As mentioned previously, one of the main issues in bilingual research con cerns the relation of the two language systems: To what extent are the two languages functionally independent, and to what extent do they constitute a sin gle functional system? Although there have been several attempts to investi gate this problem with different types of bilinguals, we do not seem to know enough yet to describe what really goes on in a multilingual’s mind. Durgunoglu and Roediger (1987) argued that this question is indetermin able because “the varying retrieval demands of different tasks produce dif ferent patterns of results and lead to opposite conclusions” (p. 377). In this book the focus of attention is on adult speakers of two or more languages. We find it important to emphasize this because approaches to bilingual de velopment differ to a great extent depending on which period of develop ment researchers are interested in. As mentioned earlier the extremist posi tions (Cook, 1992; Grosjean, 1989) state either that the two languages form separate systems (De Houwer, 1990; Genesee, 1989; Lambert, 1990; Neufeld, 1976) or that they make up a single unified system (Redliger & Park, 1980; Swain, 1977). There are also opinions that the truth lies somewhere in be tween. Bilinguals have two separate systems that interact constantly and overlap at certain points (Cook, 1992; Paradis, 1985; Perecman, 1989; Sridhar & Sridhar, 1980).