THE SIBLING EFFECT IN THE DEVELOPMENT OF PERSONAL REFERENCE - A LONGITUDINAL-STUDY INCLUDING 27 GERMAN SINGLETONS AND 20 GERMAN SIBLINGS DURING AGE-2 AND AGE-3

The central issue of this paper is the transition from nominal to pron ominal expressions for personal reference in early first language deve lopment. Two groups of children are being compared: on the one hand si ngletons and on the other hand siblings who are growing up in close co ntact with a sibling of the same sex one or two years older. Is the co urse of time different in the development of personal reference in tho se two populations of children? In their normal home-environment 47 Ge rman children (20 siblings and 27 singletons) aged 1;7-2;7 at the begi nning of the longitudinal study, were investigated three times with an interval of about three months between consecutive visits. The childr en's task was to identify and name photographs in a picture-booklet sh own by their mothers as addressees of the personal reference. The phot ographs depicted the children themselves or their mothers and were tak en shortly before each of the three sessions. All sessions were videot aped and transcribed according to CHILDES (Mac Whinney, 1991). In each transcript, the children's personal references were categorized accor ding to mode of reference (verbal vs. missing), semantic correctness a nd (nominal vs pronominal) form. Statistical comparisons involved log- linear analyses of the relationship between the status of the child (s ingleton vs. sibling) and the direction of reference (self = speaker v s mother = addressee) on the one hand and three different aspects of t he references produced across three sessions on the other. The results show that the transition from nominal to pronominal personal referenc e was speeded up in siblings compared to singletons, when the referenc e fulfilled only a descriptive function. However, the effect was limit ed to the pronominal form of the reference, since the increase of corr ect semantic references over time and the cognitive processes underlyi ng the production of these references appeared to be highly similar in siblings and singletons. We discuss how the sibling-effect in the dev elopment of personal reference can be explained by differences in ling uistic input and social interactions of 2 and 3 year old children.