Variation and language engineering in Yoruba-English code-switching

This study deals with the identification and characterization of the variable features of code-switching as used by Yoruba-English bilinguals in Lagos, Nigeria. Against the background of the domestication of English in Nigeria and the reality of language variation across individual, social and generational perspectives, we explore the unique linguistic strategies with which the Yoruba-English speaker reinvents urban speech to meet the challenges of contemporary dynamics of communication in a fast changing world. 1.0 Introduction The concept of language variation has remained a prominent theme in sociolinguistic enquiry by virtue of its centrality to the explication of the social context of language use. Since no speech community can be said to be completely homogenous, the fact of language variation remains a glaring reality as exemplified in everyday uses of language in different societies. Firth (1951: 78) had stressed the fact that language must be as varied as the groups who use it and the multiplicity of functions to which it is applied. Similarly, Coates (1990: 24) in delineating the domain of sociolinguistics as the social context of language use, argues that ―the study of language in its social context means crucially the study of linguistic variation‖. Consequently, sociolinguistic studies have been largely characterized by the exploration of the systematic relationship between language and socio-cultural organization of speech communities. The basic assumption behind this is that speakers functioning as members of a particular speech community, and within the ambit of a particular culture, have internalized not only the rules of grammar but also the rules of appropriate speech usage. These rules which are broadly shared by other members of the speech community are applied daily in speech behaviour (Sankoff, 1989). To this end, Chambers (1995: 15) defines Emmanuel Adedayo Adedun and Mojisola Shodipe 104 sociolinguistics as the study of the social uses of language, encompassing a multitude of possible enquiries. These include questions about personal, stylistic, social and sociocultural patterns of language use in society. In this regard, sociolinguistics can be said to share the goals of the ethnography of communication (Saville-Troike, 1982) which takes language as a ̳socially situated‘ cultural form. This direction gives prominence to the analysis of the code and the cognitive process of its users. Furthermore, sociolinguistics and the ethnography of communication are united in their focus on the speech community, the systematic organization and patterning of communication within it, as well as the interaction of these communicative events with other systems of culture. The scope and focus of ethnography of communication underlie the significant contribution of these perspectives to sociolinguistic research, particularly in the description and analysis of naturalistic speech in various social contexts. Thus, while the sociolinguistic perspective of this study explores the dynamics of language variability, the ethnographic dimension deals with the socio-cultural organization of communicative events such as ways of speaking, social norms and values (Saville-Troike, 1982, 1989). Our ethnographic approach in this study will also reveal how socialization processes influence the social differentiation of language behaviour from one generation to another. In this regard, there is need to emphasize the relevance of the internal structure of the urban Lagos Island speech community to the linguistic material which emanates from it. The internal structure of the Lagos speech community includes the various modes of socialization (e.g. greetings, forms of address, dressing, and inter-group relations) as well as cultural world view as expressed at both individual and communal levels of social interaction. Against the background of the notion of communicative competence (Hymes, 1964, 1974), naturalistic data of language alternation emanating from the Lagos Island speech community are used in this study to account for differential performance at the individual and group levels. Moreover, in view of contemporary directions in sociolinguistic studies of bilingual behavior (Auer, 1991, 1990, 1984; Sankoff 1989; MyersScotton 1993;), the social context of speech usage finds prominence in this study in terms of the patterns of bilingual speech production in the daily interactions of Yoruba-English bilinguals. Therefore, as a Variation and language engineering in Yoruba-English 105 sociolinguistic study, this paper explores the systematic interrelations between the language users, the linguistic form and the social context of speech usage. Our sociolinguistic perspective in this study thus deals with the significance of the social context as a crucial component of language alternation in this cosmopolitan speech community. The term ̳language engineering‘ refers to the potential of a language to express new and emerging ideas, notions or concepts. Capo (1990) defines language engineering as: that domain of applied linguistics concerned with the design and implementation of strategies (i.e. conscious and deliberate steps) toward the rehabilitation and optimal utilization of individual languages (1990:1). Language engineering is therefore conceived as a conscious attempt to influence the form of a language and this implies three phenomena that are related to lexical change (Ammon, 2005: 26). These are: 1. Standardization of pronunciation, spelling, and the meaning of words. 2. Creation of new names from organizations whose acronyms create easily pronounceable words and are semantically related to the organization‘s aims. 3. Public use of language. However, language engineering in this article involves both the conscious and unconscious use of two bilingual languages for communicative interactions and the resultant lexical change arising from everyday communication. Thus, languages are constantly engineered to meet the challenges of everyday communication often necessitated by changes in the social, economic or political life of a speech community. Dadzie (2004: 68) notes that every human language is subject to change and several factors responsible for this may range from the historical to the cultural and the linguistic. The English language underwent significant changes as a result of successive invasions of English territories by Saxons, Normans, Danes and the French. So great is the influence of these incursions that the English which was spoken in the 9 th century bears no resemblance to the present day English. For example, Old English seo eaxl is a far cry from its modern equivalent shoulder (Dadzie, 2004). The Nigerian situation typifies what obtains in many Anglophone West African countries where English gained ascendancy over the numerous ethnic languages as an official lingua franca. The people acquire it as a means of responding to several sociolinguistic needs which include the use of English as a medium of education and as the Emmanuel Adedayo Adedun and Mojisola Shodipe 106 language of politics, commerce and even religion. Naturally some localization must occur, since the language must reflect its new environment and portray ideas which did not exist in its original home. The Nigerian experience is thus characterized by the development of a variety of English which has unique local colouring in the form of the infusion of ideas and concepts from the indigenous languages. Thus Nigerian English (Dadzie 2004a, 2004b; Okoro 2004; Adetugbo 1976, 1980; Adegbija 1987, 1988, 1989, 2004) has become established as a variety of ―English as used by Nigerians‖ (Okoro, 2004). Okoro (ibid: 169) further categorizes the features of Nigerian English into four as follows: (1) Common-core features: These are features shared with other English varieties worldwide and include syntactically and semantically neutral sentences like ―Good morning‘, ―I am going home‖, etc. (2) Peculiar Nigerianisms: These include loan words such as ‘agbada‘, ̳iroko‘, ̳garri‘; coinages such as ̳bride-price‘, ̳boysquarters‘, ̳cash-madam‘, ̳head-tie‘ and ̳area-boy‘; category shifts such as ̳to flit a room‘ (to spray with insecticide), to Xerox a document‘ (to make a photocopy of), ̳to tipp-ex an error‘ (to cover with correcting fluid), and meaning broadening in words such as ̳customers‘ (referring to both buyer and seller in Nigerian English whereas in native Speaker English, it is restricted to the buyer).We must however point out here that usages like ̳to tipp-ex‘ and ̳to xerox‘ are also common in British and American English and are therefore not peculiar to Nigerian English usage. We can however argue that these usages represent examples of the influence of Americanisms in contemporary social interactions of the Nigerian speaker. (3) Local Idioms: (including modifications of existing nativespeaker idioms). These include the following: – Don’t put sand in my garri. (don‘t ruin my chances) Variation and language engineering in Yoruba-English 107 – You met me well/your legs are good. (inviting someone who has just arrived to join in a meal) – She used long leg to obtain the job (she obtained the job through undue influence and favouritism) – One tree can‘t make a forest. (BrE: One swallow cannot a summer make) – Cut your coat according to your size (BrE ... according to your cloth). (4) Characteristic breaches of the code (i.e. characteristic errors): Here, Okoro distinguishes between random errors (those that occur as part of an individual learner‘s interlanguage and are not necessarily shared by other users) and characteristic errors (those that are so regular and so widespread that they have come to be identified as part of the unique features of the language variety being described).


