The Sociolinguistic Dynamics of Rural Multilingualism in Africa. the Case (2024)

1

University of Buea

Faculty of Arts Department of Linguistics

The sociolinguistic dynamics of rural multilingualism in Africa The case of Lower Fungom

A thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy (PhD) in Applied Linguistics

Rachel Ojong

Supervisors

Ayu'nwi Ngwabe Neba, PhD

Pierpaolo Di Carlo, PhD

December, 2018 1

2

University of Buea

Faculty of Arts Department of Linguistics

The sociolinguistic dynamics of rural multilingualism in Africa

The case of Lower Fungom

A thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy (PhD) in Applied Linguistics

Rachel Ojong

Supervisors

Ayu'nwi Ngwabe Neba, PhD

Pierpaolo Di Carlo, PhD

December, 2018 ii

DEDICATION

To the entire Lower Fungom

iii

UNIVERSITY OF BUEA

FACULTY OF ARTS

______

CERTIFICATION This is to certify that the work described in this thesis entitiled “The sociolinguistic dynamics of rural multilingualism in Africa, the case of Lower Fungom”, was carried out in the Department of Linguistics under the supervision of

Signature ______

Ayu'nwi Ngwabe Neba (PhD)

Date ______

and

Signature ______

Pierpaolo Di Carlo (PhD)

Date ______

iv

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to my supervisors, Prof Ayu'nwi Neba and Dr. Pierpaolo Di Carlo, for all their contributions: time, enriching ideas, patience and immense knowledge that made my PhD experience productive and stimulating. Their desire to see me make the best of this journey and soar in my career pulsated me through.

The National Science Foundation funded project, KPAAM-CAM (Key Pluridisciplinary Advances in African Multilingualism Cameroon), funded the research leading to this thesis. Prof. Jeff Good, the PI of the project, and Dr. Pierpaolo Di Carlo, along with the KPAAM-CAM team in Cameroon, kindly accepted me as one of their students in the project. I was able to do fieldwork, travel overseas to present parts of my thesis in conferences thanks to the funding and support of this project to which I will forever be indebted.

My immense gratitude goes to the Linguistics department of the University of Buea for the opportunity it gave me to study for a PhD, to Anna Belew for reading through the draft of this thesis, providing invaluable input and kindly providing invaluable literature which led to the improvement of the thesis. She equally allowed me to tag along with her during one of her field trips to Campo, a trip that broadened my perspective on the topic.

I would like to thank all my language consultants and the entire Lower Fungom community for their hospitality, for offering their invaluable time and efforts, for the realisation of this endeavour.

Finally, I my sincerest gratitude go to my family, siblings, for supporting me unwaveringly in several ways throughout my PhD programme. v

ABSTRACT

This thesis investigates and describes multilingual practices in an intensely linguistically diverse locality in the North West Region of Cameroon. It offers a case study on an under-researched topic examining the daily linguistic habits of a small group of highly multilingual individuals in Lower Fungom, a rural highly linguistically diverse community in the North West Region of Cameroon. Exploiting the central question "how do rural multilinguals in Lower Fungom use their linguistic repertoire" the thesis argues that localist ideologies having to do with participants, settings and the covert intention of the speaker, condition the code choice of individuals in this area. Using recorded naturally occurring conversations, semi-structured interviews and non-participant observation, it emerges that multilingual individuals still maintain and deploy in daily interactions their rich linguistic repertoires, with each indigenous language given equal value in ways that suggest the existence of a language ideology that radically differs from that dominating in urban centres. Data obtained reveal that (with their ideologies) participants conscientiously refrain from code swtiching and that a concept that has been used to explain how multilingualism functions elsewhere (di/polyglossia) does not hold grounds in Lower Fungom. The data thus reveals an inclination to the use of local languages to index relational rather than essentialist conceptions, a crucial addition to scholarly perpectives on African multilingualism. This ethnographic qualitative study serves as a contribution to sociolinguistic studies on rural multilingualism in Africa.

vi

TABLE OF CONTENTS

DEDICATION ...... ii

CERTIFICATION ...... iii

ACKNOWLEDGEMENTS ...... iv

ABSTRACT ...... v

TABLE OF CONTENTS ...... vi

CHAPTER ONE

GENERAL INTRODUCTION

1.1 Introduction ...... 1

1.2 Background to the problem ...... 1

1.3 Problem of the study ...... 6

1.4 Aims of the research ...... 6

1.5 Objectives ...... 6

1.6 Research questions ...... 7

1.7 Research approach and design ...... 7

1.8 Method of data collection ...... 8

1.9 Instruments and tools of data collection ...... 8

1.10 Sampling and participant selection ...... 9

1.10.1 Sample size ...... 10 vii

1.10.2 Setting ...... 11

1.11 Procedure for data collection ...... 12

1.12 Data analysis ...... 13

1.12.1 Coding ...... 14

1.12.2 Theoretical saturation ...... 14

1.12.3 Constant comparison ...... 14

1.13 Theoretical framework ...... 14

1.14 Delimitation of scope ...... 16

1.15 Organization of the study ...... 17

CHAPTER TWO

LITERATURE REVIEW ON CONTEXTUAL AND CONCEPTUAL

BACKGROUND

2.1 Introduction ...... 19

2.2 Literature on Lower Fungom ...... Error! Bookmark not defined.

2.3 Terminologies related to multilingualism ...... 25

2.3.1 Diglossia ...... 25

2.3.2 Triglossia ...... 31

2.3.3 Code switching/mixing ...... 57

2.3.3.1. Definitional issues ...... 57 viii

2.3.3.2. Code switching in Africa – urban and rural ...... 59

CHAPTER THREE

LITERATURE REVIEW ON THEORETICAL FRAMEWORK

3.1 Introduction ...... 70

3.2 Language ideology ...... 70

3.2.1. Indexical order ...... 73

3.2.2 Giles’s Accommodation Theory ...... 84

3.2.3 The ‘markedness’ theory of code-switching ...... 85

3.3. Review of literature that informs the methodology ...... 87

3.3.1 Grounded Theory Model ...... 87

3.3.1.1 Using the GTM for data analysis and interpretation ...... 89

3.3.1.1.1 Coding ...... 89

3.3.1.1.2 Constant comparison ...... 90

3.3.1.1.3 Theoretical saturation ...... 90

3.4 Research design ...... 91

3.5 Ethnographic approach ...... 94

3.6 Instruments of data collection ...... 99

ix

CHAPTER FOUR

METHODOLOGY

4.1 Introduction ...... 102

4.2 Research design and approach ...... 102

4.3 Method of data collection ...... 104

4.4 Instruments and tool of data collection ...... 105

4.4.1 Sociolinguistic interview guide ...... 106

4.4.2 Video and audio recordings ...... 108

4.4.3 Observation (Narrative field notes) protocols ...... 110

4.5 Procedure of data collection ...... 112

4.5.1 Audio and video recording ...... 112

4.5.2 Non-participant observation ...... 117

4.6 Piloting ...... 119

4.6.1 Collective versus individualistic perspectives ...... 119

4.6.1.1 Schools ...... 120

4.6.1.2 Churches ...... 122

4.6.1.3 Njangi houses ...... 123

4.6.1.4 Market ...... 123 x

4.6.1.5. Homes ...... 124

4.7 Sampling strategies ...... 126

4.7.1. Setting ...... 126

4.7.2. Population sample ...... 128

4.7.3. Participants ...... 131

4.8 Data analysis ...... 132

4.8.1 Grounded theory ...... 133

4.8.1.1 Coding in grounded theory ...... 133

4.8.1.2 Constant comparison ...... 135

4.8.1.3 Theoretical saturation ...... 135

4. 9 Data presentation ...... 137

4.10 Ethical consideration ...... 138

CHAPTER FIVE

PRESENTATION OF PARTICIPANTS AND THEIR DAILY LINGUISTIC

ACTIVITIES (SELF REPORTS)

5.1 Introduction ...... 140

5.2 The sociolinguistic profile of primary consultants as reported in the sociolinguistic questionnaire ...... 140

5.2.1 Janet ...... 140

5.2.2 Boniface ...... 146 xi

5.2.3 Kulo...... 150

5.3 Minor consultants ...... 157

CHAPTER SIX

PARTICIPANTS’ DAILY ACTIVITIES (NATURAL INTERACTIONS)

6.1 Introduction ...... 165

6.2 Presentation of findings ...... 165

6.2.1 Kulo...... 165

6.2.1.1 Data obtained through observation: languages used by Kulo ...... 166

6.2.1.2 Data obtained through recordings ...... 175

6.2.2 Janet ...... 177

6.2.2.1 Data based on observation ...... 177

6.2.2.2 Data obtained through recordings ...... 179

6.2.3 Boniface ...... 180

6.2.3.1 Data gathered through observation ...... 181

6.2.3.2 Data obtained through recordings ...... 181

6.3 Characterization of the three consultants ...... 183

CHAPTER SEVEN

MULTILINGUALISM WITHOUT CODE SWITCHING?

7.1 Introduction ...... 185 xii

7.2 Definitional issues ...... 185

7.3 Code switching around the world ...... 187

7.4 Code Choice in Lower Fungom ...... 189

CHAPTER EIGHT

LANGUAGE CHOICE IN LOWER FUNGOM

8.1 Introduction ...... 211

8.2 The notion of di/polyglossia in Lower Fungom ...... 212

8.3 What motivates language choice among Lower Fungom multilinguals? ...... 218

8.3.1 The co-existence of humans and the supernatural ...... 218

8.3.2 Accommodation ...... 221

8.3.3 The desire to communicate in secret ...... 229

8.3.4 The need to mimicking ...... 230

8.3.5 The need to obtain a favour ...... 231

8.3.6 Avoidance ...... 234

8.3.7 The need to convey a thought ...... 238

8.3.8 Communication ...... 239

8.3.9 Protection ...... 241

8.3.10 Topic ...... 243

8.3.11 Gender of the interlocutor ...... 244 xiii

8.4 Conclusion to the thesis ...... 246

8.4.1 Recap of the problem ...... 246

8.4.2 Aims of the research ...... 246

8.4.3 Summary of main parts ...... 247

8.5 Limitations of the study ...... 250

8.6 Significance of the study ...... 251

8.7. Recommendations ...... 253

References ...... Error! Bookmark not defined.

APPENDIX 1 CV ...... Error! Bookmark not defined.

APPENDIX 2 Interview guide ...... 262

Subject Index ...... 264

Author Index ...... 266

1

CHAPTER ONE

GENERAL INTRODUCTION

1.1 Introduction

This chapter is an overview of the thesis. As an introductory chapter, it explains how the interest in rural multilingualism developed. It discusses the background to the problem, the problem, the aims, the methodology, significance of the study, research questions, the theoretical framework and the scope of the study.

The structure of the thesis is then outlined at the end of the chapter before a conclusion.

1.2 Background to the problem

Multilingualism is considerably familiar and widespread nowadays although it still tags along a series of arguments. The first bone of contention comes with the distinction between the terms multilingualism and linguistic diversity. We use the term multilingualism here as in Di Carlo, Good & Ojong Diba (2019) to refer to individual multilingualism, where “a multilingual individual is anyone who can communicate in more than one language, be it active (through speaking and writing) or passive (through listening and reading)” (Li, 2008). In this sense of the term, a geographical space such as a country wherein individuals know and use several languages is rather refered to as linguistically diverse and not multilingual. 2

Secondly, distinguishing between “urban” and “rural” environments in Africa has also been long debated. Polome (1982) in his study of multilingualism in

Tanzania asserts that it is very difficult to establish a clear-cut distinction between what is considered as rural and urban. According to Di Carlo et al.

(2019: 3) and for this study, we understand rural environments to be characterized by relative lack of demographic pressure and industrial development, where most inhabitants are engaged in food production.

These arguments have impeded a comprehensive study of multilingualism in

Africa, especially because rural settings are erroneously considered by many to be linguistically homogeneous. This is however not the case with Lower

Fungom. Lower Fungom is an area of about 240 square kilometers located in

Menchum Division, in the North West Region of Cameroon (Di Carlo &

Pizziolo, 2012). According to these authors, its topography is characterized by numerous steep hills. They describe the environment as the forest-savanna mosaic type, dominated in wooded areas by palm trees and by elephant grass.

Lower Fungom comprises thirteen villages in which Di Carlo (2016: 77) identified at least eight different Bantoid languages, or small language clusters.

According to him, a rough account of language density would yield a figure of about one language per 34 square kilometres. Consequently, Lower Fungom, located at the northwestern edge of the Cameroonian Grassfields, an area itself 3 long known for its striking degree of linguistic diversity (Stallcup, 1980: 44) would seem to be one of the linguistically most diverse micro-areas of the

African continent (Di Carlo & Pizziolo, 2012). Lüpke (2016) goes further to say that Lower Fungom is one of the most linguistically diverse regions in the world, with patterns of multilingualism that predate the colonization, the spread of European languages and the emergence of Pidgin and Creole languages.

Lower Fungom’s thirteen villages each claim to be speaking a different language. They admit that some of them are similar in some respects. Agwara

(2013) says that “what a linguist will define as dialects of the same language is, for the people of Lower Fungom, separate languages, though similar to one another. Good, Jesse, Mve, Nganguep, Voll, & Di Carlo (2011: 105) emphasize that although some of these villages have mutually intelligible languages, each of these languages is “unambiguously distinct”. 4

Figure 1: Maps showing the distribution of languages and thirteeen village of Lower Fungom (Di Carlo, 2016: 78) 5

SUBGROUP LANGUAGE VILLAGE POPULATION Yemne-Kimbi Mungbam [mij] Abar ca. 800 Munken ca. 600 Ngun ca. 150 Biya ca. 100 Missong ca. 500 Mufu-Mundabli [boe] Mufu ca. 100 Mundabli ca. 400 Buu [no ISO code] Buu ca. 100 Fang [fak] Fang ca. 4,000 Koshin [kid] Koshin ca. 4,000 Ajumbu [muc] Ajumbu ca. 300 Beboid Naki [mff] Mashi ca. 200 Central Ring Kung [kfl] Kung ca. 700

Table 1: Languages of Lower Fungom (Di Carlo, 2016)

Multilingualism practised in urban areas is indeed different from that practised in rural areas like Lower Fungom. In urban areas, multilingualism is intense, typically involving at least one lingua franca, a home language and a colonial language. In rural areas the situation is very different, with the exclusion or minimal use of colonial languages. It is remarkably common for people to have a repertoire of over ten languages which they use very frequently daily.

6

1.3 Problem of the study

Although there have been several studies on the dynamics of multilingualism and linguistic diversity over the years, many of these studies have examined this phenomenon in urban areas. Rural multilingualism, as the present thesis reveals, is very different from urban multilingualism. Also, worthy of note is that about

60% of Africans live in rural settings (FINCA International, 2015). With many studies on multilingualism focused on urban areas and on colonial languages, very little is known about rural multilingualism where mostly local languages are in active use daily. For multilingualism and linguistic diversity to be fully understood, data and perspectives from rural areas too, are of utmost importance.

1.4 Aim of the research

This is a small scale sociolinguistic study which first and foremost set out to examine and describe how multilingualism functions in a rural setting. This study set out to investigate the kinds of linguistic activities attested in Lower

Fungom. It also sought to explore the language ideologies governing language use in Lower Fungom.

1.5 Objectives

To achieve the aim stated above, this study set out to:

1) Describe individual rural multilingualism in Lower Fungom; 7

2) Account for the kind of sociolinguistic dynamics of multilingualism- code

switching and language choice attested in this area;

3) Investigate the social motivations and impact of the sociolinguistic

dynamics of multilingualism in Lower Fungom.

1.6 Research questions

Based on the aims and objectives of this study, three research questions were formulated:

1) How do typical rural multilinguals in Lower Fungom use their linguistic

repertoire?

2) What kind of linguistic activities exist in Lower Fungom?

3) Why and with what impact do multilingual individuals in Lower Fungom

choose one language over another?

1.7 Research design and approach

This research sought to provide understanding on a sociolinguistic phenomenon- rural multilingualism. To provide a fine grained description of the topic, a qualitative design was selected. The qualitative design aimed for a complete, detailed description of multilingual practice events including their context and circumstances. Concomitant to this design is the inductive approach. The inductive approach which does not involve the formulation of any hypothesis was chosen for this study. The study started once we had formulated the 8 problem, aims and objectives, research questions. We then went to the field to collect data to answer the questions in order to attain the set objectives. After collecting the data, we looked for patterns, resemblances and regularities in the linguistic practices and then reached conclusions, making the research hypothesis generating.

1.8 Method of data collection

In addition to selecting an inductive approach, the inquiry was a case study which according to Dornyei (2011), allowed the researcher to dig deep into the lives of the selected participants and collect extensive data for the period available for the study. In conformity with Dornyei (2011) the case study led us to collect and organize data to maximize our understanding of the multilingual dynamics of a limited selected highly multilinguals of Lower Fungom. Case studies by default make use of an eclectic approach to data gathering. These include observation, interviews and recordings.

1.9 Instruments and tools of data collection

Using an eclectic approach, we made use of an ethnographic sociolinguistic interview guide (see the appendix 2) to elicit demographic variables and individual ideologies. An observation protocol was also used where necessary.

