Skip to content
BY 4.0 license Open Access Published by De Gruyter Mouton August 6, 2021

Translanguaging patterns in everyday urban conversations in Cameroon

Eric A. Ambele EMAIL logo and Richard Watson Todd


This study analyses the translanguaging patterns of urban Cameroonians’ linguistic choices (e.g. lexical or phonological) in everyday conversations in Cameroon. Using observation and audio-recordings of 20 naturally occurring conversations as data, a descriptive corpus-based methodology was adopted for analysis. The quantitative approach utilises AntConc (Version 3.5.8) with descriptive analytical tools to identify the speakers’ idiolectal choices in meaning-making translanguaging patterns. The results revealed salient patterns of the speakers’ deployed lexical, grammatical, morphological, phonological and syntactical forms as an integrated system of language. It revealed the speakers preference for polysemous words (e.g. repe) over less polysemous words (e.g. father); choice for shorter lexical words (e.g. man) over longer words (e.g. manpikin); a preference for specialised gender-neutral markers (e.g. ih, which refers to both male and female) over gender-specific forms (e.g. he/she); a preference for voiceless interdental fricatives (e.g. dem, dey) over voiced interdental fricatives (e.g. them, they) and where the choice of inflectional morpheme expressing tense (e.g. ed) is one that can either be omitted or added to a word, the presence of this inflectional morpheme is sometimes fairly used. Such results have practical implications for understanding peoples’ language use as a translanguaging act in bi/multilingual contexts.

1 Introduction

Cameroon is a highly multilingual country with over 285 languages being spoken (Atechi 2011; Ayafor and Green 2017). Of these, however, three dominate: Standard English (SE) and French as the official languages used for formal purposes, and Cameroon Pidgin English (CPE) as a de facto lingua franca in everyday informal interactions (Nkengasong 2016). This lingua franca operates as a bridge between the socio-politically defined linguistic groups in Cameroon, cutting across ethnicities, social classes and educational levels. Nevertheless, Cameroonians’ real-life language use, particularly in the Northwest and Southwest Regions of Cameroon “does not seem to consider context when conversing” (Ambele 2020: 89). The manner in which people in this context use language, especially in urban areas, blur the distinctiveness between languages.

As it has been observed in previous studies (Ambele 2020; Ambele and Watson Todd 2017), such Cameroonians interlard their urban conversations with what the society calls pidgin and English forms. The following utterances from two speakers in the data set illustrate this interlarding of societally described pidgin and English forms in this context. The utterances are presented verbatim as they were transcribed from the conversation recording, with an English translation text following in brackets to ease comprehension for non-Cameroonian readers.

Speaker A: Mami a dohn hear ya complain ih be very important but you no see say you fit take da ya complain go fo mayor, ih fit support gee you any financial support. (Mother, I’ve heard your complaint and what you’re saying is very important but I think you should take your worries to the mayor to see if he can give you any financial support.)

Speaker B: Financial weti? (What financial support?)

Speaker A: Support nor (Financial support)

Speaker B: Masa, leave da tihn, mayor wey ih stronghand so tee, you wahn make mayor sell me na sellam foh nyongo. (Please, just forget about that financial support. The mayor is very stingy and the only kind of support he can give me is to use me for money rituals.)

According to Otheguy et al. (2015) and Saraceni (2017), from the translanguaging perspective, people’s linguistic choices are made from a mental repertoire that is unitary, not subdivided into compartments corresponding to the socioculturally defined languages. Thus, their repertoire choices constitute their linguistic behaviours in speech. With this understanding, the multilingual context, Cameroon, has been described as a complex setting where people tap into their repertoires in dynamic ways to achieve their communicative needs (Ambele 2020). Studies have established that translanguaging occurs in free interactions and other discourses without the speakers adhering to any named language system (Makalela 2016; Ndhlovu 2018; Saraceni 2017). Thus, the reason for this line of research in everyday conversations in Cameroon since it can provide insights into the translanguaging phenomenon in postcolonial settings. Moreover, in analysing translanguaging in different discourses, the concept has been widely applied in educational research and related bi/multilingual studies (García and Wei 2014; Gorter and Cenoz 2015; Prinsloo and Krause 2019). However, translanguaging research into real-life conversations remains relatively scarce, with fewer studies that directly investigates peoples’ linguistic choices in conversations. From previous research, we expand our focus of translanguaging here to urban speakers’ linguistic practises in everyday conversations. Furthermore, most previous research into translanguaging has been highly interpretative and qualitative because of its epistemological orientation. While we acknowledge the value of such research, we also believe that taking a quantitative approach to translanguaging may lead to a newer understanding of the translanguaging practises of people in actual linguistic encounters, describing their idiolectal usage in everyday discourses. To achieve this, the current study takes a quantitative corpus-based approach in analysing the translanguaging practises and linguistic behaviours of urban Cameroonians, using multiple conversations.

