English Doorway Academy

Asslaam o Alaikum!
Anyone here from Islamia University of Bahawalpur....

Happy New year
Happy New Resolutions
Happy New Opportunities

Factors in Language Mixing/Switching

The term code switching (CS) to refer to the use of various linguistic units (words, phrases, clauses, and sentences) primarily from two participating grammatical systems across sentence boundaries within a speech event. In other words, CS is intersentential and may be subject to discourse principles. It is motivated by social and psychological factors.

The term code mixing (CM) to refer to the mixing of various linguistic units (morphemes, words, modifiers, phrases, clauses and sentences) primarily from two participating grammatical systems within a sentence. In other words, CM is intrasentential and is constrained by grammatical principles and may also be motivated by social-psychological factors.

Unlike the monolingual, who has access to only one set of mutually intelligible styles, the bi-/multilingual may exercise a range of styles in two or more varieties that are not mutually intelligible. The bilingual’s pragmatic competence enables him or her to determine the choice of one language over the other in a particular interaction. On the basis of a number of factors such as with whom (participants: their backgrounds and relationships), about what (topic, content), and when and where a speech act occurs, bilinguals make their language choice.

For most bilinguals, overt or explicit language negotiation is rather unnatural and counterproductive and can yield communication mishaps and sometimes alarming results. In most cases, the unmarked choice is made almost instantaneously and smoothly, which in turn results in language matching by participating bilinguals.

The rule of thumb is that LM/S marks a socio-psychological change of some kind. This change could be prompted by the presence of an out-group member, by constantly changing social relations or identities or by the need to create a special effect (in terms of socio-psychological, stylistic effects) within the sentence or beyond the sentence, i.e. within a discourse unit. In short, the following four factors determine language choice and mixing on the part of bilinguals: (1) the social roles and relationships of the participants; (2) situational factors: discourse topic and language allocation; (3) message-intrinsic considerations, and (4) language attitudes including social dominance and security.

(1) The social roles and the relationships of the participants

Participants’ roles and the dynamics of their relationships play a crucial role in bilinguals’ unconscious agreement or disagreement on language choice. Based on a mutual understanding of the obligations and rights of participants agreement leads to language matching, which in turn reflects the nature of their perceived social relationship.

(a) Language mismatching and repair

There are instances when the unconscious process of language negotiation does not work smoothly. This results in language mismatching and such mismatching occurs primarily under two conditions: (1) bilinguals are uncertain of each other’s language identity/backgrounds and are thus still in an exploratory mode, or (2) their preference for mutual identity through language shows a differing preference for identities, or degree of formality and power relations, among other possibilities.

Situation: For instance, in a South Asian restaurant, a bilingual waiter might choose to speak in English in spite of the fact that the customer has already made a move by choosing a shared South Asian language. The waiter might show preference for English rather than matching the customer’s language. This linguistic mismatch is the result of failure of the waiter and the customer to negotiate a preferred identity. The waiter’s choice of English shows his or her preference for educated identity and other perks (dominance, image of the restaurant, etc.) as opposed to seeking ethnic and regional solidarity. What is important is that the waiter’s choice is not a marked choice; it exhibits a difference in personal choice like that of choosing coffee over tea on a particular occasion or as an attempt to calibrate the dominance relation between the two participants. If the process of repair is not carried out quickly by arriving at a mutually agreeable code, the consequence is likely to be a widening distance between the two participants which can ultimately lead to deteriorating relations or even hostility, if each of the two participants sticks to his or her own choice.

(b) Dual/multiple identities, social distancing, and speech accommodation
Language mismatching does not necessarily reflect a failure of language negotiation. On the contrary, it may exemplify an accommodation to circumstances. Consider the following situation. Suppose two bilinguals, one Telugu-English and the other Hindi-English bilingual, are conversing with each other. The language of the conversation must naturally be English, the shared language of the participants. A third bilingual, a mutual friend of the two participants, joins them by greeting the Telugu-English bilingual in Telugu. The greeted bilingual does not greet back in Telugu, but chooses to respond in English instead.

