Ch.+23+by+Kees+de+Bot


 * CHAPTER 23**
 * COGNITIVE PROCESSING IN BILINGUALS: LANGUAGE CHOICE AND CODE-SWITCHING**
 * By Kees De Bot**

The main questions of cognitive processing are the following:
 * How we go from a communicative intention to articulation in language production?
 * How do we go from auditory and visual information to message interpretation ?

For cognitive processing in bilinguals, there is an addition question In bilingual language use there are two important mechanisms: language selection and language separation.
 * What mechanisms are needed to acquire more than one language?

In the study of cognitive processing in bilinguals there are different sources of data. One of them is data from code-switching study (CS). The study of CS provides us with an information on the way languages are used together and can help to understand the cognition of bilingualism.


 * Language production in bilinguals**

Levelt`s “speaking “ model (1989) is taken as a starting point in discussion of language production. It includes different modules//.// One of the modules is the //conceptualizer,// in which communicative intentions are translated into messages that have the role of input to the speech production system. Another module is //the formulator,//in which the preverbal message is converted into a speech plan by selection of lexicon and the application of grammatical rules. The last module is //the articulator//, which convert the speech plan into actual speech. **Speech Production in Bilingual Speakers** Poulisse (1997) mentions the following factors that should be considered in a bilingual model: Poulisse (1997) argues that the incomplete L2 knowledge base and the lack of automaticity of L2 speakers can be solved by existing monolingual production models but the L2 traces in L2 speech cause problems. However, Paradis (1998) claims that neither switches to the L1 nor cross-linguistic influence (CLI) phenomena draw attention to the adaptation.
 * Basically L2 knowledge is incomplete as L2 learners lack certain vocabulary and rules as compared to L1 speakers. This leads them to avoid the structures they are not familiar with.
 * L2 speech is more uncertain and is full of errors and slips. Cognitive skill theories focus on the development of automatic processes that are difficult to acquire. Moreover,
 * L2 speech carries traces of the L1. L2 speakers have developed L1 system and may switch to their L1 deliberately (“motivates switches) or unintentionally (“performance switches”).

**Keeping Language Apart** Code-switching and keeping language apart are different aspects of the same phenomenon. Paradis (1981) proposed the //subset hypothesis// which accounts for most of the data found. He also suggests that words form a subset of the total inventory. Each subset is activated independently. The subsets are formed and kept by the use of words in specific settings. The main advantage of this hypothesis is that the set of lexical and syntactic rules is reduced due to the fact that a particular language/subset has been chosen. Thus, the subset hypothesis can explain how languages in bilinguals can be kept apart. Hence, this hypothesis proposes that bilingual speakers have stores for //lemmas//, //lexemes//, //syntactic rules, morphological rules// and //articulatory elements// that are not different from those of monolingual speakers. Within these stores there are subsets for different languages as well as for different varieties, styles and registers.


 * Language choice and CS**

CS employed as a meaningful discourse strategy and CS resulting from the lack of knowledge should be distinguished. Matching of parts from the preverbal message with the meaning part of lemmas is the most fatal as it is at this point that a shift from conceptualization to language-peculiar coding occurs. Due to Levlet (1993), there are three parts in lemma; //semantic specification, syntactic information// and //pointer to a certain lexeme//. The series of conceptual circumstances under which the lemma may be appropriately applied form the //semantic specification//. //The syntactic information// has to do with the syntactic category and the grammatical functions of a lemma. Most likely, the activated lemmas will “seek for” other lemmas which will fit. The //pointer to a lexeme// is the last kind of information in the lemma. The precise relation between the lexeme and the lemma is rather obscure. Tip-of-the-tongue research findings reveal that sometimes even after retrieving the lemma speakers are unable to find the lexeme related to it. Most studies on Cs deals with morphological and syntactic constraints, and just very few of them pay attention to switching phenomena at other levels in the process. Here I discuss the switching sites briefly. Here are some different sources of CS theories and data in order to know how this findings and models work with the idea of cascade of switching sites.
 * Language production and switching sites**
 * 1) The lemma selection: A misuse of misuse of the word at the conceptual level will cause th speaker to select the word from the other language.
 * 2) The syntactic structures activation
 * 3) The selection of lexemes : in this case morphemes should be combined e.g stem of the verb + marker (aspect or tense) for example, a verb stem from one language and the marker from the other language like “climbeden” (English V stem + Dutch marker).
 * 4) The change from address frame to phonological words. There are many examples of this case. The common one is the use of bilabial English /W/ instead of labiodentals /W/ by Dutch migrants in Austrailia.( De Bot and Clyne)