Introduction
The concept of language variation has remained a prominent theme in sociolinguistic enquiry by virtue of its centrality to the explication of the social context of language use.Since no speech community can be said to be completely homogenous, the fact of language variation remains a glaring reality as exemplified in everyday uses of language in different societies.Firth (1951: 78) had stressed the fact that language must be as varied as the groups who use it and the multiplicity of functions to which it is applied.Similarly, Coates (1990: 24) in delineating the domain of sociolinguistics as the social context of language use, argues that -the study of language in its social context means crucially the study of linguistic variation‖.
Consequently, sociolinguistic studies have been largely characterized by the exploration of the systematic relationship between language and socio-cultural organization of speech communities.The basic assumption behind this is that speakers functioning as members of a particular speech community, and within the ambit of a particular culture, have internalized not only the rules of grammar but also the rules of appropriate speech usage.These rules which are broadly shared by other members of the speech community are applied daily in speech behaviour (Sankoff, 1989).To this end, Chambers (1995: 15) defines sociolinguistics as the study of the social uses of language, encompassing a multitude of possible enquiries.These include questions about personal, stylistic, social and sociocultural patterns of language use in society.In this regard, sociolinguistics can be said to share the goals of the ethnography of communication (Saville-Troike, 1982) which takes language as a ‗socially situated' cultural form.This direction gives prominence to the analysis of the code and the cognitive process of its users.
Furthermore, sociolinguistics and the ethnography of communication are united in their focus on the speech community, the systematic organization and patterning of communication within it, as well as the interaction of these communicative events with other systems of culture.The scope and focus of ethnography of communication underlie the significant contribution of these perspectives to sociolinguistic research, particularly in the description and analysis of naturalistic speech in various social contexts.Thus, while the sociolinguistic perspective of this study explores the dynamics of language variability, the ethnographic dimension deals with the socio-cultural organization of communicative events such as ways of speaking, social norms and values (Saville-Troike, 1982, 1989).Our ethnographic approach in this study will also reveal how socialization processes influence the social differentiation of language behaviour from one generation to another.In this regard, there is need to emphasize the relevance of the internal structure of the urban Lagos Island speech community to the linguistic material which emanates from it.The internal structure of the Lagos speech community includes the various modes of socialization (e.g.greetings, forms of address, dressing, and inter-group relations) as well as cultural world view as expressed at both individual and communal levels of social interaction.
Against the background of the notion of communicative competence (Hymes, 1964(Hymes, , 1974)), naturalistic data of language alternation emanating from the Lagos Island speech community are used in this study to account for differential performance at the individual and group levels.Moreover, in view of contemporary directions in sociolinguistic studies of bilingual behavior (Auer, 1991(Auer, , 1990(Auer, , 1984;;Sankoff 1989;Myers-Scotton 1993;), the social context of speech usage finds prominence in this study in terms of the patterns of bilingual speech production in the daily interactions of Yoruba-English bilinguals.Therefore, as a sociolinguistic study, this paper explores the systematic interrelations between the language users, the linguistic form and the social context of speech usage.Our sociolinguistic perspective in this study thus deals with the significance of the social context as a crucial component of language alternation in this cosmopolitan speech community.
The term ‗language engineering' refers to the potential of a language to express new and emerging ideas, notions or concepts.Capo (1990) defines language engineering as: that domain of applied linguistics concerned with the design and implementation of strategies (i.e.conscious and deliberate steps) toward the rehabilitation and optimal utilization of individual languages (1990:1).
Language engineering is therefore conceived as a conscious attempt to influence the form of a language and this implies three phenomena that are related to lexical change (Ammon, 2005: 26).These are: 1. Standardization of pronunciation, spelling, and the meaning of words.2. Creation of new names from organizations whose acronyms create easily pronounceable words and are semantically related to the organization's aims.3. Public use of language.However, language engineering in this article involves both the conscious and unconscious use of two bilingual languages for communicative interactions and the resultant lexical change arising from everyday communication.
Thus, languages are constantly engineered to meet the challenges of everyday communication often necessitated by changes in the social, economic or political life of a speech community.Dadzie (2004: 68) notes that every human language is subject to change and several factors responsible for this may range from the historical to the cultural and the linguistic.The English language underwent significant changes as a result of successive invasions of English territories by Saxons, Normans, Danes and the French.So great is the influence of these incursions that the English which was spoken in the 9 th century bears no resemblance to the present day English.For example, Old English seo eaxl is a far cry from its modern equivalent shoulder (Dadzie, 2004).
The Nigerian situation typifies what obtains in many Anglophone West African countries where English gained ascendancy over the numerous ethnic languages as an official lingua franca.The people acquire it as a means of responding to several sociolinguistic needs which include the use of English as a medium of education and as the language of politics, commerce and even religion.Naturally some localization must occur, since the language must reflect its new environment and portray ideas which did not exist in its original home.
The Nigerian experience is thus characterized by the development of a variety of English which has unique local colouring in the form of the infusion of ideas and concepts from the indigenous languages.Thus Nigerian English (Dadzie 2004a(Dadzie , 2004b;;Okoro 2004;Adetugbo 1976Adetugbo , 1980;;Adegbija 1987Adegbija , 1988Adegbija , 1989Adegbija , 2004) ) has become established as a variety of -English as used by Nigerians‖ (Okoro, 2004).Okoro (ibid: 169) further categorizes the features of Nigerian English into four as follows: (1) Common-core features: These are features shared with other English varieties worldwide and include syntactically and semantically neutral sentences like -Good morning', -I am going home‖, etc.
(2) Peculiar Nigerianisms: These include loan words such as 'agbada', ‗iroko', ‗garri'; coinages such as ‗bride-price', ‗boysquarters', ‗cash-madam', ‗head-tie' and ‗area-boy'; category shifts such as ‗to flit a room' (to spray with insecticide), to Xerox a document' (to make a photocopy of), ‗to tipp-ex an error' (to cover with correcting fluid), and meaning broadening in words such as ‗customers' (referring to both buyer and seller in Nigerian English whereas in native Speaker English, it is restricted to the buyer).We must however point out here that usages like ‗to tipp-ex' and ‗to xerox' are also common in British and American English and are therefore not peculiar to Nigerian English usage.We can however argue that these usages represent examples of the influence of Americanisms in contemporary social interactions of the Nigerian speaker.
(don't ruin my chances) -You met me well/your legs are good.
(inviting someone who has just arrived to join in a meal) -She used long leg to obtain the job (she obtained the job through undue influence and favouritism) -One tree can't make a forest.
(BrE: One swallow cannot a summer make) -Cut your coat according to your size (BrE … according to your cloth).
(4) Characteristic breaches of the code (i.e.characteristic errors): Here, Okoro distinguishes between random errors (those that occur as part of an individual learner's interlanguage and are not necessarily shared by other users) and characteristic errors (those that are so regular and so widespread that they have come to be identified as part of the unique features of the language variety being described). Examples:

Random Error
My father told me to told you to come.
This is considered random because the pattern is not widespread and the speaker is not likely to be consistent in his/her faulty double marking of tense, and can be easily corrected.

Characteristic Error
Buy your stationeries here.
This is characteristic in the sense that it displays the peculiar Nigerian English feature of pluralizing non-count nouns.Similarly, many African languages have had their fair share of colonial influence occasioned by contact with European languages like French and English.In the Nigerian situation, Yoruba along with Hausa and Igbo has the status of a national language.The history of contact between English and Yoruba accounts for the process of assimilation and acculturation (Akere, 1987).This led to a ‗reinventing' of the linguistic repertoire of the Yoruba-English bilingual speaker to include English loan words and assimilated forms.Ufomata (1991) in her article ‗Englishization of Yoruba Phonology' observes that the adoption of certain loanwords from English has effected a fundamental change in the phonological system of Yoruba.This includes the violation of the restriction on the occurrence of high tone on the first syllable of Yoruba vowel-initial words as in the examples: In these examples, it is evident that stress in English words is converted to a corresponding set of tonal patterns when borrowed into Yoruba.
Another feature of the influence of English on Yoruba in the use of loan words is the establishment of pitch and segment correspondences between the two languages.According to Ufomata (2004), in most instances, loans simply take on these correspondences while consonant clusters which are absent in Yoruba phonological system are resolved by epenthesis or deletion as in the examples: barber (bábá), soldier (sójà), half penny (pronounced /eipni/) (éékpìnì), street (títì), kettle (kétù), bicycle (báísíkù) (2004:580-581).
At the semantic level however, there are notable exceptions to the correspondences discussed above.In such cases, Yoruba tonal patterns actually keep meaning apart in homonymous English loans e.g.