The language mismatch in this situation signals the intention on the part of the greeted bilingual to stick to the code already agreed upon between the Telugu-English and Hindi-English bilingual. The choice of English in this instance is viewed as a “neutral” strategy and reflects speech accommodation on the part of the greeted bilingual, unlike the case of speech divergence in the restaurant example. Since the three participants are mutual friends, the phenomenon of language mismatching does not result in any ill-feeling toward the greeter. The mismatch is the result of an optimization process: while ethnic identity is already achieved through Telugu, the switch to English on the part of the greeted is an invitation to join in the formation of the mutually beneficial collective identity among all three participants. The greeter accepts this by responding in English. This shows how effectively and efficiently bilinguals can modulate between two identities, with the result that social distance is narrowed as desired by the participants.

The language mismatch in such instances occurs in a group of diverse participants and usually it takes place at either the beginning or the closure phase of the interaction.

(2) Situational factors: discourse topic & language allocation

In bi-/multilingual societies, languages generally do not overlap each other’s discourse domain. Some languages are viewed as more suited to particular participant/social groups, settings or topics than others. For instance, very often bilinguals organize their two languages according to their public vs. their private world. The public language often serves as the “they” code and code can be used to perform a range of functions, from creating distance, asserting authority, and expressing objectivity, to suppressing the tabooness of an interaction. The private language often serves as the “we” code and it conveys a range from in-group membership, informality, and intimacy, to emotions.

Example: Richard Rodriguez (1982) notes that Spanish was the language of his private world. His Spanish voice insisted, “We are family members. Related. Special to one another,” whereas English sounded loud, booming with confidence. The day the family decided to use English at home, family intimacy was not the same. In professional domains such as advertising, different languages carve their topical and social-psychological domains.

Social variables

Social variables such as class, religion, gender, and age can influence the pattern of LM/S both qualitatively and quantitatively. The social cueing of mixing is exemplified in (1) and (2).

(1) Hindi-Persian-English
(a) A. are cacaa jaan. aadaab., sab xairiyat hai, na?
‘Hey uncle. Hello. All is OK, isn’t it?’
(b) B. xush raho, sab xairiyat hai aur tumahaara kyaa haalcaal hai?
‘Be happy. All is fine. And how are you?’
(c) A. aapkii duaa hai
‘It is your kindness.’ (d)
B. kidhar jaa rahe ho?
‘Where are you going?’
(e) A. bhaaii jaan ne bulaayaa thaa. unke daftar jaa rahaa huun.
‘My brother called me. (I) am going to his office.’

(2) (a) A. are ankal jii namaste, sab Thiik hai, na?
‘Hey uncle. Hello. All is OK, isn’t it?
(b) B. xush raho, sab Thiik hai aur tumahaaraa kyaa haalcaal hai?
‘Be happy. All is fine. And how are you?’ (c) A. aapkii kripaa hai
‘It is your kindness.’ (d) B. kidhar jaa rahe ho?
‘Where are you going?’
(e) A. bhaiyaa ne bulaayaa thaa. unke aafis jaa rahaa huun.
‘My brother called me. I am going to his office.’

The exchanges in (1) and (2) mark LM/S with Persian and English, respectively. The two styles in (1) and (2) are in general not compatible. For example, though one might say cacaa jaan (with the Persian-Arabic honorific jaan) or cacaa jii (Hindi honorific jii) or ankal jii, the form *ankal jaan is not possible.

Even if the identity of the participants (A and B) is not given explicitly in (1) and (2), the pattern of LM/S provides sufficient indexical cues about the participants and their roles. The conversational roles played by participants are given in table 13.2 together with other roles played by the two aspects of mixing.

