ABSTRACT
MacSwan provides an historical overview of research on code-switching (CS) from early constraint-based (CB) approaches, which posited explicit rules governing language mixing, to current constraint-free (CF) solutions, which seek to derive CS from independently motivated properties of grammar. MacSwan argues that early lexical insertion in the Minimalist Program resolved technological limitations associated with prior models which prevented implementation of a CF solution to CS. Illustrations, offered as proof of concept for the utility of the CF approach, are presented. Current debates around non-lexicalist approaches to CS are discussed.