9

Also, we made use of non-participant observation for this study. In a bid to avoid controlling the data, to maximize its natural flow, we observed without telling the consultants the specific activities they needed to do or where they needed to go. We let them live their lives as they did prior to our arrival in

Lower Fungom. In the case where the consultant was mobile, like on his/her way to one of the other villages or just walking around the market, observation was used, and unstructured interviews employed for further information. If the participants were at their homes or if they visited a home in their villages or another stable setting, then a video or audio recording was prioritized.

We did not want to work only with audio recordings for fear that we would lose the nonverbal aspects of the original communication situation such as facial expressions and gestures. The respondent’s emotional overtones can be crucial with regard to the real meaning of the message. For this reason, relevant non- verbal expressions were also noted through video recordings.

1.10 Sampling and participant selection

Participants for the study were selected using a multistaged approach. First, the researcher interviewed fifteen participants whom we were told spoke multiple languages. The goal was to identify multilinguals suitable for the study. After the analysis of the data from the sociolinguistic interviews, the researcher identified real multilinguals suitable to be main participants in the study. It was 10 crucial to select only those who were multilinguals in the community given that the investigation was designed to study rural multilingualism.

Participants were therefore selected based on the reason that they best informed the research questions and enhanced understanding of the phenomenon under study. We selected three consultants from different villages. These consultants were requested to wear a digital audio recorder as they went about their regular daily activities. We listened to the recordings with each of the main participants and discussed the secondary participants with whom they interacted and their linguistic behaviours. From this, we further selected another set of participants; those participants with particular characteristics, and with a more complex, unexpected linguistic habit. To these participants, we had unstructured interviews to understand their language choices (A sample of the interview guide is found at the appendix).

1.10.1 Sample size

Once a sampling technique has been determined, the researcher must consider the sample size. Since this research was qualitative, the sample size was not predetermined. The number of participants depended upon the information required to fully inform the phenomenon under study. We drew a line of completion on the participants when additional interviews did not result in identification of new concepts (saturation). According to Dornyei (2011): 11

Qualitative inquiry is not concerned with how representative the respondent sample is or how the experience is distributed in the population. Instead, the main goal of sampling is to find individuals who can provide rich and varied insights into a phenomenon under investigation so as to maximize what we can learn (p.126).

1.10.2 Setting

As far as setting is concerned, we set out without establishing boundaries as well. We followed our consultants or simply recorded their conversations regardless of where they were located. Geographical space was not of primary concern; consultants were. We followed them even when they travelled out of

Lower Fungom for some time during our field trip. However, after our first field trip, we found major consultants to work with from Abar, the centre of the thirteen villages, the village with the lone secondary school in the area, the only major market and the only health centre. Abar is consequently more heterogeneous than the rest of the villages. We also worked with an individual from Missong, another village of Lower Fungom not far from Abar. As a matter of fact, we worked with individuals from Abar and two of its neigbouring villages: Buu and Missong. From our feasibility study we discovered that the farther one went from the centre- Abar, the less multilinguals suitable for the study one would find. We found people with repertoires of over eight languages in some of the other villages of Lower Fungom such as Munken, but they were elderly; their ability to recall facts, or just sit down, listen and discuss the 12 lengthy recordings was limited. Abar, Buu and Missong were areas we were convinced would generate the quality of data suited for the study. (For a detailed description of the location of the research site, see map 1 in chapter 2)

1.11 Procedure for data collection

We scheduled specific periods of daily continuous monitoring with the three major consultants, for three months; one month per trip. We expected that one month would allow participants time to become accustomed to our being with them and to become less self conscious. This worked perfectly because our presence did not always immediately suggest work. We also expected that when interacting with others whom the participants know, the social obligations and linguistic ideologies would lead participants to disregard the fact that they were being monitored.

In sum, we programmed a feasibility study as a first trip during which we administered a general sociolinguistic questionnaire to several individuals from several villages of Lower Fungom. This was meant to single out those who were multilingual the most, suitable and available for the study. We also visited areas where multilingualism would definitely be practised; schools, churches and

Njangi houses. During the second trip, we tagged along the chosen participants for close to a month and collected data to examine how these multilinguals live their daily lives with their complex linguistic repertoire. Our aim was to observe 13 as many factors as possible, those which may be related to their code choices.

We observed patterns, behaviours and explanations. During the third trip, we gathered as much data as we did during the second trip but this time the analysis of it was different. We translated the data more meticulously, examining the multilingual dynamics such as language choice and code switching found in it.

1.12 Data analysis

Analysing qualitative data entails reading a large amount of transcripts looking for similarities or differences and subsequently finding themes and developing categories. Dornyei (2011: 246) asserts that “the first step in data analysis is to transform the recordings into textual form.” Qualitative research yields mainly unstructured text-based data. For this study, these textual data constituted interview transcripts, audio and video recordings of natural occurring conversations and observation notes. The languages/codes the consultants use, the persons with whom they discussed, the transcription and free translation and the researcher’s comments in relation to specific events were examined, significant patterns were identified and meaning drawn from them. Grounded theory was used in analyzing the data. Grounded theorists are concerned with or largely influenced by understandings of the world. To Strauss and Corbin (1998:

12), the main facets of this theory are coding, theoretical saturation and constant comparison. 14

1.12.1 Coding

Bryman (2004: 408) holds that “coding is the starting point of most forms of qualitative analysis.” It is the key concept in grounded theory. Here, data was reviewed and given names to component parts that seemed to be of potential theoretical significance, such as distancing and accommodation.

1.12.2 Theoretical saturation

This is a process that relates to two phases in grounded theory: the coding of data and the collection of data. Data was collected here and analyzed in a cyclical fashion, till the data stopped yielding new information.

1.12.3 Constant comparison

This is an iterative process in data collection and analysis. We examined and compared the occurrences of events in the speech of the consultants to find a pattern. Just like the other aspects of grounded theory above, comparison was very important for this study. The details of the methodology employed for this study are borne out in chapter four.

1.13. Theoretical framework

The theoretical framework is the foundation on which all knowledge is constructed for a research study. The theoretical framework provides a grounding base, an anchor for the data analyis. Thus, our choice of theories here was not arbitrary. The first theory used is that from frame semantics. Neubert 15 and Shreve (1992) define frames in terms of organization of experience and knowledge repertoire. This organization of experience maybe referred to as framing and the knowledge structures themselves as frames. Honig (1991) holds that frames are a combination of prior knowledge, generalization and expectations. In this same vein, Katan (1999) asserts that even the simplest of messages come with another message. This is called the meta message and it is this meta message which carries the force of the message and provides a clue to its interpretation. Katan cites Bateson (1972) who holds that frames are an internal psychological state that makes up part of our map of the world; they help listeners infer information. Our perceptions are always mediated by our assumptions, our beliefs and by the language we speak. Frames, to Neuman

(1992), are conceptual tools which media and individuals rely on to interpret and evaluate information. There are several frame types, speakers and listeners adopt these frames and see meaning through these frames as their culture predisposes.

Frames help make sense of experiences that are hidden deep within the language and traditions of the society. Katan (1999) adds that frames are embedded in culture.

Giles’ (1982) accommodation theory holds that people are motivated to adjust their speech styles or accommodate as a means of expressing values, attitudes and intentions towards others. It is proposed that the extent to which individuals 16 shift their speech styles towards or away from the speech styles of their interlocutors is a mechanism by which social approval or disapproval is communicated. A shift of speech style toward that of another is termed convergence and is considered often a reflection of social integration, whereas a shift away from the other’s style represents divergence and is considered often a tactic of social dissociation.

These theories helped us examine and ground our discussions on language choices and code switching.

1.14 Delimitation of scope

This study examined and described multilingualism in rural Lower Fungom.

Using self-reported data, video and audio recordings and non-participant observation, it shines a light on sociolinguistic practices in Lower Fungom.

There have been other sociolinguistic studies in Lower Fungom, which include an assessment of the level of multilingualism in Lower Fungom (Agwara, 2013).

Consequently, we did not test the level of multilingualism as a priority in this study; neither did we include phonology of the languages of Lower Fungom.

This study was strictly sociolinguistics; studying individuals, their daily linguistic practices and the language ideologies surrounding these practices.

17

1.15 Organization of the study

This thesis is divided into eight chapters. The first chapter is the general introduction of the thesis; it introduces every chapter or section in brief. Chapter two and three comprise the literature review. Chapter two examines, summarizes and evaluates what other authors such as Di Carlo and Agwara have written on sociolinguistics and the language situation in Lower Fungom. Literature to justify the aims of the research, the theoretical framework and the methodology of the study is reviewed in chapter three. The fourth chapter consists of the methodology of the study and chapter five comprises a detailed discussion of all the participants involved in this study and their linguistic habits.

Chapter six is a presentation of the first part of the data. It discusses how typical rural multilinguals use their linguistic repertoire daily.

Chapter seven examines instances of language use in Lower Fungom in comparison with instances of language use elsewhere in Africa and beyond. We investigate and discuss Lower Fungom as multilingual setting void of code switching.

Chapter eight presents another portion of the data. It discusses social motivations for code choices and their impact on the participants. It also discusses the limitations of the study, makes a number of reccommendations before concluding the entire work. 18

Conclusion to chapter one

This chapter has introduced the entire thesis. The background to the study, its aims and objectives, the problem of the study and research questions have all been discussed here. The chapter has also presented, briefly, the theoretical framework and the methodology of the thesis. The following chapter examines works that influenced this study. It provides information on the area of study

(Lower Fungom) and discusses works that demonstrate the need for the study. 19

CHAPTER TWO

LITERATURE REVIEW ON CONTEXTUAL AND CONCEPTUAL BACKGROUND

2.1 Introduction

The literature review, according to Cahn and Barnard (2012: 84) demonstrates that you are “fully aware of what has been published about the topic you are investigating, so that you can conclude the review by identifying the ‘research space’ that your thesis intends to occupy." In this chapter, literature is reviewed in sub-sections. The first section discusses literature on Lower Fungom in order to acquaint the reader with the area of research and also to highlight the gap this study fills. Subsequently, it examines works bearing some fundamental concepts related to this study such as di/polyglossia. Then, in section 2.3.1, scholarly works focusing on language ideologies is presented. Concepts such as code switching, mixing and language choice are also explored and then related to the realities of Lower Fungom and other areas around the world.

20

We begin by looking at works on Lower Fungom, in order to acquaint the reader with the environment as far as its linguistic system and geographical scenery is concerned and also with what has already been done in the region. Detailed pioneering studies in this area were undertaken by Jeff Good and Pierpaolo Di

Carlo while researching in this area since 2010. It is only appropriate for their studies to take the initial spot in this review. Good et al. (2011) and Di Carlo

(2011) are influential works that have guided this study. Both works orientated us towards choosing our area of study, with regard to the regions or villages that would be most rewarding for this thesis. The approximation of the population of the different villages is very important for this work. We needed the linguistic situation of the area and the map presented in the first work cited above to choose the villages from which to collect data. Their description of Lower

Fungom as an area of “astonishing degree of language density” (Di Carlo, 2011) convinced us of the relevance and weight of the study. Given that more elaborate and recent research on the complex linguistic system, historical and geographical background of Lower Fungom has already been provided (Good et al 2011; Di Carlo 2011and Di Carlo 2016), it is unnecessary to provide the very same, detailed information in this thesis. As a result, reference is simply made

when and wherever necessary.

21

Di Carlo and Good (2014: 2) explored the language dynamics of Lower

Fungom, and declared that the area offers an extreme case of linguistic diversity within the already exceptionally diverse Cameroonian Grassfields. Their study elaborately presents the linguistic classification of the languages of Lower

Fungom. We include just some parts of it in this thesis as a review, just to acquaint the reader with its area of focus. Di Carlo and Good hold that the languages of Lower Fungom have all been classified in the Bantoid group. "This puts them among the closest relatives to the well-known (Narrow) Bantu group of languages, which dominate southern Sub-Saharan Africa.” (Di Carlo & Good

2014: 7). They go further in saying that the primary basis for this classification is their Bantu-like systems of noun classes (Good et al., 2011). Their work attempts a critical description of the endangered languages of Lower Fungom. It presents the villages of Lower Fungom and their languages in great detail. Good and Di Carlo (2014) declare that in Lower Fungom, there is about one language per thirty-four square kilometers. They compare Lower Fungom to the famously linguistically diverse country of Vanuatu, which has about one language for every hundred square kilometers. We therefore realize that Lower Fungom although previously unknown to the world is more linguistically diverse than some of the areas that are even famous such as Vanuatu.

22

Di Carlo and Good (2014) look into what accounts for multilingualism in Lower

Fungom. They feel that the fact that this area constitutes autonomous villages which do not join into larger units such as kingdoms is connected to this phenomenon. Particularly relevant to the present study is their investigation of the feelings and attitudes of Lower Fungom residents with regard to this diversity. Di Carlo and Good examined some key aspects of the local language ideologies of Lower Fungom. From their study, we learn that what appears to make the languages of Lower Fungom special, “is not their value as self- contained storehouses of a culture but, rather, their utility as tools for the flexible construction of multiple identities" (Di Carlo and Good, 2014: 30).

Although, Di Carlo and Good (2014) say, this is not particularly new or uncommon in Africa. Multiple case studies have demonstrated that multilingualism is ‘the African lingua franca’.

Worthy of note is the fact that the population of these thirteen villages is not all long-time inhabitants of Lower Fungom; some of them migrated to this area at different times due to wars. We learned from our trips that most of the villages in Lower Fungom migrated into the area at different times due to wars. The original indigenes lived in just three villages: Abar, Buu and Ngun. The latter is the oldest. Every other village migrated to this area at some point in the distant past. Some of these villages nevertheless share traditions, while others, such as 23

Mashi and Kung, do not share traditions with any other village of Lower

Fungom.

In addition, Di Carlo and Good discuss how naming children is carried out in

Lower Fungom. This is important for this study, as it shows that the manner in which children are named suggests that the indigenes of Lower Fungom are in favour of the negotiation of multiple identities. In mixed marriages, Di Carlo and Good say that “the father’s language is the exclusive code to be used for communication with their paternal kin, whereas the mother’s language must be used with their maternal kin" (Di Carlo and Good, 2014: 23). To them, “this is the clearest … instance of the significance of multilingualism for the region’s traditions. It indicates that the local culture acknowledges (and prizes) the possibility for an individual to develop multiple social identities, stressing language as a major means to symbolize them” (Di Carlo & Good, 2014: 24).

They point out that this naming style is “merely one prominent, linguistically- oriented example of a more general tendency of maintaining networks of solidarity groups apparently common to much of sub-Saharan Africa. They cite

Kopytoff (1987: 24) who says that quintessential Africans typically attach themselves to several groups for several reasons; expecting support and offering it too when necessary.” Di Carlo & Good (2014: 25) hold that especially in times of conflict, traditional African individuals sought survival by “being under 24 the protective umbrella of one or another such group, and the larger and more powerful it was, the safer one was”. Kopytoff asserts that the most immediate and most secure groups of support were those based on ties of kinship

(Kopytoff, 1987).

Agwara's (2013) Master's dissertation is of great help to the present study. Her work first and foremost brings to light the root problem of this thesis. She declared that although a lot of work has been done on multilingualism around the world, much of it dwells on the urban sector, with the focus being on the interaction between colonial languages and local languages. She affirms that the rural sector has not yet been fully explored. Her study examined multilingualism in Lower Fungom; she joined her thoughts to those of Di Carlo & Good (2014) and dug deeper to determine what accounts for multilingualism in the area. She used ethnographic questionnaires to investigate the reasons for the linguistic diversity of Lower Fungom. She also examined this diversity in terms of the gender divide. As mentioned already in this study, there are individuals who have been identified as being able to speak all the languages spoken in all thirteen villages. Agwara found Lower Fungom as a suitable context for the investigation of the social motivations behind these unusual dimensions of multilingualism. Her dissertation principally sought to identify social motivations for high rates of multilingualism in Lower Fungom. One of her 25 hypotheses was that social motivations such as favoritism, friendship keeping, spouse pleasing, security, and pleasing in-laws are some of the reasons for multilingualism in Lower Fungom. Agwara (2013) is the first detailed sociolinguistic study that has been carried out in Lower Fungom. With the degree of multilingualism described by Agwara, this thesis adds to it by examining how this multilingualism is exploited.

2.3 Terminologies related to multilingualism

A basic fact about language is that no one can use two or more languages simultaneously. A pioneer study on the use of two or more languages was conducted by Ferguson (1959). He understood that linguistic diversity would work in a community wherein “two varieties of a language exist side by side throughout the community with each having a definite role to play” (Ferguson,

1959: 232). He termed such a situation diglossia.

2.3.1 Diglossia

Diglossia is a notion that is typically evoked when mutilingualism is concerned.

It is crucial to the study of language use in different situations and thus very pertinent to this study. Ferguson (1959) introduced this term and explained that there are many speech communities wherein two or more varieties of the same language are used by speakers under different conditions. He sheds light on this by giving examples of a country like Iraq where in Baghdad; Christian Arabs 26 speak a ‘Christian Arabic’ dialect when talking to one another but speak the general Bagdad dialect ‘Muslim Arabic’ when talking to a mixed group (p.

