Mandarin-speaking Children’s Acquisition Of Passives:Comprehension&Production

The apparent cross-linguistic delay of the passive compared to other constructions has caused worldwide concern because passives not only occupy an essential position in grammar, but also symbolize the maturity of children’s language development. Much evidence has shown that children have great difficulty comprehending and producing sentences in the passive voice until at least five years of age. Such difficulties were originally reported in English by Bever (1970) and have been repeated numerous times in English (Fox&Grodznisky1998; Gordon&Chafetz1990; Maratsos et al.1985; Stromswold et al.2002etc.), as well as in a diverse set of other languages, such as Spanish (Pierce1992), Japanese (Sano2000; Sugisaki1999), German (Bartke2004), Russian (Babyonyshev&Burn2003), Greek (Terzi&Wexler2002), and Chinese (Liu&Ning2009; Xu&Yang2008).This study investigates Mandarin-speaking children’s acquisition of passives. Based on the project of DREAM, the present study selects698Mandarin-speaking children of2;06to7;11to investigate their comprehension and production of passives with different verb types (actional passives and non-actional passives) and different syntactic structures (unmarked patient-subject sentences, short passives and long passives) based on the observations of performance in the picture identification task and the syntactic priming production task. We also attempt to explain the underlying reasons for the various developmental stages displayed in Mandarin-speaking children’s passives. Based on the discussions of the results of both the comprehension and production experiments, the general conclusion can be wrapped up as follows:(A) The acquisition of actional passives is better and earlier than that of non-actional passives.(B) The production part of the experiments also provide evidence showing that young Mandarin-speaking children subject to Universal Phase Requirement (henceforth UPR) seem to interpret the actional passives as resultative-like structures. Actional verbs allow a resultative reading since they involve a target state, and non-actional verbs for the most part do not involve such a target state.(C) The acquisition of patient-subject sentences prior to that of short passives and the acquisition of short passives proceeds that of the long passives.(D) We argue that the discrepancy of Mandarin-speaking children’s acquisition of patient-subject sentences, short passives and long passives stems from their different syntactic structures. Since the derivation of patient-subject sentences includes neither movement violating PIC nor other redundant process, it proved to be the earliest and easiest passive structure for children to acquire. Although the PRO movement in the short passive does not violate PIC, the moved PRO needs to be controlled by the base-generated subject, which results in the later and harder acquisition of short passives. The typical violation of PIC in the null operator movement and the existence of predication process in the derivation of long passives make them to be the latest and hardest passives for children to acquire.