An overview of Myers-Scotton’s Matrix Language Frame model

1 Background The alternation of two languages inside and outside of sentences –codeswitching (CS hereafter) has been studied mainly from two perspectives, i.e. grammatical/syntactic and pragmatic/discourse (Romaine, 1995). Although these two aspects are related and a knowledge of one helps understand the other (Myers-Scotton, 2002:Chapter 2), because of a limit in space, I will focus on the structural perspective here. A number of researchers have proposed descriptive grammatical constraints in CS, rather than theoretical. Poplack's study on Spanish/English bilinguals (1980) was one of the most influential studies. The equivalence constraint and the free morpheme constraint were proposed after a large bilingual data was examined. Codes may be switched after any constituent in discourse provided that constituent is not a bound morpheme. The equivalence constraint Code-switches will tend to occur at points in discourse where juxtaposition of L1 and L2 elements does not violate a syntactic rule of either language, i.e. at points around which the surface structure of the two languages map onto each other. Her study is specifically based on syntactically close languages, i.e. English / Spanish but counterexamples can be found in syntactically distant languages. In the 1980s some other models were proposed to define grammatical constraint, e.g. Chomskians' Government Binding model by Di Sciullo, Muysken and Singh (1986). Myers-Scotton and her associate Azuma developed a model based on speech production theory, i.e. Azuma's frame-content hypothesis (1993) and Myers-Scotton's Matrix Language Frame (MLF) model (1993). Myers-Scotton's MLF model is an abstract theoretical model but it has been employed to examine contact phenomenon among a variety of languages snce the 90s. Her MLF model was revised in Myers-Scotton, 1997 and extended sub-models were proposed. The concept of the MLF model is influenced by psycholinguistic theories. The most significant of the three are the differential activation of base language and guest language (Grosjean,1988 quoted in Myers-Scotton, 1993), the different retrieval process of closed class items and open items in Garret's speech error study (1975 quoted in Myers-Scotton, 1993), and lemmas in the mental lexicon linking conceptual information and grammatical function in Levelt's language production model. 2 2. MLF model CS can be structurally divided into inter-sentential CS and intra-sentential CS. The inter-sentential CS has been broadly studied in the socio-linguistic field but grammatical constraints are not a major focus there. With the intra-sentential CS, grammatical constraints directly affect the behaviour of two, or more participating languages. The MLF model is …