2 Theoretical framework

The clearest definition of translanguaging is the one provided by Otheguy et al. (2015, 2018. They define translanguaging as “the deployment of a speakers’ full linguistic repertoire without regard for watchful adherence to the socially and politically defined boundaries of named (and usually national and state) languages” (2015: 283). This concept focuses on how speakers move beyond what society calls different languages “as an integrated system” in their cognitive linguistic repertoire (Canagarajah 2011: 242).

In distinguishing translanguaging from code-switching, Lewis et al. (2012: 59) assert that the latter considers languages as distinct structures while the former “celebrates and approves flexibility in language use and the permeability of using language”. To clarify, translanguaging extends beyond concepts like code-switching “in that it refers not simply to a … shuttle between two languages, but to the speakers’ construction and use of original and complex interrelated discursive practices that cannot be easily assigned to one or another … language, but that makes up the speakers’ complete language repertoire” (García and Wei 2014: 22). In other words, code-switching entails a process that goes “between languages”, but translanguaging goes “over languages”, contesting cognitive boundaries between socially named languages.

In support of the translanguaging view, Wei (2018) stressed that language users are only aware of languages as a social reality that is distinct from the cognitive reality of how languages are enregistered in the mind, which is a single language. So, it is the speakers’ translanguaging instinct that causes them to go beyond narrowly defined social reality of language boundaries to a cognitive practise of a single language use in order to achieve effective communication. In fact, Otheguy et al. (2015: 293) stressed that “… in accepting terms like ‘language’, ‘a language’, ‘monolingual’, and ‘bilingual’, we are using categories that have nothing to do with individuals when seen from their own internal linguistic perspective, categories that have nothing to do, that is, with the billions of the world’s idiolects, which exist in a separate, linguistically unnamed and socially undifferentiated mental realm”. This quotation justifies why the translanguaging framework was deemed appropriate to examine the kind of conversation data in this study since it deals with lexical and structural matters in the speakers’ mental realm from which they make linguistic choices without adhering to any particular language norm. Besides, García and Kleyn (2016) and Saraceni and Jacob (2018) have asserted that in lingua franca multilingualism, languages are so deeply intertwined and fused into each other that the level of fluidity renders it difficult to determine any boundaries that may indicate that there are different languages involved.

Recently, there has been a continuous increase of inter-ethnic mobility in urban Cameroon. As a result, people tend to utilise their respective linguistic resources to effectively communicate with their interlocutors. This practise results in stronger bonds between interlocutors as they freely use their repertoire resources, showing patterns of linguistic choices from their repertoire, in ways suitable to them. This phenomenon is highly pertinent in the context of the current study.

3 Research purpose

From the external and social point of view, there appears to be two languages in the speakers’ repertoire in the context of this research, however, in reality, it is only a single language from the translanguaging point of view. This manner of language use appears natural to the speakers with a range of intentions and metalinguistic awareness during interactions (Trudgill and Hannah 2013; Wei 2011). The speakers’ language behaviours are thus expressed through their morpho-grammatical or lexical choices as a translanguaging act. This study, therefore, aims to analyse the linguistic choices and practises of urban Cameroonians in how they deploy their repertoire linguistic forms in everyday conversations. The study examines the frequency of their linguistic choices in the data set to induce patterns of their translanguaging practises within a wider space of the conversations based on the following research question: Are there any underlying patterns in how urban Cameroonians use language in everyday conversations in Cameroon?

To exemplify this purpose, what we assumed was that there are alternative realisations for the different linguistic forms that are available in the speakers’ repertoire, from which a speaker strategically makes a choice of which form to use during a conversation. Taking an utterance from the data set – ‘dat house di bore’, for example – illustrates the alternative linguistic forms available in the speaker’s repertoire (see Table 1).

Table 1:

Example of alternative realisations.

Form used dat house di bore
Alternative form used that lohng is boring
Language category pronunciation lexis grammar Morphology (inflection)

The example in Table 1 compares two actual utterances that the speakers used in the corpus by tapping into their respective mental linguistic systems, naturally and spontaneously, in the course of the conversation. We can see that the speakers deployed a full range of their repertoire resources by selecting certain linguistic forms in order to communicate. For example, one speaker selects the voiceless interdental consonant sound (‘dat’) while the other speaker uses the voiced alternative (‘that’). These forms are all in the cognitive repertoire of the speakers and they individually deploy any form as they converse in a way that appears natural to them (without necessarily being aware of their associations to any societally named language). So, this quantitative analysis focuses on the proportions of the speakers’ linguistic forms used (where there are alternatives) in realising a target meaning.

