The perception and learnability of english prosodic phonology by Luo speakers: A cross linguistic experimental study.

Abstract

The present thesis is an experimental study that examined the prosodic phonological perceptual strategies that Luo L2 learners of English use as they process and assign meaning to English aural stimuli in the process of language learnability. The study’s scope was limited to senior three secondary school Luo listeners’ perception and learnability of English word, sentence and discourse prosodic phonology. The study was conducted in Gulu, Lira and Tororo districts of Uganda and Homa Bay District in Kenya. The present study adopted the Solomon-Four group experimental design. There were two independent variables: treatment Prosody Oriented Approach (POA) vs. Control group Segmental approach (SA) and pre-test / no pre-test. The Solomon four group design involved the random assignment of subjects into one of four groups. Two groups are pre-tested and two are not. One of the pre-tested groups and one of the non-pretested groups receives experimental treatment. All the four groups were post-tested with the independent variable. If the pre-tested experimental group performed differently on the post test than the non-pre-tested, there is probably a pre-test treatment interaction. There were a total of twenty experimental tests: four constituted the pre-test and sixteen the post. The multi-trait multimethod approach was used to disentangle the method effect in the testing situations. The results of five experiments on the word-level prosody show that Luo listeners were not able to use English prosodic cues reliably in locating English primary word stress. The experimental POA had a significant effect in improving the listeners’ perception of primary word stress. The L1 on the listeners has a significant effect on their perception of English primary stress, while pre-testing did not have any effect. The results of ten experiments on syntactic disambiguation show that Luo L2 learners of English could not use segment length reliably in locating word boundary. Secondly, they couldn’t use prosody reliably in syntactic disambiguation. POA and L1 had significant effect in syntactic disambiguation. The five discourse coherence experiments show that xi POA has a significant effect on the accuracy of interpretation of English discourse coherence by Luo listeners. The study concludes that English prosodic phonology is learnable and that an effective approach such as POA should be adopted in the teaching of English as a Second language. Second, there is a clear learnability path that can be modelled within an optimality theoretic framework. This thesis shows that a Robust Interpretive Parsing/Constraint Demotion (RIP) (CD) learning algorithm proposed by Tesar and Smolensky (2004) can be used to model the learnability path of English prosodic phonology by Luo L2 learners of English within Optimality Theory.The present work has significant pedagogical implications.