Emmanuel Adedayo Adedun

Visiting Scholar, King’s College,University of London


Mojisola Shodipe

 Department of English, University of Lagos, Lagos





This paper examines the phenomenon of language attrition as one of the major components of generational differences in the use of Yoruba-English code-switching in the Lagos speech community. It is argued that while L1 loyalty may be associated with the members of the older generation, the younger generation tends more towards a preference for English, the L2, leading to L1 attrition. Our study also shows  that this generational difference can be accounted for within the context of the social environment of bilingual speech behavior in this cosmopolitan Nigerian setting.

Key Words: Attrition, code-switching, variation, Yoruba-English, cross- generational 


In all parts of the world, there has been an increasing tendency among educated urban dwellers to bring up their children in a language other than their mother tongue, thus increasing the risk of abandoning the indigenous language. The typical Lagos dweller, whether educated or not, is no exception.  Various sections of the Nigerian society use English for varied communicative purposes. An interesting trend in many Nigerian  urban speech communities is the ascendancy of English  as  the first language in many homes whereby children grow up with English and only learn their indigenous languages  as a second language at different stages later in life. The mother tongue is thus reduced to a medium for communication with uneducated family members ( Akere, 1982; Omoniyi, 2004). The institutionalized status and prestige value attached to English as an indispensable tool of personal and global identity, as well as the multi-ethnic linguistic diffusion of urban existence have combined to establish English as the logical choice for most communicative interactions. As Greenberg (1963,1971) observes, once a language is established as a lingua franca, it rapidly acquires a momentum which puts other languages out of business. Myers-Scotton (1988a, 1988b, 1990, 1993) however opines that English retains prominence in many African countries largely as a result of its international status and an Anglophone colonial heritage. But more importantly, she points out that English and other former colonial languages (French, Portuguese) gained ascendancy in African communities as a neutral solution to the multiplicity of indigenous languages, none of which possesses sufficient dominance  to function as an official language. Myers-Scotton (1993) reports that one of the most interesting findings in a survey conducted in Lagos, Nigeria was that only 5 percent of the population are monolingual.  While most people speak more than one language, (usually an indigenous Nigerian language and a  foreign language) the vast majority speak  English as a  second language.


Language attrition

One of the commonest observations about bilingual situations where one language has more functionality and greater prestige than the other is the phenomenon of language attrition in the case of the second or ‘inferior’ language. Gal (1979) documented the sociolinguistic basis of second language attrition in Oberwart, Austria where two languages, German and Hungarian coexisted for centuries before a massive shift to the use of German by young people, especially women. This pattern of generational attrition is however not peculiar to stable bilingual situations such as that which exists in the urban Lagos speech community. By definition, stable bilingualism is a condition where two languages coexist, not necessarily as equals, but as complementary, each with its own functions, or as contextual equivalents (Chambers, 1995; Preston, 1989).

Although not much is documented on the neurological bases of language attrition, empirically, the phenomenon has been recorded in children who stopped using a language before a certain period (cf. Olshtain and Barzilay, (1991)) as well as in adults. Schmid (2002) had linked the notion of attrition of less-used languages to the rule of Pitres which predicts aphasia recovery for adult aphasics – at a significantly greater than chance level. Historically, the notion of attrition is rooted in the word of Ribot’s (1881) who propounded the theory that a language to which an individual is exposed at an early age may reveal itself in adulthood even when that language has ceased to be spoken for years. Similarly, Goral, Levy and Obler (2002) also affirm that anecdotal evidence suggests that re-immersion in an attrited language can result in relatively fast reacess to it.

Indeed, many linguists have linked language attrition to emotions. According to Kim and Starks (2008), bilinguals’ code-switching is often viewed as a discourse strategy to mark social or affective stance. Documented studies of code-switching behaviour have shown that bilinguals use their L1 as an unmarked code for intimacy, solidarity and emotional expressions, while their L2 is used to signal distancing and emotional detachment (Gumperz, 1982, Myers-Scotton, 1993).

Similarly, Dewaele and Pavlenko (2002) have also recorded observations of highly proficient bilingual authors who write in their L2 in order to avoid emotional associations with their L1. Bond and Lai (1986) have also reported on L2 learners who use their L2 to discuss emotional topics. Kim and Starks (ibid) point out that these instances of the linkages between bilingual code-switching and emotions are strong indications of subtle and complex relationship concerning language choice. In their view, the situation is more complex for late bilinguals undergoing both L1 attrition and L2 acquisition as their emotional/cultural attributes of L1 and L2 will naturally be in the process of on-going change (Dewaele, 2004).     

Furthermore, it has been observed  that cases of language attrition border on the bilingual continuum. Myers-Scotton (1993) also opines that bilingual speech showing language attrition coincides with code-switching in several ways, with the resemblances fading over time. According to her, speakers showing attrition typically will show a change in the language setting and the grammatical  frame – the matrix language.  

Code-switching and language attrition

Code-switching has been established as one of the most dominant features of bilingual behaviour in language contact situation worldwide (Gumperz, 1971, 1982; Myers-Scotton, 1993, 1995). When Gumperz (1982) first introduced the study of the phenomenon he referred to as code-switching, he defined it as ‘the juxtaposition within the same speech exchange of passages of speech belonging to two grammatical systems or subsystems’. But Gafaranga and Torras (2002:1) argue that the concept of code-switching has traditionally been understood to mean any occurrence of two languages within the same conversation.

