Language contact phenomena involving Cantonese and a “Western” language, in particular, English, have received much scholarly attention from professional linguists and linguistics students alike, if most works to date have focused on the forms and functions of lexical borrowing and code-mixing/code-switching, mostly taking a synchronic perspective that is, analysing a set of collected data without much consideration of change and variation within the datasets. This paper attempts to extend the field by proposing alternative perspectives and exploring areas that could be further investigated; in particular, it proposes an evolutionary perspective in which lexical borrowing and code-mixing/code-switching are seen as connected phenomena in the course of language contact, and it discusses a number of constructions which have been taken as “code-switching” but are perhaps better described as “structural borrowing” (Thomason and Kaufmann 1988) in such a perspective.