In many traditional societies, where gender roles are clearly demarcated, i.e. men work outside the home and women are engaged in domestic activities, LM/S in women is qualitatively different from that in men (e.g. lack of LM/S with English among women). Similarly, age plays an important role in determining the nature of LM/S; see, for example, the case of second- generation Turkish immigrants

Consideration of situational factors such as shifting personality, thoughts, audience and topic can further promote language alternation. This behavior can be well exemplified here by Kipling’s portrayal of a Tibetan Lama in his novel Kim, which is renowned for the author’s mastery of insights into the verbal repertoire of multilingual India.

(3) Message-intrinsic factors

LM/S is also a function of additional linguistic and pragmatic considerations such as those outlined below.

(a) Quotations
Direct quotation or reported speech triggers LM/S among bilinguals cross- linguistically. This function has been attested by a wide variety of empirical studies. The following example illustrates this function.

Urdu-English

From a conversation between two women discussing about their daughters study. One while referring to her daughter says the following:

My daughter was not good in Maths during mid-term exams, but she says, wo final term mei cover kr lay gi.

My daughter is not good in Maths, but she says that she will cover it in the final term exams.

(b) Reiteration

Reiteration or paraphrasing marks another function of mixing. The message expressed in one language is either repeated in the other language literally or with some modification to signify emphasis or clarification. The following examples illustrate the emphatic and clarificatory role of mixing, respectively.

English-Spanish: Chicano professionals (Gumperz 1982, p. 78)
A: The three old ones spoke nothing but Spanish. No hablaban ingles.
‘. . . They didn’t speak English.’

English-Chinese (Mah, 1997, p. 104)
The moment my aunt saw the jacket, she knew that Miss Chen was dis- honest and had fu zhong lin jia (scale and shell in her belly).

(c) Message qualification

Frequently, mixing takes the form of a qualifying complement or argument as exemplified by the disjunctive argument and the adverbial phrase respectively in the following sentences.

English-Urdu
‘Will you take coffee? ya chay?’

(d) Topic–Comment/relative clauses
Related to the function of message qualification is yet another function – the Topic–Comment function.

For example, in the following sentence English LM/S revealed that the topic is introduced in English and the comment is given in Hindi, as shown in the following example:

English-Hindi
The boy who is going meraa dost hai.
my friend is
‘. . . is my friend.’

In English-Hindi LM/S, the English sequence the boy is understood to be a Topicalized NP with two associated Comments – the English who is going and the Hindi meraa dost hai.

(e) Hedging

LM/S serves an important function in hedging (e.g. taboo suppression, de-intensification, or a vague “sort of” expression). Although the formal and functional range of hedging is quite wide and both languages of a bilingual can contribute, the language which is allocated as the “they” code is often used for this purpose, particularly when hedging performs the function of taboo suppression. This aspect of LM/S is often deliberate and is by and large a conscious process. The passage below concerns a veterinarian doctor’s attempt to explain to villagers in rural India the process of artificial insemination. When the listeners fail to understand the English term, the doctor attempts to explain the concept by paraphrasing it figuratively into Hindi. The dilemma regarding how to suppress the tabooness via Hindi is clear from his hesitation and halting speech.

Doctor to villagers (Hindi-English)
“artificial insemination.” Dekho ise kyaa kahte hain hindi men . . . barii aasaan ciiz hai . . . jab bhains garam ho rahii ho . . . to use AI Center le jaaiyee aur uskaa AI karva Daaliye
“Artificial insemination.” Look, what do people call it in Hindi . . . It is very easy (to explain). When a buffalo is in heat, take her to the AI Center and have her artificially inseminated.” (Literally: have AI done on her)

(f) Interjections

Another function of LM/S is to mark an interjection or sentence filler. Bilinguals in Singapore are well known to exploit this function by mixing a number of particles, as in (12). Interlocutors A, C, and D are as described for (2) above.

English-Hokkien (Tay 1989, p. 416) D: Do what?
A: System analyst la
‘System analyst, what else?’ C: hà
‘Is that so?’
A: Programmer la.

(g) Idioms and deep-rooted cultural wisdom

English-Chinese (Mah, 1997, p. 238)
The solution is simple – yi dao liang duan
‘The solution is simple – let’s sever this kinship with one whack of the knife.’

