Classificare il code mixing: una reinterpretazione dei parametri di constituency del modello di Muysken

Aim of this paper is to discuss the problem of classification and analysis of code mixing, in particular by observing the different conditions of constituency in the three classes of phenomena described by Muysken (2000): insertion, alternation and congruent lexicalization. After a close reading of Muysken’s parameters concerning constituency (as detailed in Bilingual Speech), the paper will propose a partial revision of his classification model, by introducing two scales combined in a matrix for the evaluation of the conditions of constituency. This partial revision will then be tested on different cases of code mixing already analyzed by Muysken (2000) and on selected examples from an Italian-German corpus collected in South Tyrol. The proposed model aims at observing code mixing phenomena in relation to the sociolinguistic context in which they occur, in order to achieve a better understanding on the process and factors involved in the interaction between grammars in language contact situations.