232). He labels the Christian Arabic variety (the superposed) as High or simply

H and the regional dialect- Muslim Arabic as Low or L. With this, Ferguson goes further to discuss some of the major features of diglossia. He states that one of the most important features of diglossia is the specialization of functions for

H and L. He asserts that in a certain situation only the H is appropriate and in another only the L is appropriate (Ferguson, 1959: 236).

It is generally agreed by scholars concerned with diglossia that it is characterized by prestige. As a matter of fact, the foremost characteristic of diglossia is prestige. Ferguson declares that speakers regard the H as superior to

L and that sometimes they even think that H is the real variety and L does not exist. He adds that there is usually the feeling that H is more elegant, more expressive and more logical.

What is pretty glaring in his discussion of diglossia is a community with two language varieties, with one of them highly considered. H is used out of the home and away from kindred while the other which has no particularly desired prestige and is typically considered substandard is used with kin. He also says that diglossia is a stable condition with varieties having separate and distinct functions. Ferguson provided an early perspective on linguistic diversity and 27 how it works. Diglossia according to Ferguson (1959) is concerned with variables such as prestige, the High and Low distinction and a stable society.

Ferguson through this inaugural study inspired several scholars to look into linguistic diversity. Joshua Fishman is one of these scholars.

After Ferguson advanced the notion of diglossia in 1959, the notion became popular and was extended by Fishman (1967). Ferguson’s study uses diglossia in connection with a society that uses language varieties for internal communication; it refers to separate codes within a society, each of these codes serving distinct functions. In Fishman’s (1967) case, reference is made to two languages. Fishman asserts that “Whereas one set of behaviours, attitudes and values are supported and were expressed in one language, another set of behaviours, attitudes and values are supported and expressed in the other” (p.

74). He amplifies Ferguson’s (1959) thoughts on H and L, as he too says that the separation between these languages “was most often along the lines of High on the one hand and utilized in conjunction with religion, education and other aspects of high culture and a Low language on the other hand.” (p. 74)

Fishman (1967) also discusses other scholars such as Gumperz, who explored diglossia further. He explained that it does not only exist in multilingual societies which officially recognize several languages. He presents several kinds of societies which practice diglossia to cope with multilingualism. One of these 28 societies is Paraguay where almost the entire population speaks both Spanish and Guarani. Paraguay practices diglossia in a form of officially recognized bilingualism. Fishman explains that:

The formerly monolingual rural population has added Spanish to its linguistic repertoire in order to talk and write about education, religion, government, high culture and social distance… whereas the majority of city dwellers … maintain Guarani for matters of intimacy and primary solidarity even in the midst of Spanish urbanity. (p. 75)

Fishman also distinguishes the above form of diglossia from the kind which involves bilingualism which is not widespread. He says that there are societies with national diglossia but not widespread bilingualism. Fishman explains that in such a situation, speech communities came together functioning as a single unit but with a socio-cultural distinction separating them.

The last but one kind of community he discusses is one which experiences bilingualism without diglossia, and finally that with neither bilingualism nor diglossia. The latter, he says, is easier to hypothesize than to find.

Eckert (1986) also discusses diglossia as propounded by Ferguson and Fishman.

Eckert (1980), just like Fishman (1967), defines diglossia as “the use in one community of two languages: a superposed variety, referred to as the high (H) language, which is reserved for use in more public, formal, learned domains and 29 a vernacular or low (L) language used in more popular and intimate domains”

(1980: 1054). We are interested in her work because she too, just like the scholars above, considers diglossia as an “organizing principle in bilinguals and bidialectal communities: a linguistic division of labour whereby each language is limited to its own domain” (p. 1054). She sees diglossia as a structured means of reserving the vernacular for in-group use while speakers use the standard language for entrance into the wider society. Eckert asserts that diglossia is a democratic arrangement insofar as it allows the vernacular to coexist with a high language. She also states that “the notion of diglossia brings language choice into the framework of structural linguistics by providing a structural-functional account of behaviour at the community level.” She adds that this functioning is a complementary distribution type; one which eliminates the possibility of a random language choice (p. 1054). She cites others (e.g. Fishman (1971: 87) who feel that this structural functioning (division of labour) allows the speakers to keep the two linguistic systems separate, and thus retain the structural integrity of each language. Particularly important for the present study is

Eckert’s aforementioned assertion that diglossia is frequently seen as a structured means of reserving the vernacular for in-group use and using the standard language for entrance into the wider society. It is important in this study to consider the implications of diglossia, with its “in-group” and “wider society” hallmarks, in rural settings. In Eckert’s view, “diglossia does not arise; 30 it is imposed from above in the form of administrative, ritual or standard language. By virtue of its political and economic status, this language becomes requisite for access to power and mobility within the society" (p. 1057). By this declaration, one would wonder once more how to apply this framework to a rural setting, where there is largely the absence of a quest for an access to power

(official or institutional) and also an absence of a single language that provides mobility within the society.

Ferguson (1959), Fishman (1967), Eckert (1980), Jaspers (2016) and Dorian

(2002) define diglossia and provide its characteristics in diverse settings. In fact, diglossia to these scholars is concerned with a high language and its low language/variety with well-known, clear-cut functions and different domains.

From these authors, it is also understood that diglossia is characterized by stability. Ferguson, as pointed out by Eckert (1980), declares that diglossia can be extremely stable, but only on the condition that the High language is restricted to the literate elite. In other words, this stability is dependent on rigid social stratification (1980: 1054). It can thus be assumed that communities or countries that lack a rigid social stratification may not be adequately explained through the framework of diglossia. Worth remembering throughout this study is the idea of "separateness" required of a diglossic situation that these scholars stress, and the fact that in diglossic situations, the choice of language is not an 31 individual's preference but paramount decisions from the powers that be.

Fishman (1967) talked about cases outside Africa, like Paraguay and the Swiss-

German Canton. Ferguson described the situation with Arabic, Greek, Haitian and Swiss German communities in which there exist compartmentalized H and

L varieties. What is interesting is that even in Africa similar situations have been found as seen below.

2.3.2 Triglossia

There are communities in Africa which are awefully multilingual so that a two language/variety sturcture would not work. Some scholars have thus proposed a three-language function system which they refer to as triglossia. Mkilifi (1972) is one of such scholars; he draws a comparison between the diglossia that

Ferguson and Fishman discuss with the situation in Tanzania. In his "Triglossia and Swahili – English Bilingualism in Tanzania," he explores the speech behaviour of bilingual speakers of English and Swahili in Tanzania. More specifically, he is concerned with the variables affecting language maintenance, code switching and language mixing among bilingual speakers of Swahili and

English in Tanzania. He describes this linguistic situation as triglossia or trilingualism (1972: 202). Mkilifi observes:

Three languages with both varying and overlapping roles interact, creating a triglossia situation: first the vernacular or mother-tongue of each particular ethno-cultural group; secondly Swahili, the local lingua franca and national 32

language; thirdly English, the predominant language of higher learning and to a certain extent of official and commercial business.

For his study he interviewed respondents living in the capital city of Tanzania-

Dar-es-Salaam. As such, this is a study based on urban multilingualism.

According to Mkilifi (1972), unlike that discussed by Ferguson, the relation between the languages in Tanzania is not to maintain strict segregation; this would be a problem because of the prevalence of code-mixing in conversation, as well as code-switching between Swahili and English. Thus, in Tanzania the function of languages is not strictly observed as suggested by proponents of diglossia. Mkilifi sees functions as playing a complementary distribution function. This mixing, in Mkilifi’s view, seems to be due to the “habit perhaps arising from lack of constraint to language mixing” (p. 203). Mkilifi says that although the vernacular is adequately expressive of its own socio-cultural context it may still be considered a restricted code. He goes on to say that some bilinguals found themselves assuming new socio-economic roles such as executives, technologists and politicians which affected their linguistic habit.

Urban life, he says, tends to impose its own socio-cultural influences on the bilinguals. He emphasizes that these are roles that the colonial government did not in their educational system specifically train these bilinguals to assume. As a result of this, there is free shifting and mixing between Swahili and English, 33 based on interlocutors, topics and setting. Cultural adjustments have constantly been made as roles changed. While the simultaneous use of both English and

Swahili demonstrates an imbalance in education and employment, the vernaculars demonstrate an ethnic identity and solidarity. As Mkilifi puts it, “the vernacular manifests a rural African sociocultural life characterized by a subsistence peasant/pastoral socio-economic organization” (p. 203), it is learned and used almost wholly in face-to-face situations. In this paper, Mkilifi also asserts that Swahili in Tanzania is based on a more urban type of African culture, and he claims that although Swahili in Tanzania has a greater range of syntactic and lexical differentiations than the vernacular it is still in its basic form and is thus representative of a pre-industrialized non-technological culture.

In this sense it is somewhat a restricted code. English in Tanzania is generally the dominant language of official business, commercial and legal transactions, and higher education. It is also a language through which almost all world literature and technological and scientific information is acquired. According to

Mkilifi, it is an elaborated code in a complementary relationship with Swahili.

From his paper, we learn (among other things) that the linguistic scenery in

Tanzania is not based on a strict compartmentalized, role-oriented language system like Ferguson (1959), Fishman (1967) and Eckert (1980) advanced. This also implied instability in triglossia in contrast with the stability maintained by scholars discussed earlier in this chapter. As far as similarities are concerned, 34 there is still the presence of H and L which are referred to in this paper as elaborate and restricted codes respectively.

v Criticisms to di/polyglossia

Jaspers (2016) says that when the notion of diglossia became much more popular, several scholars examined Fishman’s (1967) matrix with their own data and realized that the notion or term needed some modification in certain contexts. This modification was at the level of the prefix for –glossia; to some scholars, this term became known triglossia. Aside Mkilifi above, another one of these scholars is Johnson (1975). We move from the language situation in

Tanzania to that in Ghana with Johnson (1975). He defines triglossia as

A type of language situation characterized by a similar division of communicative functions among three language varieties, a vernacular and two superposed varieties, one of which is an indigenous lingua franca and the other an introduced world language. The three languages are not necessarily even related” (p 93).

Johnson (1975) describes the language situation at Larteh, a town in Southern

Ghana, 35 miles north of Accra. He holds that indigenes of Larteh are multilingual; they have Larteh, Twi and English in their repertoire. Like many other scholars, Johnson feels that language choice is meaningful. He explains that a switch from English to Twi or Larteh or from Twi to Larteh could mean loyalty to home town or an ethnic group. In addition, a switch from Larteh to 35

Twi or Twi to English can signal education, because these languages are formally learned, although English is learned in school and Twi before formal schooling begins. Larteh is the first and best language of the Larteh while

English is universally regarded as a prestigious useful language (1975: 99).

Johnson goes further in saying that triglossia suggests the existence of three levels of communication in Africa. These levels are “local (Larteh), regional

(Twi) and (Inter) national (English)” as Johnson explains; the above divisions are associated with vernaculars, indigenous lingua francas and introduced world languages, respectively. This according to Johnson (1975) naturally turns toward triglossia. Nider and Wonderly (1971) agree with this claim. They assert that every language situation consists of three “major communication rules” which may be realized through 1, 2 or 3 languages. Johnson (1975) declares that this three-level pattern is widespread in Africa and it is largely due to two factors: linguistic diversity and a high degree of intercommunication. He goes on to cite

Whiteley (1971: 13) who holds that “it is role and code versatility that is rewarding and role and code limitation that impoverishes” (1975: 100). He thus claims that in most African multilingual societies like Larteh in Ghana, the division of communicative function is shared among three languages; that which has to do with education will naturally come with prestige while the lingua franca and vernacular will be used for accommodating or distancing.

36

Dakubu (1997) also shares her thoughts on diglossia and triglossia; she understands that a trilingual model of multilingualism as proposed by Johnson is particularly applicable in Africa. She cites Mkilifi (1972) who, as discussed above, proposed that modern East Africa is typically "triglossic," with communicative functions shared among three language varieties that need not be closely related or even related at all. She however has some critical thoughts on triglossia. Dakubu makes reference to Johnson (1975), discussed above, who used triglossia to explain how linguistically diverse Larteh, a small hilltop town in Ghana copes with its languages. Dakubu (1997) admits that it is true that in many places in Africa today, observable language practice may be described by the term trilingualism, for example: "English-Swahili-Luganda in Kampala … or English-Akan-Ga in Accra” (P. 33). She even cites Myers-Scotton (1975,

1976) who also presents a trilingual model of language use in Kampala

(English-Swahili-Luganda), Nairobi (English-Swahili-Kikuyu), and Lagos

(English-Pidgin English-Yoruba).

Nevertheless, she has some criticisms as far as triglossia is concerned, and these are of interest to this study. It is also important to point out that just like

Johnson; Dakubu also worked in an urban community in Ghana: Accra, not far from Larteh. She holds that one glaring difficulty with the use of the notion of diglossia or trilingualism to describe language use in Africa is the existence of 37 several languages used simultanaenously as a second language, often by the same people in a community. She uses Accra as an example as she states that

English, Akan and Ga are used as second languages and all three languages are in competition as lingua francas especially among migrants, who speak none of them as their community languages (1997: 34). In Larteh for example, Larteh can be used domestically, but so can Akan. It is English that is primarily used in schools. As a result of the fact that several languages can be used at the same time as a second language (that is, occupying the same domain as a communicative function), Dakubu finds the popular “triadic configuration” questionable. She states that the "stable trilingualism" model not only misses the essence of a situation that is neither strictly trilingual nor inherently stable but also fails to give adequate explanation as to why a particular language is preferred (p. 35).

Dakubu asserts that trilingualism sheds no light on the fact that one individual could use “more than one language in the same or very similar domains, sometimes in combination, like with the phenomena of code switching and mixing”. One of the bases of her argument is the Guang speakers who use Akan with their children, even though they know the Guang language. She finds it difficult to see how a model that is based on “generalized theories” can adequately explain language choice. If Dakubu has a problem with generalized 38 theories, she could be advocating for more meticulous studies that would generate local theories. Despite this model, Dakubu notes that, we are still left to wonder why people learn more than one language, only to use one of them, depending on the occasion.

Jaspers (2016) is also concerned with diglossia. His paper describes diglossia’s provenance, its major conceptualizations by Ferguson and Fishman, and its major contribution to the study of language and society and ends up with some of the cracks in diglossia and triglossia. First and foremost, it should be mentioned that he shares the same definition of diglossia with those mentioned above and also admits that diglossia is very popular in sociolinguistics. He says that one of the reasons for this is that people talk and write differently even in the most homogenous of communities. What is particularly important for this study is Jasper’s emphasis on the different but principled ways in which people use language. He adds that these principles are so important that those who fail to observe them have to face consequences. He believes that one must be well versed with the linguistic norms of a society to function properly in it.

However, Jaspers believes just like Dakubu (1997) that diglossia is not the most appropriate concept to explain the social use of language for the same reason that it is preoccupied with universal causality principles, so much so that the ideological nature of the linguistic divisions of labour tends to be overlooked. 39

Jaspers, in his quest to better elucidate this concept, provides a history of how it has been used in talks about Katharevousa – a ‘purer’ version of Greek, and

Dhimotiki, also known as ‘popular’ Greek. He says that Katharevousa was a language that had been manipulated so that it could bear a resemblance to classical Greek, and it was for the elite who could afford education. This turned the majority of Greeks into “impure” speakers. Dhimotiki was “the language of the people”; it outclassed the “artificial” Katharevousa in speaker numbers, vitality and perceived authenticity. As movements of emancipation grew across

Europe during that period, some intellectuals began to kick against this linguistic defilement as having detrimental effects. Meanwhile, supporters of the prestigious variety kicked against this resolution and continued to stigmatize

Dhimotiki. This difficult linguistic scenario pushed Psichari (1929) to come up with the term “Diglossie” in French which he used to label the internal bilingualism that characterized the Greek linguistic situation at the time. Jaspers

(2016) adds his voice to that of Fishman (1967) who stresses that the term

“bilingualism be reserved for individuals’ linguistic versatility and diglossia for wide spread norms guiding the allocations of codes across functions, regardless of these codes’ genetic relationship.” Jaspers continues to cite Fishman’s “short but persuasive” article. He says that Fishman was highly concerned with protecting indigenous and minority languages as a result of decolonization of the

1950s and 60s. Fishman’s argument for diglossia suggests that if each language 40 had its own exclusive domain, “its own space in which it alone is normatively expected” (Fishman, 1991: 85), they could exist side by side.

Jaspers (2016) also discusses Ennaji and Sadiqi (1994) who talked on the quadriglossia existing in Maghreb and the Arab world. Jaspers also points out that diglossia has been used to draw attention to the fact that people routinely associate ways of speaking with typical settings, activities and social personae.

He says that these ways of speaking as well as setting, activities and people with whom they are associated are differentially valued; they come in higher and lower versions. Jaspers is preoccupied with the fact that diglossia is typically focused on universal patterns of causality. Another reason why he is not fully in support of the division of communicative functions is that their logic resides in strict observation of domain and rigorous code compartmentalization. This is a problem to him as he feels that human behaviour can hardly be kept in check or restrained like studies of diglossia suggest. This leads us to what Jaspers calls

“leaky diglossia” which is a situation where “one variety ‘leaks’ into the function formerly reserved for another. He even discusses further how leaking is not haphardzardly done and that this exercise is also meaningful. Jaspers holds that if diglossia considered local perceptions, it would have been a much better principle.