In contemporary sociolinguistic research however, perhaps the most dominant feature of code-switching is the diversity of researchers’ opinions about an adequate definition for the phenomenon. While some (cf. Carol Myers-Scotton, 1993) consider the term ‘code-switching’ as being synonymous to ‘language switching’, others like Romaine (1993) insist on using the term in the same sense as Gumperz (ibid) initially defined it. Poplack (1980) however defines code-switching not only in relation to discourse but also as inclusive of the twin phenomenon of code-mixing. Code-switching refers to the mixing, by bilinguals (or multilinguals), of two or more 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 (ibid) notes that linguistic attention has however been focused on the occurrence of code-switching within the confines of a single sentence, constituent or even word. Valdes (1981) defines code-switching as the alternative use of two languages at the word, phrase, clause or sentence level. Some researchers have also called for a distinction among the different language contact phenomena. In this regard, Auer (1984, 2005) distinguishes code-alternation from what he refers to as ‘the new code’ and in turn distinguishes code-switching from transfer within the ambit of code alternation.

A crucial aspect of the study of code-switching remains the relationship between sociolinguistics, psycholinguistics and second language acquisition. The central concern of SLA is the description of inter-language, that is, the systems which develop during language acquisition. Preston points out that the psycholinguistic tradition itself suggests the centrality of linguistic description to the field of SLA. The psycholinguistic tradition states that the best model of the mental representation of a language is the best grammar of that language. Therefore the details of grammatical description must have psycholingustic relevance. Similarly, Cook (1993) subscribes to the view that linguistics and indeed sociolingusitics have relevance for the study of SLA. According to Cook, linguistic approaches to second language research involve dealing with minds that are acquiring, or have acquired, knowledge of more than one language, a state she describes as ‘multi-competence’. According to her:

The importance of second language research lies not in its account of the knowledge and acquisition of the L2 … isolation, but its account of the second language present and acquired in a mind that already knows a first….| (1993:3).

More recent studies of code-switching have focused on the psychological implications of knowing and using two languages. A major direction in this regard has been the upsurge in studies of language attrition. Kim (2005a, 2005b, 2005c, 2007) has examined first language use of young Korean-English bilinguals in a second language environment with specific emphasis on the issue of attrition.  Jaspaert, Kroon and Hout (1986) explored points of reference in L1 language loss while Hakuta and D’andrea (1992) investigated some properties of bilingual maintenance and loss in Mexican background high-school students. Although much work in this area tend to focus on such topics as attitudes, identity and motivation, the importance of emotions in L1 attrition is generally acknowledged. Kim and Starks (2008) have explored the role of emotions in L1 attrition and L2 acquisition in a group of 30 Korean- English L1 dominant late bilinguals in New Zealand. Their findings point to a shift from L1 to L2 and thus an increase in L2 fluency and a decrease in L2 accuracy.

In 1953, Weinreich suggested that emotional stress may cause interference during linguistic performance in situations involving language contact. Subsequent studies have confirmed a link between language proficiency and the bilingual’s emotion-related language choice (ERLC) (Pavlenko, 2003). Several studies on second language acquisition have addressed the relationship between language attrition and emotion empirically (cf. Dewaele & Pavlenko).  For instance, Schmid’s (2002) study found that German Jews exposed to a higher degree of dramatization had a greater degree of attrition in their L1 proficiency. Similar results are documented in Ben-Rafael and Schould (2007) research on Israeli immigrants. Those immigrants who had strong ideological motivation for the revival of Hebrew and considered their immigration as a ‘major life accomplishment’ were found to code-switch to Hebrew more frequently than those who had immigrated to Israel for life style and economic reasons.

Furthermore, Nicoladis  and Grabois (2002)  also report that there are few documented  cases of complete loss of a first language. Previous studies of language  loss have focused on children of immigrants( as in most studies highlighted in this section) and specifically  detail the loss of a minority language ( or some aspects of it)  while learning a majority language by school-age children. This phenomenon has been described as ‘language arrest’, a situation whereby children stop learning their minority L1. Pan & Gleason also argue that while the maintenance and development of a language from an early  age is essentially a function of exposure, at some point in human development it becomes difficult to lose  a first language completely as a result of lack of exposure. Moreover, in the case of adult bilinguals several studies have established effects of long term exposure to a second language on the mother tongue. Pavlenko (2003) equally reports the observation that adult bilinguals may undergo changes in their understanding of culturally specific concepts. In this regard, Nicoladis and Grabois (2002) advance a number of  variables  crucial for the determination of first language loss experienced by bilinguals who have had extensive exposure to a second language. These include the degree of competence in the L1  prior to exposure to the L2, importance of the L1 for the creation and sustenance of personal identity, as well as a number of affective variables which include  motivation for L1 sustenance and valorization of the L1.

More relevant to this paper are studies which have suggested that late bilinguals may experience changes in the way they view their L1 over time. It is argued that the L2 learning process in an L2 dominant environment may result in conceptual restructuring in the late bilinguals mental lexicon and that this may lead to a feeling of inadequacy when using the L1 to express certain feelings. To this end, Kim and Starks (2008) observe that due to conceptual transfer from the L2 to L1, bilinguals may feel that their L1 is inadequate to express their thoughts in an L2 linguistic frame hence a preference for L2 and the risk of L2 attrition.       


