Language mixing is a common practice in bilinguals' daily lives. Previous studies have focused on discovering the constraints that govern language mixing possibilities (Poplack 1980, Myers-Scotten 1993, Muysken 2000, 2013, among others). These works were based on the "insertional" approach—the idea that different languages are separate systems between which speakers shift when mixing. In recent years, research has shifted toward developing a language mixing grammar within a single computational system under the integrated model (MacSwan 1999 for the Minimalist Program, Alexiadou and Lohndal 2018, Lopez 2020 for the Minimalist Distributed Morphology (MDM), among others). An integrated model proposes that speakers have just one computational system and one linguistic competence governing all languages. Under this view, primitive features (e.g., definiteness) come from a universal inventory and can be expressed by linguistic items from either language or both (doubling).
However, work on the integrated model has mostly focused on inflectional language pairs. The mixing patterns of other language types, such as agglutinative and isolating languages, remain limited and open for exploration. This study employs MDM to investigate mixing patterns in the nominal domain of three language pairs: Cantonese-English, Japanese-English, and Cantonese-Japanese. The project aims to answer the following questions and contribute to our understanding of language mixing and universal grammar: