Cross-linguistic Differences in Children’s Syntax for Locative Verbs

Learning a verb’s meaning and its associated syntactic structures pose a number of difficult problems for a learner. However, it is widely assumed that there are consistent correspondences between verb meanings and verb syntax, and that knowledge of these correspondences may provide important help to the learner (Gleitman 1990; Grimshaw 1981; Landau & Gleitman 1985; Pinker 1989; Gropen et al. 1991a,b). To take a simple example, English mental verbs like "think", "know", and "believe" take sentential complements, as do many other mental verbs in English and other languages. Accordingly, if this connection is universal, and if the learner knows this, then it could be very useful. If the learner already knows that a verb is a mental verb, then she can infer that it allows a sentential complement. If the learner hears an unfamiliar verb that takes a sentential complement, then this can be used as a clue that the verb might be a mental verb. In this paper we are primarily concerned with locative verbs, such as “pour”, “fill”, “load”, and “stuff”. Locative verbs encode the relationship between a moving object the “Figure”and a location the “Ground”. Although locative verbs all show this semantic similarity, they fall into at least four different syntactic subclasses based on their syntactic possibilities (Pinker 1989), as shown in (1-4). In addition to the Figure and Ground Non-alternating classes, Alternating verbs are divided into two subclasses, the Figure Alternating verbs in (3) and the Ground Alternating verbs in (4), based on which argument is obligatory in both syntactic frames (see Pinker 1989).