This study is a sociolinguistic investigation of the dynamics of bilingual behavior in the Lagos Island speech community, Central Lagos, Nigeria. The study is motivated by the imperatives of seeking empirical basis for  the explanation of first language attrition, one of the most dominant consequences of  second language acquisition and usage in a cosmopolitan setting. The phenomenon of first language attrition attracts our attention based on the significant insights it provides into the problem of linguistic imperialism occasioned by the imposition of English as an official language in Nigeria. The overwhelming influence and prestige status of English has given rise to numerous sociolinguistic and psycholinguistic  consequences for the various ethnic constituents of the Nigerian linguistic setting. Perhaps the most compelling of these issues is the imperative of the   preservation of indigenous languages worldwide  and the sustenance of the unique  African indigenous culture  richly exemplified in the s sociolinguistic situation of urban Lagos



A total of 159 Yoruba-English late bilinguals participated in this study. This figure accrued from a combination of respondents across two generations – youths (101) and adults (58). All the participants were born and raised in Central Lagos and spent most of their childhood in the community. They all had contact with English during elementary education and sustained the contact through adolescence (12-18) and young adulthood (19-30 years) while retaining their L1. L1 performance however varies from middle adulthood (31-49 yrs) where bilingualism is characterised by L2 solidarity which is maintained till late adulthood (50-65yrs). Conversely, the younger generation of bilinguals (12-30 yrs) exhibit divergence features by maintaining more English content in code-switching than their older counterparts. Generally, it is assumed that all respondents had mastered the command of Yoruba, the L1, at an early age and also  acquired the basic skills in English from secondary school age. The point at which bilingual behaviour is characterized by increased L2 use and decreased L1 use constitutes the point of language variation across these two generations of Yoruba-English speakers.

All participants were selected through a complex methodological procedure of probabilistic sampling involving multi-stage, systematic and stratified sampling methods. Research instruments used included a 40-item questionnaire designed to elicit demographic and linguistic information, spontaneous and structured interviews as well as participants observation during a short story telling task to test for variation in L1 usage. The bilinguals ages ranged from 15 to 35(youths) and from 36 to 69 years(adults) years and most of the respondents have lived in the community for at least ten years before moving to other parts of Lagos.


Data collection and analysis

All the participants completed a questionnaire to elicit personal and sociolinguistic data. They were also engaged in interview sessions which included questions on the traditional and contemporary life in Lagos. Essentially, the interviews were designed to elicit vernacular speech. Part of the interviews was the test of lexical usage to elicit data on L1 proficiency. All verbal data were audio-recorded. The instruments for the lexical test were:

i)                          A list of words for parts of the body written in English. Respondents were asked to provide equivalents of these words in their indigenous language. Responses of members of the two generational groups were compared to extract indices of L1 attrition.  Both versions were recorded in separate sessions. 

ii)                        A list of Yoruba words representing technical ideas or untranslatable items. Respondents were required to provide English equivalents of these words and the responses of the two groups were compared to extract indices of L1 attrition or possible loss. Both versions were recorded in separate sessions.

However, for logistic purposes, the word lists could only be administered on a small group of 20 respondents – 10 youths and 10 adults. This was to enhance ease and speed of data gathering as a larger group would have proved cumbersome especially as most of the interviews were conducted in crowded and rowdy locations.

Interview sessions were designed to elicit vernacular speech from respondents. Recorded sessions of youths and adults were compared for assessment of English versus Yoruba content of code-switching discourses of members of the two generational groups.  Proficiency measures were obtained by comparing percentage scores for Yoruba-English bilingual performance for youths and adults, against social indices like age, social mobility and social class. Percentage scores were used to establish youths versus adults L1 and L2 performance for lexical choice, lexical preference and lexical loss.

The questionnaire

The questionnaire was designed to elicit different kinds of information about the respondents. The questionnaire thus included questions about demographic characteristics of speakers, social mobility, language characterisation/ attitudes, among others.

Social mobility

Social mobility is considered an important determinant of language behavior in speech communities since speakers’ language contact and personal experiences have a role to play in the establishment of linguistic characteristics. In this regard, one of the requirements for the selection of respondents for this study is that they must have strong personal affiliation with the community by virtue of their birth and length of stay in the community. Generally, respondents must have been born or have resided in the community for not less than five years. Our frequency table shows that Central Lagos records the highest percentage for residence (61.6%) and place of birth (77.7%). Information on social mobility also indicates low mobility rate for both categories of respondents which indicates a level of cultural isolation and strong affinity with community norms. Linguistically, this factor lends credence to the speech of our respondents since it is quite possible  to deduce that their speech  would  exhibit less of external linguistic forms   and more of the  Lagos linguistic character. In this regard, our respondents can be regarded as ‘ideal informants’ (Chambers, 1995) whose bilingual  speech can be described as uniquely Lagosian. To further press home the issue of ideal informants and to further establish the respondents’ natural attachment to their indigenous environment, respondents were asked to identify the sociolinguistic parameters by which they define themselves as Lagosians. The percentage scores are as follows:  those who were born in Lagos and spent their childhood there (40)%; those born outside Lagos but have lived there since childhood (5.7%); those born in Lagos and have lived there all their lives (30.8%); those born outside Lagos and have only spent their adult lives there ( 5%) and those who have spent their lives in Lagos but no longer live there (8.8%).   From these, the core of ‘ideal informants’ is formed by two groups of Yoruba-English speakers: those who were born there and spent their childhood there (40%) and those who were born in the community and have lived there all their lives (30.8%). Hence the common denominator is being born in the community and having considerable affinity with the community from childhood. This is expressed as follows: a) community affinity established from childhood and sustained into adulthood in spite of external exposure and; b) cultural affinity established from childhood and sustained by community norms.

