Developmental pathways for first language acquisition of Mandarin nominal expressions: comparing monolingual with simultaneous Mandarin-English bilingual children.

Abstract

This comparative study explores the first language acquisition of Mandarin nominal expressions by monolingual and simultaneous Mandarin-English bilingual children, with the aim of contributing new data relevant to the autonomous vs. interdependent debate in bilingual first language acquisition. More specifically, it focuses on the developmental sequences of complex definite and indefinite nominal expressions in the forms of [demonstrative + numeral + classifier + noun] and [numeral + classifier + noun] respectively, comparing both qualitatively and quantitatively the developmental pathways and time progression observed in corpora derived from two longitudinal studies. The bilingual corpus analyzed consists of an extensive daily record of utterances from age 10 months to 33 months, while the monolingual corpus is derived from a set of eight video recorded sessions spanning the age range from 20 months to 26 months. The outcomes provide support for the Autonomous Development Hypothesis, to the extent that the developmental pathway of Mandarin nominal expressions as elaborated in the bilingual child's case is found to be highly comparable with that of the monolingual counterpart and that up to the end of the observation period no apparent transfers from English nominal syntax to Mandarin were observed.

Key words

autonomous development

bilingual first language acquisition

Mandarin nominal expressions

Introduction

A major issue, which has been in the center of debate in research on simultaneous bilingual first language acquisition (BFLA) is whether the developmental pathway and time progression of morphosyntactic development of simultaneous bilinguals who have been regularly and frequently exposed to two languages from birth are the same as that of monolinguals of the respective languages. Substantial extant literature has indicated that in terms of the general course of morphosyntactic development, simultaneous bilinguals proceed through the same developmental pathway and eventually arrive at the same kind of grammatical competence as their respective monolingual counterparts (see De Houwer, 1990, 2005; Deuchar & Quay, 2000; Genesee, 2001; Meisel, 2001 for reviews; Genesee & Nicoladis, 2006. For differing conclusions, see e.g. Dopke, 1997, Yip & Mattews, 2007). It has been generally reported that simultaneous bilingual children exhibit also the same rate of morphosyntactic development as that of the monolingual counterparts, at least in their dominant language (De Houwer, 1990, 2005; Paradis & Genesee, 1996; Nicoladis & Genesee, 1997).

The concepts just outlined are encapsulated by the Autonomous Development Hypothesis (as termed in Meisel, 2001: 16) for simultaneous BFLA, which proposes that the two grammatical systems of simultaneous bilingual acquisition develop independently of each other, comparable with their acquisition by monolingual children of the respective languages. (De Houwer, 1990; Meisel, 1990, 1994; Paradis & Genesee, 1996, 1997). However, the question as to how this might be affected by different degrees of input for each language is an important empirical issue. Variables such as the input patterns and relative quantity of input for each language, which may alter substantially from one case to another, should be considered. De Houwer (1990, 2005) considers specific features in order to assure a successful development of two distinct target-language-like grammatical systems. Such features are proposed to include (1) frequent and regular exposure to the two languages, (2) exposure from birth, and (3) adoption of a one-person one-language principle.

A fundamental theoretical assumption within the generative framework of language acquisition is that all humans are born with a genetic endowment, the language faculty, that enables them to acquire the language(s) they are exposed to in a relatively short period of time. Ensuing from this, grammar is supposed to develop in principled ways on the basis of the language faculty and its interaction with the speech input of the target language (the primary linguistic data). In other words, the language faculty provides the child with a procedure for analyzing the primary linguistic data in such a way as to derive the grammar of the target language. Assuming further (cf. Meisel, 2001) that the human language faculty supports multilingualism, there seems to be no apriori reason to postulate that simultaneous acquisition of two (or more) languages from birth should be qualitatively different from that of monolingual acquisition. Consequently, both languages in simultaneous bilingual first language acquisition should, in principle, develop in the same fashion as for monolingual acquisition.

Detailed empirical data from longitudinal case studies can provide crucial benchmarks to guide, challenge and validate hypotheses and models for language acquisition. Systematic comparisons of monolingual versus bilingual development are crucial since they bear upon fundamental issues at the heart of BFLA research. Detailed comparisons on the developmental course of morphosyntactic phenomena in one specific language between bilingual and monolingual children have been undertaken for Basque, Dutch, English, French, German and Spanish (Meisel, 1985; De Houwer, 1990; D6pke, 1997; Almgren & Barrefia, 2000; Almgren & Idiazabal, 2001). Among the language pairs that have been studied in BFLA to date, however, the vast majority of language combinations involve Indo-European languages. In view of the significant population of bilingual children globally who speak one (or more) Chinese language such as Mandarin, Cantonese, Southern Min and so on, partnered with another language, it is an appropriate observation (if not indeed an understatement) that bilingual development involving Chinese languages has been understudied.

