, see Branner 1999 “A Neutral Transcription System for Teaching Medieval Chinese”, in T’ang Studies 17, available at the following link: http://yintong.info/yintong/public/Branner_MedievalSpelling.pdf Also see the dedicated database for the transcription “Yīntōng: Chinese Phonological Database” at: http://yintong.info/yintong/public/index.php © Nanjing University Press 2020 1 Q. Gu and R. V. Simmons, Common Phonology of the Chinese Dialects, https://doi.org/10.1007/978-981-15-3102-6_1 2 Introduction So we have brought them all together under the final *o in our common phonology. Another example is the case of tā 他 {the-1}‘he’, nà 那 {neH-1}‘that’, ná 拿(拏){na-2}‘hold’ and lā 拉 {lap-1a}‘pull’ in the Qièyùn system belong separately to guǒ-shè kāikǒu, jiǎ-shè 假摄 kāikǒu, and xián-shè 咸摄 kāikǒu in the rù 入 tone. We have assigned them all to a single category, the final *a in our common phonology, on the basis of their actual dialect forms. (2) While phonetic and phonological change is generally fostered and spread through lexical diffusion, rime book categories and the distinctions reflected therein often make it difficult to clearly discern the course and evolution of the changes. For instance, shuāi 衰 {srwi-3c}‘decline’, shuài 帅 {srwiH-3c}‘commander’, and shuǐ 水 {sywiQ-3c}‘water’, shuì 睡 {dzywiH-3b}‘sleep’ all belong to zhǐ-shè 止摄 in the Qièyùn system. But on the basis of the dialect forms, in which the former pair shows some affinities for the xiè-shè 蟹摄 category finals, we respectively assign the two pairs to the finals *uai and *ui in our common phonology. (3) There are words and morphemes in the modern dialects that are not accounted for in the traditional Qièyùn system. Some of them may have existed early on but were not incorporated by the tradition; some arose later in the dialects. An illustration is found in the word(s) for dǎ 打 {treingQ-2a}‘to hit’, which is listed in the Qièyùn with the fǎnqiè 反切 spelling dé-lěng 德 冷切, corresponding to the traditional categorization of gěng-shè 梗摄 kāikǒu in the second division (èrděng 二等), shǎng 上 tone gěng 梗 rime (yùn 韵) with initial duān 端. This Qièyùn categorization corresponds only to the Wú 吴 dialect word for ‘to hit’, for example Sūzhōu 苏 州 [taŋ3] and does not correspond to the northern word dǎ. But the latter, dǎ is widely found in a great number of dialects. For this reason our common phonology includes an entry for the northern word dǎ 打 under the final *a, and provides the common form for it as *ta3. The most authoritative and influential works on the traditional Qièyùn system include Bernhard Karlgren’s Études sur la phonologie chinoise (1915–1924), Dǒng Tónghé’s 董同龢 Hànyǔ yīnyùnxué 汉语音韵学 [Chinese Historical Phonology] (1965), Wáng Lì’s 王力 Hànyǔ yǔyīnshǐ 汉语语音史 [History of Chinese Phonology] (1985), and Lǐ Róng’s 李荣 Qièyùn yīnxì 切韵音系 [The Phonological System of the Qièyùn] (1952). They all based their phonological reconstructions on the traditional Qièyùn categories. Later Yuen Ren Chao departed from the Qièyùn based treatment in his Tōngzì fāng’àn 通字方案 [Proposal for General Chinese] (1983), which outlines phonological categories based on a broad cross-dialectal analysis with the aim of developing a set of Chinese characters that can be used as a pan-dialectal syllabary, with a single character to represent each intersection of initial, final and tone categories. Chao also developed a pan-dialectal romanization for his system. His goal was to work out a writing system that would work across dialects but that would be based on a restricted number of graphs (approximately 2, 000) so as to be relatively easy to learn. Chao also noted at a possible use for the system would I. The theory and method behind the Common Phonolgy 3 a be to design “type lists for linguistic survey based on General, rather than Ancient Chinese.” Though Chao’s General Chinese did not receive much attention from the scholarly community, the phonology that it outlines contains the essential characteristics of a common Chinese phonology which provides a solid foundation for us to compile a common phonology of the Chinese dialects. Jerry Norman in his “Common Dialectal Chinese” (2006) also pointed out that it is possible to avoid using the Qièyùn system and compile a common phonology for the modern Chinese dialects based purely on a comparative methodology. Norman advocated excluding the Mǐn 闽 dialects from the first stages of the comparison, as they include elements and distinctions that are outside of the common system of the other dialects, elements that derive from an older, b higher level phonology. Wáng Lì (1981) also noted that “It is possible to undertake very c valuable dialect research entirely without reference to the Guǎngyùn 广韵” . The present volume proceeds from the foundation established by Chao’s General Chinese, using the comparative method to develop a phonology that includes all the distinctions found in the Chinese dialects, while excluding extraneous distinctions that are not found in dialects, such as those that are indicated in the Qièyùn system. The common phonology represented in these pages is thus primarily based upon dialect data, with initials that preserve