Language identification/ attitudes

The language identification enquiries sought to determine the status of the two languages employed in the bilingual speech of Yoruba-English Lagosians. The two distinct varieties were identified as Lagos Yoruba and Lagos English respectively based on their cosmopolitan peculiarities in the bilingual speech of Lagosians. The respondents established that Lagos Yoruba is a variety of Yoruba spoken all over Lagos cosmopolitan area (40.9%). Furthermore, since English functions as L2 and thus cannot be said to be domiciled in a non-native context, the logical question was to ask respondents which variety of English is actually spoken in this community. Lagos English (30.2%) was identified as the variety characterized by the use of Standard British English mixed with Nigerian English. Others are Nigerian English, Pidgin English and Nigerian dialects (17%) and Standard British English spoken by educated people (24.5%). This indicates that respondents characterize Lagos English as an indigenized or localized variety of Standard British English.

Language choice and language attitudes

Questions about language choice were designed to elicit information about everyday domains of English usage. For instance, we asked respondents how frequently they use English outside formal or official circles. Those who use English Always (28.9%), Often (31.4%) and Sometimes 28.3%) recorded the highest frequencies, indicating varying testimonies of the communicative uses of English for formal purposes. There is also the indication here of the varying social background of speakers. It is however pertinent to note that those respondents who claim to use English often (31.4%) outside formal situations far outnumber those who use English Always and Sometimes. This further buttresses respondents’ earlier assertion of the importance of English for social interaction in this community. What is gleaned from all these is that English is generally considered an indispensable tool for both formal and informal communication. Moreover, the very low frequency of those who Never use English (1.3%) informally shows that virtually all respondents acknowledge the communicative facility of English in all interactive situations. Language attitudes such as these are largely responsible for language choice in most communicative situations. Against the background of the widely acknowledged importance of English, we also asked the respondents to provide information on the motivation for the pervasive infusion of English in most Yoruba-oriented interactions. Respondents proffered differing views about social and psychological motivations for the infusion of English into Yoruba speech. Expectedly, 23.71% of respondents attribute this trend to the pervasive speech norm in the Lagos community whereby it is normal for speakers to interchange codes in bilingual situations.  In this regard, the issue of non-translatability of certain Yoruba expressions has been adduced as a major factor in language mixing and switching. Some respondents (32.8%) believe that it is virtually impossible to speak undiluted Yoruba because English fills the vacuum of untranslatability. This of course raises the serious issue of possible language loss or language attrition. It also poses the question: why would bilinguals have difficulty accessing versions of English expressions in their indigenous language? What category of speakers falls into this group? Are there factors in the language acquisition process responsible for this? Why should some speakers display better facility in their L2 than their L1 even for isolated lexical or syntactic expression? What does this portend for the future of the indigenous languages? Our frequency scores show that Yoruba-English code-switching is attributed to speech norms (23%) and the highest frequency (67%). It is also interesting to note that the issue of non-translatability shares prominence (21.6%) with speech norms, a motivating factor for language switching.

Social identity

The need for people to define themselves either narrowly or generally in society has been identified as a fundamental underlying cause of sociolinguistic differences (Chambers, 1995). In the Lagos speech community, there is the instinctive consciousness of every individual to establish and maintain a sense of belonging, to be relevant and to forge some bonding with others, usually people with whom they share similar traits, views and ideals. In this regard, several avenues exist for people of different demographic identifications to come together and express their social identity through group affiliations. Speech is a veritable tool in the establishment and sustenance of these social linkages. Group belonging is not just social practice for many members of this community, but a significant emblem of an individual’s social relevance and a demonstration of communal identity.  To this end, among the constituents of communal identification, intimations of social identity were a major interest in our questionnaire administration. Percentage scores for social identity shows that 50.7% of respondents belong to social groups. It was also established that religious affiliations (21.3%) are more prominent than political (8%), and cultural affiliations (8%). Generally we submit here that social identity is a significant tool of self-definition in this community.



The interviews focused on different aspects of social life in both the traditional Lagos setting and in contemporary times. They were mostly unstructured as the researcher realized the need to blend into the easy-going, informal lifestyle of the average  Lagosian. Many interviews were conducted on the streets and in the markets where other members of the neighbourhood could easily drop by and participate. Some of the interviews, especially those with the older respondents were however conducted in home settings.


There is considerable evidence in the literature of first language attrition to support the argument that variations in the bilingual speech of different groups are largely attributable to the loss of crucial aspects of the language. In the case of the Yoruba-English bilingual of Central Lagos, these attrition features are predominantly exhibited in the area of lexical variation and possible lexical loss as highlighted below.  

Lexical Variation

Investigations of first language attrition have indicated that one of the influential factors in L1 attrition in majority-minority language contact situations is the restricted domains of the use of the first language.  As a result of diverse interests of youths and their older counterparts, the social network of adults tends to be more complicated and multi-faceted, with the attendant implications for bilingual communication. Many findings on second language acquisition (Preston, 1989; Cook, 1993) support the claim that adults proceed through the early stages of syntactic and morphological development faster than younger speakers, a situation which is accounted for by cognitive advantage. On the other hand, childhood exposure to second language generally results in higher L2 proficiency than adolescent or adult exposure. To this end, language learning difficulties after puberty may be related to the social and psychological changes which an individual undergoes at that life-stage. Equally made clear is the fact that linguistically, adults are more influenced by the standard language as they jettison their peer group influences for the more conservative norms of the wider society. Since adulthood is essentially a period of social stability and is associated with task-oriented mentoring and leadership roles, the linguistic markers of this social group tend to be more towards the strong influence of socially conservative factors. This includes the need to achieve and sustain social and economic progress. According to Preston (1989), as language speakers get older and advance to adulthood, they move into wider and less cohesive social networks and are more influenced by mainstream social values and consequently greater convergence with standard language.      