41

Jaspers continues in his criticism to this linguistic situation by adding that merely observing that H (High) is H and L (Low) is L falls short of explaining how these varieties came to be H or L in the first place; it disregards their implication in a social struggle over what comes as appropriate or best language to be used. He, like Eckert (1980) asserts that the stability that diglossia in a number of cases exemplified, the stability that Fishman was so attracted to, works best in societies with rigid social stratification (p. 1054).

From the above, one would retain that diglossia has been extended since

Ferguson and used remarkably to explain how communities deal with the variety of codes with which they are faced. Di/polyglossia revolves around the notion of prestige, hence the use of the label H and L for languages and their varieties.

Prestige, per the literature discussed above, is the mainstay behind the production and reproduction of the hierarchy of languages in people’s ideological organization of their linguistic space. By this, it is also clear that if in a given society prestige can be attained through language use and not just by birth, then, there is a possibility for upward social mobility in that society.

However, there is need to be cautious with the assumption that diglossia can lead to social mobility-for example, one cannot just speak English and forge their way into wealth in Tanzania if they were born poor and uneducated.

Instead, it is rather probable to speak English if one were born rich and went to 42 fancy schools; it is a give-and-take relationship where the use of the H variety signals the prestige you already have, in addition to accrued prestige.

As progress is made in this study, the language ideologies motivating language choice and use in Lower Fungom will be examined and analogized with the communities discussed above in greater detail. It is important to examine linguistically diverse communities such as Lower Fungom through the lens of ideas, such as prestige and compartmentalization, embodied in diglossia.

In discussing multilingualism, we examined some scholars who handled it in terms of a division of labour of communicative functions. Scholars have been discussed above who purport the notion of diglossia, and those who modified it according to their data and referred to it as tri- or polyglossia. We have also discussed scholars who, as per the communities in which their studies were conducted, proved that these divisions in communicative functions are not always practical; such divisions do not address local or particularized perceptions of language use, but rely on generalizations. Even assuming validity of these generalizations, they fail to account appropriately for language choices and the labels (H and L) these languages have accrued. Below are more thoughts and reflections as concern these notions supported by methodical research. The studies discussed below examine linguistic diversity using methodologies that 43 constitute detailed case studies. We move from universal theories on bi/multilingualism to more localized theories.

Alternatives to di/polyglossia

It is worth mentioning that the group of scholars discussed below has undertaken studies similar to the present thesis (a study of rural multilingualism in Africa).

This is in contrast to the studies discussed above, which were carried out in urban settings.

First and foremost, François (2012) examined the dynamics of multilingualism in the Torres and Banks Islands, two small archipelagos located in the northern parts of Vanuatu. These areas are home to 9,400 inhabitants and 17 distinct languages. He observes that with an average of 550 speakers per language, the

Torres and Banks Islands constitute an extreme case of the linguistic fragmentation which is typically observed throughout Melanesia. This macro- region consists typically of small-scale, egalitarian societies among which cultural and linguistic diversity is the norm. François (2012) aimed at observing the linguistic diversity of this “socially coherent cluster of islands” (2012: 86).

François asserts that “social networks in the world are potentially subject to two kinds of pressures”. He calls the first kind of pressure “centripetal”, and explains that this type results in convergence (where differences are disregarded) and the 44 second type of pressure he terms “centrifugal” which results in divergence where differences are emphasized (2012: 85).

He draws a comparison between societies such as modern European societies, which tend to value and promote homogeneity and some other small-scale societies in the world that show a greater tolerance for diversity. François cites

Evans (2010: 14) who was of the opinion that such small-scale societies just like

Lower Fungom are “economically self-sufficient, and proudly form the center of their own social universe without needing to defer unduly to more powerful outside groups.” Evans adds that the independence of these societies, their small and manageable sizes, and their valuing of variety keeps outsiders at a suitable distance without the need to side linguistically with large numbers of other people in the world. (2010: 86)

For better understanding it is important to make a connection between

François’s area of study and that of this thesis. The Torres and Banks Islands of northern Vanuatu form a relatively small area, comprising 10 islands and approximately 50 villages. François studied two of these islands. These 50 villages speak 17 different languages. These languages, François says, are all closely related, yet have historically diverged so much that they have lost mutual intelligibility (François, 2012: 88). Note particularly that there is no mutual intelligibility between these languages unlike in Lower Fungom where some of 45 the lects/languages are mutually intelligible. Many of these languages are spoken in a single village or two; the maximum is six villages for one language, and the average is three (fifty villages for seventeen languages). Lower Fungom is a region with thirteen villages with one language (indigenous language) per village, as aforementioned, some of which are intelligible. Fifty villages in

Francois’s setting use seventeen languages whereas the geographically smaller community of Lower Fungom uses at least eight distinct languages and five dialects in its thirteen villages.

Relations of trade, exchange and interisland marriages, have defined a social network in which cultural and linguistic contact have sustained multilingualism in this area. Just like in almost every part of Lower Fungom (with the exception of Mundabli). Intermarriages and movement of families to language communities different from their own promote diversity; parents tend to raise their children in more than one language. Consequently, kinship networks extend from one part of the community to another. François explains that what maintains this region’s linguistic diversity is the “particular manners of cooking, mat-weaving, dancing or singing” which he says may differ, “sometimes quite subtly, between two local groups” (p. 92). François argues that while linguistic diversity is arguably triggered by the desire for disparate social emblematicity, it needs egalitarian multilingualism to be maintained over generations. These 46 small-scale societies are marked by egalitarianism in the fact that each village community’s political power is distributed horizontally across families, with little village-internal hierarchy. In addition, the relationship between local communities is one of mutual respect and peaceful alliance, with virtually no relationship of dominance or prestige of one community over the others. It is our aim to stress here with François that this egalitarianism between social groups is mirrored in the balance of power between languages. According to François no language in this region is ever represented as more prestigious, useful, or important than another. As a result, there is no pressure for any of these villages

“to align its language to that of another one …all languages of the region are deemed equal in this respect.” (p. 93)

In this paper, François (2012) used two isands in Melanesia to discuss what causes linguistic diversity and what keeps a community linguistically diverse over a long period of time. François talks about egalitarian multilingualism and the absence of a lingua franca as factors that promote and maintain multilingualism in the Torres and Banks Island. He also mentions the lack of prestige and linguistic stratification in his area of study. This is the first rural area seen in this study where linguistic diversity is handled without any sort of linguistic division of labour.

47

François is not alone in his exploration of multilingualism minus di/polyglossia;

Friederike Lüpke’s research is also focused on a rural environment. The findings of her study significantly showcase the difference between urban multilingualism and rural multilingualism. These differences highlight the need for more localized models for accounting for the functioning of linguistic diversity. Like the works of Alexandre François discussed above, Lüpke (2016) is also focused on small-scale multilingualism. She studied communicative practices in a linguistically diverse community, one which happens to be a textbook example of a community in which “multilingual interaction is not governed by domain specialization and hierarchical relationships of the differently named languages and lects used in them, but by deeply rooted social practices within a meaningful geographic setting.” She also explores the language ideologies governing them and their known patterns of language use.

Lüpke advocates for more studies of small-scale multilingualism, as this in her opinion, will greatly advance our understanding of language contact and of the social conditions that have shaped language use in human history.

What is of great interest to us is that just like François (2012), Lüpke examines small-scale multilingualism in non-polyglossic settings not only in Africa, but beyond: in Oceania, South America and Australia (Lüpke, 2016: 37). Lüpke has similar preoccupations to those of this thesis; she is concerned with under- 48 researched roles of multilinguals as individual agents of linguistic diversity. She emphasizes that the role of individuals as agents of multilingualism, their creativity, and the social meaning they attach to language at an ideological level, have generally been ignored although these are fundamental for an understanding of multilingualism. Lüpke frowns at approaches to the study of language use that disregard speakers’ opinions about their intentions. She feels that they typically do not show a complete and authentic reality on this concept

(2016: 38).

Lüpke (2016) adds her voice to Di Carlo (2016) discussed in subsequent paragraphs, to call attention to the fact that each and every society is different and peculiar in its own way. Evidence of this fact is the degree of contact with the other languages in a given community, “the different patterns of language acquisition and socialization, individual movement and migration patterns and different types of social networks in an individual’s life span.” (2016: 39) Lupke affirms that all these will most certainly affect language use in subtly different ways. Worthy of note is that a study on small-scale multilingualism like the present one pays strict attention to subtle details.

We should recall that François (2012) made mention of the absence of a lingua franca in Melanesia which was one of the reasons for the maintenance of multilingualism in that area. Lüpke talks here of the absence of a standard 49 language; she declares that in a cultural context where there is an absence of a standard language in use, “speech becomes so fluid and ephemeral that it is misleading to talk about trans- or polylanguaging ... as this would presuppose two separate systems.” This means that it would be difficult for such a community to observe strict and stable divisions in communicative functions, as the diglossia model would dictate. Lüpke also discussed concepts such as nonce borrowing and code switching, which are typically at play where there is such fluidity in language use. She goes further and insists that “the dynamic nature of language use requires us to account for the non-static nature of multilinguals’ language use, flanked by detailed demographic and sociolinguistic information on the speakers.” (2016: 40)

Lüpke believes that “small-scale multilingualism is attested mainly in areas that have not had or have been relatively only recently exposed to western ideas of nation states and standard language ideologies” (p. 41), like Lower Fungom which is at the moment still void of basic corollaries of westernization like electricity. The fact that Lower Fungom is somewhat detached in terms of connectivity has resulted in its practicing the particular form of multilingualism, which could be called indigenous multilingualism. We saw above that because communities vary from one to another, some rural communities lack 50 stratification, and so their language practices do not fit generalized models of divisions and domains of language use as upheld by Fishman (1967).

Lupke (2016) goes on to talk about diglossia and polyglossia which, in her opinion, “is a situation in which the languages in a given society exhibit a clear division of labor going hand in hand with difference in prestige based on official literacy standards for the high varieties and the absence of (prestigious) writing cultures for the low varieties.” Lüpke cites Evans (2010: 14) who asserts that

“small-scale societies […] are economically self-sufficient, and proudly form the center of their own social universe without needing to defer unduly to more powerful outside groups.” Lüpke (2016) also cites François (2012) who holds that these small-scale societies practice “egalitarian multilingualism.” They also refer to it as “balanced multilingualism” (Aikhenvald & Dixton, 2007) or

“traditional multilingualism” (Di Carlo 2016). These are terms we will use to describe the type of multilingualism practiced in Lower Fungom. Lüpke (2016) discusses “pioneering” case studies that showcase the differentiation of power- sensitive and prestige-insensitive multilingual settings. Some communities have language practices which are not in her view primarily motivated by power relations or prestige. The examples she discusses in her work include settings in the Amazon area of South America, precisely the Upper Xingu and Vaupés basin areas. In West Africa, she inspects the renowned Casamance area of 51

Senegal, and also the area of focus of the present study: Lower Fungom. She also considers examples from the Melanesian Archipelago of Northern Vanuatu, and in the Northwest Arnhem Land in Australia. She insists that “less is known about actual communicative practices in most of them” (2016: 47). Lupke feels that three of the multilingual settings presented in this paper appear to be driven by egalitarian and (at least to some extent) reciprocal ideologies of multilingualism. The societies in question are the Lower Casamance region of

Senegal in West Africa (Cobbinah, 2010; Lüpke 2010; Lüpke & Storch 2013), the Lower Fungom area in the Grassfields area of North Western Cameroon (Di

Carlo 2016; Di Carlo & Good 2014), and the Torres and Banks Islands of

Northern Vanuatu (François 2012).

We examine and include Lüpke’s discussion of Lower Casamance, as it is very helpful to analogize this to Lower Fungom. The Lower Casamance area of

Senegal is a highly multilingual area where individuals report speaking 5 to 10, or even more languages. These individuals, being extremely multilingual, assume accommodating language practices. She states that the main motivation for a highly versatile linguistic behaviour and complex repertoire lies in the need for assuming changing identities in order to create manifold alliances with other small groups in the area. According to Lüpke (2016), in the Baïnounk language areas, languages do not express identity in essentialist fashion, as in western 52 language ideologies. “Rather, languages are used in indexical fashion and multilingualism is a social strategy that enables speakers to index different identities to different stakeholders” (p. 4). (See also Di Carlo 2016 for details of indexicality and essentialism) These multilingual speakers are often very accommodating and they see multilingualism as an integral part of their identities. Lüpke believes that these traits are shared across the region and they create a particular language attitude that supports multilingualism.

Lüpke already cited Di Carlo (2016) as a study that focuses on small scale multilingualism with unparallel language ideologies to westernized societies. Di

Carlo has another terminology for small scale multilingualism- “endogenous multilingualism”. With this terminology, Di Carlo asserts that “speakers’ language repertoires and ideologies are largely localised.” He believes that this is the situation of Lower Fungom.In his words, his study stems from a strongly ethnographic approach, and allows him to reconsider basic notions in mainstream sociolinguistics. Di Carlo aimed at finding language ecologies and markets where European languages play less of a central role than they do in cities. Like François and Lüpke, Di Carlo is against ready-made notions scholars tend to bear in mind as they carry out studies in multilingual communities. He does not agree with those who think that this should be a default lens through which most sociolinguists view multilingualism. He sees the term polyglossia as 53 deriving from “the well-known concept of diglossia in the wider sense of

Fishman (1967) it essentially refers to a situation in which, in a given multilingual speaker community, there is consensus that the main spheres of social interaction need be accessed via specific languages, or registers, or codes.” He declares that “this compartmentalised view of the relationship between polyglossia and multilingualism is exemplified by Stroud (2007: 511, emphasis added) when he writes that “multilingualism involves some form of functional ideological division of labor between languages”. He explains that

“compartmentalization usually co-occurs with the assumed existence of an underlying ideological High/Low incline.”

It is important for us to emphasize here that, like Lüpke (2016), Di Carlo is of the opinion that it may be a mistake to view the polyglossia model as being of universal applicability. He says that societies are invariably different: it is plausible that in culturally different societies, “some of the key values to be negotiated through linguistic means are either just not found in our Western tradition or pattern in unexpected ways with the embedded cultural matrix.” We consequently will not obligatorily expect Lower Fungom to observe di/polyglossia and with Lüpke (2016) and Di Carlo (2016), we now bear in mind that the absence of such notions in this area is not unheard of. Di Carlo himself does not toss the notion of di/polyglossia in the trash. He acknowledges that this 54 is a commendable theory sociolinguists must master, but encourages them not to uncritically view linguistic diversity through this lens as they carry out their research. Di Carlo (2016) also elaborates on a term that is crucial in a discussion on language choice and use- indexical order. This term incorporates two notions indexicality and essentialism. Indexicality to Di Carlo simply means "belonging to a group" without personal features inferred from the identity of the group, without further implications of the "essence" of the speakers: no "certain kind of person" is indexed. Di Carlo says that indexical order with reference to Lower

Fungom is as if I signal my being "Abar" by using Abar, but this doesn't mean anything about who I am, what kind of person I am, but just that I am affiliated with Abar." He added that, "While essentialist claims allow hearers to "deduce" what kind of person one is by the language he uses, pure indexicality prompts other kinds of deductions.

We also present here Di Carlo’s systematic methodology which validates that of this study. He used semi-structured interviews and contacted a total of 97 individuals (54 men and 43 women; only 17 respondents aged 40 or less) from six villages following an interview guide which was made up of three sections: biography/ethnography, self-reported multilingual competence, and motivations for learning each of the languages mentioned. In the first section, Di Carlo says that the aim of the study was to obtain as much detail as possible about the 55 biographical factors that could allow us to evaluate the number and nature of the social networks the respondent was part of, therefore assessing the degree of exposure to different languages. The second part was intended to produce a list of all the languages or lects the respondent claimed to be competent in, accompanied by self-evaluative remarks about the claimed competence in each language or lect. Di Carlo notably states that he did not aim at assessing their

“actual” competence in one or the other language. This is part of what the current research addresses, as it considers the actual language use of the consultants; their natural speech is of prime importance to this thesis.

What is striking and relevant for this section is the result of Di Carlo’s systematic inquiry. His findings draw our attention to the fact that in Lower

Fungom, language choice is made without any consideration of prestige. He stresses that prestige is only an issue when colonial languages are involved. His study of two of the smallest villages in Lower Fungom, Ngun and Ajumbu (each comprising about 100 inhabitants), points to the fact that “the Lower Fungom’s

‘linguistic market’ (Bourdieu, 1991: 37) is structured in a way that largely escapes the model of polyglossia scales, dominant in mainstream sociolinguistics, where each language or variety is found at a given ‘rank’ reflecting the degree of prestige attributed to its speakers" (2016: 83).

56

He parallels his study to Connell (2009) who also found the notion of prestige to be absent in the linguistic market of another rural area along the Nigeria-

Cameroon borderland, namely Mambila. Connell’s study and Di Carlo’s share a common ideological background in non-urban, more traditional social contexts of this part of the world, in opposition to urban ones. Also, the lack of prestige as a reason for language choice only suggests the absence of a stratified linguistic ecology and consequently the impracticability of di/polyglossia in the area.

