The Language Acquisition Laboratory has since its inception dedicated itself to the study of how the languages of Hong Kong and China are acquired by children. Issues of how language takes shape in the evolution from the babbling stage to the one-word stage, and from there to early syntax, have been investigated from the vantage point of linguistic theory and cross-linguistic comparison.
(1) Construction of three child language corpora
We have computerized the longitudinal records of two child Mandarin corpora (one of Beijing Mandarin and another of Changsha Mandarin) and a child Cantonese corpus, involving data transcription, standardization of formats, and parts of speech tags suited to the project aims. Research on Beijing Mandarin (BJCELA) was carried out in collaboration with Fang Li (Beijing Language and Culture University) and Yang Xiaolu (Tsinghua University), and ressearch on Changsha Mandarin (HNCELA) was supported by Ning Chunyan (then Hunan University, now Tianjin Normal University). The Beijing corpus consists of 196 hours of transcribed data from 4 children aged between 10 months and two and a half years of age. The Changsha corpus consists of 93 hours of transcribed data from two children aged between 9 months and 2 years and 3 months of age. The Hong Kong Cantonese corpus contains 86 hours of transcribed data from two children aged between 1 year 1 month and 2 years and 7 months of age. These data have been used for the study of early vocalizations, gestures, early words and utterances, and semantic development.
(2) Analysis of the language production of infants and toddlers
In our analysis of the language of toddlers in their second year of life, we have identified some phonetic characteristics of infant vocalizations in Cantonese and Mandarin, and demonstrated that the acquisition of Mandarin tone is not an all or none process, but is one affected by the lexical dimension. The point of word spurt has been observed for Mandarin, occurring at around 1 year 5 months of age, associated with the naming insight and entrance into the two word stage. Our analysis has delineated milestones on the development of verbs and nouns, classifiers and nominal structure, word order and argument structure, as well as functional categories and negation. We argue for certain theoretical positions such as the early availability of abstract categories and the productivity of early sentence patterns.
(3) Cross-sectional experiments carried out with pre-school children
We have experimented with the visually reinforced infant speech discrimination paradigm making use of a head-turn procedure to demonstrate the perception of Cantonese tones in Hong Kong infants aged between 6 and 8 months. We have also tested the perception of the vowel length contrast in Cantonese by pre-school children in Guangzhou and Hong Kong, showing that the distinction was not fully acquired even at three years of age, and that the perception of the distinction depended on children’s word knowledge and the syllable ending.
With respect to syntactic development, we have performed a series of experiments in Shenzhen testing 3- to 6-year-old Mandarin-speaking children’s comprehension of the individuating and quantificational functions of classifiers in relation to nouns, isolating these two functions in children’s language development. We have also performed a set of experiments in Shenzhen investigating 2- to 4-year-old Mandarin-speaking children’s interpretation of word orders that are consistent with the target language and those that are not, demonstrating the dominance of SVO word order.
(4) Establishing milestones with large scale testing on the Hong Kong Cantonese Oral Language Scales (HKCOLAS) Project
The lab through its director has been involved in the assessment of Cantonese grammatical knowledge as part of the HKCOLAS project, based at the City University of Hong Kong, and funded by the Department of Health of HKSAR. This effort resulted in the publication of the assessment instruments in 2006, now widely used by local speech therapists.