The bilingual speech of youths is largely characterized by urbanized, metaphoric usages while the speech of adults generally reflects the original, conservative forms of the same expressions. In this regard, the speech of youths can be said to represent divergence norms and that of adults is essentially a reflection of convergence.  Our submission is that this convergence-divergence dichotomy has implications for L1 attrition. In the first instance, data from our study of the Yoruba-English bilinguals of Lagos would tend to indicate, with considerable lexical evidence, that the creative instinct of bilingual youths in this community is largely a response to the loss of significant features of the first language. Apart from long association with and prolonged use of English, many young bilingual speakers have lost touch with the cultural /ethnic nuances of speech usage in the contemporary world.

In this regard, language constrains access to different activities and to the formation of social relationships. Language can thus be seen as vital in the formation of group boundaries. In addition, patterns of language use reflect the shared experiences and background knowledge that underlie group membership and ethnic identity. In the following bilingual discourse patterns, our study shows distinct features of individual and group identification across generations:

Text 1:

  (Two young men are engaged in a discussion on extortion by the police)

Speaker A: Awon olopa Island yii ma ti wa terrible gan-an o! Mo kan park moto siwaju filling station, ki n to seju, won ti mu mi. Won ni mo wa ni obstruction. Ni won ba ni ti mi o ba fe lo si Station , kin lo mu October wa.

These policemen have become so terrible these days. Just because I  packed my bus in front of the filling station, they said I was constituting an obstruction and arrested me. They then demanded ‘October’ or they would take me to the station.

Speaker B: October ke?! Oppression niyen now. Ki lo wa se to to yen?

What? ‘October?’ But that’s oppressive! Is your offence that grave?

Speaker A: O ya mi lenu o. Gbogbo isu to wa lowo mi ko ju twenty faiba lo. Ni mo ba ni : “ Oga. A beg o, I no fit get October o, make I give you February. In the end, won sa ri  pe won gba  March lowo mi  ki won to release moto fun mi. mo lo ya isu yen ni. Omo, bimo se wa yii, mo gentle gan-an ni..

I wonder.! All I had on me was twenty ‘faiba’. So I said : “Oga, please I cannot get ‘October’. Let me give you ‘February’. In the end, they made sure they took ‘March’ from me. I had to go aborrowing. Right now, am so ‘gentle’.

There are many language features in the above extract which clearly suggest  that much of the content of this discussion have  restricted usage. Words like February, March and  October  do not stand for months of the year but are in fact examples of common slang  which are a major  feature of in-group language behavior of members of the younger generation. In the peculiar social context of speech usage in this community, names of months are used as pseudonyms for money . Since there is considerable secrecy involved in the extortion racket between motorists and the police, it is necessary to device   restricted nomenclatures for such communicative purposes. In this regard, ‘February’ being the second month of the year stands for two thousand naira, ‘March’ represents three thousand naira, while ‘October’ stands for ten thousand. This kind of usage can also be described as form of register variation where different segments of the community have different terms for referring to money or monetary transactions. Thus the slang word faiba is the in-group code for a particular naira denomination- ten naira note. Twenty faiba therefore translates to two hundred naira. Yet slang for money used in the above extract is the Yoruba word isu which literally means yam. The sociolinguistic import of such usage is reflected in expressions like: ‘Awon politicians nri isu bu ni Council’. (Politicians are making a lot of money in the Council).

The last sentence of the extract also expresses similar restricted usage in the expression; “Bi mo se wa yii, mo gentle gan-an ni” (literal translation: as you see me, I am very gentle). The idea of being ‘gentle’ represents the emotional state of someone who is in a poor financial condition. Thus being ‘gentle’ means being stripped of the usual animation and vibrancy of someone who is ‘loaded’ i.e. rich.

Apart from the cultural and generational indices of language use which are expressed in the above extract, it is pertinent to state more importantly that they also have serious implications for L1 attrition. In the first instance, the tendency of the younger bilingual speakers to diverge from adult speech norm is a significant indication of the possibility of language loss. It has been observed earlier that adults exhibit greater influence of standard language features. The farther the younger generation move away from established standard usages, the greater the likelihood of losing those features after prolonged usage. Moreover, the younger speaker who prefers to jettison conventional usages and create new restricted codes may have limitations expressing themselves effectively outside their social groups. This singular factor accounts for the wide disparity in the lexical usage of these two generational groups. It appears that the more the youths consciously separate themselves from adult speech norms, the farther they detach themselves from their own linguistic realities since the speech of adults generally represent the conventional speech norms of the community.    

Text 2:

( A group of adult speakers in their fifties are discussing happenings on the political terrain)

Speaker A: He came, o loun fe se politics……. With all those Danfo boys. He was just coming; just like that.

He came and announced that he was in politics…with all those ‘Danfo’ boys. He was just coming; just like that.

Speaker B: Sé àwon òré è niyen?

Are those his friends?

Speaker C: Ta ló mò fun? O dìgbà tí real race bá bèrè ka tó separate the wheat from the chaff.

Who knows? We shall separate the wheat from the chaff when the real race begins.