‗baby'
[bèbí] pretty young lady [bébì] baby ‗cocoa' [kókò] cocoyam [kòkó] cocoa ‗party' [pátì] political party [patí] party, social gathering ‗father' [fádà] male parent [fadá] reverend father ‗sister' [sístà] reverend sister [sistá] older female From the foregoing, it is shown that the connection between language variation and language engineering in bilingual situations is essentially a function of the socio-cultural context of language use.In this paper, we are concerned with the various ways in which the Yoruba English bilingual in the Lagos cosmopolitan setting explores the linguistic potentials of the two codes in the task of reinventing bilingual speech to accommodate contemporary nuances of everyday communication.

Background
Much has been written about the forms and functions of English and indigenous languages in Nigeria.Scholars from both literary and linguistic realms have expressed differing views about the status of indigenous Nigerian languages vis-à-vis the overwhelming influence of English (Osundare, 2004;Ufomata, 2004;Bamgbose, 2004).Against the background of this important linguistic principle of language equality, many scholars have lamented the hegemony of English which according to Bamgbose (2004) is characterized by, among other things, increased power and prestige of English at the expense of other languages; the spread and domination of Anglo-American culture and positive attitudes and preference for English at the expense of one's own language.Bamgbose (2004: 2) further explains that all languages cannot possibly possess equal status in view of differences in language function.He then poses the question: if all languages are equal, why are some languages used in a wider range or domain?According to Bamgbose, differential values may be assigned to languages depending on a combination of sociolinguistic and economic factors.In order to consider these differential functions our study of Yoruba-English bilingualism explores the relative facility of the two languages to use linguistic innovation to meet the demands of contemporary usage.In this regard, Bamgbose (ibid) posits that: by concentrating on language structure and potentiality of language use, the linguist emphasizes language equality, while by concentrating on language function and language attitudes, the educationist and sociolinguist emphasizes language inequality.(2004:2) Thus the situation in Nigeria (like many African countries which are former British colonies) is characterized by the retention of English as an official language and the language of post-primary education.While the indigenous languages are restricted to primary education and the cultural domain, Yoruba, alongside Igbo and Hausa, has the status of a national language as prescribed in the 1999 Constitution of the Federal Republic of Nigeria.Along with Igbo and Hausa, Yoruba is also prescribed in the National Policy on Education (1977) as a medium of instruction in primary schools and to be studied as a second language in junior secondary schools in Nigeria.The Yoruba language is the mother tongue of a substantial number of speakers in South West Nigeria and is also spoken outside Nigeria in places like Republic of Benin (where it has the status of a national language) and Togo.Moreover, the Yoruba language retains its presence in the oral literatures of Yoruba descendants now domiciled in Brazil, Cuba and parts of the West Indies, Trinidad and Tobago.In Nigeria, where a majority of speakers live, it is spoken mainly in Lagos, Ondo, Kwara, Ogun, Ekiti, Oyo and Osun states as well as in parts of Edo and Kogi states.According to the 2006 census, the population of Yoruba speakers within Nigeria was estimated at thirty million, close to one fifth of the population of Nigeria.The same 2006 census put the population of Hausa speakers at about fifty four thousand and the speakers of Igbo at about seventeen thousand.
It is pertinent to say at this juncture that despite the entrenched functions of English and its subsequent preeminence in Nigerian polity, the Yoruba language has had an interesting history of growth and literary development of instruction at all educational levels.Yoruba has been explored in rigorous academic research at the tertiary level.Literary works abound in Yoruba in the three genres of poetry, drama and prose while all the genres of oral literature have equally been documented in the language.Furthermore, the use of Yoruba language in publishing, journalism and broadcasting has been a major boost to indigenous communication in Nigeria and beyond.Presently, there are efforts towards the development of a Yoruba language based computer system.
Like English, the Yoruba language has however had a chequered history of development as a dynamic medium of communication.Babalola (1972) had observed that one of the major problems of Yoruba is that of expressing new items or ideas introduced into the language through other languages in contact situations such as English, French, Arabic and Hausa.He suggested the use of neologisms to solve this problem as in the examples of: minute (ìséjú), lesson (èkó), glass (ife), matches (ìsáná), pencil (léèdì), etc. Yusuff (2008) however observes that it is part of the natural developmental process for the speakers of a language to devise means of expressing ideas and concepts which are alien to one's culture.He further notes that apart from deliberate efforts at lexical developments for formal use, the Yoruba language has the grammatical resources to create lexical items when faced with the challenge of innovativeness.Part of the dynamism of the Yoruba language is the shift from the original Oyo dialect base into what is now regarded as standard Yoruba (SY), the variety which is being presently used as a medium of instruction in schools, in literature and broadcasting.
According to Yusuff (ibid) SY is not a regional dialect but it is needed for harmony among the regional dialects.This variety can easily be described now as Common Yoruba (CY), that is, the spoken form which is moving away from the norms of Standard Yoruba.Common Yoruba is the variety being widely spoken in cosmopolitan settings like Lagos, Ibadan, Abeokuta where users possess native speaker intuitive knowledge of Yoruba and where linguistic challenges exist in everyday communication.Thus, Common Yoruba is widely spoken among students, artisans, teachers, broadcasters, writers, musicians, housewives, etc.
From the foregoing, it is evident that the urban variety (Common Yoruba) which operates alongside English (Standard English forms or Nigerian English variety) can be described as a variety of Yoruba which has the potential for being reinvented in contemporary usage.Moreover, the use of an urban variety of any language with its characteristic feature of marking the speaker as civilized or sophisticated and willing to embrace the dynamics of an evolving world, clearly places the Yoruba-English bilingual in Lagos as an innovative language user.