4 Methodology

This study adopts a quantitative analytical approach using AntConc (Version 3.5.8) (Anthony 2019), to understand the translanguaging patterns of Cameroonians when they engage in conversations. Taking such a quantitative approach within a previously qualitative and interpretative area might seem difficult for scholars to see why this might help; nevertheless, taking the case of literature studies as an example can show why this may be possible. Traditionally, literature is highly interpretative with researchers aiming at interpreting authors’ intentions in their writings. However, corpus techniques have allowed a different approach to literature which can be exemplified through Culpeper’s (2009) article on Shakespeare’s Romeo and Juliet, where he created a corpus of the characters’ talk throughout the play to identify differences in how they spoke which can reveal things about their personality. In this way, we can see how even in areas which are traditionally highly qualitative, a quantitative approach can help to provide useful insights.

4.1 Data and participants of the study

The data for this study consisted of 20 conversations from Cameroonians in four urban cities in the Southwest and Northwest Regions of Cameroon, namely: Kumba, Buea, Limbe and Bamenda. The participants were Cameroonians from these regions (between the ages of 20–35), fluent in the languages that society calls SE and CPE, and residing in the selected urban cities where the coexistence of these socially described languages have been observed to be widely spread and used. The participants were contacted by word of mouth to observe their language behaviours through modified random sampling (Milroy and Gordon 2003) and social networking (Tagliamonte 2012). The collected data consisted of 20 elicited conversations (see Appendix A), each about 25–30 min with full participant metadata (e.g. age, education, social status/role, frequency of SE, CPE and other language use), collected mainly through audio-recording of the conversations and field notes. The conversations were naturally occurring and spontaneous interactions in which the participants were observed to effortlessly deploy their linguistic choices from their repertoire to create meaning and accomplish goals (Ambele and Watson Todd 2018).

We observed and audio-recorded the participants’ interactions at a place and time that was suitable and convenient for them. In the meantime, detailed notes were taken of the participants’ behaviour as they were conversing. The participants, it should be noted, have a shared socio-political and ethnic identity, however, it is their language practises from their own internal repertoire that is linguistically described here in terms of their idiolects and not in association with any societally named language in the context. Such real-life data is necessary to base the translanguaging arguments in this paper on actual linguistic examples.

4.2 Data analysis and procedures

A quantitative approach was employed using descriptive statistics. The statistical analysis was based on general descriptive information of the participants’ language use in the conversations, such as overall frequencies and percentages. The choice of a descriptive statistical design was to present a concise quantitative analysis of the speakers’ linguistic choices. The conversations were analysed based on five language categories (i.e. lexis, grammar, morphology, pronunciation and syntax). The choice of these categories was based on the fact that they capture the salient units of linguistic description where peoples’ linguistic compositions can be easily observed and accounted for (Ambele and Watson Todd 2017; Watson Todd 2016).

The guidelines for identifying the linguistic forms in each category were then set up into lexical items, function words, inflectional affixes, phonemic choice and syntax. Thereafter, the data was divided into identifiable analytical units based on the five categories (i.e. lexis and grammar were restricted to a word as the unit for analysis, morphology was restricted to a morpheme, pronunciation was restricted to a syllable and syntax to a clause as the unit for analysis). All 20 conversations were quantitatively processed on AntConc (Version 3.5.8) (or manually for syntax) for key linguistic forms from the participants’ repertoire. These forms were selected from a much wider range of forms in the analysis based on their high frequency. Then, the frequently used forms, alongside their alternative realisations in the conversations, were again processed using AntConc for concordance hits. To keep the analysis manageable, we created a list of the key linguistic forms from the AntConc wordlist and concordance hits analysis and grouped them. Using lexis as an example, 38,206 different words were sorted; analysing all of these was impractical. Therefore, an initial step was to conduct filtering to reduce the number of words to a manageable level. This was done by sorting out only the topmost frequently used forms from the wordlist. Similar processes were conducted for grammatical words, morphological affixes, syllables and clausal types. For each of the key linguistic form identified, alternative realisations in the interlocutors’ repertoires were identified in the conversations and calculated as frequency-percentages, allowing for the identification of the participants’ patterns of language use.