Speaker B: Se o ro pe politics je nnkan ti awon bonje le maa dabble si  ni?

Does he think politics is a game for social miscreants?

Speaker A: Ma daa lohun. Ki lo kan an lati maa contest fun Governor? Instead ko rora ko owo e, oun pelu iyawo ati omo , ki won lo fun summer holiday ni London, Paris ati Brazil, ki won lo gbadun ara won.  O kan n waste owo, o l’oun n se politics. Oju e maa to clear!

Don’t mind him. What is his business contesting for the post of the Governor? Instead of putting his money to good use by taking his family for summer holidays in London, Paris and Brazil and enjoy themselves, he is just wasting his money, calling himself a politician. He will soon learn a bitter lesson!

There are glaring features of lexical variation between Text 1 and Text 2 above. The pervasive use of slang and lexical creations is absent in this extract i.e. Text 2.  Rather, the bilingual speech of adults is exemplified by the use of conventional lexical items on their ordinary sense.  The Yoruba word for money, owó is used twice in this extract without an attempt by any of the speakers to give it any other name although the code-switching patterns are quite similar, with the characteristic infusions of English words in the Yoruba discourse. However, other elements of adult bilingual speech are apparent in the use of certain lexical items which are no longer in wide usage in the general sociolinguistic content of Lagos speech. The expressions Danfo boys and bonje both have the same referent: social miscreants. Usages like these are essentially characteristics of adult speech and serve as an indication of the older generation’s reliance on conventional usages. These were the original terms for present day slang area boys. Thus the use of these kinds of lexical items sets the speech apart as adult speech.     

Lexical change and lexical loss:

It was observed that lexical variation relies on diachronic influence in both youth and adult use of language. In other words, bilingual behaviour across the two generations is susceptible to language change in progress, and this is mainly a function of socio-cultural, political or ideological changes in and around the community. In terms of lexical change, we observed that certain lexical items undergo changes over time as a result of the demands of social expressiveness and the imperatives of day-to-day interactions. For instance, in the Lagos socio-cultural milieu, words like bonje (meaning: area boy, a local slang for social miscreants), koroma  (crewman), eepini ( half penny), and faria (farina) used to have wide currency but are now obsolete and mostly confined to restricted contexts such as the speech of the elderly.

Similarly, in the case of lexical loss, we observed that this trend is quite common in the speech of youths and conversely minimal in that of the older generation. This can be largely attributed to the convergence-divergence dichotomy between speakers in the two generations. Thus in an attempt to break away from community norms in some morpho-lexical speech patterns, the younger generation have lost many indigenous terms for referring to certain words, for example, parts of the body.  Our investigation shows that the younger generation speakers could not readily provide Yoruba words for body parts like eye-lashes, eyebrow, spinal cord, ankle, chin, wrist and jaw while members of the   older generation were more forthcoming with appropriate responses like:  ipenpeju (eye-lashes), irun oju (eye-brow), opa ehin (spinal chord), kokose (ankle), and agbon (jaw), respectively. This has serious implications for the future of the indigenous language, especially as most of the younger respondents had to extract the correct answers by ‘cross-checking’ with the adults’ responses during the questionnaire administration sessions (see tables in the Appendix). This is a clear indication of cultural distance because names of parts of the body have emotional features in their connections to idiomatic expressions in many Nigerian indigenous languages.

Furthermore, we asked our respondents to provide English equivalents of certain Yoruba words which are believed to be untranslatable but which actually have terms allocated to them in contemporary Yoruba language use. Many young respondents could not provide Yoruba words for terms like engineering, aviation, finance, and journal simply because these words are represented not by single lexical items, but explanatory phrases in Yoruba. Thus engineering is translated as èkó ìmò èro ( ‘the knowledge of machines’), aviation is translated as igbokegbodo oko ofurufu  (‘the activities of airplanes’)  and finance is translated as ètò ìsúná owó. However, the same respondents had Yoruba words for English  terms like:  newspaper (iwe iroyin), commerce (eto oro aje),  expressway (ojú ònà márosè) and committee (ìgbìmò alábe sékélé). This was not surprising because these expressions are freely used in grassroots radio and television broadcasts, and have therefore being transported to domains of wide usage in the urban Lagos sociolinguisticrepertoire.     

 Lexical variation:

Fig. 1 below shows varying statistical implications of the differences in lexical choices of youths and adults in bilingual situations. The result of comparative assessment of youth versus adult language performance at the lexical level has implications for many social variables like social network and social mobility.  In terms of social networks, we observe that people tend to inculcate the speech norms of the social groups towards which they would normally gravitate, hence the younger generation generally favours the use of slang and relexified forms while the older generation retains the indigenous versions of lexical items as in the following.
















Fig 1. Give the English equivalents of the following Yoruba expressions:



Fig.2:  Give the Yoruba expressions for the following:



Surprisingly however, Fig.2 shows more preference for indigenous usages than slang, another indication of the disparities in respondents’ claims and actual speech behaviour. This however does not negate the fact that youth and adult speech usage is characterized by differences in lexical choices as shown in Table 4.3.3. (see Appendix)


Discussion and conclusion

This paper reports on bilingualism and L1 attrition in code-switching discourse of two generational groups in Lagos, Nigeria. We have shown with considerable evidence from naturalistic speech of Yoruba-English bilinguals in this community that there is indeed observation of variation in the use of code-switching strategies by youths and adults in this urban community. The different aspects of this variability are found to be a direct consequence of L1 attrition in the younger generation.

