Early Chinese classifier acquisition and the predictors : a corpus-based comparison of Mandarin- and Cantonese-speaking preschoolers in Beijing and Hong Kong

Abstract

Chinese classifiers (Liángcí) quantify and enumerate objects by their explicit and inherent characteristics, and are often regarded as a window on cognitive development. Previous Chinese-classifier acquisition studies have either adopted different typologies, or been conducted with small sample sizes of different age groups, making it difficult to draw conclusions on the patterns of early classifier development. Moreover, very few studies have investigated the influence of home language input on Chinese-classifier acquisition. Thus, this thesis addresses the research gaps by examining early Chinese-classifier development of preschoolers in Beijing and Hong Kong via one test and three studies. Test One was a typology test to examine the practicability of the proposed typology for two corpora, the Early Cantonese Corpus and Early Mandarin Corpus (Li and Tse, 2010), which comprise, respectively, spontaneous utterances produced by 144 Mandarin- and 144 Cantonese-speaking preschoolers in Beijing and in Hong Kong ranging in age from 2;6 to 5;6 years old during a 30-minute free-play task. All classifiers produced by the Beijing and Hong Kong preschoolers were classified according to the proposed typology. Study One and Study Two were a pair of cross-sectional, corpus-based studies on Mandarin and Cantonese classifier acquisition and its predictors respectively. The Mandarin and Cantonese classifiers produced by the respective preschoolers were extracted and examined. Their parents’ responses to the Home Language Environment Survey (HLES) were also considered to investigate relationships between child classifier acquisition, and family language inputs and schooling experiences. Study Three was a comparison study of Studies One and Two. Repertoires, classifier types and the critical periods for Mandarin and Cantonese classifier development were compared. The predictors of the family and school language input on early classifier acquisition in Beijing and Hong Kong were also scrutinized to draw conclusions on common developmental trends and patterns of Chinese-classifier acquisition. The findings showed that both Beijing and Hong Kong preschoolers in the 2;6-year-old group had already acquired the basic syntax for classifiers. An age-related developmental trend was also found for the repertoire sizes and the types of classifiers used. Repertoires of Mandarin and Cantonese classifiers were 82 and 69 classifiers respectively. No significant difference for noun classifier types was observed, but a noticeable difference in the verb classifier types used was found between the two groups of preschoolers. Additionally, the critical periods for early classifier development in Mandarin and Cantonese were between 3;6 and 5;6 years old, and 4;6 and 5;6 years old respectively. Furthermore, Mandarin/Cantonese storytelling was a positive predictor on early Chinese-classifier acquisition, whereas English storytelling and the use of English by foreign domestic helpers were negative predictors on early Chinese-classifier acquisition. As the first large-scale comparison study of early Mandarin and Cantonese classifier acquisition, this study provides cross-linguistic evidence to support the proposed classifier typology. The study concludes by illustrating the cognitive process of Chinese-classifier acquisition according to Piagetian and Vygotskian theories, and proposing an ecological-cognitive model of Chinese-classifier acquisition. The findings have practical implications for early Chinese-language learning research, teaching and curriculum development.