The acquisition of theget-passive

Abstract

This thesis addresses children's comprehension of the get-passive, examining (i) how early getpassives are acquired in development, (ii) how children perform with get-passives relative to bepassives, and (iii) what structure children initially assume for the get-passive. These questions are addressed through two different experimental methodologies, a Picture-Selection Task and an Act-out Task, with children ages 3 to 6. Results from the Picture-Matching Task suggest 3-yearolds initially assume a control/causative analysis of the get-passive, enabling them to perform above chance with only those that have animate subjects. Around the age of 4, however, children understand all (actional) get-and be-passives. The Act-Out Task additionally confirms that those 4;06 and older interpret the by-phrase as containing the agent, like adults. These results suggest there is no "advantage" to the get-passive after the age of 4. Children, rather, acquire a raising analysis of actional passives around the same time, contra previous claims. 1.0 The Syntax of the Passive in Adult GrammarAlthough English has two types of passives (exemplified in 1-2), they differ in systematic ways; in this chapter I will review syntactic analyses of both types of passives. Be-PassiveChomsky's 1981 analysis derived the passive via NP-movement, where the object moves to the matrix subject position, whereas the subject is base-generated in the by-phrase (if there is one), or absent.(3) John was pushed (by Bill) John.Jaeggli (1986) argues that the external theta-role is absorbed by the passive morphology, and then transmitted to the argument in the by-phrase. Jaeggli also assumes that the external argument is implicit in the structure when there is no overt argument in a by-phrase (following Manzini 1983). In this case, the external theta-role is retained by the passive morphology.Baker, Johnson & Roberts (1989) propose that the by-phrase is always projected, if only implicitly. For instance, it is possible to use agent-oriented adverbials (see 4) with short passives (i.e. those without a by-phrase) and long passives, indicating that there is an understood agent.(4)The vase was broken on purpose.This assumption of an implicit by-phrase is not uncontroversial (see Bhatt & Pancheva 2006); however, as will be discussed in Section 3, there is support for this notion from child grammar (cf. Orfitelli 2012). I will be assuming that the by-phrase is implicit, and hence functions as an intervening argument that the object must move around in order to get to the subject position (cf.Collins 2005). I will return to this issue in Section 3.