Our study shows that while some bilinguals prefer to use their L2 as a means of social sdistance, others simply have lost considerable proficiency in their own indigenous language and tend to make up for their language loss by exhibiting a preference for English. While this portends an uncertain future for the status of indigenous African languages, it is however significant to note that language attitudes have a role to play in the choices made by speakers in different interaction settings.

Since our results suggest that L1 attrition is a function of generational differences and by implication varying perceptions of the L1 by speakers, it is logical to conclude that code-switching and L1 attrition are influenced by generational differences  characterized by   varying perceptions of the L1 by speakers which is a significant feature  of the social consequence of language use. Adedun’s (2010) comments are instructive about the survival of Nigerian languages in view of the variation trend in the language use of youths in Nigeria.


Adedun, Emmanuel Adedayo (2010): ‘The Sociolinguistics of a Nollywood Movie’,    Journal of Global Analysis, vol. 1, no. 2

Akere, Funso. 1982: ‘Sociocultural constraints in         the emergence of a Standard Nigerian English.’ In J.B. Pride (Ed.) New Englishes. Newbury House Publishers.

Auer, Peter J.C. 1984: ‘On the meaning of conversational code-switching’ in J.C.P. Auer and A.O. Luzo (eds) Interpretative Linguistics: Migrants-Children-Migrant- Children. Turbingen: Newmeyer. pp. 87-108.

Auer, Peter J.C. 2005: ‘A postscript: code-switching and social identity.’ Journal of Pragmatics. Vol. 37. pp.403-410.

Ben-Rafael, M. & Schmid, M.S. 2007: ‘Language attrition and ideology: two groups of immigrants in Israel’. In B.  Köpke., M.S. Schmid, M. Keijer & S. Dostert (Eds.) Language attrition: theoretical perspectives.(pp. 205-220). Amsterdam: John Benjamins.

Bond, M.H. & Lai, T.M.  1986: ‘Embarrassment and codeswitching into a second language’. The Journal of Social Psychology. Vol. 126, No 2. pp. 179-186.

Brown, Spencer 1974: A History of the People of Lagos: 1852-1886.  Ph.D. Dissertation, Northwestern University.

Chambers, J.K. 1995: Sociolinguistic theory: linguistic variation and its social significance. Oxford: Blackwell.

Cook, Vivian. 1993: Linguistics and second language acquisition. London: Macmillan Press Ltd.

Dadzie, Koffie.A.B. 2004: ‘The Concept of Nigerian English’. In A.B.K. Dadzie & S. Awonusi (Eds.) Nigerian English: Influences and Characteristics. (pp.85-990). Lagos: Concept Publications.

Dewaele, Jean-Marc & Pavlenko, Aneta. 2002: ‘Emotion vocabulary in interlanguage’. Language Learning. Vol. 52 No 2. pp.263-322.

Dewaele, Jean-Marc. 2004: ‘Perceived Language Dominance and Language Preference for Emotional Speech: The Implication for Attrition Research’, In  Schmid, M. S., Kopke, B; Keijer, M. & Weilemar, L. (Eds.) First Language Attrition: Interdisciplinary        perspectives on methodological issues. (pp.81-104). Amsterdam: John Benjamins.

Dewaele, Jean-Marc. 2004: ‘Blistering barnacles! What language do multilinguals swear in?! Estudios de Sociolinguistica, Vol. 5, No 1. pp. 83-105.

Dewaele, Jean-Marc. 2006: ‘Expressing anger in multiple languages’. In  A. Pavlenko, (Ed.) Bilingual minds: emotional experience, expression and representation (pp.118-151). Clevedon, UK. Multilingual Matters.

Gal, Susan. 1979: “Language shift: social determinants of linguistic change in bilingual Austria.’ New York: Academic Press.

Gal, Susan. 1989:  ‘Lexical innovation and loss: the use and value of restricted Hungarian’. In N.C. Dorian (Ed.) Investigating obsolescence: studies in language  contraction and death. Cambridge: Cambridge University Press. pp. 313-331.

Goral, M., Levy, E. S. and Obler, J. K. 2002: ‘Neurolinguistic aspects of  bilingualism’. The International Journal of Bilingualism. Vol. 6, No. 4. pp.411-440.

Greenberg, J.H. 1963: The Languages of Africa. Special Supplement to International Journal of Linguistics, Vol. 29:1 Cambridge: Cambridge University Press. pp. 55-99.

Greenberg, J.H. 1971: ‘African languages’. In Language, Culture and Communication. Stanford, California: Stanford University Press. pp 126-36.

Gumperz, J.J. 1971: ‘Bilingualism, Bidialetalism and Class-room Interaction’. Language in Social Groups, Dil, A. S. (ed.), Stanford, CA: Stanford University Press.

Gumperz, J. J. (1982): Discourse Strategies. Cambridge: Cambridge University Press

Hakuta, K. & D’andrea 1992: ‘Some properties of bilingual maintenance and loss in Mexican background high-school students’. Applied Linguistics. Vol. 13. No1.pp.72-99.

Jaspaert, K., Kroon, S. & Van Hout, R. 1986: ‘Points of reference in first language loss research’. In B. Weltens., K. De Bot T. van Els (Eds.) Language attrition in progress. (pp. 37-49). Dordrecht: The Netherlands: Foris Publications.

Kim, Sun Hee Ok. 2005a: ‘At the crossroads of psycholinguistics and sociolinguistics: Issues in researching first language attrition’. Paper presented at the New Zealand Linguistic Society Conference 2005. Auckland, New Zealand.

