Gestural expressions of spatial information in L1 and L2

This paper reports on the analysis of gestures in the expression of static locative relationships in French and Dutch, for L1 speakers as well as for francophone learners of Dutch (L2). The data analysed is drawn from video-taped picture descriptions where subjects were asked to talk about the location of certain entities on these pictures. Tutton (2012) has observed that in spatial descriptions gestures often express information that remains unexpressed in the verbal production and that typically the information that is gesturally expressed is directional (cf. also McNeill 2000; Gullberg 2009, 2010). Our data only partially confirm his findings: in most cases, gestures express information that is also expressed verbally. In addition, while gestures are indeed well-suited to express direction, we argue that a clearer distinction is needed between directional and (purely) locational gestures. We suggest that the crucial factor identifying a locative gesture is the fact of the gesture being anchored in the representational gesture space, an issue that hitherto has not been discussed in the literature. While all gestures are necessarily made in the gesture space, anchored gestures are those that receive a clear representational location. These can be pointing gestures, but often they are not (e.g., an anchored shape-, size- or manner-gesture). Functionally, they are not unlike what Liddell (2003) has called buoys in ASL, i.e., clearly located and stationary signs that function as conceptual landmarks while the discourse continues. The difference with anchored gestures is that the latter are not stationary. Non-anchored gestures do not have such a precise location. For example, directional gestures are not really anchored to a specific point, but merely indicate a direction. Similarly, some iconic gestures express locative relations (e.g. BETWEEN, EVERYWHERE), but are made without being anchored in the representational gesture space (e.g., just in front of the speaker, in centre space). We argue that despite their locative semantics, they are not locative gestures. In fact, anchored locative gestures could thus be seen as grounding predications, i.e. "an instance (but not a type) is thought of as having a particular location in the domain of instantiation" (Langacker 1991:57). In addition, typological differences are manifest in gesture. In line with Talmy’s (2000) typological distinction between verb-framed and satellite-framed languages, Dutch can be described as a “location-rich” language and the descriptions of the native Dutch speakers abound with locative descriptions, through the highly grammaticalised use of posture verbs but also via other linguistic means (prepositions, adverbs, etc.). French, in contrast, is “location-poor”: the French narrations have significantly fewer locative descriptions and the locative information is much more general. Instead, they add narrative detail and meta-linguistic comments to their descriptions. The francophone learners of Dutch (with 3 levels of proficiency) use more gestures revealing the challenge that free expression in a second language poses, especially for the lowest proficiency levels: they use more shape gestures, more enactment gestures (e.g., pulling a drawer, brushing one’s hair, etc.), more reality-anchored gestures (e.g., pointing at one’s shoes when talking about shoes), and more meta-communicative gestures indicating their lexical shortcomings, e.g., word- search gestures (see Ladewig 2011). Overall, and as can be expected, the low proficiency L2 speakers use almost more gestures than words, which can be seen as a visual compensation for their lack of lexical accuracy; the gestural expression of advanced learners, in contrast, is much more locational in nature, in line with the target language (cf. also Gullberg 2009, 2010, Alferink & Gullberg 2014)