5 Results

In line with the research purpose, the participants’ idiolects are presented here as structured lists of five linguistic categories, i.e. lexical items, function words, inflectional affixes, phonemic choice and clausal types. The term idiolect is used here in the same understanding as in Otheguy et al. (2015: 289), who succinctly operationalise this term as “a mental grammar that is acquired primarily through, and deployed mostly in social and personal interaction”. Thus, all the participants’ idiolectal choices (which we call linguistic forms) are presented in Tables 2 6 to reflect their actual linguistic behaviours, and should therefore be understood as pairings of the participants’ full range of idiolectal resources that are deployed in the conversations. This pairing in the tables, it should be noted, are not necessarily by meaning but by reference.

Table 2:

Frequency of lexical choices for the same form.

Form used Freq. (%) Alternative form Freq. (%) Other alternative form Freq. (%)
bike 100.0 byciko 0.0
close 100.0 closeam 0.0
debt 100.0 owam 0.0
fit 100.0 fitam 0.0
house 100.0 lohng 0.0
man 100.0 manpikin 0.0
money 100.0 douh 0.0
price 100.0 hamush 0.0
school 100.0 lewah 0.0
see 100.0 seeam 0.0
start 100.0 startam 0.0
wise 100.0 wiseam 0.0
work 100.0 bolo 0.0
go 99.5 shake 0.5
talk 99.0 tok 1.0 tokam 0.0
know 98.6 knowam 1.4
pay 98.6 payam 1.4
person 98.6 pipo 1.4 people 0.0
like 98.5 likeam 1.5
tell 98.4 te 1.6 tellam 0.0
how 98.0 ha 2.0
pass 97.6 passam 2.4
get 97.4 getam 2.6
self 97.4 sef 2.6
want 97.3 wantam 2.7
try 96.0 tryam 4.0
open 94.7 openam 5.3
give 94.5 geeam 5.5
call 93.6 callam 6.4
cook 93.5 cookam 6.5
change 93.3 changeam 6.7
pikin 92.9 child 7.1
write 92.7 writeam 7.8
force 92.3 forceam 7.7
need 91.9 needam 8.1
drink 89.3 drinkam 10.7
send 88.6 sendam 11.4
say 87.8 weti 12.2
repe 87.5 father 12.5
ask 86.4 askam 13.6
small 85.5 petit 14.5
bring 85.3 bringam 14.7
morning 85.3 sharp 14.7
make 85.0 makeam 15.0
take 83.2 takeam 16.8
one 78.1 first 21.9
do 78.0 doam 22.0
girl 77.8 bae 22.2
buy 77.8 buyam 22.2
comot 74.8 leave 25.2
mimbo 72.2 alcohol 27.8
please 61.8 abeg 38.2
struggle 61.1 hussel 38.9
sell 57.1 sellam 42.9
waka 54.3 walk 45.7
mommy 52.4 mami 47.6
Table 3:

Frequency of grammatical choices for the same form.

Form used Freq. (%) Alternative form Freq. (%) Other alternative form Freq. (%)
be 100.0 bime 0.0
right? 100.0 na 0.0
all 100.0 altin 0.0
but 100.0 beh 0.0
some 99.0 soh 1.0
di 97.6 am 1.4 is 1.0
da 97.5 that 2.5
dohn 96.2 have 3.9
no 94.0 never 6.0
my 93.9 ma 6.1
a 93.4 I 6.6
weti 87.0 wey 10.0 what 3.0
whosai 75.0 where 25.0
ih 70.2 he 29.8
uh 50.0 wuna 31.2 we 18.8
Table 4:

Frequency of morphological choicesa for the same form.

Form used Freq. (%) Alternative form Freq. (%) Other alternative form Freq. (%)
-ly 67.9 #ly 32.1
-ing 57.7 #ing 42.3
dem 54.7 -s 30.8 #s 15.3
-ed 52.8 #ed 47.2
  1. aThe minus sign before the affix (e.g. –ly) stands for the presence of such affix in a word, and the hashtag sign (#) against a word represents the absence of the affix in a word.

Table 5:

Frequency of phonological choices for same form.

Form used Freq. (%) Alternative form Freq. (%)
know 100.0 nooh 0.0
make 100.0 mek 0.0
dey 99.8 they 0.2
tihn 99.7 thing 0.3
dem 99.4 them 0.6
geh 98.9 get 1.1
de 98.2 the 1.8
work 98.1 wok 1.9
da 96.0 that 4.0
bet 94.7 but 5.3
how 94.5 ha 5.5
a 93.9 I 6.1
time 92.6 tam 7.4
e 91.7 air 8.4
be 90.9 bi 9.1
just 90.7 joh 9.3
ver 70.9 ve 29.1
tok 70.6 talk 29.4
ih 69.9 he 30.1
ing 65.3 ng 34.7
my 62.0 ma 38.0
for 38.1 foh 61.9
Table 6:

Frequency of syntactic choices for the same structure.