Kim, Sun Hee Ok. 2005b: ‘First language use of young adult Korean-English bilinguals in a second language environment: Are there any signs of attrition?’ In S. May., M. Franken & R. Bernard (Eds.) First International Conference on Language, Education and Diversity (LED 2003). Hamilton, New Zealand: University of Waikato.

Kim, Sun Hee Ok. 2005c: ‘I know the word but…’ Korean-English late bilinguals’ vocabulary knowledge in first language and second language’. Australian Review of Applied Linguistics. Vol. 28, No 2 . pp 15- 27.

Kim, Sun Hee Ok. 2007: ‘First language attrition in a second language learning environment: the case of Korean-English late bilinguals’. Unpublished doctoral thesis, the University of Auckland, New Zealand. 

Kim, Sun Hee Ok & Starks, Donna. 2008: ‘The role of emotions in L1 attrition: The case of Korean-English late bilinguals in New Zealand’. In The International Journal of Bilingualism. Vol. 12. No 4. pp.303-319.

Kloss, H. (1966): ‘Types of Multilingual Communities: A discussion of Ten Variables’, Sociolinguistic Enquiry, 36, pp.241- 258

Myers-Scotton, Carol. 1988a: ‘Patterns of Bilingualism in East Africa.’ In C.B. Paulston (ed.) Handbook of Bilingualism and Bilingual Education. Westport, Conn: Greenwood. pp. 203-24.

Myers-Scotton, Carol. 1988b: ‘Code-switching and types of multilingual communities.’ In P. Lowenberg (ed.) Language spread and language policy. Washington, D.C: Georgetown University Press. pp.61-82.

Myers-Scotton, Carol. 1990: ‘Code-switching and macrolevel meaning’ In Jacobson (1990) Code-switching as a worldwide phenomenon. NY: Peter Lang.

Myers-Scotton, Carol. 1993: Social motivations for code-switching: Evidence from Africa. Oxford: Clarendon Press.

Nicoladis, E. & H. Grabois.  2002: ‘Learning English and losing Chinese: A case study of a child adopted from China.’ In The International Journal of Bilingualism. Vol. 6. No 4. pp.441-454.

Omoniyi, Tope. 2004: ‘The acquisition and use of English as a second language in Nigeria.’ In A.B.K. Dadzie & S. Awonusi (Eds.) Nigerian English: Influences and Characteristics. Lagos: Concept Publications Ltd. pp.100-118.    

Pavlenko, Aneta. 2003: ‘I feel clumsy speaking Russian’: second language influence on first language in narratives of Russian second language users of English’. In V. Cook (Ed.) Effects of the second language on the first. (pp. 32-61). Clevedon: Multilingual Matters.

Parkin, D.C.  1974: ‘Status factors in language adding: Bahati Housing Estate in Nairobi.’ In Whiteley (1974) Language in Kenya. Nairobi: Oxford University Press.

Pitres, A. 1895: ‘Etude sur laphasie chez les polyglottes’. Revue de Medecin. Vol. 15. pp. 873- 899. (Translated in M. Paradis (Ed.), Readings on aphasia in bilinguals and polyglots. pp. 26-48.

Preston, Dennis. R. 1989: Sociolinguistics and second language acquisition. Oxford & New York: Basil Blackwell.

 Olshtain, E. and Barzilay, M. 1991: ‘Lexical retrieval difficulties in adult language attrition’. In H. Selinger and R. Vago (eds) First Language Attrition. New.York: Cambridge University Press.

Pakir, A. 1989: ‘Linguistic alternates and code selection in Baba Malay.’ Montreal: Marcel Didier. World Englishes Vol. 8: pp. 379-88.

Poplack, Shana. 1980: ‘Sometimes I’ll start a sentence in English Y TERMINO EN ESPANOL: Toward a typology of code-switching’ Linguistics: Vol. 118, pp. 585-618.

Ribot, T.C. 1881: Les malades de la memoire . (2nd ed) Pans: Bailleres.

Romaine, Suzanne. 1993: Bilingualism (2nd ed). Oxford: Blackwell.

Rubin, J. 1968: National Bilingualism. (2nd ed.) Oxford: Blackwell.

Schmid, M. 2002: ‘First language attrition, use and maintenance: The case of German Jews in Anglophone countries’. In Studies in Bilingualism. Vol. 24. Amsterdam/Philadelphia: John Benjamins.

Scotton, Carol. 1975: ‘Multilingualism in Lagos: What it means to the social scientist’ Columbia: Ohio State University, Department of Linguistics. pp. 78-90.

Shodipe, Moji. 2010: ‘Sociolinguistic variation in bilingual discourse: A cross-generational study of Yoruba-English code-switching and code-mixing in Central Lagos’. Unpublished doctoral thesis, the University of Lagos, Nigeria.

Shunkal, A. and L. Marchese 1983: ‘Creolisation in Nigerian Pidgin English: A progress report.’ English Worldwide. Vol 5. No 2.  

Sridhar, S.N. and Sridhar, K. 1980: ‘The syntax and psycholinguistics of bilingual code-mixing’, Canadian Journal of Psychology. Vol. 34. pp. 407-16.

Valdes, G. and Pino, C. (1981): ‘Muy atus ordenes: Complement Responses among Mexican-American Bilinguals’, Language in Society, vol. 10, pp.53-72.


Weinreich, Uriel. 1953: Languages in contact.  The Hague: Mouton.