Comparing multilingual children with SLI to their bilectal peers: evidence from object and action picture naming

Abstract

Against the background of the increasing number of multilingual children with atypical language development around the world, this study reports research results on grammatical word class processing involving children with specific language impairment (SLI). The study investigates lexical retrieval of verbs (through picture-naming actions) and compares performances for the same children with noun retrieval (through picture-naming objects). It was found that nouns (object names) were better retrieved than verbs (action names) in the multilingual group with SLI, a finding similar to bilectal peers with SLI and typically developing language-matched controls. The results suggest that grammatical class is an organising principle shared across languages. Moreover, when tested in first and third language, multilingual children with SLI revealed a comparable verb–noun dissociation both in terms of the direction of the effect (V < N) and major error type. These findings are discussed in relation to the delayed acquisition hypothesis for SLI and psycholinguistic models of multilingualism.

Keywords:

• (a)typical language development
• bilectalism
• grammatical word class
• Greek
• nouns
• verbs

Acknowledgements

Part of the research reported here was presented by M. Kambanaros at the 2010 Bloomsbury Round Table on Communication, Cognition and Culture with the theme ‘The multiple faces of multilingualism’ (Birkbeck College, University of London, 24–25 June 2010). We would like to thank the audience for their valuable comments and input. All of us gained tremendously from feedback within our research group, so many thanks to all members from the Cyprus Acquisition Team (http://www.research.biolinguistics.eu/CAT). We acknowledge financial support from the University of Cyprus for the Gen-CHILD Project (UCY project no. 8037–61017) and the Cyprus Research Promotion Foundation for the CySLI Project (RPF project no. ANΘPΩΠI∑TIKR∑/ΠAIΔI/0609(BIE)/11), both awarded to K.K. Grohmann (see http://www.research.biolinguistics.eu for more).

Notes

1. This assessment is corroborated by the obvious need that led to the creation of the European research network ‘Language impairment in a multilingual society: Linguistic patterns and the road to assessment’ (COST Action IS0804), a 4-year activity which started in June 2009. See the Action's website for more information and resources at http://www.bi-sli.org.

2. The ‘Curriculum for the Modern Greek Language’ or — (http://www.moec.gov.cy/analytika_programmata/nea-analytika-programmata/nea_elliniki_glossa.pdf) was put to generalised experimental application in September 2011 and is meant to encourage the use of CG even in primary education. It is part of a project commissioned by the Cyprus Government to develop recommendations to the education curriculum (Hadjioannou, Tsiplakou, & Kappler, 2011). This did not have any bearing on our data, which were collected prior to the start of the new curriculum.

3. Throughout we will use the following conventions: actions and objects in the real world are not marked in any way (e.g. a brush or painting). The lexical items referring to these are italicised, with the noun appearing in its bare form and the verb in its participial form (e.g. brush or painting). Responses from children are put in double quotes, especially when different from the target (e.g. ‘that which you paint with’ instead of brush). Modern Greek does not have an infinitive anymore, which is one reason why we use the present participle for the English citation form; as described below, all elicitations required the third person singular form in present tense, which is the second reason for our convention.

4. In the context of the linguistic situation in Cyprus (see the section ‘Multilingualism in Cyprus’ under the ‘Background’ section earlier), one may be tempted to classify the bilectal children as ‘monolingual’ (bilectal speakers of the low and high varieties) and subsequently the tri- or multilingual children as something like ‘bilingual bilectal’ speakers, speaking two distinct languages (such as English and Greek) and two lects of one (CG and SMG). We leave such finer distinctions for future research, which should also address the precise role of such varieties in language development: lects, dialects, languages. Relevant for such research will also be a finer-grained incorporation of ‘second dialect acquisition’ (e.g. Siegel, 2010) in both (early/child) second language acquisition and bi- and multilingualism studies (see also Aronin & Singleton, 2012, among others).

5. See note 1.

6. See note 1.