As the eight-year-old Sophie has noticed, her sister Alicia at age four has her own way of dealing with tense. Sophie is puzzled by Alicia's use of is die with the adverb already. The form is die already is not among those used at school, where the concept of tense has recently been imparted to Sophie; nor is it obviously ‘Chinese English’ (to use Sophie's term) as some phenomena discussed in this book are. Where, then, does Alicia's own ‘grammar’ come from? This and similar questions are the subject of our penultimate chapter.

In chapter 1 we discussed how developments in bilingual individuals parallel, and ultimately underlie, those taking place in the course of contact-induced change. In bilingual individuals we can observe processes such as code-switching, transfer and other forms of grammatical interaction; in languages in contact we observe processes such as lexical borrowing, calquing and contact-induced grammaticalization, while the outcomes include language shift, pidginization and creolization. The relationship between the individual and language-level processes is not well understood, as we shall see in the domain of grammaticalization. It is, however, widely recognized that ‘the bilingual individual is the ultimate locus of language contact’ (Romaine 1996: 573).