When we talk of multilingualism, we think of a linguistic division of labour, where we assume that, logically, there should be a most important and standard code which is appropriate in certain circumstances and must be used as such; we commonly think of notions such as diglossia and tri/polyglossia. The literature above has shown this as being untrue with traditional multilingualism; this kind of multilingualism, from the above discussions, is not predictably divided into standard and substandard. It is predictable here that language choice indexes social identity, but the choice of which language to use is unpredictable and nuanced. It follows from considerations different from prestige: this study aims to contribute to the understanding of what these "other" considerations may be. 57

2.3.3 Code Switching/mixing

2.3.3.1. Definitional issues

Basically, the term "code" here is a relatively neutral conceptualization of a linguistic variety, be it a language or a dialect. Poplack (2015) holds that although code switching has been exceptionally well researched, controversy continues to reign over its identity, structure and the rules governing its use.

Auer’s (1999) well-articulated paper discusses a continuum of language alternation phenomena. This continuum ranges from code switching, through code mixing to fused lects, with code switching and fused lects marking the polar extremes of the continuum and code mixing the point in between. He explains the continuum in simple terms, stating that code switching is a language alternating phenomenon. Auer holds that in code switching (hereafter

CS) the contrast between one code and the other is meaningful and can be interpreted by participants as indexing situation, participant or topic (Auer

1999:310). CS in this sense has been dealt with by many scholars including

Poplack (2015) and Myers-Scotton (1993).

According to Myers-Scotton (1993: 47), code switching refers to “the use of two or more languages in the same conversation, usually within the same conversational turn, or even the same sentence of that turn.” It is worth emphasizing that she treats code switching and mixing as a continuum, with the 58 two concepts only distinguished by the degree or position (inter-sentential or intra-sentential) of the alternating varieties.

Poplack (2015) asserts that languages may in principle be combined at any level of linguistic structure. However, only three of these levels have been favoured by many studies. It could be a tag, word or clause, within and between sentences. Poplack pays even stricter attention to one of the types of CS: intra- sentential CS. In her well-grounded opinion, code switching is not just the practice of moving back and forth between languages or dialects. She explains that intra-sentential switching tends to occur at points in discourse where the juxtaposition of L1 and L2 elements does not violate the surface syntactic rule of either language. The inserted element has to remain faithful to its original language and also fit in the recipient language.

Milroy and Muysken (1995) define CS as “alternative use by bilinguals of two or more languages in the same conversation” (p. 7). They use CS as a cover term under which different forms of bilingual behaviours are subsumed. Myers-

Scotton (1993) defines it as “alternations of linguistic varieties within the same conversation. Other scholars also emphasize that switching can occur not only between languages, but also dialects of the same language. Gumperz (1982), the pioneer scholar of CS, is one of those who maintain this belief, as he refers to 59 the term as the “juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or subsystems” (p. 59).

All in all, the common trend is to see CS as an umbrella term encompassing many other related phenomena. In this study we also use the term CS as an umbrella term to cover the phenomena of alternating between two languages or varieties within the same conversation, especially within the same sentence.

However, it is not the purpose of this study to outline in a refined way the complexity of CS and its related notions. Rather, it is the purpose of this study to establish a clear definition of what we consider as CS in order to further discuss its practice in the setting of this thesis. One term to be remembered here in code switching is “alternating”, or a back and forth movement in language use. In addition, Poplack says that for “another language” to be considered code switched, there must be an asymmetrical relationship in which one language dominates and items from the other language(s) are inserted. Again, she stresses that the said languages are said to alternate.

2.3.3.2. Code switching in Africa – urban and rural

Myers-Scotton (1993) principally sought to understand what bilinguals gain when they choose to conduct a conversation in two languages instead of just one. She thinks that there are socio-psychological motivations for such an action and she presents a theoretical framework for explaining CS. Just like the present 60 thesis does, Myers-Scotton uses naturally occurring conversations as data for her study. Worth mentioning is that it was conducted in Africa, specifically in

Nairobi, Kenya and Harare, Zimbabwe. Her book is imperative for the present study as Myers-Scotton sheds much light on the African linguistic scenery. Two

African nations with English and indigenous languages sharing official language status are brought to limelight here. It is worth noting that in both countries,

English for obvious reasons has more roles in the socio-economic domain. She examines what counts as code switching and she argues that it could be intersentential and intrasentential (p. 4).

Myers-Scotton’s work tilts towards the model of compartmentalized communicative functions discussed above. She says “there are three different spheres of language use in much of urban Africa: home, neighbourhood and work.” These, she says, are “still relatively compartmentalized” (p. 40). She too argues for language choices being indexes as discussed above; to her, all linguistic varieties are indexical. She uses indexicality in the sense proposed by

Pierce (1955), who opines that indexicality is a pointer; it ‘points to something’.

The example Myers-Scotton uses is smoke which points to the fact that there is fire. She insists that “all linguistic choices stand for attitudes and expectations of participants towards one another.” Language choices to her are indexical of particular relationships. Myers-Scotton says that choosing one language over 61 another, points to the character or temperament of the speaker toward the addressee and how he or she expects to be treated in return. It is important for us to remember this assertion from Myers-Scotton: linguistic choices to her are social strategies as speakers choose languages to express their “intentional meaning” (1993: 57). To support this claim she proposes a ‘markedness theory’, whose objective is to explain the social motivation for code-switching. The theory behind this model holds that speakers have a sense of markedness as far as choosing codes is concerned. Speakers choose their codes “based on the persona and/or relation with others which they wish to have in place.” She believes that speakers know the consequences of making marked or unexpected choices. She calls the unmarked choice ‘safe’, which “conveys no surprises because it indexes an expected interpersonal relationship.” Speakers sometimes go for this choice but they also sometimes “assess the potential costs and rewards of all alternative choices before making their decisions." (1993: 75)

Myers-Scotton repeatedly suggests that code choices are indexical of the rights- and-obligations sets (RO sets) between participants. An RO set is ‘an abstract construct, derived from situational factors’ (Myers-Scotton, 1993: 85). Myers-

Scotton further suggests that speakers have an innate knowledge of this indexicality or mental representations of a matching between code choices and

RO sets. Linguistic codes come to index a particular RO set because they are 62 regularly used in a particular interaction type. Myers-Scotton explains in this paper that linguistic choices are not haphazard; for both the interlocutor (s) and the speaker, linguistic choices are pointers. Sometimes they are expected and, like Silverstein (2003) suggests, these are not socially meaningful, but at other times these linguistic choices are not expected and consequently are socially meaningful.

From Myers-Scoton's paper we learn that the linguistic choices of speakers are most often than not calculated and that they are also made in response to cultural prescriptions which Myers-Scotton termed sets of rights and obligations.

Sometimes these choices are expected and at other times they come as a shock to the interlocutors. What is also clear here is that whether it is marked or unmarked, the interlocutors have an understanding of these linguistic choices.

Similarly, Tabouret-keller (1997), upholds the idea of convergence and divergence as far as language choice is concerned, as does Franceschini (1998) who asserts that “language serves to differentiate the speaker from others and by marking the speaker’s belonging to a certain group by means of a similar use of language” (p. 63). The converging use of language, to Tabouret Keller creates confidence and cohesion whereas divergent use of language tends to be dissociative and is easily associated with either ‘personal’ or ‘foreign’.

“Speakers have the option of moving around in their repertoire… according to 63 their competence.” In line with one of the premises of this study, Franceschini declares that “a speaker’s choice is regulated by his/her aims, interlocutors such as parents, playmates, friends and superiors.” She insists that speakers have learnt how to use these codes and they are well aware of which values are transported via these codes.

Franceschini declares that “a certain degree of social and political flexibility of norms at the macro social level is necessary for code switching to come into being.” She says that “the coexistence of several languages and varieties has to be appreciated socially instead of being fundamentally excluded or stigmatized”

(p. 65). She further predicts that in social situations where speakers of different languages are hostile towards each other (for social or political reasons) code switching is less easy. Franceschini is of the opinion that “code switching displays the underlying and social flexibility of speakers in conversation” (66, also see Good & Di Carlo (2012).

Myers-Scotton (1993) seen above sets out to examine what speakers gain when they begin conversations in one language and end in another. In another paper of hers (Myers-Scotton, 1982), she sets out to examine why multilinguals with the same repertoire tend to switch languages when they converse, even when it is easy to finish the conversation in a single language. She says that it is easily justifiable why communities are multilingual, but the question is: why do these 64 communities remain multilingual? Myers-Scotton hypothesizes that

“multilingual communities remain multilingual because of the function of the different languages as tools of both positive and negative identification for the subgroups within the community; that is; different codes are maintained because they serve as social markers for different subgroups” (P. 432).

The importance of the preceding discussion and research work to this thesis cannot be over emphasized. Myers-Scotton’s (1982) data on code switching is drawn from a multilingual African speech community, in Nairobi, Kenya. In her work, Myers-Scotton provides an example of participants in a code switching situation, they are involved in a heated argument and one of them who happens to be of the dominant group calls the other participant “uncivilized”. In this example, the participants switch from a neutral code (Swahili) to their own individual first languages as their dispute continues and they use their first languages to (a) show positive symbols of their own group identities and (b) as a means of imposing a negative identification on the opponent. Myers-Scotton

(1982) adds that such examples are frequent in any multilingual community between members of different subgroups. Her hypotheses are supported by data from natural conversations among members of the Luyia speech community in

Kenya. Myers-Scotton illustrates that Lwidakho, the Luyia variety is the mother tongue of all participants present, and it is the unmarked choice, yet one speaker 65 continually switches to English. She says that the particular speaker was an educated person, who had a high-level civil service position in Nairobi. He had only returned to his home area on leave. Myers-Scotton asserts with help from the subjects of that community that his continuous switch to English does not demonstrate a lack in ‘linguistic virtuosity’ but a display of his unbalanced interpersonal relationship vis-à-vis his listeners. Her examples demonstrate a use of language to mark superiority, a switch to English or Swahili to invoke authority, a phenomenon which Di Carlo (2016) mentioned above has called which in turn represent behavioral, moral, and existential features).

Her findings support the claim of her study, that intragroup code-switching primarily encodes interpersonal dynamics rather than linguistic deficiencies.

This paper presents “evidence that the possibility of code-switching is a motivation to maintain multilingualism in a community, even for intragroup contacts” (p. 440). We understand from this that multilingual speakers in urban as well as rural areas take delight in using more than one language; they enjoy the liberty they have to do this and the benefits this act holds persuades them to nurture multilingualism.

Myers-Scotton (1976) is another indispensable literature for this thesis as it illuminates our understanding of how a multilingual maneuvers his/her linguistic repertoire. In this paper, Myers-Scotton (1976) aims at examining the languages 66 used among peers in inter-ethnic work situations in three African cities. What is clear here is that multilinguals always have to make language choices and sometimes the situation coerces them to choose a language that seems to have no particular tie with an ethnicity, a social class or any such variable. Myers-

Scotton (p. 919) says that individuals seek a “linguistic variety which avoids commitment to certain socially meaningful attributes; ethnicity, education, authority, or communality.” With this concept of neutrality in mind we immediately imagine such instances within the context of this thesis wherein the participants would prefer the language of wider communication in Cameroon

(CPE) to their indigenous languages.

Myers-Scotton believes that uncertainty regarding language choice is commonplace, not because the components of the situation are unknown, but because more than one variable may apply and it is tricky knowing which one is most important. She observes that “society provides norms, but it does not always order them. When more than one norm applies, the speaker may be uncertain which are dominant” (p. 919). Here the speaker may choose a linguistic variety which has the value “neutral”. Myers-Scotton says that choosing a neutral language is such an “avoidance strategy.” Speakers may not want to give preference to a particular attribute if its value is unfavourable.

Myers-Scotton talks about individuals from mixed ethinic backgrounds at the 67 same workplace. In such a workplace reciprocal accommodation is a priority.

She says that communication in 'nobody's language' is the order of the day in such a setting. Myers-Scotton presents data on work encounters with peers from three African capitals: in East Africa, Kampala (Uganda) and Nairobi (Kenya); in West Africa, Lagos (Nigeria). Her primary data is from Kampala and Lagos; the one from Nairobi is presented more as supportive than as primary evidence.

Accommodation is important because the speaker normally has long-term goals in the relationship. He/she wants to maximize his own rewards and minimize his costs, but also to allow his companions to maximize their rewards to a degree that the relationship is maintained on an amicable basis. If it were not for the long-term nature of many work relationships, Myers-Scotton declares, strategies of neutrality would not be required. Uncertainty arises in the work situation because it is unclear what strategies a speaker should use to establish such a relationship.

Although we are both studying the social motivations for code switching in

Africa, our methods of going about it are entirely different. Consider for example our sample sizes. While Myers-Scotton relies mainly on the administration of an oral questionnaire to 187 adults in Shomolu, Lagos, and to 223 adults, representing about 25% of the heads of households in two areas in

Kampala, our study was rather "individual-based" comprising recorded and 68 transcribed hours of individual-based interactions. In other words, Myer-

Scotton’s work was a survey while ours was an ethnographic case study.

However, her study highlights important information about why a multilingual would choose a particular language; one of such reasons is neutrality.

Conclusion to chapter two

This chapter is the first part of the literature review. It has presented background information on the geographical setting and the distribution of languages and language varieties in Lower Fungom, including earlier works on sociolinguistics in the research area. These studies mentioned the need to carry out sociolinguistic study in rural areas as much to meet up with the large number of studies available on urban sectors. It also looked at terms related to multilingualism such as di/tri/polyglossia and code switching/mixing. This first part of the literature review reveals a lack of significant existing literature on rural multilingualism in Cameroon, in comparison with urban multilingualism. It also showed a restraint towards generalized theories (di/tri/polyglossia) proposed by earlier studies, associated with multilingualism in favour of a more 69 meticulous studies that would generate local theories. Localized theories is not the most appropriate to explain the social use of language than theories that are preoccupied with universal causality principles for a localized theory, small- scale multilingualism studies are highly recommended. The next chapter is a continuation of the literature review. It discusses the theoretical framework that guided the design of the study and the interpretation of the findings, together with its methodology. 70

CHAPTER THREE

LITERATURE REVIEW ON THEORETICAL FRAMEWORK

3.1 Introduction

This chapter discusses additional literature related to the study. It reviews literature on its theoretical underpinings; the purpose of which is to describe and explain propositions advanced in the study. The chapter ends with a review of literature on the methodology of the thesis.

3.2 Language Ideology

Buda (1991) urges us to recognize that language choice events do not exist in a vacuum. He explains that language is after all a medium for interaction and communication between people and the use of language will reflect the infinite complexity of human relationships. Language choice and use embody varied and numerous abstractions. These abstractions could be termed loosely as ideologies. The aforementioned sections have in one way or another touched on how people perceive languages and how this influences their use of these languages in their communities. To better understand these perceptions and how they play out in the social and cultural systems, it is important to have clear ideas of what is meant here by language ideology.

71

Ideology is a complex construct to define. In simpler terms, it is a way of thinking that is influenced by history and social facts. Schiffman (2005) in his class handout for a course on language policy uses the term ‘linguistic culture’ for language ideology. Linguistic culture to him describes and delineates the sum of ideas, values, beliefs, attitudes, prejudices, myths, religious strictures and any other cultural information that speakers bear in mind when they use a language. Ideologies of language are important for a study in sociolinguistics such as this for the simple reason that they are part and parcel of language itself.

Williams (1977: 21) could not have put it better when he wrote in Marxism and

Literature that "a definition of language is always, implicitly or explicitly, a definition of human beings in the world."

Schieffelin, Woolard & Kroskrity (1998: 3) maintained that language ideologies are cultural representations, whether explicit or implicit, of the intersection of language and human beings in a social world. They examined definitions and conceptions of language and how these affect social and linguistic behaviour.

These scholars just like Raymond Williams mentioned above point out that language ideologies are not about language alone; they mediate between social structures and conversations linking language to identity, power, aesthetics and morality.

72

They cite Rumsey (1990: 346), who used the term linguistic ideology and language ideology interchangeably. He defines linguistic or language ideologies simply as "shared bodies of commonsense notions about the nature of language in the world." Silverstein (1979: 193) on his part defines linguistic ideologies as

"sets of beliefs about language articulated by users as a rationalization or justification of perceived language structure and use". On the other hand, with a greater emphasis on the social facet, language ideology has been defined as "the cultural system of ideas about social and linguistic relationships, together with their loading of moral and political interests" (Irvine, 1989: 255).

Schieffelin, Woolard and Kroskrity (1998) argue that language ideologies have a direct link with power struggles: social, political, or economic. In their study, ideology is seen as ideas, discourse, or signifying practices in the service of the struggle to acquire or maintain power. However, unlike Schieffelin, Woolard and Kroskrity (1998), the present study does not use the term ideology as a tool, property, or practice of dominant social groups; it is our belief that it does not pertain only to dominant groups. The study rather adopts the unconstrained definition of both Rumsey and Michael Silverstein cited above. More explicitly and according to this study, language ideology is a theory whereby people in general are defined or judged by the language they speak based on ideas that are not necessarily factual. This term could also be just any information that informs 73 someone about an individual’s behaviour or their manner of living such as hospitality. It is clear that the interaction of people from different backgrounds in a social setting invariably leads to the assignment of people to different categories. One method of classifying people has always been through the language they speak. As a matter of fact, the theory of language ideology states that when any member of a given group uses a particular language/lect, all the assumptions and prejudices harboured against the group as a whole will automatically be transferred to that person. Studies on language ideology are obviously woven into an ethnographic study on linguistic diversity such as the present study. It is vital to review them so that we are familiarized with the concept.