Methodology
The thrust of this study is the exploration of the linguistic resources at the disposal of the Yoruba-English bilingual and the various ways these resources are used to reinvent the two languages for the expression of contemporary dynamics of urban existence.In this regard, this study relies on a corpus of naturally occurring bilingual speech of Lagos city dwellers in different social settings.The speakers cut across all strata of society but are all united by the common trait of being Yoruba L1 and English L2 speakers with common history of acquisition of English at an early age.However, levels of proficiency in both languages vary based on level of education and length of exposure to both languages.Using the spontaneous interviews and the non-participant observation method, the researcher isolates forms (lexical, morphological or semantic) which exhibit the features of language engineering in the use of both Yoruba and English.
It has however been argued (Wolfson, 1976) that in the absolute sense, there is no such thing as natural speech such as the type obtained from tightly controlled structured interviews and the spontaneous interviews.Wolfson's argument is that the respondent is usually constrained to operate within the performance context of the questionanswer pattern of the interview format.Therefore, the respondent's responses in terms of content and style of delivery are considerably influenced by the cues given by the interviewer.Some scholars have also observed that Wolfson's position appears to take little cognizance of the great potency of the natural setting as a determinant of language or variety choice.In this kind of situation, emotions such as joy, fear, tension, anger or anxiety (which are normally exhibited in speech acts such as exclaiming and interjecting) are usually demonstrated in the more naturally occurring of the two codes or varieties in a speaker's repertoire.This kind of language use can be said to underlie not only patterns of codeswitching in bilingual speech, but also the choice of the elements used by the speaker.