In recent years the first systematic longitudinal case study in BFLA involving Mandarin and English as a pair has appeared (Chang-Smith, 2005). This involved the collection of a comprehensive daily record of utterances as well as regular video recordings of the bilingual child Ralph, whose first three years of development was captured. A larger-scale longitudinal study is also available involving six Cantonese-English bilingual children (Yip & Matthews, 2007). Also the first of its kind, the study covered the first three years of language development. Interestingly, the Cantonese English bilingual subject children of the Yip and Matthews study (2007) are shown to develop quite differently from monolingual children. The authors demonstrate significant cross-linguistic interactions between the children's developing grammars and ascribe this to Cantonese language dominance in their bilingual development. These two BFLA studies provide significant and very interesting contrasts, not only with reference to Mandarin instead of Cantonese, but also because the subject child Ralph appeared to be a very balanced bilingual throughout the course of the case study. His morphosyntactic development in Mandarin, as elaborated in the earlier study and later in this article, was highly comparable with that of his monolingual counterpart.

This comparative study explores the first language (Ll) acquisition of Mandarin nominal expressions by monolingual and simultaneous Mandarin English bilingual children, with the aim of contributing new data relevant to the autonomous vs. interdependent debate in BFLA. It builds upon an earlier-mentioned study (Chang-Smith, 2005), comparing both qualitatively and quantitatively the developmental pathways and time progressions of the Mandarin nominal syntax acquisition. In the earlier study, qualitative comparisons between the two sets of observation data one for the bilingual subject Ralph and the other for a monolingual subject Bing--were made for the middle stages of their developmental pathways. However, analysis of the later developmental stage of the bilingual child was not conducted. This work extends the data analysis of the bilingual child's development of Mandarin nominal expressions for an extra 10 and a half months, beyond the point at which the full-fledged structural representation for definite and indefinite nominal expressions manifested. New quantitative information is also provided for both corpora. Collectively, this facilitates a more complete comparison of the developmental pathways of the two subjects and reveals a robust picture of the bilingual subject Ralph's acquisition of Mandarin nominal expressions in relation to the broader issues outlined earlier.

The structure of the rest of the article is as follows: Section 2 briefly outlines relevant aspects of Mandarin grammar. Methodology of the longitudinal case studies is summarized in Section 3. The findings and accompanying discussion are provided in Section 4, followed by the conclusions in Section 5.

2 Nominal expressions in adult Mandarin

Nominal expressions can either have a specific or a non-specific referent. Briefly, a nominal expression is regarded as specific when it is used by the speaker to refer to a particular entity. A non-specific nominal expression is not used by the speaker to refer to any particular entity (see Li & Thompson, 1981: 126-132). For the present study, intention on the part of the speaker to refer is a simple and relevant criterion for identifying, or interpreting, the child's nominal productions (see Lyons, 1999 for formal definitions). The terms specific and non-specific hence are used in this study only to refer to the semantic interpretations of the nominal expressions.

Mandarin does not have the equivalent words that correspond to the English definite article the and the indefinite article a, hence the numeral 'one' and demonstratives take up the coding of (in)definiteness. In Mandarin, a numeral such as yi 'one' is used prior to a classifier to mark the indefiniteness of the following noun phrase. For example, yi-zhi mao 'one-Cl cat' is indefinite. On the other hand, the demonstratives zhe 'this' and na 'that' are used to mark the definiteness of their following noun phrases. As a result, a [demonstrative + numeral + classifier + noun] string appears. For instance, na yi-liang che 'that one-Cl car' is definite. The numeral in a definite nominal expression can be omitted, if it is yi 'one' (cf. Tang, 1990b: 404).

It is worth noting that Mandarin bare nouns (although with no overt marking of (in) definiteness) can either have a specific or non-specific reference depending on the context and on the grammar of the rest of the sentence. Hence, it is common to use bare nouns interchangeably with both definite and indefinite nominal expressions in discourse.

2.1

Mandarin numeral classifiers

In Mandarin, common nouns can be quantified and …