Form used Freq. (%) Alternative form Freq. (%)
s+aux+v 69.2 s+v 30.8
s+v+conj+v+n 69.0 s+v+v+n 31.0
v+conj+v 61.2 v+v 42.7
s+v+v+adv 58.6 s+v+adv 41.4
s+v+art+n 56.6 s+na+n 43.4
s+v+prep+n 54.5 prep+n 45.5
s+v+0 54.1 s+di+v+o 45.9
s+di+v 53.2 s+trans v 46.8
s+v+conj+v 53.0 s+v+v 47.0
s+adj 50.9 s+aux+adj 49.1

The ensuing Tables 2 6 are primarily sequenced based on what society calls English or Pidgin and, importantly, on the highest-lowest frequency value sorting of the linguistic forms in order to induce patterns of linguistic choices. It should be noted that the columns in the tables do not have named language labels because this study adopts the unitary conception of linguistic competence sponsored by many scholars in translanguaging studies. More importantly, beyond the labels, the columns do not, in fact, correspond to the two socially named languages, since both columns contain items from both.

From Table 2, the following underlying patterns can be induced: (1) some pairs of alternative realisations concern reference words. For example, the words ‘repe’, ‘father and ‘old man’; ‘girl’ and ‘bae’; ‘please’ and ‘abed’, etc. share a referent, even if they may not, in all cases, share a meaning; (2) where there is a lexical variance based on word length for words with the same root, there is a choice for shorter lexical forms (for example, ‘sell’, ‘send’, ‘man’), over longer forms (for example, ‘sellam’, ‘sendam’, ‘manpikin’) and (3) another similar but opposite pattern is that where the word (for instance, ‘pikin’, ‘abeg’, ‘pipo’) and their alternative forms (‘child, ‘please’, ‘person/people’) are distinct with dissimilar roots, there seems to be a variableness in the choice of the preferred words as a result of the nature of the conversations and the interlocutors involved.

From Table 3, the following underlying patterns can be induced: (1) certain grammatical forms have to do with pre- and post-monosyllabic and polysyllabic modifiers. Where a pre- and post-monosyllabic and polysyllabic modifiers provide possible alternative realisations, there is a choice for shorter pre- and post-monosyllabic realisations, for example: ‘be’, ‘uh’ and ‘all’, as opposed to longer pre- and post-polysyllabic realisations, such as: ‘bime’, ‘wuna’ and ‘altin’, respectively; (2) where grammatical forms denoting quantifiers provide possible alternative realisations, there seems to be a choice for certain grammatical forms denoting imprecise quantity (e.g. ‘some’), over other grammatical forms, for example, ‘soh’, denoting the same imprecise quantification that seems to be less preferred and (3) the words ‘he’, ‘she’ and ‘ih’ share a referent, that is, ‘male or female’, even if, in some cases, they may not share a meaning. Where the grammatical form is one that seeks to denote gender differences, there is a choice for specialised gender-neutral markers, for example, ‘ih’ (which refers to both male and female). However, where the grammatical particle is one that seeks to denote gender-specific forms, such as, ‘he/she’, such gender-specific forms seem to be less preferred.

Based on Table 4, the following underlying patterns can be induced: (1) where there are alternatives in inflectional affixes predicated on the absence or presence of such affixes (such as #ing’, ‘#ed’ [absence] and ‘-ing’, ‘-ed’ [presence]), irrespective of the word class, there seem to be a variableness in the choice of the preferred morphological forms as a result of the nature of the conversations and the interlocutors involved in the discourse; (2) where the third person plural morphemes provide possible alternative realisations (such as ‘-s’, ‘#s’), a specialised morpheme (‘-dem’) seems to be the preferred realisation instead of the presence of the inflectional affix (e.g. ‘-s’) or the absence of it (e.g. ‘#s’) in a word, and (3) where the choice of inflectional morpheme expressing tense, modality and aspect (for example, ‘ed’) is one that can either be omitted or added to a word, the presence of this inflectional morpheme is sometimes fairly used.