Conceptual Consideration
The description of language, whether spoken or written, is the primary business of linguistics.Similarly, studies of language use in human societies belong in the realm of sociolinguistics.This paper explores the relationship between language variation and language engineering within the framework of bilingual behaviour in a non-native English environment.Our exploration of these areas also embraces the issue of code-switching and language change since bilingual behaviour, codeswitching and borrowing represent some of the crucial features of the sociolinguistic influences which promote language change worldwide.
By the term ‗code-switching', we refer to any kind of discourse in which words originating in two different language systems are used sideby-side.Gumperz (1972Gumperz ( , 1982) ) had described the phenomenon of codeswitching as ‗the juxtaposition within the same speech exchange of passages of speech belonging to two grammatical systems or subsystems'.In the same vein, Li Wei (2003) describes bilingual codeswitching as the alternation of languages in the same interactional episode.Gafaranga and Torras (2002) however argue that the concept of code-switching has traditionally been understood to mean any occurrence of two languages within the same conversation.They further identify code-switching as a specific type of language alternation which they describe as ‗interactional otherness', a term which refers to a kind of internally motivated deviance (that is, serving a specific interactional function) (Gafaranga and Torras 2002).
In contemporary sociolinguistic research however, perhaps the most dominant feature of the literature on the subject is the diversity of opinions about what constitutes an adequate definition of the phenomenon.While some linguists (cf.Myers-Scotton, 1993) consider the term ‗code switching as being synonymous with ‗language switching', others like Romaine (1989) insist on using the term in the same sense as Gumperz (ibid) initially used it.Poplack (1980) however defines code-switching not only in relation to discourse but as also being inclusive of the phenomenon of code-mixing.According to Poplack, -code-switching refers to the mixing, by bilinguals, (or monolinguals) of two languages in discourse, often with no change of interlocutor or topic‖.While asserting that such mixing may take place at any level of linguistic structure, Poplack notes that considerable linguistic attention has however been focused on the occurrence of code-switching within the confines of a single sentence, constituent or even word.Similarly, other sociolinguistic scholars have defined code-switching as the alternative use of two languages at the word, phrase, clause or sentence level and have also called for a distinction among the different language phenomena.For instance, Auer (1984Auer ( , 1991Auer ( , 1998) ) distinguishes codealternation from what he refers to as the ‗new code' and in turn distinguishes code-switching from transfer within the ambit of code alternation.
Among scholars who use the term code-switching to describe any instance of language alternation, there has been the need to establish its different types based on the dynamics of its usage.To this end, Gumperz (1982) identifies situational, metaphorical and conversational codeswitching while Myers-Scotton (1993) distinguishes between marked and unmarked code-switching.From a discourse perspective, Romaine (1989) also identifies tag-switching, inter-sentential code-switching and intra-sentential code-switching and states that all three types of codeswitching may be found within one and the same discourse.
These various perspectives underscore the fact that the diverse postulations of scholars on the appropriate definition of code-switching should be viewed beyond the realm of differing terminologies.These different views and categorizations of code-switching are actually based on different theoretical orientations and on different views about the notion of language and that of the code in social interactions.Gafaranga and Torras (2002) assert that these viewpoints reflect different epistemological orientations.To this end, they describe the views of Gumperz and Myers-Scotton as ‗identity-related explanation' since they are interested in the social values of language and the social motivations for code-switching, respectively.Auer's view on the other hand represents what is described as a sequential perspective, a viewpoint which is based on Auer's conversation analysis approach to the subject.
At this point, it is worth noting that linguists have observed some note-worthy situations in the social interactions of bilinguals worldwide.It is quite possible for instance, to find that social interaction among bilinguals may not always be conducted in two languages.In fact, many researchers have observed that talk among bilinguals may be conducted in one language only.This has been variously described as preference for same language talk (Auer, 1984) and as the preference for same medium talk (Gafaranga and Torras, 2002).When this occurs, the interaction is usually not considered worthy of notable accounting.Rather, it is seen as the norm among both monolinguals and bilinguals.This has been referred to as the monolingual bias (Gafaranga and Torras, 2002) which has been accounted for in terms of the monolingual medium.
The second situation which has been observed involves cases where bilinguals use both of their languages without any obvious motivation (Labov, 1972).This differs from monolingual language use in the sense that it is usually noticed by both researchers and community.Linguists commonly refer to this practice as ‗a new code' (Auer, 1984), as ‗codemixing' and more technically as ‗unmarked codeswitching' (Myers-Scotton, 1993).This kind of talk has also been recognised by members of speech communities as a ‗stable' practice (Garfinkel, 1972) and thus speakers have been known to assign labels like Spanglish, Chinglish, Franglais and Yorubanglish to such speech patterns as a recognition of the specific nature of these different types of language alternation.This takes us back to the issue of the monolingual bias as the underlying factor for speakers' recognition of language alternation.The phenomenon of language ‗switching' or ‗mixing' is noticed because it is generally assumed that the norm is to speak in only one language.In this regard, Auer notes: ….In many bilingual communities, there is a preference for same language talk, codeswitching runs counter to this preference which of course only heightens its signalling value … (1991: 28-29) Our exploration of bilingual behavior in this paper also embraces the issue of code-switching and language change.It has been observed that bilingual behaviour, code-switching and borrowing represent some of the crucial features of the sociolinguistic influences which promote language change worldwide.Backus (1996: 27) defines ‗language change' as ‗contact-induced structural change'.In other words, changes in the structure of a language as a result of language contact.One of the major structural changes in contact situations can be observed in word order.Thus it operates on the hypothesis that word order change in bilingual situations is caused by frequent code-switching (Backus 2005)  However, some linguists believe that formal changes in a language can be stated at various levels of abstraction.Backus (ibid) thus argues that it is rather imprecise to say that -word order‖ has changed in a given language.We do not know for instance whether it is -basic word order‖ (cf.Whaley 1997) that has changed or word order in a specific construction, such as topicalization or the change from declarative to interrogative.Therefore, before accepting a form as ‗new', it is necessary to show that it was not part of the language all along.According to Backus, the clearest evidence of this of course would be the absence of the feature in the pre-contact variety, but this type of evidence is not always easy to obtain in syntax.Moreover, Nettle and Romaine (2000) observe that many languages which are currently heavily dominated by another language (such as Yoruba) do not have monolingual speakers anymore who can act as a yardstick, while pre-contact records have often disappeared or are non existent.
Many scholars have noted that all theories of language change are skeptical about what can be achieved by way of predicting the future of languages (Croft 2000, Weinreich 1953).To this end, Matras (2000) poses the question: -how predictable is contact-induced change in grammar?Matras' answer is that one can predict for certain subsystems of grammar what course change will take, or at least make ‗an intelligent guess'.Johanson (2002) however argues that the idea of looking for causes of language change has its pitfalls chiefly because causation of language contact takes place at various levels and many factors interact.Thus, while some description of change address structural factors (e.g.gaps in the system), others address the type or mechanisms of the change (e.g.reanalysis), yet others the social context in which the change arose (e.g.language contact or desire for upward mobility).
Backus (ibid) further observes that a relevant consideration in this regard is the distinction between ultimate and proximate causes.Ultimate causes of language change are likely to be socio-cultural (Kontra, 2001;Thomason and Kaufman, 1988) while proximate causes include cognitive, attitudinal, motivational and probably, purely structural factors.
Similarly, Thomason and Kaufman (1988) make a distinction between ‗predictors' (social and linguistic factors that drive change), ‗mechanisms' (the processes through which change is effected) and ‗results' (the visible effects on the changing language.Backus (ibid) observes however that the distinction between predictors and mechanisms though useful, is not always easy to make.
In view of all the factors involved, it is evident why it is difficult to arrive at a generally accepted theory.In addition to this there is the issue of whether contact induced change should be separated from a general theory of change, mainly because bilingualism is one of the more important social factors promoting change and internal and external factors often conspire in this type of change.
Furthermore, Backus (ibid) notes that in addition to direct importation (borrowing), there is also indirect change induced by the circumstances of contact, but this does not involve actual borrowing.Indirect changes are especially likely to involve a combination of internal and external causes (Romaine 1989;Thomason 2001).

