A dissertation submitted in partial fulfilment of the requirements for the award of the degree of master od Arts in linguistics in the institute of Languages

Résumé,

This thesis is a case study of Kirundi loanword phonology, with a focus on the two-way interaction of perceptual and production factors in resolving foreign structures which are illicit in Kirndi. I frame my investigation within the view that loanword adaptation results from attempts to match the non-native perception of the L2 input, wthin the confines of the L1 grammar, Neither a purely perceptual nor a purely grammatical account can explain the facts.

Based on a corpus data of 239 French loans and 44 English loans, this study examines loanword adaptation at two levels; the phonemic and the phonotactic levels. It analyses the strategies that the language uses in adaptng the foreign phonemes to he native phonological. It also examines the way foreign syllable structures are adaped to ft the Kirundi phonotactics.

On the basis of adaptation pattems in Kirundi, we prove how the constraint-ranking theoretical framework (OT) can account for the phonological adaptations of loans but with limitations. The adaptation cannot be fully understood unless perceptual similarity and auditory factors are integrated in the grammar. The results of this research corroborate the hypothesis that perceptual similarity plays an important role in loanword adaptation if they are integrated in the grammar. (steriade 2002).

Besides, this research suppors cross-linguistic findings about the preference for faithfulness of manner over faithfulness of other features such as voicing and place; Broselow (1999), seriade (2002) and the preference for segment preservation over deletion in loan adaptation, Paradis &. La charite (1997). We prove along this line that during segmental adapation, changes occur in the least perceptible feature to achieve maximal similarity between the input segment (French or English) and its Kirundi substitute. The present research therefore enriches our understanding of the role of perceptual similarity and perceptual salience in phonology and their relationship to constraint ranking.