In Chinese, as in many languages with long written histories, textual sources tend to reflect koines or regional lingua francas rather than dialects as such. The recording of ordinary regional vernaculars as a matter of scholarly or scientific interest is in the main a modern phenomenon. As a result, Chinese dialects are known mainly through their modern spoken forms. And, has been pointed out by Norman (1988:42), the primary expedient for recovering the phonological history of families of dialects is the application of the comparative method. Early attempts in this direction involved work on large, very old dialect groups. More recently, attention has focused on smaller groups and subgroups, for it is well-known that the comparative method is most effective when applied at shallow time depth to closely related languages. The effectiveness of this approach has been clearly demonstrated in recent publications, a prime example being that of Akitani (2003), who skillfully applies the classical comparative method to a single sub-group of the W? dialect family. The present monograph is intended as a study of the type espoused by Norman and implemented by Akitani. It deals with the Hu?ng-Xi?o group, a sub-group of upper Yangtze Watershed Mandarin dialects.

Earlier versions of this work were circulated for several years in the form of a long paper, which was read and assessed by a number of friends and colleagues. Particularly useful and interesting were thoughts offered by by Chen Zhongmin, Jerry Norman, Laurent Sagart, and Ting Pang-Hsin. In some cases I have adopted ideas of theirs, while in others, perhaps at my peril, I have taken my own tack. All weaknesses which remain are of course my own responsibility. The earlier paper was also delivered at the 35th International Conference on Sino-Tibetan Languages and Linguistics, held at Arizona State University in November, 2002. I am indebted to those who offered responses and suggestions there.

As mentioned in Chapter I below, two of the unpublished dialect sources consulted here represent fieldwork by Mr. Z?ng Gu?xi?ng, whose kindness in allowing me to use them I gratefully acknowledge. My contact with Mr. Z?ng was arranged by Dr. G? Qi?n of Nanking University. Without her help, access to these important sources would not have been possible. Dr. G? also sent me other materials of of great interest and importance, as will be mentioned below. I remain eternally indebted to her for this.

Finally, I should like to thank Professor Ho Dah-an of Academia Sinica for encouraging me to complete this monograph, and for his editorial advice and assistance.

W. South Coblin
Iowa City, Iowa
August 2004In Chinese, as in many languages with long written histories, textual sources tend to reflect koines or regional lingua francas rather than dialects as such. The recording of ordinary regional vernaculars as a matter of scholarly or scientific interest is in the main a modern phenomenon. As a result, Chinese dialects are known mainly through their modern spoken forms. And, has been pointed out by Norman (1988:42), the primary expedient for recovering the phonological history of families of dialects is the application of the comparative method. Early attempts in this direction involved work on large, very old dialect groups. More recently, attention has focused on smaller groups and subgroups, for it is well-known that the comparative method is most effective when applied at shallow time depth to closely related languages. The effectiveness of this approach has been clearly demonstrated in recent publications, a prime example being that of Akitani (2003), who skillfully applies the classical comparative method to a single sub-group of the Wú dialect family. The present monograph is intended as a study of the type espoused by Norman and implemented by Akitani. It deals with the Huáng-Xiào group, a sub-group of upper Yangtze Watershed Mandarin dialects.

Earlier versions of this work were circulated for several years in the form of a long paper, which was read and assessed by a number of friends and colleagues. Particularly useful and interesting were thoughts offered by by Chen Zhongmin, Jerry Norman, Laurent Sagart, and Ting Pang-Hsin. In some cases I have adopted ideas of theirs, while in others, perhaps at my peril, I have taken my own tack. All weaknesses which remain are of course my own responsibility. The earlier paper was also delivered at the 35th International Conference on Sino-Tibetan Languages and Linguistics, held at Arizona State University in November, 2002. I am indebted to those who offered responses and suggestions there.

As mentioned in Chapter I below, two of the unpublished dialect sources consulted here represent fieldwork by Mr. Zēng Guóxiáng, whose kindness in allowing me to use them I gratefully acknowledge. My contact with Mr. Zēng was arranged by Dr. Gù Qián of Nanking University. Without her help, access to these important sources would not have been possible. Dr. Gù also sent me other materials of of great interest and importance, as will be mentioned below. I remain eternally indebted to her for this.

Finally, I should like to thank Professor Ho Dah-an of Academia Sinica for encouraging me to complete this monograph, and for his editorial advice and assistance.

W. South Coblin
Iowa City, Iowa
August 2004