Looking at Table 5, the following underlying patterns can be induced: (1) there are certain phonological choices that have to do with consonant differences. This is particularly the case, as seen in the data, where the difference is that of an interdental fricative sound (such as the –th sound). To illustrate this pattern, where there are alternative realisations in the choice of a consonant interdental fricative sound (such as –th sounds), there is a general preference for voiceless interdental fricatives (e.g. ‘dem’, ‘tihn’ and ‘dey’) over voiced interdental fricatives (e.g. ‘them’, ‘think’ and ‘they’); (2) some pairs of alternative realisations concern vowel sound differences. To make clear, where the difference between the alternatives is a vowel sound difference (e.g. ‘bet’, ‘tok’ and ‘but’ and ‘talk’), respectively, in syllables where the consonant sounds are the same, there is a preference for a vowel sound realisation (that is more akin to what society calls SE) with the rest of the consonant sounds in the syllable unchanged and (3) some other pairs of alternatives have to do with monophthongised and diphthongised vowel realisations. For example, where a syllable provides possible alternative vowel realisations, there seems to be a preference for diphthongised realisations (e.g. ‘know’) over monophthongs (e.g. ‘nooh’) that are dispreferred.

From Table 6, the following underlying patterns can be induced: (1) some pairs of alternative syntactic forms concern simple and complex clausal structures. Where there exist simple and complex clausal types, the tendency is for the speakers to prefer a more complex structure (for example, ‘s+v+conj+v+n’ and‘s+v+prep+n’) over the use of a simple one (for example, ‘s+v’, ‘s+adj’); (2) in the same vein, some pairs of alternative clausal types have to do with flexible versus rigid structural patterns. To illustrate, there is a general preference for, and a more frequent use of, the choice of flexible structural patterns which allow speakers to move constituents into different clausal positions to convey different thoughts. However, where the structure is fixed and regular, then such clausal structures are less likely to be preferred by the speakers and (3) where the pairs of alternative forms concern pre-verbal clausal structures, there is a preference for NP and VP structural patterns over structures containing the insertion of a specialised pre-verbal particle (for example, ‘na’) that are dispreferred and less frequently used by the speakers.

6 Discussion

It follows from the results in Tables 2 6 that the participants’ translanguage in everyday conversations as the linguistic choices from their repertoire show an intense usage-practise of their idiolects in their everyday translanguaging practise. From this quantitative translanguaging analysis, the participants naturally overlap their linguistic choices beyond socio-politically defined systems for a smooth flow of meaning (García and Wei 2014). In fact, the participants’ linguistic choices are so deeply intertwined, fused into each other, and deployed in the conversations as a single resource without any regard, whatsoever, for linguistic boundaries. Salient idiolectal choice patterns were therefore observed from the results (see Tables 2 6). This indicates that the linguistic knowledge of the available resources in peoples’ repertoire is often embedded and revealed in interactions (Ambele 2020; Makalela 2016).

When conversing, as we observe, the participants naturally and spontaneously deployed a full range of their linguistic resources. The participants have a wide range of resources available to them in their repertoire, and so, they translanguage by deploying the resources in conversations in ways they believe are effective for their communication goals (Wei 2011). This shows that the participants tend to strategically merge the different linguistic forms of what society calls English and Pidgin in their repertoire as a single resource in making sense of their worlds (García and Kleyn 2016; Saraceni 2017). Makalela (2016: 192) asserts that “for speakers in complex multilingual zones, all socially defined language entities become present and they are used simultaneously for meaning-making and engagement in deeper thought processes”.

Otheguy et al. (2015: 291) observe that “languages are not true linguistic entities because their boundaries are established on non-linguistic grounds. Rather, they are groupings of idiolects of people …” This quote reflects the complex and fluid repertoires of the participants in this study. Such linguistic fluidity disrupts any boundaries that could be assigned to the linguistic choices in Tables 2 6. In this light, the participants in this study do not speak English or Pidgin, rather they effortlessly use the language in their repertoire to communicate in ways that appear natural to them without regard for exiting language boundaries. This means that all the participants’ linguistic choices in the results can be viewed as part of their idiolectal translanguaging practise from their own mental grammar.

This conversational translanguaging practise, by extension, seems to be common among most speakers in Africa given the multilingual and multicultural nature of the African context (Alimi and Matiki 2017; Ndhlovu 2018). It is therefore unsurprising that translanguaging is an everyday common practise and part of the linguistic life of most Cameroonians. Even though it may seem difficult to determine the linguistic identity of the speakers based on our data, however, from the results, we can see that what the participants do with the linguistic resources in their repertoire in making meaning enact a sense of who they are. Understood in this way, the participants’ linguistic identities are a complex unbounded and overlapping one that is deployed in the way they speak.

