In many of the world's bilingual communities, fluent bilinguals sometimes engage in code-switching by producing discourses which, in the same conversational turn or in consecutive turns, include morphemes from two or more of the varieties in their linguistic repertoire.1 Thanks to a plethora of publications and conference presentations on code-switching since the late 1970s, an overview of CS in the middle 1990s can offer a rich characterization of CS itself, as well as comparing it more precisely with other language contact phenomena involving two or more languages.