The nature of two trilingual children's utterances

This thesis is an exploration of the nature of utterances produced by two children of the same family, who are growing up with three languages - Croatian, English and German - from birth. The period of study covers ten months, and the children are aged between 1;4 (one year and four months) and 2;1 and between 2;9 and 3;6 respectively. During this period, the children produce mono-, bi- and trilingual utterances. The focus in this thesis is on utterances involving more than one language. Such utterances involve two types of mixing: (i) whole-word mixing, in which whole words are contributed from at least two of the participating languages, and (ii) word-level mixing, in which individual words are made up of constituent parts belonging to different languages.
In order to gain an understanding of the occurrence of such utterances, a close account is provided not only of the circumstances in which these utterances are produced but also of the frequency with which they are recorded. Attention is also paid to the level of language development, which, it is thought, can have an influence on the production of mixed utterances.
The application of existing analytical frameworks (Myers-Scotton, 1993; 2006; Poplack, 1980) to the selected corpus of data from the present thesis demonstrates their limitations with regard to accounting for the reported variety of (mixed) utterances. A degree of modification is proposed in this context, but the thesis calls for additional empirical research in the search for more appropriate analytical frameworks for multilingual children¿s language productions.