Prosodically constrained syntactic changes in early archaic Chinese

In diachronic studies of Chinese syntax, an interesting question is why the two SOV structures, [wh-object V] and [Neg Pro-object V] in Early Archaic Chinese (EAC, 1000-500 B.C.), disappeared after the Han Dynasty (206 B.C.). This paper proposes that these two notable OV orders in EAC are structurally distinct and that the structure of [wh V] is also different from that of [wh Neg/Aux V]. Furthermore, it is argued that Proto-Chinese is an SOV language and that the change from SOV (Proto-Chinese) to SVO (EAC) caused a stress shift from preverbal to postverbal position. According to the theory developed here, some problems that have remained in the syntax of Classical Chinese cease to exist, including the following: Why did the two OV structures remain in EAC? Why did the [wh-object V] order disappear later than the [Neg Pro-object V] order? Why did monosyllabic wh-words (e.g., he 'what') but not disyllabic wh-expressions (e.g., he-shi 'what thing') immediately precede the verb? And why was the disappearance of the [wh-object V] structure followed by a development of disyllabic wh-words (e.g., he wu 'what thing' > hewu 'what')? Each of these questions is answered in terms of prosody. The arguments made here claim that prosody is very important in resolving questions of how syntactic changes take place. © 1996 Kluwer Academic Publishers.