(4) Language attitudes, dominance, and security

Other factors such as individual and social attitudes, language dominance, and linguistic security determine the qualitative and quantitative properties of language mixing. On the basis of differences in language attitudes, one can postulate primarily four types of language-mixing communities.

For instance, Puerto Rican bilinguals in New York City unconsciously have positive attitudes toward bilingualism and LM/S; consequently, they tend to code-mix/switch 97 per- cent of the time and this mixing/switching is smooth, as observed by Poplack.

English-Hindi mixing in India and Punjabi-English mixing in Great Britain follow this pattern. When such mixing becomes the mark of cultural or social identity, the speech community begins to view mixing positively at both the unconscious and the conscious levels. This may further promote the incidence of LM/S both quantitatively and qualitatively. As a case in point, extensive Arabic-French-English LM/S occurs in Lebanon, where trilingual mixing is viewed as a distinctive feature of Lebanese culture.

The third type of attitude toward mixing can be exemplified by the Ottawa- Hull speech community. In Hull the English-French mixing is one-third to one-fourth as frequent as in Ottawa. The metalinguistic commentary offered by bilinguals and accompanied by flagged mixing provides evidence that bilinguals view mixing covertly in negative terms although consideration of linguistic accommodation at the conscious level leads to flagged mixing.

The last type of bilingual, who considers bilingualism negatively on both conscious and unconscious grounds, is exemplified in the speech patterns of Flemish and French bilinguals in Brussels. Because of the long history of linguistic rivalry and conflict in Belgium, one witnesses Flemish-speakers and French-speakers entrenched in their own language while talking with each other and no attempt is made to switch or mix the two linguistic systems.

Within each community, variables such as gender and topic add further intricacy to language mixing. In India, male and female attitudes toward mixing are quite divergent. Males prefer mixing with English whereas women do not. Women’s role in the preservation of Indian culture goes largely uncontested by men. Therefore, the incidence of language mixing with English is much lower in women than in men. Unlike the Western English-speaking speech community, where men’s speech is regarded as standard, in South Asia, women’s speech is considered more standard than men’s. However, in the discussion of certain topics such as children’s education, one can witness the phenomenon of over-compensation of mixing with English among women. That is why female Indian writers often poke fun at excessive or odd mixing on the part of middle-class women. A case in point is women’s use of phrases such as childrens kii education ‘the education of children,’ and childrenon kii education ‘the education of children.’ In the first form, the speaker effectively reduplicates the English plural marker; in the second example, the speaker adds -on-, the Hindi oblique marker, to English children. Men will tend either to use children kii education or to prefer not to use English in the modifier position.

The speaker’s language proficiency and language dominance also determine the incidence and nature of LM/S. For instance, in South Asia English educators tend to code-switch with English more than balanced Hindi-English bilinguals.

Code Switching

Code-switching is the linguistic activity when bilingual speakers use more than one language within one, and the same conversational context (Appel & Muysken, 1987:117; Grosjean ,1982; Romaine, 1989). Researchers have come up with various reasons for what motivates bilingual speakers to code-switch. Two areas which have attracted widespread research are code-switching as a result of linguistic motivations, and the socially or psychologically motivated code-switching.

Sometimes Bilingual speakers deliberately use words or lexis from another of the known languages when they lack a particular word in the language being spoken, to compensate for a linguistic need. In other words the easier accessibility of a word lexicon in the other language motivates them to use it. This is what is referred to as the 'most available word phenomenon' (Grosjean 1982:151), as speakers consciously or unconsciously fall back to the easily accessible lexicon . Studies of second language speakers' (L2) use of communication strategies have shown that bilinguals often resort to their first language (L1) intentionally to solve lexical communication problems in the L2 (Bialystok, 1983; Poulisse 1990). Olsen (1999) cites several instances where Norwegian children learning English unconsciously code-switch to Norwegian as a compensatory strategy due to lack of appropriate words.