7 Conclusion

The goal of this study was to investigate the linguistic choices of urban Cameroonians in everyday conversations with a focus on identifying underlying patterns of the linguistic choices from their repertoire. Translanguaging focuses on speakers’ in-depth sense-making and their strategic linguistic behaviour in speech, nevertheless, as a limitation in this study, the size of the data set without interviews with the speakers on their intentions makes it difficult for us to focus on speakers. Also, we were not looking at individual speakers but an aggregation of multiple speakers to see overall patterns.

The study has noted that the participants deployed a full range of the linguistic resources from their repertoire in a manner that makes their association to existing socio-politically named languages in Cameroon difficult and inappropriate (Ambele 2020). The salient patterns of linguistic choices observed in the results (e.g. their choice for shorter forms where the alternative realisations share the same root [e.g. –am verbs]; and the choice for polysemous words where the alternative realisations have distinct roots [e.g.‘repe’]) are evidence of the effortless linguistic crossing of the participants in the data without regards for supposed language boundaries.

Makalela (2016: 191) summarises this approach to understanding peoples’ use of language in a complex context like Cameroon as “multilingualism from the translanguaging point of view does not emphasize on the enumeration of languages, rather on the degree to which simultaneous use of the languages in line with the linguistic competence of the speakers is valorised”. From the participants’ linguistic practises in conversational discourse, the study has confirmed that people do not always recognise, linguistically, socio-political language boundaries (Ambele 2020; García and Wei 2014; Turner and Lin 2020; Wei 2016; Zhao and Flewitt 2020).

Corresponding author: Eric A. Ambele, King Mongkut’s University of Technology Thonburi, Bangkok, Thailand, E-mail:

Appendix A: Capturing variables in data set.a

Recording Education Role-relationship Setting Formality distance Topic
Rc1 H-H Neighbours Home Formal University admission & problems
Rc3 ML-MH Friends Street Informal Friend’s ordeal
Rc4 H-MH Friends Home Informal Personal life story
Rc5 H-ML Friends Home Informal Personal lifestyle
Rc6 H-H Mates Faculty lobby Formal University study & personal life
Rc8 MH-MH Friends School Formal Teaching practicum
Rc10 H-H Neighbours Neighbourhood Informal Welfare
Rc13 L-ML Traders Bar Informal Livelihood
Rc17 H-H Classmates Street Informal Course registration
Rc31 H-H Lovers Home Informal Love relationship
Rc40b H-MH Friends Business centre Formal Well-being
Rc40c H-H Friends Home Informal Talent/career
Rc45 H-H Roommate Home Informal Family
Rc48 H-H Passengers Bus station Informal Mobile phone
Rc56 H-H Colleagues Church Informal Well-being
Rc60c H-H Siblings Home Informal Personal ordeal
Rc60e H-H Colleagues Office Formal Work
Rc69 ML-ML Friends Street Informal Choosing series
Rc74 MH-ML Choristers Church Informal Singing groups
Rc80j MH-H Friends Street Informal Football
  1. aRc stands for conversation recording, Rc1 (number after recording) stands for the number of conversation collected in the field, H stands for High education level (Tertiary degrees), MH stands for Mid-high education level (High school certificates), ML stands for Mid-low education level (Ordinary level certificates), L stands for Low education level (First School Leaving Certificate).


Alimi, Modupe & Alfred Matiki. 2017. Translanguaging in Nigerian and Malawian online newspaper readers’ comments. International Journal of Multilingualism 14(2). 202–218. in Google Scholar

Ambele, Eric. 2020. Variations in language patterns in Cameroon Pidgin English. Bangkok, Thailand: King Mongkut’s University of Technology Thonburi Doctoral thesis.Search in Google Scholar

Ambele, Eric & Richard, Watson Todd. 2017. Adding rigour to language variety continua. In Proceedings of Doing Research in Applied Linguistics 3 & 19th English in South-East Asia Conference. Bangkok: King Mongkut's University of Technology Thonburi, 22–24 June.Search in Google Scholar

Ambele, Eric & Richard, Watson Todd. 2018. Challenging accepted practices in questionnaire design. In Proceedings of Multidisciplinary Research, Innovation and Sustainability RMUTT International Conference on Social Sciences and Service Industry. Thanyaburi: Rajamangala University of Technology Thanyaburi, 30 June.Search in Google Scholar

Anthony, Lawrence. 2019. AntConc. (Version 3.5.8) [Computer Software]. Tokyo, Japan: Waseda University. Available at: in Google Scholar

Atechi, Samuel. 2011. The plurality of English in Cameroon and the international intelligibility question. Syllabus Review 1. 199–218.Search in Google Scholar

Ayafor, Miriam & Melanie Green. 2017. Cameroon Pidgin English: A comprehensive grammar. Amsterdam: John Benjamins.Search in Google Scholar