Lexical
The Yoruba-English bilingual situation in cosmopolitan Lagos represents the kind of speech where speakers alternatively use Yoruba and English words or phrases for a variety of reasons.This kind of linguistic ability may not incessantly be available to speakers with less proficiency in both languages.At the lexical level, the Yoruba English bilinguals adapt their urban speech for the achievement of the following: 1.

Personal identity
Personal or individual identity implies identification of the self within the large space of the society, at specific period and within groups to which people belong.This is most exemplified in the personal naming system whereby these bilinguals reinvent themselves through the prevalence of code-mixed personal names, a trend which cuts across age and gender divides.For instance, people are generally identified by the use of a Yoruba first name followed by an English appellation which indicates professional or occupational identification.Thus the common sociolinguistic pattern of personal naming in this community values names like: Ibrahim Casket, Taiye Vegetable, Sule Mandilas, Wale Teacher, Musibau Escort, Shamsideen Developer, Azeez Councillor, Musiliu Coach, Tunde Parking, Nurudeen Sergeant, Fausat Cellular, Kunle Entertainment.
Apart from indicating an acknowledgement of the prestige norms associated with English as an international language, this naming pattern captures the ways of life of speakers in terms of their specific preoccupations.Thus, the code-mixed names capture the essence of contemporary usage where speakers attempt to attach some level of prestige or sophistication for themselves through personal naming.
However, some code-mixed names also serve different functions for the users.Certain members of the community come to have certain names or nomenclatures attached to them by others by virtue of their individual personal traits, behaviour or idiosyncracies, whether positive or negative.Such distinctive personal traits are usually embedded in the English component of these bilingual names.They include: Gbolahan Computer: a smart person, a cunning smooth operator who always seems to have everything figured out (like a computer).
Tunde Banana: a slippery, deceitful person.(slippery as a banana peel) Mustapha Emotion: a gentle, lovable and genial person.
Bisi Dollar: a high flying society lady who travels abroad frequently for business and is well-known for her penchant for spending foreign currency.(e.g.dollars) Fatai Always: a man who is ever so resourceful and willing to explore any means to make money.
Names such as these operate on the principle of social acceptance or rejection, approval or disapproval as the case may be.But implicit in all these is the fact that the naming system provides insights into individual character in relation to the social norms of a specific period and in some cases, neighbourhood dynamics.The practice of identifying people by catchy, trendy or symbolic names also includes the prevalence of nicknames.These include: a.
Relexicalised versions of personal names such as: Rosco for Rasaq, Owoblow for Owolabi, China for Shina and Murphy for Mufutau.

d.
Clipping: This involves using the shortened forms of personal names as in the examples of FASH for Fashola, KUSH for Kushimo, LAI for Olayiwola and TOKS for Tokunbo.
Generally, these coinages serve many communicative functions which include the quest for social identification, group solidarity or the need to disguise one's identity.They represent the various ways in which individuals and social identities are mediated by bilingual behaviour.

Group Identity
Language features and the various identities they portray generally imply a boundary function.The peculiar sociolinguistic structure of the Lagos speech community means that individuals find their relevance in group identity.This means that people perceive themselves and are perceived by others as more relevant individually when they are also recognized as part of a group.Therefore everyone strives to be defined within a group structure.The bilingual communal norm is thus also reflected in the naming of social groups to indicate sophistication and a willingness to follow the trend of contemporary usage.Thus, we have the preference for Yoruba-English group names against monolingual Yoruba or monolingual English names as in the examples: Ricca Gents, Inabiri Ladies, Waka Club, The Great Offin Gents, Balogun Yuppies, Fila Connection, New Generationext of Olowogbowo, Upper Olowogbowo Gents, etc.Some group names are however retained in English but with phonological and/or orthographic modifications as in the examples: Sunday Skool for Sunday School, Faya One for Fire I.

Semantic
At the semantic level of usage, the Yoruba-English bilinguals of Urban Lagos attempt to reengineer their speech with the following strategies:

Devernacularisation
This involves the use of lexical items in code-switching discourses in forms which deviate from their original meaning and usage in the indigenous Yoruba language.Devernacularisation in this case is a strong indicator of generational variations in bilingual behaviour whereby the members of the younger generation exhibit a preference for inventive, creative usages while the older generation prefers more conservative usages.Thus, devernacularisation represents a feature of outer marking (Mdiase Tham, 1990) or in-group marker for youths.It takes the form of a restricted code or in-group slang which is used to exclude others not considered members of the group.For example, Yoruba words like -isu‖ (yam) and ‗eran' (meat) are devernacularised when used as pseudonyms for money in expressions like: (i) Awon boys yen se isu seriously ni council.

literal translation:
(Those boys make yam seriously in the council)

Meaning:
Those boys are making a lot of money in the council.
(ii) A ni ki leader ya eran fun awon party members.

literal translation:
(We asked the leader to cut some meat for party members)

Meaning:
We asked the leader to give money to the party members.
Such expressions however retain their original meaning and usage in the speech of the older generation of Yoruba-English bilinguals-e.g.
(Those boys made money today) A ni ki leader gbe owo fun party members.
(We asked the leader to give money to the party members).
Other examples of devernacularised usages in the bilingual speech of youths include the following: Example 1: Ojà 1 : (Youth: hard drugs) Ó gbé ojà lo si London ni wón bá mu ni airport.
(He/she travelled to London with hard drugs and was arrested at the airport).
Translation: Is there any (illicit) business today?
Translation: We are about to engage in some illicit business.Meaning: He/she was arrested by the police.
Meaning: He/she was arrested by the twins.

