The Gesture-speech Combinations Of Mandarin-speaking Children Around The Onset Of The Two-word Stage

The theoretical nature of one-word utterances has been controversial in the literature. Bloom (1973) argues that children's one-word utterances do not have an underlying linguistic structure, but the single-word utterances express complex meaning that are expressed in adult speech. However, McNeill (1970) suggests that one-word utterances have the underlying representation of a full sentence but that only one element of this full sentence is realized in the surface structure. An alternative view is the innate semantic structure hypothesis proposed by Greenfield and Smith (1976), who claim that children's one-word utterances have underlying semantic structure, involving with semantic roles like Agent, Action, Object, etc, rather than syntactic structure. However, Dore (1975) argues that one-word utterances do not have underlying syntactic or semantic structure but only involve particular communicative intentions.Recent studies have found that certain gesture-word combinations express a variety of basic semantic relations similar to those that multi-word combinations typically express (Morford et al. 1992; Capirci, Iverson, Pizzuto, and Volterra 1996). Other researchers have found that certain gesture-word combinations exhibit particular types of syntactic structure such as "argument + argument"; "argument + predicate" and "predicate + predicate" ((O|¨)z(?)ali(?)kan, and Goldin- Meadow 2005). Further, children express particular semantic relations in gesture-speech combinations before they express the same relations in multi-word combinations (Morford et al. 1992; Capirci 1996; Goldin-Meadow 1999; (O|¨)z(?)ali(?)kan, et al. 2005; Kelly 2005)The present study analyzes gestures, especially gesture-word combinations used by two Chinese children in the six months before and three months after the onset of two-word combinations to see if these combinations are precursors to linguistic constructions and if one-word utterances have underlying syntactic or semantic structure.Two Mandarin-speaking children living in Changsha were observed approximately weekly from 10 to 24 months, and hourly audio and video recordings were made of their naturalistic interactions with caretakers. As two-word combinations emerged for both children around 17 months, we analyzed 18 video sessions which covered the period 11 months to 20 months for each subject. Video clips of single gestures; gesture-gesture; gesture-vocalization; gesture-word and two-word combinations were made and coded following the methodology of Iverson et al. (1994) Gestures were classified into deictic and representational ones. The gesture and word combinations were divided into reinforcing, disambiguating and supplementary combinations, with the latter broken down into argument-predicate, argument-argument and predicate-predicate combinations.