This paper examines the characteristics of intrasentential language mixing in Korean–English bilingual children, with special attention to the distinction between codeswitching and borrowing. The findings suggest that intrasentential language mixing is determined by the bilingual abilities and preferences of both speaker and addressee. The results of the variable analysis of case marking demonstrate that most of the English-origin objects and subjects of Korean verbs are best considered as borrowings and not as codeswitches.