Meaning:
We shall open fire on those people.
Meaning: We shall switch on the light (electricity) for those people.

Relexification
This occurs when youths jettison original usages and replace them with new expressions in relexified form.This represents a form of divergence from adult speech which is usually a reflection of community norms.Like devernacularisation, the use of relexification in the bilingual speech of younger people serves as an in-group marker and is motivated by factors like secrecy, group solidarity, self protection and peer influence.
It occurs in either of the two codes.e.g.Meaning: He arrived the venue with power dressing.

Nativisation
The speech of many educated young bilinguals often exhibit the use of certain code-mixed expressions which have the form of nativised English usages.The original English expressions usually have some of their elements translated to Yoruba to create an entire new code. e.g.

English Yoruba
Have a nice day/weekend (e) ni nice day/weekend Are you alright/ok?Se o wa alright/ok?Some of these nativised expressions have a surface structure of Yoruba and an English deep structure having being derived from core English fixed expressions or idioms.Otherwise known as slang, these usages are common features of the speech of the younger generation e.g.Ó ń bé: He /She jumped/ took a leap forward.

Morphological
This involves the adaptation of English loan words to the morphology of the borrowing language (i.e.Yoruba) e.g: (i) Yoruba lai (meaning negative) + English verb as in: A ko ni lo lai discuss oro naa.
We shall not leave without discussing the matter.
Conversely, we also observe some cases of the adaptation of Yoruba words and expressions to the morphology of English whereby an English morphological feature is affixed to a Yoruba word to form a new Yoruba-English expression.Some speakers have described such usage as ‗Yorubanglish' (Garfinkel, 1972)  (As the robbers stopped his car, he stepped out of the car meekly and surrendered his car keys to them).
Most of these expressions are often confined to slang usage, in-group registers or relexicalisations for the purpose of solidarity or to show emotive content of bilingual speech.

Phonological
Some lexical features of Yoruba-English code-switching also exhibit considerable linguistic integration at the phonological level.Established loan words are marked by phonological integration mostly in the speech of the educated and non-educated speakers.This is essentially a function of speakers' bilingual ability or the peculiar sociolinguistic situation e.g.

Error Features in Yoruba-English Engineering
It is worth noting that most of the examples cited above represent bilingual language use of educated speakers.Speakers of English in Nigeria have been classified into three groups on the basis of their linguistic competence.Speakers are classified along a language continuum as acrolectal, mesolectal and basilectal respectively.This categorization is important for the assessment of the linguistic content of bilinguals' social interaction in terms of the notion of communicative competence highlighted earlier.While many Yoruba-English bilingual of urban Lagos fall into the category identified as acrolectal Nigerian speakers, some can be classified as belonging to the mesolectal group.Acrolectal speakers are those with tertiary education while mesolectal speakers generally possess school leaving certificate.The group identified as basilectal are at the extreme of the continuum and are characterized by relatively low educational attainment such as primary school education or in the extreme cases, no education at all.Yet everyone acquires the second language, English, and use it communicatively as part of the general bilingual speech norm of the community.Differential levels of usage can however be established when the notion of communicative competence is applied.It has been observed that different categories of speakers draw from the community speech repertoire based on their linguistic competence.As a result, aspects of the naturalistic speech of our respondents exhibited varying features of basilectal speech.Using Okoro's (2004) typology, Some of these can be categorized as follows: a. Characteristic errors: -A n lo si night vigil lola.(We ought to have witnessed many new innovations by this democratic government by now).
-A insist pe ki awon soja return back si barracks won.Democracy la fe! (We insist that soldiers return back to their barracks.We prefer democracy!).
(Can you borrow me some cash?I will pay back tomorrow).
(If you come late again, you shall be relieved of your job) -Landlord ti da properties wa jade.A need accommodation badly bayii.
(The landlord threw out our properties.We desperately need accommodation now) b.Random Errors: This is exhibited in varying ways.They include wrong use of words as in the example: -Eelo ni e lè avoid (afford) láti san fún aso yen? (How much can you afford to pay for the dress?) -Mo try best mi lati convince (persuade) obinrin naa lati pada wa.(I tried my best to persuade the woman to come back) -Nígbàtí àwon olópàá dé, anybody (everybody) bé dànù ni! (When the police arrived, everybody disappeared!)

Conclusion
In this paper, we have examined some of the salient sociolinguistic features of language engineering by Yoruba-English bilinguals at the lexical, semantic and morphological and phonological levels.Essentially, we have demonstrated that both English and Yoruba exert considerable influence over one another and that the various strategies for language engineering are supported by factors in the speech usage.Moreover, our examples have shown clearly that Yoruba English bilingual engineering finds a place in the explication of the unique sociolinguistic components of language change in contemporary communication.
ń bé (he is too brash) To jump the gun.As in relexification and devernacularisation, these usages retain their vernacular meanings in the speech of older literal translation of Yoruba-English speakers as in the examples: Má si/ Má be: Do not open.ó já: It broke loose.
characterized by the speaker's superimposition of indigenous pronunciation patterns on the phonic representation of English loan words.

(
we are going to a night vigil tomorrow) -O ye kí á ti máa rí orísirísi new innovations ni government alágbádá yìií.