3.2.1. Indexical order

In the literature below, we examine characteristics of ideologies, and we explore what they convey. We begin by stressing that linguistic homogeneity hardly possesses significant social implicatures; it is heterogeneity that is significant. In diversity there is meaning; indexical order and meaning are inseparable terms.

Indexical order resides in the ways in which instances of communication point towards socially and culturally ordered norms, traditions and expectations. It is worth noting that indexicality, even though operating at an implicit level, is not unstructured but ordered. 74

To illustrate the use of the term indexicality as per this thesis, we begin by exploring linguistic heterogeneity or variation, as we noted earlier, social meaning in communication lies in variation. Eckert (2012) considers the development of variation studies and implicatures on meaning. She begins her study by stating that linguistic variation studies are all about social meaning. She talks about the renowned Martha’s Vineyard study by Labov (1963: 87) wherein he discussed the diphthong /ay/ as an indexical resource. Eckert says that the mainland (or outsider) trend was to lower this diphthong to a single vowel [a] and the island (the original inhabitants) speakers had started following this trend.

However, some speakers, in a bid to recapture and maintain the distinctiveness of the island started revising this lowering tendency. To such people, especially the fishing community, this revival of a traditional local pronunciation constituted a claim to island authenticity and a resistance to mainland invasion.

Eckert asserts that this very local construction of meaning in variation, the recruiting of a vowel as part of a local ideological struggle, suggested that variation can be a resource for the construction of meaning and an intergral part of social change. (2012: 87)

Eckert (2012) asserts that this is a quintessential example of what Silverstein

(2003) termed “indexical order”. Indexical order to Michael Silverstein is a notion by which, like in the context of Martha’s Vineyard, “a feature that had 75 simply marked a speaker as a Vineyarder came to be used stylistically within the island to index a particular kind of Vineyarder, foregrounding a particular aspect of island identity” (p. 88). This is clear evidence of that fact that speakers typically exploit the diverse and multiple ways of speaking that exist to add a layer of social meaning. This is language being used for one of the characteristics it possesses-purity or the opposite of it.

Eckert (2012) explored studies in linguistic variation and claimed that these studies can be partitioned into three phases which she called ‘waves’. These waves can be recapitulated as follows:

1. Wave one- Correlations with macro-sociological factors;

2. Wave two- Ethnographic information;

3. Wave three- Style and identity;

The first wave of variation studies began with Labov’s (1963) study of the social stratification of English in New York City. His study and many others established a regular pattern of socioeconomic stratification of linguistic form.

Eckert declared that central to this approach to variation studies is the notion of the standard. This pertains to the language variety that is native to the upper middle class, that shows the least ethnic and regional distinctiveness and is maintained by powerful societal institutions. The global prestige of the standard makes it the target of upward mobility and sets a contrast with the stigmatized 76 varieties as one moves along the socioeconomic hierarchy. Labov held that certain forms of regional and ethnicity were stigmatized on the standard language market and decreased in frequency as one moved upward through the class hierarchy. Eckert (2012: 89) considered the vernacular as a "speaker’s first acquired and most automatic, hence maximally systematic linguistic production.” Linguistic production bearing sound change originating from the lower end of the hierarchy by virtue of their local origins indicated regional ethnic distinctions, while the standard which did not emanate from a specific place indexed class positions and sophistication.

This first wave studies also included gender stratification; women’s speech was considered to be consistently more standard than men’s. Women’s patterns are typically compared with class and other macrosocial categories; this shows class position as the central indexical focus of variation. The first wave viewed variation to be based on socioeconomic hierarchy. That is, “variables were taken to mark socioeconomic status, stylistic and gender dynamics…." (p. 90). In sum, in the first wave, linguistic variation was mostly linked to the socioeconomic categories such as age, sex, social class and ethnicity.

The second wave of variationist studies was mainly based on ethnographic approaches that were used in order to determine the factors that lead to these categories. Eckert reports that a good number of early variation studies, 77 including Labov's, proposed that the vernacular had positive indexical value. He often referred to the vernacular as having a local value. The second wave began with the attribution of social agency to the use of vernacular as well as standard features and a focus on the vernacular as an expression of local or class identity.

Studies in rural communities brought local issues into the understanding of the relation between variation and occupation. Eckert cites Rickford’s (1986) work in a sugar plantation in Guyana who found a major division between those who worked the sugar (the estate class) and those who worked in the offices (the non- estate class). These groups, she explained, showed major differences in verbal culture in their use of Standard English variants as far as the use of pronouns is concerned. Rickford’s study emphasized that although the vernacular may be stigmatized on a global level, its association with local values and practices gave it positive value on the local level. There were similar ethnographic studies which Eckert discussed which sought to investigate the link between ways of speaking and the local meaning as exemplified in Labov’s (1963) Martha’s

Vineyard study. What is even more relevant here is the idea that these studies concentrated on smaller communities for a long period of time and used the ethnographic approach of data collection to obtain their research objectives.

Their aim like Eckert says was to discover, not to presuppose locally prominent social categories brought about by linguistic practices. Their studies proved that linguistic variables do not always index categories. Studies also showed that 78 adolescents are also able to lead in sound change in the vernacular. This raised the question of the role of class in adolescent variation and it pushed Eckert

(1989a, 2000) to conduct an ethnographic study of adolescents in high schools from the predominantly white Detroit suburban area. Her findings suggested that social stratification is actually realized in local ways and this implies that the broader class relations are not a direct consequence of education, income, sex and occupation but by local dynamics and practiced based on local ideologies.

The real revelation in this study was that the teens were choosing to index their chosen class identities through language. In the past, variation was seen as kind of an automatic unconscious thing, and the “jocks and burnouts” study was one of the first to show that variation arises through individuals’ choices about their style and identity.

The ethnographic studies of the second wave tried to draw the link between the macro-level categories and the more specific local categories. However, Eckert asserts that the second-wave studies did not explain the indexical relations between varieties and social categories.

The third wave moved the focus from structure to practice. Researchers became interested in how speakers are constrained by social structures in their everyday life. This third wave brought the notion of indexical order in full view. Eckert 79

(2012: 95) holds that indexical order resides in situations where linguistic forms indicate social categories indirectly. The third wave deals with the intentional use of variation by speakers to create personal and social styles that are associated with social class. Variation in linguistic features of all sort are endowed with a variety of meanings. The emphasis on stylistic practice in the third wave places speakers not as passive and stable carriers of dialect, but as stylistic agents, tailoring linguistic styles in ongoing and lifelong projects of self-construction and differentiation. In a nutshell, the third wave studies in

Eckert’s view brought the following points to the limelight:

v Variation expresses the full range of social concerns in a community;

v The use of variation does not only reflect but constructs social meaning; it

is thus the driving force of social change;

v The meaning of variables is basic and becomes more specific in the

context of discourse and the construction of styles.

Each wave added something to the preceding one. The studies involved in each wave tended to inspire the following wave type. Eckert still believes that a detailed study of current variations is the absolute and variations keep changing.

She also states that not all variations are meaningful.

Johnstone & Kiesling (2008: 5) offer further clarity on indexicality. In their hypothesis-driven study they explored the meaning assigned to a sound, they 80 considered this feature as associated with local identity in Pittsburgh. This sound feature /aw/ is a diphthong but which gets to be pronounced as the monophthong

/a/ and this indexes ‘localness’ to some people. Johnstone and Kiesling observe that those who consider this sound change as ‘localness’ are unlikely to have this feature in their own speech. They claim that most of the people who do this

‘monophthongization’ do not associate this variation with localness. It is in this context that Johnstone and Kiessling argue that the social meaning of speech features certainly varies widely within a community such that /aw/ monophthongization can be interpreted using a wide range of indexical schemas, some of which attribute it to local social meaning, some of which do not. They claim that indexical order or social meaning is indeterminate; it cannot be known in advance as it encompasses a multiplicity of human experiences. This is because indexical order is created and reinforced constantly in local practices in which different people participate in different ways; people experience the society in which they live in differently.

Having established these characteristics of indexicality, Johnstone and Kiesling move on to examine its sociolinguistic roles. They view indexicality according to the linguistic anthropologist Michael Silverstein’s (2003) work wherein he considered social meanings that occur at various abstract levels as orders of indexicality. Indexicality may be hard to pin down but Johnstone and Kiesling 81 cite Silverstein who emphasizes that an indexical correlation only becomes socially meaningful when the correlation is endowed with meaning drawn from local ideology. This means that indexicality presupposes that the context in which it is normally used evokes some form of appropriateness; it should evoke an image that is characteristically true. In other words, Johnstone & Kiesling

(2008: 8) add their voices to Silverstein to argue that indexicality is presupposing; it requires an antecedent. A feature can only be interpreted with reference to a pre-existing social or semantic space. Indexicality here is seen as a metapragmatically driven native interpretation, it has meaning based on native speakers’ ideologies.

Also in this literature, indexicality has to do with variation in speech, for example sound changes. Speakers notice and attribute meaning to variants. The meanings they attribute to an utterance are shaped by ideologies about class, purity, correctness or localness (Johnstone & Kiesling, 2008: 8). They use

Silverstein’s ‘orders of indexicality’ to discuss the types of social meaning that linguistic features can carry. They regard indexical order - the kind of correlation between a form and a socio-demographic identity as ‘indicators’, for example the monophthongization of the diphthong /aw/ which appears in the speech of people from a particular part of southwestern Pennsylvania and rarely elsewhere. 82

They hold that second order indexicality occurs where people begin to achieve certain goals using the first order indexicality for example when people desire to sound educated or cosmopolitan. Johnstone & Kiesling (2008) on their part view indexicality also as indicators or as markers.

Their sociolinguistic interviews and matched-guise experiments revealed that indexicality is fluid; for a speaker to use a particular feature does not automacally make it an ‘indicator’ or a ‘marker’. Indexical meaning in their study is proven simply to be “not predictable”. Their study suggests that it is tricky to speculate a speaker’s intentions. It is important to have clear independent interview data to support the claims we make (p. 19). Johnstone and

Kiesling (2008) call attention to the fact that some speakers may use a feature without attributing a ‘marker’ meaning to it; they may equally not use that feature but know that it marks maybe localness. Speakers may even be unaware of these features as being indicators of any phenomenon. Consequently, the researcher can only know what a feature indexes to a particular speaker at a particular moment by asking those concerned. In their compelling study, they used a matched guise task and interviews to test indexicality.

Johnstone & Kiesling (2008: 24) cite Silverstein’s (2003), who with his work sheds new light to the fact that “the existence of a simple correlation - a first- 83 order indexical relationship, in other words-between a social identity and the occurrence of a particular feature is a necessary but not sufficient precursor to higher-order indexicality.” They emphasize that not all indexes or variants have social meaning. Consequently, we need to be careful not to assume that the people we are studying draw on the same theories as we do. Their findings serve as a reminder that indexical meaning carried by a linguistic form from the point of view of a hearer may not be the same as the meaning of that feature to the speaker. Thus there is need for the researcher to specify whose perspective is being considered when studying indexical meaning.

They assert that although it is possible to make generalizations, it is difficult to speak for each and every person in a community in specifying the meaning of a particular variant.

In a nutshell, Johnestone and Kiessling dealt with variants of a language. It is important to note that mutual intelligibility is assured in this community, while, the variants only have different social values and different social meaning. They argued that indexicality deals with the meanings that variants assume; these meanings vary from speaker to speaker. Johnstone and Kiessling (2008: 25) however stress that this does not mean that finding patterns in the various meanings and how they articulate with larger social categories, practices and ideologies, or explaining why those meanings arise and how they spread is 84 impossible. This only means that researchers need to employ a range of methods in their search for social meaning of linguistic variation. They suggest that a phenomenological approach to indexicality should be promoted.

Phenomenological approaches require two kinds of work: case studies of individual’s sociolinguistic worlds as they experience them and historical research about the sociolinguistic landscapes of the past.

3.2.2 Giles’s Accommodation Theory

We also make use of Giles’ (1982) accommodation theory. The British language psychologist Howard Giles examined why speakers change the manner in which they speak. The two concepts he advanced to explain this pheomenon are convergence and divergence. These two linguistic strategies are typically applied in communication. According to Giles’accommodation theory, an individual tends to shift his/her speech styles to resemble those they would like to associate with. Giles calls this convergence while he uses the term divergence for a process in which a speaker makes his/her speech sound more unlike that of his/her interlocutor. Divergence occurs when speakers wish to maintain social distance from others while convergence demonstrates a conscious or unconscious desire for social intergration, seeking or showing approval, identification or communication effectiveness with another. Giles’ theory gives strength to the thoughts expounded in this research; it goes hand in hand with 85 the notion of indexical order. This theory proves that one of the assertions made by this thesis that the choice of which language to use is (inter alia) guided by the intention of the individual. Giles affirms that people are motivated to adjust their speech styles or to accommodate, as a means of expressing values, attitudes and intentions towards others. A person will choose a variety that best suits his/her needs or those of his/her listener. Giles’ accommodation theory has been quoted in several studies on language attitudes. Fasold (1995) uses this theory and this motivates us to choose it as a baseline for the study of rural multilingualism.

3.2.3 The ‘markedness’ theory of code-switching

This theory and Giles' accommodation theory discussed above share one underlying idea: that the choice of a particular code is hardly ever random. First and foremost, the objective of the ‘markedness’ model, as its proponent Carol

Myers-Scotton points out is to explain the social motivation of code-switching.

In the ‘markedness’ theory of code-switching, Myers-Scotton repeatedly and strongly suggests that code choices are indexical of the rights-and-obligations sets (RO + sets) between participants in a given interaction type. An RO set is

‘an abstract construct, derived from situational factors’ (Myers-Scotton 1993:

85). Myers-Scotton argues that conversations are to a large extent conventionalized in all communities. Individuals follow particular regulations in 86 conversations that are typical and familiar to a community and which also have been around for a long time. All of this forms a set of norms for appropriate social behaviour, including linguistic behaviour. Sometimes, these regulations are considered “unmarked RO sets” and at other times they are “marked RO sets”. Myers-Scotton goes further to say that these rights and obligations are indexical. Indexicality in her opinion therefore is "a property of linguistic varieties", which "derives from the fact that the different linguistic varieties in a community’s repertoire are linked with particular types of relationships, because they are regularly used in conversations involving such types" (Myers-Scotton

1993: 85). Myers-Scotton further suggests that speakers have an innate knowledge of this indexicality or mental representations of a matching between code choices and RO sets. In other words, speakers know that a certain linguistic choice will be the normal, unmarked realisation of an expected RO set for a particular conventionalized exchange, while other possible choices are more or less marked because they are indexical of something other than the expected RO set (p. 158). Linguistic codes come to index a particular RO set because they are regularly used in a particular interaction type. Myers-Scotton (1993) asserts that an individual carries his social position around his head and puts it into action when the appropriate occasion arises. Not only does the speaker carry this around, others carry it as well, because social positions are matters of reciprocal expectations and must be publicly and commonly conceived. We agree with 87

Myers-Scotton as we conclude this part while looking forward to using this theory as we analyse our data. For this thesis, we believe that habitual language choice in multilingual speech communities or speech networks is far from being a random matter of momentary inclination.

3.3. Review of Literature that informs the methodology

3.3.1 Grounded Theory Model

This thesis is guided by the Grounded Theory Model for data analysis. The

Grounded Theory Model is the backbone of this study as it completely accommodates every detail as far as the data collection, analysis and interpretation of the data for this study is concerned. For this thesis, we work with Gorra's (2007) thoughts, together with those of Glaser & Straus (1967,

1995) on grounded theory. Glaser and Strauss suggest that the researcher starts collecting data with a "blank mind", that is, without reviewing the existing literature in order to carry out a truly inductive study. This perspective assumes that the researcher would see and understand the data from an authentic point of view. Gorra (2007) urges researchers to take a passive stance and "let the data emerge". This is the position we assumed for this thesis. Consider these words filled with truth from Strauss and Corbin below.

If someone wanted to know whether one drug is more effective than another, then a double blind clinical trial would be more appropriate than grounded theory study. However, if someone 88

wanted to know what it was like to be a participant in a drug study [...], then he or she might sensibly engage in a grounded theory project or some other type of qualitative study." -Strauss and Corbin, 1998: 40 The quote above from Strauss & Corbin encapsulates the essence of when it is best to use grounded theory methodology (henceforth GTM) for a research project. Gorra (2007) asserts that GTM provides useful tools to learn about individuals’ perceptions and feelings regarding a particular subject area. She says that quantitative data may be useful in measuring attitudes across a large sample but GTM offers a powerful methodological framework if the aim of the study is to learn about individuals’ perceptions. GTM shares the following characteristics with other qualitative methods, which correspond to those of this study:

• Focus on everyday life experiences

• Valuing participants' perspectives

• Enquiry as interactive process between researcher and respondents

• Primarily descriptive and relying on people's words

(Gorra, 2007)

All these characteristics describe this thesis and its aspirations completely.

