Possessive Constructions and Code-Switching among Speakers of Pakistani English

Abstract

This study aims to investigate the preferred possessive agreement forms among Pakistani English speakers in code switched mode, examining whether preferences align with the Matrix Language (ML) or Embedded Language (EL) of the possessor or possessum. It also investigates the relationship between reaction time and agreement patterns in ML vs. EL contexts, alongside the influence of extralinguistic factors such as language proficiency and code-switching habits. Finally, it assesses the role of animacy, particularly with human possessors, in shaping agreement preferences. This study examines possessive agreement preferences in code-switching among native Urdu speakers in Pakistan with English as an L2, aged 18-25. Using the Matrix Language Framework as its theoretical foundation, the study employs a language background questionnaire and a Forced Choice Task (FCT) to examine participants agreement patterns, reaction times, and influences of language proficiency and animacy. The findings from the Forced Choice Task offered strong support for the prediction that participants overwhelmingly adhered to the Matrix Language (ML) agreement patterns when the possessive structure was expressed in the ML. Faster RTs were also observed when participants followed the agreement patterns of the ML compared to the Embedded Language (EL). The results also reveled that animacy strongly influences gender agreement patterns, with English as the ML showing object-based agreement for animates and Urdu showing stronger adherence to gender norms for animate nouns but variability for inanimates. Lastly, the study of agreement patterns with extra linguistic variables such as proficiency, frequency of language mixing and code switching habits revealed that while these variables plays a role in determining possessive agreement patterns, individual variability and factors such as cross-linguistic influence are also significant.