Analyzing the case markers of Persian noun on Azerbayjani Turkish speakers’ speech, based on 4-M model

This paper attempted to study the case markers of Persian noun in Azerbaijani speakers’ speech on the basis of the 4-M model. Based on the understanding that in bilingual  speech, one of the languages is normally dominant over the other, Myers-Scotton (1993, 1997) formulated the Matrix Language Frame (MLF) model, in which she classified morphemes in complementizer phrases (CP) into two kinds: content  morphemes, such as nouns and verbs,  and system  morphemes, such as inflections  and most function words. A major difference between content morphemes and system morphemes is that most content morphemes either assign or receive thematic roles, while system morphemes do not.
    The contact languages play different roles. The language with the dominant role, that  is the one that is used for the system morphemes, is called the matrix language (ML). The embedded language (EL) is the language from which content morphemes are inserted  into the ML. The MLF model claims that in mixed ML+ EL constituents, only the ML is used to build the frame. That is, the ML determines the morpho-syntax of ML+EL constituents. Modified versions of the model have appeared since then (Myers-Scotton, 1997; Myers-Scotton, 2002; etc.), and the definitions of some of the terms of the MLF model have been elaborated on or revised. 
    By extending the MLF model, a new morpheme categorization model called the 4-M model has been proposed (Myers-Scotton, 2002, 2006; Myers-Scotton & Jake, 2000). There are four categories of morphemes in the 4-M model, i.e. content morphemes and three types of system morphemes. This model is a refined version of the content vs. system morpheme opposition and explains how different morphemes are accessed in different ways in speech production. The heart of the 4-M model is the fact that system morphemes are activated at two different abstract levels. They are classified as early and late. Furthermore, late system morphemes are in turn divided into two types: bridges  and outsiders. Early system morphemes are activated at the lemma level together with their content morpheme heads for their maximal projection. Levelt defines “lemma” as the “nonphonological part of an item’s lexical information.” He also states, “lemmas are the driving force behind the speaker’s construction of the surface structure. It is in the lemmas of the mental lexicon that conceptual information is linked to grammatical function” (1989, p. 162). Unlike content morphemes and early system morphemes, late system morphemes depend on other types of information for their activation, and this information is only available at the level of the formulator, where language specific morpho-syntactic patterns must be realized. Information about all types of morphemes is present in lemmas, information about content morphemes and early system morphemes is salient at the conceptual level, and information about late system morphemes becomes salient at the positional level of the formulator. The model has been tested in many language contact situations with positive results. In some studies of language contact phenomena, these different system morpheme types have been shown to play different roles in language production. For example, in second language acquisition, early system morphemes are acquired first, and late system morphemes are more difficult to acquire. Of those few system morphemes occurring in mixed constituents in codeswitching, early system morphemes are most likely to appear in the EL (Myers-Scotton & Jake, 2000).   
    However, while no paper has focused explicitly on Persian- Azerbaijani  contact using the 4-M model; this paper attempts to do so. The research method of this study is descriptive-analytic. First, on the descriptive level, it characterizes the case markers of Persian noun in terms of their entries in the mental lexicon (i.e., lemmas). Evidence indicates that morpheme appearance order in receptive language would not be explained without characterization of morphemes themselves, and such a characterization depends on the formalization of a connection between the underlying abstract lexical entries in the mental lexicon and surface realizations. Second, the study goes beyond describing the nature of different types of morphemes to investigating the morpheme appearance order in receptive language. The data were collected from Azerbaijani resources and its speakers in Ardabil province. The case markers of Persian noun were classified into three types on the basis of 4-M: content morphemes (adposition), early system morphemes (vocative articles) and late bridge system morphemes (genitive marker). The data analyzing indicates some Persian vocative articles have been borrowed by Azerbaijani while some adpositions appear as code-switched elements in bilinguals’ speech and genitive marker lies between them. It seems there is a relation between different morphemes nature and their order of entrance into other language. Unlike system morphemes, content morphemes assigning theta roles are rarely borrowed by other languages. Early system morphemes are borrowed more easily than late system morphemes. Late system morphemes are activated later in the production process as required by the grammatical frame of the matrix language, while early system morphemes are indirectly elected at the same time that content morphemes are directly elected by the speaker’s intentions. This paper’s findings indicate the 4-M model efficiency in explaining the order of morphemes appearance (as borrowed or code-switched) in receptive language.