Canagarajah, Suresh. 2011. Translanguaging in the classroom: Emerging issues for research and pedagogy. Applied Linguistics Review 2. 1–28. in Google Scholar

Culpeper, Jonathan. 2009. Keyness: Words, parts-of-speech and semantic categories in the character-talk of Shakespeare’s Romeo and Juliet. International Journal of Corpus Linguistics 14(1). 29–59.Search in Google Scholar

García, Ofelia & Tatyana Kleyn (eds.). 2016. Translanguaging with multilingual students. New York and London: Routledge.Search in Google Scholar

García, Ofelia & Li Wei. 2014. Translanguaging: Language, bilingualism and education. UK: Palgrave Macmillan.Search in Google Scholar

Gorter, Durk & Jasone Cenoz. 2015. Translanguaging and linguistic landscapes. Linguistic Landscape 1(1–2). 54–74. in Google Scholar

Lewis, Gwyn, Bryn Jones & Colin Baker. 2012. Translanguaging: Origins and development from school to street and beyond. Educational Research and Evaluation 18(7). 641–654. in Google Scholar

Makalela, Leketi. 2016. Ubuntu translanguaging: An alternative framework for complex multilingual encounters. Southern African Linguistics and Applied Language Studies 34(3). 187–196. in Google Scholar

Milroy, Lesley & Mathew Gordon. 2003. Sociolinguistics: Method and analysis. New York: Blackwell Publishing.Search in Google Scholar

Ndhlovu, Finex. 2018. Omphile and his soccer ball: Colonialism, methodology, translanguaging research. Multilingual Margins: A Journal of Multilingualism from the Periphery 5(2). 1–2. in Google Scholar

Nkengasong, Nkemngong. 2016. A grammar of Cameroonian Pidgin. Cambridge: Cambridge Scholars Publishing.Search in Google Scholar

Otheguy, Ricardo, Ofelia García & Wallis Reid. 2015. Clarifying translanguaging and deconstructing named languages: A perspective from linguistics. Applied Linguistics Review 6(3). 281–307. in Google Scholar

Otheguy, Ricardo, Ofelia García & Wallis Reid. 2018. A translanguaging view of the linguistic system of bilinguals. Applied Linguistics Review 10(4). 1–27. in Google Scholar

Prinsloo, Mastin & Lara-Stephanie Krause. 2019. Translanguaging, place and complexity. Language and Education 33(2). 159–173. in Google Scholar

Saraceni, Mario. 2017. World Englishes and linguistic border crossings. In Ee Ling Low & Anne Pakir (eds.), World Englishes: Re-thinking paradigms, 154–171. UK: Routledge.Search in Google Scholar

Saraceni, Mario & Camille Jacob. 2018. Revisiting borders: Named languages and de-colonization. Language Sciences 76. 101–170.Search in Google Scholar

Tagliamonte, Sali. 2012. Variationist sociolinguistics: Change, observation, interpretation. Oxford: Blackwell.Search in Google Scholar

Turner, Marianne & Angel Lin. 2020. Translanguaging and named languages: Productive tension and desire. International Journal of Bilingual Education and Bilingualism 23(4). 423–433. in Google Scholar

Trudgill, Peter & Jean Hannah. 2013. International English: A guide to the varieties of Standard English. UK: Routledge.Search in Google Scholar

Watson Todd, Richard. 2016. Discourse topics. Amsterdam: John Benjamins.Search in Google Scholar

Wei, Li. 2011. Moment Analysis and translanguaging space: Discursive construction of identities by multilingual Chinese youth in Britain. Journal of Pragmatics 43(5). 122–135. in Google Scholar

Wei, Li. 2016. New Chinglish and the post-multilingualism challenge: Translanguaging ELF in China. Journal of English as a Lingua Franca 5(1). 1–25. in Google Scholar

Wei, Li. 2018. Translanguaging as a practical theory of language. Applied Linguistics 39(1). 9–30. in Google Scholar

Zhao, Sumin & Rosie Flewitt. 2020. Young Chinese immigrant children’s language and literacy practices on social media: A translanguaging perspective. Language and Education 34(3). 267–285. in Google Scholar

Received: 2020-11-06
Accepted: 2021-02-10
Published Online: 2021-08-06
Published in Print: 2022-01-27

© 2021 Eric A. Ambele and Richard Watson Todd, published by De Gruyter, Berlin/Boston

This work is licensed under the Creative Commons Attribution 4.0 International License.

Downloaded on 30.1.2023 from
Scroll Up Arrow