the voiced obstruents found in the Wú 吴 dialects, while finals are broadly reflective of the Chinese dialects in general, and including the three-way set of consonant syllable-codas found in the Cantonese (Yuè 粤) dialects. The common phonology that we have derived from the comparison, can be considered to comprise the least common multiple for the categories shared by the dialects. The dialect data that is the base for the comparison is included under each individual syllable or morpheme for reference. This represents the results of the comparison dynamically in relation to the more static results of the comparison. Readers will be able to compare the data for themselves to discern the origin of each of the phonological categories, as well as the comparative relationship between the dialects and their common phonology and discern therein the traces of the evolution of the individual dialects. We selected sixteen dialects for the comparison. The selection includes five northern dialects: Běijīng 北京, Xī’ān 西安, Tàiyuán 太原 (representing the Jìn 晋 dialects), Yángzhōu 扬 州 (representing the Jiāng-Huái 江淮 Mandarin ), and Tàixīng 泰兴 (at the boundary of Jiāng- Huái Mandarin and the Wú dialects); the latter three of the northern set have the rù tone category. The selected Wú dialects include Sūzhōu 苏州, Húzhōu 湖州 (which has a full set of eight tonal categories, helping to clearly delineate the yángshǎng 阳上 category), and Wēnzhōu 温州. The a Yuen Ren Chao [Zhào Yuánrèn 赵元任] 1983 Tōngzì fāng’àn 通字方案 [Proposal for General Chinese] (Běijīng: Shāngwù yìnshūguǎn 商务印书馆), p. 9. b Jerry Norman 2006 “Common Dialectal Chinese”, in David Prager Branner, ed, The Chinese Rime-Tables: Linguistic Philosophy and Historical-Comparative Phonology (Amsterdam: John Benjamins, 2006), pp. 233- 254. c Wáng Lì 王力 1981 Zhōngguó yǔyánxuéshǐ 中国语言学史 [History of Chinese Linguistics] (Tàiyuán 太原: Shānxī rénmín chūbǎnshè 山西人民出版社), p. 203. 4 Introduction Xiāng 湘 dialects are represented by Chángshā 长沙 and Shuāngfēng 双峰. Nánchāng 南昌 and Líchuān 黎川 represent the Gàn 赣 dialects. Méixiàn 梅县 represents the Hakka (Kèjiā 客 家) dialects. Yuè dialects are represented by Guǎngzhōu 广州 and Yángjiāng 阳江. Finally, the Mǐn dialects are represented by Xiàmén 厦门. In light of the fact that the Mǐn dialects reflect an older, higher level phonology the present treatment does not attempt to account for Mǐn dialect sub-varieties. The data in the Hànyǔ fāngyīn zìhuì 汉语方音字汇 was meticulously compiled and collated, and it has a big degree of comparability and reliability. Its entries and data were all carefully selected for their representativeness and serve well as the basis for the common dialect phonology. To the data from that volume, we have added data for Tàixīng (from Gù Qián 顾黔 2016), Húzhōu (from AKAMATSU Yuko 赤松佑子 1991), and Líchuān (from Yán Sēn 颜森 1993). The Tàixīng data has been checked for accuracy by the author, and supplemented where necessary. Where representative forms cannot be found in Yán Sēn (for Líchuān) and Akamatsu (for Húzhōu), they are left out of the data sets. The main entries of the volume, and a large part of the supporting data, follows those of Wáng Fútáng’s 王福堂 Hànyǔ fāngyīn zìhuì 汉语方音字汇 [A Lexicon of Chinese Dialect Character Pronunciation] (1989). Ⅱ. The nature and composition of the Common Phonology of the Chinese Dialects The common phonology outline in the present volume, which outlines the least common multiple for the categories shared by the dialects, contains all actual distinctions and categories that we were able to uncover in the dialect data. The result is a concise synthesis that is reflective of the common phonological categories of the dialects. As such, it is not representative of the dialect of any single time or place. The comparative treatment and the categories outlined take into account the variant origins of forms found in the dialects. For example, the data reveals that forms for tà 踏 {thap-1a}‘step on’ have two sources, one leading to the Wú and Yuè reflexes, and the second leading to forms found in the other dialects. So we determine there are two forms, *dapR and *tʻapR. This reveals layers or differences of time and/or geography that are not reflected in the traditional rime books. This morpheme has only one form in the Qièyùn system:{thap- 1a} reflected in the Guǎngyùn by the fǎnqiè formula tāhé 他合切, which corresponds to *tʻapR. The Jíyùn 集韵, on the other hand, has only dáhé 达合切, which reflects the voiced initial of *dapR. From our comparative perspective, it becomes clear that the two rime books reflect different origins for tà 踏. The Jíyùn version is reflective of the Wú and Yuè dialect form, while the Guǎngyùn version is reflective of that seen in the other dialects. For comparative reference, the tables in this volume include the fǎnqiè formulas, the traditional Qièyùn designations, and Branner’s Neutral Transcription for each entry. The spellings of the initials and finals in the common phonology of this volume represent only the common categories and not any phonetic values. For example, for 波 *poP, *p represents the initial category and *o represents the comparatively derived final. Superscript capital letters