Andrea Gorra makes it even more suitable as she outrightly made her PhD thesis

An analysis of the relationship between individuals’ perceptions of privacy and mobile phone location data, grounded theory based. We choose grounded theory 89 methodology because like Gorra says, it advocates formulating new theories rather than testing existing theories. Dornyei (2011) holds that grounded theory has been used too often in qualitative studies that both have become synonymous. Dornyei claims that grounded theory could be used throughout the research process, from sampling to data collection and analysis.

3.3.1.1 Using the GTM for data analysis and interpretation

3.3.1.1.1 Coding

Coding is a technique employed for data analysis under GTM. Gorra holds that under GTM, there are two kinds of coding strategies: open coding and axial coding.

Gorra explains that the distinctions made between open coding and axial coding are artificial and for explanatory purposes only. Open coding is breaking data apart and delineating concepts to stand for blocks of raw data axial coding is the act of relating concepts/categories to each other. He also adds that open coding and axial coding can be used concurrently. For this study both kinds of coding were indispensable. It was crucial for the data to be broken up and labeled as was deemed necessary. In addition, data obtained from all consultants needed to be examined in relation to one another and similarities and differences 90 discussed. In a nutshell, coding was indeed pertinent as far as data analysis for this study was concerned.

3.3.1.1.2 Constant comparison

GTM uses purposive sampling. Here participants are selected according to criteria specified by the researcher and based on initial findings. Data collection and analysis for this work took place in an iterative manner, consisting of collection of data and comparison between codes which led to further data collection. For this reason, Gorra says that the development and identification of variables does not take place prior to data collection, but instead as part of the data collection process. Consequently, the variables or concepts are initiated by the consultants and further developed and conceptualized by the researcher.

3.3.1.1.3 Theoretical saturation

Data is collected until theoretical saturation is reached, in other words until no new or relevant data emerges regarding a category and relationships between categories are established (Strauss & Corbin, 1998). Labels or codes are assigned to the concepts emerging from the data. Data collection and analysis for this study follow a cyclical process typical for GTM. We used early findings to shape the on-going data collection. More sociolinguistic questionnaires were administered after we listened to the recordings and after much observation by the researcher. This was to address issues brought up by preliminary analysis 91 such as unusual language choice by individuals who interacted with one of our main consultants.

We end this section on the discussion of Grounded Theory Model as one of the theories of this study with this incredibly accurate quotation from Charmaz

(2002: 675):

“Grounded theory consists of guidelines that help researchers to study social and social psychological processes, direct data collection, manage data analysis, and develop an abstract theoretical framework that explains the studies’ process.”

3.4 Research Design

As indicated by Trochim (2006) research design can be thought of as the structure of research; it is the "glue" that holds all of the elements in a research project together. We often describe a design using a concise notation that enables us to summarize a complex design structure efficiently. A researcher could therefore sum up his/her research under qualitative or quantitative design.

Basically, quantitative research is all about quantifying relationships between variables, while qualitative research seeks to understand the relationship between variables, typically about human behaviours. This research is thus a qualitative one as it seeks to understand human behaviours in relationship to language. This design operates throughout the process of this research, from data collection to analysis and then to the development of a theory. Silverman 92

(2001) holds that the choice between the qualitative and quantitative research design should depend on what the researcher is trying to find. He asserts that if one aims at counting, grading, measuring and offering statistical validation, qualitative methods will fail miserably, but if one aims at understanding feelings, values and perception that influence human behaviour in a social context, then qualitative methods have been proven to do extremely well. Hence, our choice of a qualitative research mainly depended on our aims and objectives.

To buttress the point, Mason (2002: 1) maintains that through qualitative research it is possible to explore a wide array of dimensions of the social world, including a weave of everyday life. This can be done by using methodology that celebrates richness and depth, nuance, context and complexities. There is a general consensus that unstructured interviews and observation are instruments commonly used in qualitative research to obtain intricate details about atttitudes, beliefs and values. The choice of the design for this study is further strengthened by Paltridge and Starfield (2007). They state that qualitative research seeks to understand behaviour and the meaning of that behaviour in its specific social context. Because this study aimed at explaining social phenomena (such as attitudes, language choices, and their usage patterns) and their interconnections, qualitative methodology is an appropriate choice. Language behaviours and their 93 meanings are better explained and understood when a great deal of descriptive detail is provided in the course of the study.

Qualitative researchers are known to probe. Lincoln & Guba (1985) maintain that qualitative methods are used because they are more adaptable to dealing with multiple realities. They are more appropriate in studies seeking to uncover phenomena that yield multiple meanings, for instance code switching. The use of qualitative methods can uncover why an interlocutor utters a certain speech act, and what possible meanings lie behind it. Moreover, a linguistic research like this one, with a natural research setting and natural occurring linguistic practices, requires a qualitative design.

This study could only be qualitative, given its goals and objectives; a study of the linguistic habits of rural individuals like the title implies can only be qualitative. Blommaert & Dong (2010: 93) on their part maintain that “people are not cultural or linguistic catalogues and most of what we see as their cultural and social behaviour is performed without reflecting on it and without an active awareness….” Consequently, it is not a thing they have an opinion about, nor an issue that can be comfortably put in words when asked about it which is typical with quantitative studies.

94

In addition, Gumperz (1982: 61) emphasizes that “Participants immersed in the interaction itself are often quite unaware which code is used at one time. Their main concern is with the communicative effects of what they are saying.” He adds that “selection among linguistic alternants is automatic, not readily subject to conscious recall” (1982: 61). In order to understand these overt language behaviours, we need qualitative methods such as observation. Dornyei (2011:

28) supports this assertion by observing that “by using qualitative methods we can uncover subtle meanings that are inevitably lost in quantitative research.”

Dornyei (2011) maintains that qualitative research, given its exploratory nature, is preferable when research aims at studying uncharted areas. In the previous chapter, we discussed Di Carlo (2016) who emphasized that theory generating studies on rural multilingualism are truly still rare. Dornyei cites Eisenhardt

(1989) who affirms that “if very little is known about a phenomenon, the detailed study of a few cases is particularly appropriate.”

3.5 Ethnographic approach

This study also makes use of ethnography whose main goal in Dornyei’s (2011:

130) opinion is to provide a thick description of the target culture. When we considered the goals of this study, we realized that they tie with those of a typical ethnographic research. Dornyei (2011: 130) emphasizes that the main goal of most ethnographic research is to provide richly and in great detail the 95 daily life of the community, and ideologies the participants attach to their activities and behaviours. As a typical ethnographic study we sought to understand the linguistic behaviours and the language ideologies guiding these behaviours through the eyes of our consultants who are considered ‘insiders’.

We prioritized the subjective interpretation of their linguistic choices.

In addition, Dornyei advocates for the prolonged engagement of the researcher with the natural setting of the study. He explains that some nuances would not be uncovered if the researcher does not spend an extended period of time living with and observing the participants.

Moreover, as we mentioned earlier, this study also exploited the case study method for data collection. A case study according to Dornyei (2011: 151) as the term suggests is the study of the 'particularity and complexity of a single case'.

He adds that cases are primarily people, but researchers can also explore in depth a programme, an institution, an organization, or a community. Dornyei believes that a case study is a method of collecting and organizing data to maximize our understanding of the character of the social being or object studied. Dornyei (2011) distinguishes three types of case study: (1) the 'intrinsic case study'- that which is undertaken to understand the intriguing nature of a particular case. That is, the case is of interest not because it illustrates something or represents other cases but because of its own value or speciality; (2) the 96

'instrumental case study' is intended to shed light on a wider issue while the actual case is of secondary interest. In order words, it facilitates our understanding of something else; and (3) the 'multiple or collective case study' where several cases are studied jointly to investigate a phenomenon or general condition. The 'multiple case study' is apt for this inquiry; it necessitates not one but also a limited number of subjects. Dornyei cites Duff (2006), who admits that most of her students conduct case studies with 4-6 focal participants in one or more sites. Dornyei (2011: 153) holds that in intrinsic case studies the case represents some unique or 'not-yet-understood' feature which might aid in the general understanding of a wider domain. Case studies tend to require ethnograhic approaches to data collection.

The instruments and procedure of data collection of this study were also ethnographically based and justified by Dornyei. We sought to explore and describe in great detail the particulars of a small group and individual members of this group, this study just like many other ethnographic studies non- participant observation, unstructured interviewing and field notes by the researcher. This data was further supplemented by video and audio recordings as tools (see detail in chapter four).

In summary, Dornyei (2011) sheds light on ethnographic studies; he maintains that if the goal of a study is to provide vivid, rich and particularized information 97 on a given phenomenon, then the ethnographic method is the absolute best shot.

He declares that ethnographic studies leave nothing unaccounted for and argues that ethnographic studies ensure comparison and contrast between what people say and what people do in a given context in order to arrive at a fuller representation of what is going on. It is by comparing and contrasting dimensions, that a realistic and multilayered description can begin to emerge.

In the same light and even more related to this study, Di Carlo, in his seminal

Multilingualism, affiliation and spiritual insecurity, from phenomena to processes in language documentation (2016) urges anyone aspiring to do documentary linguistics nowadays to bear in mind that their study must be

“tailored to suit the specific environment both social and linguistics in which they are carried out”. This reinforces the point already made above by Lüpke

(2016) that every society is unique. Lüpke and Di Carlo (in press) think that because multilingual societies are diverse only detailed ethnographic studies would greatly inform research on language contact and multilingualism.

Mobility patterns that create the interwoven fabrics of multilingual societies, marriage exchange networks, mobility of children, and ritual multilingual communication are all worth considering in studying multilingualism. They believe that “these patterns have an impact on how societies and individuals 98 conceptualize themselves and how they construe others, or in other words, what their ideologies and attitudes to languages and identity are.”

In his paper, Di Carlo (in press) showcases the best methodology needed to extract well-founded, long-lasting and dependable socio-cultural and linguistic information from a given society. Such information can most certainly be obtained through an ethnographic approach to data collection. Di Carlo (in press), which focuses on documentary linguistics, upholds the ethnographic method of data collection as he believes it to have the best practices in language documentation. To him, “ethnography provides us with the most reliable tools to get as close as possible to documenting socio-cultural (including linguistic) facts” (p. 73). He asserts that to explore a given ‘communicative ecology’ the researcher should strive to have “a clear view not only of the different languages or varieties present in the repertoire of the targeted speaker community, but also of the registers and the genres through which discourse is articulated, as well as of the language ideology.”

At this level, it is already obvious that Lüpke (2016) is one of the scholars who espouse the ethnographic approach in data collection and here she insists on collecting natural data. She justifies this choice as she explains that there are

“mismatches between language ideologies and actual communicative practices; parts of repertoires can be erased or downplayed on ideological grounds." (2016: 99

52) In order to understand communities and their nuanced multilingual practices

Lüpke advocates for a detailed sociolinguistic and ethnographic investigation.

Such investigations are needed in order to investigate the complex language ideologies at work in different interactions, including those with an outsider linguist. She joins Himmelmann (1998), Good (2013), Di Carlo (in press),

Lüpke & Storch (2013) who are of the opinion that a study based on actual language use offers a better understanding of the phenomenon and “limits the danger of relying on a priori judgments.” Lüpke (2016) concludes her paper by hitting hard on researches based on what she terms predictable models of language contact; these areas in her opinion host complex language ecologies that only an empirical, co-ordinated ethnographic study can uncover the nuances therein.

3.6 Instruments of data collection

We were guided by intruments and techniques used for similar studies in applied linguistics. Groom & Littlemore (2011) were very inspiring; it too had various consultants wear their recorders every day at work for two weeks. The researchers were able to collect approximately twenty hours of language spoken by staff in the nursery, which they say was then transcribed to make a small corpus. Groom & Littlemore (2011) admit that the researcher is always faced with the decision of whether to choose between using audio or using video 100 recordings in research; both have advantages and disadvantages but they are complementary. Groom & Littlemore (2011: 72) assert that “Much qualitative research takes place through observation.” They add that “At its simplest, observation can be defined as simply watching and recording how participants behave and interact in a certain situations” (p. 72). We chose to do both non participant and participant observation for this study, alternating from time to time. Groom and Littlemore say that to produce qualitative data, it is possible for the researcher to “quietly observe what is going on in a particular situation without being involved in the situation herself or himself” (p. 72). However, they also provide an explanation for why participant observation is even better suited for this study. They state that “closely associated with … ethnographic research, this approach has its origins in anthropology and was originally used to investigate the behaviour and characteristics of particular cultures.” (p. 76) They specify that this method involves the “researcher immersing him or herself totally in the culture under examination, living in the community and using this experience to provide a very richly detailed data about how the members of a culture behave on a day-to-day basis” (p. 76).

Conclusion to chapter three 101

In this chapter, we have demonstrated an understanding of theories and concepts that are relevant to this research. It has presented an overview of relevant theories and concepts, which would in the next chapters, give a basis for our hypotheses. With this chapter, we have created the research space which this study sought to fill. Our setting (rural), our methodology (small scale ethnography) and the time (most recent) of our fieldwork, as shown by the literature reviewed, prove very important and necessary. The relevant theoretical insights have also been examined in detail and literature which underpins the adopted methodology has also been presented with adequate justification why it was chosen. The next chapter presents the methodology of the study. 102

CHAPTER FOUR

METHODOLOGY

4.1 Introduction

This chapter provides further details of the methodology and fieldwork undertaken to collect data to analyse the research questions. As a reminder, the research problem is that many studies on multilingualism have been focused on urban areas and on colonial languages leaving very little to be known about rural multilingualism where mostly local languages are in active use daily. The chapter provides a detailed account of the research method employed, sampling techniques, tools and instruments of data collection, procedures, and data analysis techniques. It also looks at issues of ethics, reliability, and replicability before ending with a summary of its key points and a brief preview of the following chapter.

4.2 Research Design and approach

A research design is the logical structure of an inquiry. It is sometimes considered as the choice between qualitative and quantitative research methods or mixed methods. The choice between these concepts depends on the problem, aims and objectives of the study, the method of data collection applied, the nature of the collected data, and the method used to process the data and to 103 obtain results. The research design functions to ensure that the data obtained enables us to answer the question (s) set forth for the study. For this study, a qualitative paradigm was used, making it a flexible and an emergent (emic) study. We chose the qualitative design as it is best for the ontology and epistemology of the study. Relativism and emic assumptions are best for researching on topics which are largely unknown. The qualitative design is best for the kind of questions which are the concern of this study- 'why, with what impact' and 'how' questions of human attitude and actions. The choice of the qualitative method enabled the researcher to aim for a complete, detailed description of language use patterns including their context and circumstances.

Through qualitative research we explored a wide array of dimensions of the sociolinguistic life of the participants including the diversity in their behaviours and the way they were woven to fit together. The qualitative paradigm gave us the latitude to dig deep into the lives of the subjects of this study and to provide in-depth answers for the research questions. As we mentioned in chapter two and according to Mason (2002: 1) qualitative research "celebrates richness, depth, nuance, context, multi-dimensionality and complexity rather than being embarrased or incovenienced by them," and because it is not in the quest for generalization, qualitative research "has an unrivalled capacity to constitute compelling arguments about how things work in particular contexts."

104

It is also worth mentioning that in adopting the qualitative paradigm, we by implication assumed an inductive approach. This approach generally associates with qualitative research and comprises a good blend for exploring this grey area of sociolinguistics-rural multilingalism. In using the inductive approach, we set out with no hypothesis. Rather, our point of departure was the research questions, aims and objectives. Once in the field, we looked for patterns, resemblances and regularities in experiences in a cyclical manner and conclusions reached (hypothesis generated).

4.3 Method of data collection

A research method is a systematic plan for conducting research. Qualitative methods explore the perspective and meaning of experiences; they seek insights and identify the social structures or processes that explain people's behaviours.

Most importantly, qualitative research relies on extensive interaction with the people being studied and often allows researchers to uncover unexpected or unaticipated information. We used case study for data collection for this study.

We chose the case study method because we were convinced that it would allow for exploration and understanding of the complexity and the subtle variations contained in language use. We considered it as a robust (not likely to fail) research method particularly fit for a holistic and in-depth investigation. The literature elaborately reviewed in chapter two motivated this choice. We recall here Yin’s (1984: 23) definition of the case study research method as an 105 empirical inquiry that investigates a contemporary phenomenon within its real- life context. Furthermore, we chose one of the three case study types laid down by Stake (Dornyei, 2011), the "multiple or collective case study" type, where several cases are studied jointly to investigate a phenomenon or general condition. As a result, we worked with a limited number of consultants.

Through this method, we were able to dig deep in a bid to understand the linguistic behaviour and experiences of the consultants through their perspectives. We were able to closely examine the data within specific contexts.

The limited number of individuals as consultants to be examined is what gave room for idiosyncrasies. Also, as far as this limited number of individuals together with the concept of generalizability is cocerned, people typically wonder how knowledge coming from a small group could be of wider relevance.

