从清浊别义看群、匣、喻三的上古来源 = A study on the origins of middle Chinese initials of ɣ-, g-, j- in archaic Chinese : a morphological perspective

• Main Author: 郑琇瑜 Tay, Xiu Yu
• Other Authors: Phua Chiew Pheng
• Format: Final Year Project
• Language: Chinese
• Published: 2014
• Subjects: DRNTU::Humanities::Linguistics::Anthropological linguistics
• Online Access: http://hdl.handle.net/10356/55580
• Institution: Nanyang Technological University

群、匣、喻三的上古来源问题至今已被众多学者研究许久。目前只剩两种说法比较受学术界认可，其他的说法都已经被排除掉。 第一种说法是群、匣、喻三的上古来源是舌根塞音，*g-和*gw-。支持这个说法的学者包括李方桂、龚煌城和丘素郡。他们通过演变规律、汉藏语比较以及《广韵》中声母和等韵的搭配关系支持他们的立场。第二种说法是群、匣、喻三的上古来源有两个，舌根塞音和 *g-和*gw-以及*ɣ和*ɣw-。支持这个说法的学者包括丁邦新和邵荣芬。他们通过东汉谐声情况、闽方言材料、通假字、异文、《说文》读若支持他们的立场。然而，第一派学者丘素郡忽略了重纽和龚煌城对来母的上古构拟对群、匣、喻三上古构拟的影响。丘素郡的构拟认为*grj-和* *gwrj-分别是上古喻三开口和合口的读音。丁邦新和龚煌城则认为*grj-和*gwrj-有其他用途。如果丘素郡关于*grj-和*gwrj-的构拟是错的，这会导致她对群、匣、喻三的的上古构拟不成立，因此有必要研究*grj-和*gwrj-的用途。本文使用清浊别义来判断*grj-和*gwrj-的用途。研究成果显示*grj-和*gwrj-是群母重纽三等字的开口和合口读音，因此丘素郡构拟不成立。群、匣、喻三的上古来源是一个的说法被削弱。本文因此倾向群、匣、喻三的上古来源是两个的说法。之后，在结合了龚煌城的构拟后，改进了丁邦新的构拟，推出新的构拟。The study on Middle Chinese initials ɣ-, g-, j- in Archaic Chinese has been ongoing for decades. Currently, there are two main theories remaining. The first theory is that Middle Chinese initials ɣ-, g-, j- all belong to *g- and *gw- in Archaic Chinese. Scholars who support this argument include Li Fang Kuei, Gong Hwang-Cherng and Kew Sook Chiun. The second theory is that initials ɣ-, g-, j- belong to two separate groups, *g- and *gw- along with *ɣ- and * ɣw-. Scholars who support this argument include Ting Pang-Hsin and Shao Rongfen. Kew Sook Chiun supported the first theory but she ignored the effect that chongniu (重钮) and Gong Hwang-Cherng’s reconstruction of Middle Chinese initial l- in Archaic Chinese had on Middle Chinese initials ɣ-, g-, j-. Kew Sook Chiun reconstructed *grj- and *gwrj- as the origin of Middle Chinese initials j- but Ting Pang-Hsin and Fong Hwang-Cherng felt otherwise. If Kew Sook Chiun’s reconstruction of Middle Chinese initials j- is wrong, this would adversely affect her entire reconstruction of Middle Chinese initials ɣ-, g-, j- hence it is necessary to test the purpose of *grj- and *gwrj-. The test is conducted using qingzhuo bieyi (清浊别义). Test results show that the *grj- and *gwrj- are the origin of Middle Chinese initial g- chongniu Division III rimes, hence Kew Sook Chiun’s reconstruction is faulty resulting in support of the first theory being weakened. Hence, this paper believes that initials ɣ-, g-, j- has two origins in Archaic Chinese. Also, after incorporating Gong Hwang-Cherng’s theories, a modified version of Ting Pang-Hsin reconstruction is produced.