What L2 learners mean: What gestures reveal about placement verbs and event construal in L2 production
In second language (L2) acquisition research it is often assumed that when L2 speech looks target-like, meaning is also target-like. This study investigates whether L2 learners mean the same things as native speakers when they use a given form. The test domain is that of caused motion or placement events (e.g. 'put the cup on the table'), a spatial domain firmly grounded in sensory-motor experience and therefore a popular candidate for universal, language-neutral event construal. Yet, placement events are lexicalised differently cross-linguistically. For instance, Dutch uses a set of obligatory positional verbs (zetten, leggen, 'set', 'lay'), the choice of which hinges on properties of the located object and its final orientation with respect to the ground. French instead typically uses a very general placement verb (e.g. mettre, 'put'). The languages thus differ in semantic granularity. The assumption is that Dutch learners of French will easily acquire the appropriate French term, moving from finer to coarser distinctions. This study asks whether they also adjust their semantic and event representations towards a more coarse-grained perspective by examining speech and gesture in combination.
A first study examines native event construals. The results from an event description task show that the iconic gestures of native speakers of Dutch (N=12) and French (N=12) systematically reflect language-specific perspectives on placement derived from the semantic granularity of the verbs. Dutch gestures typically incorporate objects in handshapes, reflecting the fine-grained perspective necessary for the choice of the right verb. French gestures display only path in accordance with the coarse-grained perspective. Gestures thus reflect semantic representations and language-specific event construals of placement events.
A second study examines placement descriptions by Dutch learners speaking L2 French (N=12). L2 speech is overall target-like, but L2 gesture patterns mixed: one group of learners perform French-like gestures, another group both Dutch- and French-like gestures, and a third group Dutch-like gestures. Different learners thus apply different perspectives to the events and 'mean' different things by mettre in L2. While target-like speech may be easy to achieve when moving from finer to coarser distinctions, the gesture evidence suggests that re-organisation of representations in L2 acquisition is a gradual, complex process which may not be achieved even in advanced stages of acquisition. But it is possible, as evidenced by the French-gesturing group. The findings raise questions regarding the implicit acquisition of meaning and cross-linguistic influences on event representations in multilingual speakers.