In line with this argument are models of some speech production that categorise bilingual languages in terms of their structural constraints within the speaker's lexicon. Myers-Scotton (1992) makes a distinction between languages within what she referred to us 'matrix language frame model'. In this model the language providing more morphemes for the relevant interaction type than the other languages used in the same conversation is the 'matrix language'(ML)( Myers-Scotton 1992:105). The matrix language plays the more dominant role in the conversation between bilinguals. The other language(s), which by comparison to ML have relatively fewer morphemes for that particular interaction are known as 'embedded language' (EL). Myers-Scotton (1982) argues that the speaker always accesses ML lemmas and builds the morpho-syntactic frames on the basis of the relevant information contained in those lemmas. However, it is really difficult to pinpoint whether the 'matrix language framework' consciously or unconsciously comes into play when a speaker switches codes.

Socially Motivated Code-switching

The socially motivated code switching, which is the most widely researched area has exposed numerous references. Socially or psychologically motivated code-switching may be practised when a speaker aims, in some instances, to emphasise their identity or group membership, or it may be that they want to mark a change of subject, to specify a particular addressee, to draw attention to a particular part of the message, to express certain emotions or to mark 'asides' from the ongoing discourse (Grosjean 1982: 149-57; Appel and Muysken 1987: 118-20; Giesbers 1989:28).

Some scholars have argued that most code-switching is intentional behaviour albeit without the speaker's awareness ( Nortier 1989:4). There are cases, however, where unintended code-switching occur as a result of language interference . These may be referred to as incidental language switches, slips of the tongue or accidental speech errors ( Poulisse & Bongaerts 1994: 37). As mentioned already, L2 speakers often resort to L1 intentionally to solve lexical communication problems in the L2 (Bialystock 1983). They may switch unintentionally, however, when L1 words are easily accessible in the place of the appropriate L2 ones .

It would appear that there is an intersection between the linguistic and social motivations for code-switching. Myers-Scotton's (1982) 'matrix language' (ML) model imply that some languages are more dominant than others within a conversational context and, she also claims that language codes are 'indexical' of social relationships (1989). In the latter case, through language code, a speaker is established as a certain kind of person in relation to others. She claims that language code specifically indexes a particular set of rights and obligations that will hold between participants in an interaction. In this regard, a speaker will select a code that indexes the rights and obligations that he/she wishes to be in force between himself and others.

Myers-Scotton has identified different patterns of code-switching based on the notion of 'markedness'. An 'unmarked'choice means an expected choice, one that is associated with the type of interaction in which it occurs. This is an attempt to redefine relationship (Myers-Scotton 1989:334). She describes recorded instances of 'marked' and 'unmarked' choices of code-switching . The two examples, recorded in Kenya involves two friends and four young office workers. In the first instance, a Principal visits a friend who works in a car sales company. The Principal speaks Swahili to a guard at the gate, but switches to English when talking to the receptionist at the same organisation. At the friend's office the two friends, who speak one L1 switch codes 'unmarkedly' from Luhya (L1) to Swahili and sometimes to English. She argues that language in this instance is a mark of social identity. In the first instance, the Principal speaks Swahili to the guard at the gate because he 'places' the guard among the social category of those who speak Swahili but are not 'educated' enough to be able to speak English. The receptionist, on the other hand belongs to another social category, that of those who can speak English.

Marked choices, on the other hand may serve different functions. Among in-group members marked choices may, for instance, encode solidarity among a small number within the group ( Myers-Scotton 1989 :336) as the case of the young office workers illustrates. Four young office workers in the same government ministry in Nairobi are chatting. Two are Kikuyu, one is a Kisii and one is a Kalenjin. Swahili-English switching has been the unmarked choice when suddenly the two Kikuyu persons switch to their language. The conversation which was about setting up a group 'emergency fund' suddenly stops when the Kikuyu switch to their language to make a disparaging remark about what has been said. This is a marked choice communicating solidarity between the two Kikuyu but distancing them from others. This action motivates the Kisii to complain in Swahili and English, and the Kalenjin makes a switch from Swahili to English , a marked choice, to return the discussion to a more business-like plane (Myers-Scotton 1989: 336).

In other examples, marked codes may result from switching which are associated with emotion, social status or authority . In those instances, switches often encode more social distance between participants, sometimes out of anger or a desire to lower the addressee or increase one's own status. Codes-witching in this category is related to and indicative of group membership in particular speech communities (Auer 1998). The extent and the regularity with which they use two or more languages within a conversation may vary to a considerable degree between speech communities.

This marked choice is usually associated with authority, more commonly in former colonial regimes where the coloniser's language such as English was a mark of power (Myers-Scotton 1989 ). In all these activities the interlocutors are undertaking communication strategy to 'compensate' for a 'social' or 'linguistic' inadequacy.

Code-switching as Contextualisation

Code-switching studies have also looked at strategic activities of speakers in varying their communicative behaviour within a socially agreed matrix of conventions, which are used to alert participants in the course of the on-going interaction to the social and situational context of the convention. Conversation participants appear to exploit variable spoken language elements at all linguistic levels ( Local 1986; Local et al 1986) and at non-verbal level ( Duncan 1969, 1972; Kendon 1977) to contextualise their suppositions. According to Gumperz (1982:132-135) contextualisation conventions or contextualisation cues function to signal participants' orientation to each other.

As an example, Chinese/ English bilingual speakers switch languages to contextualise preference organisation and repairs ( Weir & Milroy 1995: 296). By building a contrast in language choice for two stretches of conversation , the speakers are able to draw attention to details of the projected course of conversation and to check each other's understandings. This is relevant, particularly in contextualisation preference organisation. Preference organisation refers to ranking of alternative second parts of the so-called adjacency pairs, such as acceptance or refusal of an offer or agreement or disagreement with an assessment (Levinson 1983; Pomeranz 1984).

Wei & Milroy (1995: 281-299) demonstrate this in their study of code-switching among three generations of a Chinese community in the North Eastern part of England. In one context B offers her assessment of A's new dress- ho leng 'very pretty' . A's response to this consists first of a 'reflective' question in Chinese - leng me ? 'pretty ?'. This type of question is formed by partial repetition plus question marker me and has discourse similar to English tags such as 'isn't it?' 'really?', suggesting that the interaction functions as a 'hedge' heralding a further dispreferred assessment of dress, and indicates only a qualified agreement with B's original assessment ( Pomeranz 1984). When B asks for confirmation in the following turn gua a guai a? 'expensive or not? , A's preferred response is in Chinese the same language as B's question.

Sometimes code-switching is used primarily to contextualise imminent completion of a turn or talk or topic shifts, but at other times they have the capacity to signal meanings such as irony or seriousness, and social identities and attitudes of the participants. Auer (1984, 1991) has argued that bilingual code-switching should be analysed as a contextualisation cue, because it works in many ways like other contextualisation cues. However, code-switching has some characteristics of its own in addition to those it shares with such elements as gestures, prosodies and phonological variables. In particular, the sequential organisation of alternative choices of language provides a frame of reference for the interpretation of functions or meanings of conversational code-switching.

Code-switching for Political and Economic Reasons

Language choice and shift may also be due to political and economic reasons. People recognise that the official language becomes the vehicle of political participation and socio-economic mobility (Myers-Cotton 1993a:28). The competition among groups for primacy of one language over others, or at least parity with others is based on the supposed superiority of a language. If ethnic group's language become official, its members have a head start , while others have to try and identify with it. On the other hand, many nations, particularly those which were formerly colonised have always opted for their former colonial language choice or shifts due to its diversity and the fear of domination by others (Myers-Scotton 1983a). Thus, as already been illustrated, code-switching to a language such as English, French or Portuguese, for example, installs the speaker to a position of authority, power or social superiority over others in those multilingual communities formerly colonised .

The distribution and use of language choices in multilingual communities, therefore, can reveal not only the extent of stability of intergroup relationships, but also the ways in which the regulation of access to symbolic resources is tied to the regulation of access to material ones ( Heller 1992:123). Code switching in this instance, therefore, may or may not be conventional depending on the setting or context of the conversation. For instance, we have mentioned where code-switching is an unmarked expected behaviour , for example, among peer in-groups and where it is marked and intended to put down someone considered to be inferior.

In socio-political terms code-switching may represent part of a range of linguistic resources upon which people can draw to define the value of resources they control and to regulate access to them. In line with this argument, resources are distributed by specific groups in specific situations through the provision and evaluation, among other things, of symbolic, including verbal, performances (Heller 1992:123). A good example of this use of language is the French-English code-switching in a variety of settings in Quebec and Ontario Canada ( Heller 1992), where code-switching is used as a means of drawing on symbolic resources and deploying them in order to gain or deny access to these symbolic or material resources.

The understanding being developed above builds on Bourdieu's concepts of symbolic capital and symbolic market places, and Gumperz's concepts of speech economies and verbal repertoires (Bourdieu 1977, 1982; Gumperz 1982). In these instances code-switching is a means of calling into play specific forms of linguistic and cultural knowledge, forms which conventionally possess certain kinds of value (Heller 1992: 124). The value is linked to the extent to which these forms facilitate access to situations where other kinds of symbolic and material resources are distributed, resources which themselves have value based on prevailing modes of organisation of social life in the community and who controls them. The resources in question are not just those with concrete functional value but those related in more indirect ways to the methods people have of not only acquiring the basic things they need to survive, but also of acquiring various forms of power and solidarity ( Heller 1992: 123).

Finally, in relation to the linguistic motivation to codeswitching is the grammatical theory and how this structures and explains it. Muysken (1995:178) argues that formulation of this is crucial for research in linguistics as a scientific discipline. He thus poses a number of important questions that may help to explain how lexicon and grammar of a language structure code-switching. Some of the questions relate to the extent to which we can rely on properties of individual words, when we produce and comprehend utterances, and to what extent we can rely on general rules of the language we speak. Other important questions relate to whether we can reduce the differences between languages to lexical differences. Muysken (1995) proposes a universal explanation , for instance, when sentences are built up with items drawn from lexicons from two languages. He proposes a model that believes there is a general set of constraints on code-switching, constituted, for example, by structural equivalence (Poplack 1980) or government (Discuiullo, Muysken and Singh 1986), or matrix language embedded asymmetry (Myers-Scotton 1993a).

In conclusion, it is clear that code-switching is a vast and complex linguistic area of knowledge. For instance, this discussion has illustrated the general and less complex cases of the practice whereby learners in L2 code-switch to their L1 unconsciously to compensate their poor grasp of L2 ( Olsen 1999; Grosjean 1982). The discussion then delved in greater detail into the socially motivated code-switching, where the concept of a matrix language and the idea 'markedness' is demonstrated in a conversation among bilingual speakers (Myers-Scotton 1982; 1989). Using Myers-Scotton (1982) explanation of how a matrix language (ML) dominates over embedded language within a conversation context , and the concept of 'markedness' the essay demonstrates how code-switching becomes a deliberate tool for bilinguals to perform certain linguistic 'acts', for example, that of showing their social positions of power, education or even to discriminate others. More complex sociolinguistic aspects of code-switching such as contextualisation and its use for political and economic reasons have been discussed. We have discussed how contextualisation in code-switching help to complete a conversation turn or talk or topic shifts, but how at other times they have they signal meanings such as irony or seriousness, and attitudes of the participants ( Wei and Milroy 1995).

An important sociolinguistic discussion of this essay has been how code-switching is practised for political reasons. We have seen how there is competition among groups for primacy of one language over others based on the supposed superiority of a particular language. This essay considered how when an ethnic group's language becomes official, its members are assumed to have a head start , thereby motivating others to try and identify with this language. Furthermore, many nations, particularly those which were formerly colonised have always opted for their former colonial language choice or shifts because it is believed that it is diverse, and they also fear being dominated by others (Myers-Scotton 1983a).