In Section 4.10.2 (below), we see that qualitative research is not overly concerned with generalizability but rather with idiosyncrasies. The goal is generally the peculiarity, the individualizing characteristics brought about by the sample.

4.4 Instruments and tools of data collection

Like other case studies, we combined a variety of data collection instruments and tools. They were:

v A sociolinguistic interview guide; 106

v Audio and video recorders;

v Observation protocol

4.4.1 Sociolinguistic interview guide

The interview guide used for this study was made up of loosely written questions which directed the conversation towards the topics and issues of interest to this research. We used the guide with both major and minor consultants. These guides (see the appendix) were composed of three interrelated parts. The first part was made up of questions designed to elicit, in great depth, participants’ biodata that could relate to their reported rates of multilingual competence. We included questions about (1) the various names the participant had (see Di Carlo 2016 for a discussion on the naming system in

Lower Fungom), (2) the provenance of the participants’ grandparents, parents and spouses. This information was relevant to the understanding of the multilingual nature and language choices of the consultants.

The second part of the guide sought to produce a list of all the languages/varieties (lects) in which the consultants claimed competence. It also comprised a part which allowed the consultants to grade themselves about their competence in each of the languages/varieties they had professed competence.

We also asked them their acquistion patterns and then, the third and final part aimed at gaining insights into the ideas or practices that the respondents 107 associated with each of the languages/varieties they professed to be able to understand or speak. We asked the respondents questions such as “When do you use language A”? “Why did you not use language B”? Can you perform this activity using language C”? Which language do you use when you meet X and why? This part permitted us to explore the language practices and language ideologies of the participants.

In addition, the interview guide gave us the opportunity to probe further into the linguistic choices of consultants which we had already observed, and which needed clarification, especially as far as the minor consultants were concerned.

Deeper questions varied from participant to participant and depended on their prior linguistic behaviours with one of the major consultants. Sometimes, based on our immersive observations and previous recordings, the participants were presented with previously observed scenarios during which the participant exhibited a particular behaviour we wanted to probe. This usually refreshed the memories of the participants or encouraged them to be more forthcoming. The interview guide enabled us to engage in a free-flowing conversational exchange.

These one-on-one conversational exchanges lasted no less than thirty minutes and overall, they were apt in shedding light on the communicative behaviours and beliefs of both the major and minor consultants.

108

It is important to stress here that interviewing for this study was indeed very iterative and this guide could not serve us always. Additional questions came from the behaviour of the consultants or from certain unanticipated situations that presented themselves. As such, the guide varied from one consultant to another.

4.4.2 Video and audio recordings

Following Di Carlo (2016) who recommends that anyone who legitimately aims at exploring multilingualism, especially multilingual competence, should base their study on scientific evidence rather than on respondents’ self-reported information solely, we made use of video and audio recordings. We used a camera (Samsung HMX-W300) for video recordings and voice recording, H1

Handy Recorder and a MTCM110-Tiepen microphone for the audio recordings.

The interviews discussed above were recorded using the voice recording audio device mentioned above. These recordings assisted us in accurately documenting the particiapnts' responses as it was not possible to write by hand everything the consultants said. The video recordings were chosen for the reason given by Loizos (2008), that video recording is necessary "whenever any set of human actions is complex and difficult to be comprehensively described by one observer as it unfolds" (p. 149). We understood that the recording of moving images would allow capturing aspects that would otherwise go unnoticed if only 109 observation or sound recordings were used. Futhermore, we also chose video recording as it favours the reliability of the judgments we made; there was the possiblility of revisiting the recorded material several times, skipping parts, pausing, freezing images, rewinding and fast forwarding. We viewed these recordings as many times as was deemed necessary for proper understanding and the interpretation of the data. What's more, video recordings allow other researchers to provide their own interpretations of the data. We collected data by video recording the naturally occurring conversations of three individuals.

Working with these individuals was both time-consuming and rewarding.

However, many of the recordings were audio, since consultants needed to be static for effective video recording. When a consultant was mobile, walking around the village or walking to another village in their regular daily activities, it was hard to chase them around, so overtly, with a video recorder. By contrast, when they were in the same location for a while, video recording was possible and smooth. So, we recorded when consultants were in a restricted mobile state, such as within a room at the palace, at a workplace, or at a drinking spot.

Another difficulty came with the possibility that the equipment could fail to function properly during a recording session, especially as the consultants had to manipulate the equipment from time to time by themselves. This was quickly solved because we made several recordings so that we could choose those which 110 were perfect. We always checked that there was enough space on the disc and that the batteries were fully functioning. It is true that because of the natural nature of the setting we could not control things like background noises.

However, the noises did not impede understanding of the data. These recordings, both audio and video, were transcribed into written form on ELAN files, studied in detail, linked with analytical notes or coded. The recordings were annotated and labelled as files for archiving. This material was then exported into word documents and inserted into the thesis in the findings section. Throughout this thesis we will always reference these files which can be accessed from the various archive where they were deposited should the need arise.

4.4.3 Observation Protocols

Narrative field notes came from the observation guides we used for observation sessions. For this study, we used the non participant observation technique; daily observations resulting to field notes also provided data for this study.

Observation provided us with the opportunity to learn more about the consultants, their behaviours and events first hand. During our three trips to

Lower Fungom, we almost always tagged along with the consultants, paying strict attention to their linguistic habits. Our observation could be as simple as paying a visit to one of the consultants in their homes or workplace, just discussing every-day issues. This made the observation unstructured. We did not 111 have a specific time frame during which we would observe but we made stretches of at least three hours daily, depending on the day and the activity in which the consultants would be involved. While we were observing some days, we went along with our cameras to video record, should a striking linguistic event happen. It was on such an occasion that we recorded a very important video with one of our consultants, Janet, using several languages in one setting and the same topic but with three different participants. Sometimes we tagged along with the consultants even when they went to their farms. Our constant presence made us participant observers, we were seen around everyday and so our presence did not always signify work. With observation came field notes.

We used a note book and a pen to take down notes. These notes contained:

1) The date of observation;

2) A list of participants;

3) A description of the setting and event;

4) Detailed descriptions of the actions of those being observed;

5) Who to meet to reflect upon the actions and event observed.

Inter alia, the above five items constituted the guiding variables on our observation protocol. 112

4.5 Procedure of data collection

4.5.1 Audio and video recording

We chose three individuals as aforementioned, to whom we gave audio recorders and microphones in turns. They wore the microphones on their shirts

(since these were Lavalier microphones) and attached the recorder to their belts.

We taught them how to switch the audio recorders on and off so that they could continue recording even in our absence. For example, some times, the consultants turned off the recorder, removed it from their bodies and had a bath.

They turned it back on and wore it again once they had finished with the exercise. This they did by themselves. After adjusting the necessary parameters

(on the recorder) like time, date and volume (crucial for metadata), the consultants then wore the microphone and went about their activities. For example, when we collected data in Missong, Boniface (our consultant, see section 5.2.3), wore the recorder and microphone when he went either to fetch palm wine (for instance) from the forest or to attend his Njangi meetings. While he was away, we visited homes in Missong to discuss research related issues until it was time to collect the recorder from him. We normally would have an agreement with our major consultants to come back with the recorder after a maximum period of three hours, thirty minutes. After collecting the tape recorder, we would make an appointment for another day with the consultants 113 during which we would listen to the data. While we discussed the recordings, more questions invariably came up which often led us to some other (minor) consultants.

During our discussion with these minor consultants, they would sometimes get apprehensive. Sometimes, they would hesitate to answer some questions truthfully, even when it was not on a sensitive topic per se. Once we realized this, we explicitly told tell them that we had either heard them on tape or physically behave in a particular manner. After which they would typically giggle and modify their responses.

We did a one-off event kind of interview with our consultants. The way the interview guides were designed enabled us to collect as much information as the study required. We discussed with the consultants face-to-face, took down notes ourselves while also audio recording the conversation. We recorded these organized conversations because we could not take down everything that was said, and we did not want to miss out on relevant details for the study.

Sometimes this event happened at the home of the participant, where we would draw the particpant to one corner. If their homes were noisy because of (for example) the presence of children, we would move to another quiet location with less distractions. These sessions lasted for about thirty to forty-five minutes each. 114

Audio recording extended the research access by providing consultants with recording devices to capture aspects of their lives or practices that we could not be present for, because of cultural or social reasons. As a matter of fact, we could not always follow them around throughout the day, for ethical reasons and also to limit the observer’s paradox. They wore the devices in our absence as they carried out their activities, freely. This provided for us broad comprehensive and natural data.

When we had to do video recordings, we used a camera and we were visibly present in front of the participants. Generally, the video clips provided a fine- grained record of the events we recorded, detailing gaze, expression, participants present, proximity and gestures. They recorded conversations in context.

Through videos we did a rigorous and systematic analysis considering facial expressions, gazes, gestures and even hesitations. Hesitations, for example, could indicate a lack of mastery of a professed known language which in turn, would throw more light on the linguistic habits of a particular consultant. As an ethnographic study, it was only axiomatic that we used a method that would yield in-depth information. In addition, we considered video data to be a durable, malleable, shareable record that could be repeatedly viewed and manipulated to be viewed in slow or fast motion. This made every detail very clear to the researcher. We listened and relistened to some portions of the 115 recordings in transcribing the recorded marterial. Sometimes we even told the consultant with whom we were transcribing that what they were telling us was different from what was in the recording. Whenever, the consultants said they did not hear what was said by one of the participants in the recording, we rewound the recording as much as was needed. We listened to the recordings with consultants, who told us what was said in the recording. This was usually in one of the languages of lower Fungom and so, the consultant also had to translate the material into CPE which we then translated it into Standard

English. We also noted the participants present, the topic of discussion and the setting.

Video data is limited and shaped by decisions in the field, such as the camera position and the stability or mobility of the consultant. It also presents issues of anonymity and ethics. Thankfully, there was little to no need for anonymity on the part of the consultant as far as the research process was concerned.

Linguistic habits are hardly ever considered sensitive in this community; on the contrary, in intensely multilingual communities like Lower Fungom, the need to clarify issues surrounding socio-cultural life and languages are generally open for discussion. As a matter of fact, all the consultants and most participants were always ready and happy to be recorded, whether audio or video. However, we still always sought their consent first and if their body language or expressions 116 showed hesitation, we refrained from recording. Sometimes, we were even called to record events which were unrelated to this study, especially because the participants were not entirely clear on what specifically we were studying, but were more excited to have their culture documented. For example, we were called by one of the minor consultants to record a traditional dance that was taking place in one of the villages at that time. We went to the event but did not record as it was not relevant to the study. This act of hers (the consultant) illustrates the enthusiasm expressed by the consultants towards being recorded.

The question of capturing a naturally occurring event in video usually arises as to whether the event is being provoked or stimulated. Also, issues with the observer's paradox are common in ethnographic fieldwork. For this study, because we were living in the community, we always had our recording devices handy. We recorded events only when we found them, hence most of our videos do not actually start at the beginning of the event. Our constant presence there made it possible to find out how people talked when they were not being overtly systematically studied. As for the observer’s paradox, the community had become used to us being present and the camera or audio recorder was no longer news to them. This helped in minimizing the chances of the observer’s paradox occurring.

117

4.5.2 Non-participant observation

We used this method of data collection based on our objective to gain a deep knowledge of the intricacies that could not be directly obtained with the use of an interview. Generally, non-participant observation yields insights into people’s lives and customs that they would not be able to explain if they were simply asked. We lived together in Lower Fungom, built relationships with its people, and even participated in some of their activities. We took notes to keep track of our observations when it was appropriate and necessary; we typically wrote after retiring home at the end of a session. Sometimes the writing took place in the presence of the consultant who was being observed or when we were by ourselves. Sometimes, we went an extra mile by letting our consultants deliberately go by without recorders; we preferred that the encounter be more natural, for people not to see any recording device and to act relatively normal.

During such instances, we would examine what is happening live and always indistinctly ask questions concerning the languages used if we found them unusual. This method of data collection was friendlier and more informal but worthwhile. It was a different way of being immersed into a culture and attracting more natural data. It is true we never know for certain what we would be observing; we just daily tagged along with one of the consultants while they carried out their activities. One day for example, we asked Kulo (see section

5.2.4) if we could accompany him to his farm. During the journey to the farm, 118 we made a lot of significant realizations. This first-hand information was used for comparison with those recorded using visible recording devices. Through observation we gathered the difference between what people do and what they say they do.

We considered tagging along with the consultants for observation less intrusive.

We do bear in mind the observer's paradox but we also believed that we had been in Lower Fungom for such a long time (three trips of more than twenty- five days each) such that they had grown used to us sitting close by or walking around with the participants. If the recorder or the questionnaire had the respondents apprehensive, making them unable to provide accurate information, observation filled that gap. This reduced the observer's paradox in this study.

Aside making it possible to collect different types of data, and collecting first hand information, unobtrusively, participant observation made it possible for our data to include greater detail. Living in the community over a period familiarized us to the community, thereby facilitating involvement in almost every activity in which we chose to partake. Also, we became very familiar with the linguistic ideologies of the region. The participants developed confidence in us which allowed them to provide a great deal of information.

119

As far as the procedure for data collection was concerned, emergence and naturalness were the central factors. The interviews, recordings and observations resulted in circumstance-dependent data which was unpredictable in nature and volatile in appearance. This, indeed, provided transparency and objectivity to the study.

4.6 Piloting

It is common knowledge that the researcher should always pilot their research instruments and procedures before launching the study. This is usually to ensure reliability of the instruments and validity of the outcomes. While this principle is generally endorsed by research methodologists, Dornyei (2011:75) asserts that piloting is more important in quantitative studies than in qualitative ones. He cites Richards (2005) who asserts that in qualitative studies, there is normally no real piloting stage in which the research 'tools' are tested. Dornyei however emphasizes that this does not mean that there is no use in trying out certain techniques (for example, interviewing skills). He says that in qualitative studies, initial inquiries are considered 'trials' but constitute part of the data collection.

This is justified in the paragraphs below.

4.6.1 Collective versus individualistic perspectives

As Richards (2005) holds, the researcher in a qualitative project often starts out by treating everything around the topic as potential data. To study rural 120 multilingualism, we started off with the option of visiting groups (schools, churches, homes, Njangi houses (a local co-operative, financial institution) and market where we learned multilingualism would be practiced.

4.6.1.1 Schools

We visited the only secondary school in the region (Government Secondary

School Abar) and studied their highest class: form five. We asked one of the students to wear an obvious microphone from her home as she walked to school in the morning. As she got to school we collected the microphone so that she would attend her classes normally. During break, she wore the microphone again and took it to class, this time to the end of the day; given that it was around the end of the term, when only students from examination classes go to school typically to study on their own. She stayed in class with her classmates as they tutored one another. Here, we found out that because this is the only secondary school in Lower Fungom, students come from all its villages. In

Lower Fungom, Cameroonian Pidgin English is typically used in a mixed crowd, and so this is what prevailed in this class. While the students were studying, they used CPE, outside the classroom and on her way back home, the student wearing the microphone used her mother tongue with people from Abar and Buu. She still used Abar with people from other villages who spoke to her in their mother tongues but who could understand Abar. Worth mentioning is that 121 we had this student wear the microphone from our home and walked to school while we stayed back at home. During all the time she recorded, she was only with her classmates, friends, school staff and the other indigenes of Lower

Fungom whom she met on her way to and from school.

We also visited the government primary school in Abar where we sat in class one. We imagined that, in an environment where people almost always used their mother tongues with children, the transition from these indigenous languages to English (the language of education) would be interesting. We found out that the teachers at the primary school were from other parts of

Cameroon and were sent to Lower Fungom by the government. Consequently, they did not speak the languages of Lower Fungom and could not use them in classrooms. However, we had a case where the class one teacher was teaching the English alphabet, and she was at the sound [d] when she felt like using a dog as an example (‘d’ as in dog). She did not begin by telling the children about a dog, but she tried to make them come up with the word by themselves. When one pupil finally got it, she asked the pupil to say the word ‘dog’ in his mother tongue so that the other pupils would know. Aside this, the teachers spoke to the pupils in English and among themselves in CPE.

122

4.6.1.2 Churches

In Lower Fungom, we have basically three churches: The Presbyterian Church in Cameroon (PCC), The Cameroon Baptist Convention (CBC) and The Roman

Catholic Church (RCC). It is the PCC and CBC churches that are indeed operational. There is just one pastor of the PCC who moves from one village to the other. He, like the teachers discussed above, is from outside Lower Fungom, although he is still from Menchum Division. He does not speak any of the languages of Lower Fungom. In church, he reads the bible and preaches in CPE and English, although he uses CPE most of the time. Outside church, when he is with the members of his congregation, he typically uses CPE.

The Baptist pastor is not from Menchum Division like the Presbyterian pastor, he is from Donga-Mantung division, (a much more distant administrative unit from Lower Fungom). He too does not speak any of the languages of Lower

Fungom. He had been in Lower Fungom only for two years. We attended church service at the Baptist church in Abar three times; twice within the week (as it was Easter week) and once on Sunday. We did not hear him use the languages of Lower Fungom. He read the bible in CPE and preached in CPE. However, in an interview he told us that sometimes he used words from these languages to make a topic clearer. He explained that one day when he was preaching on tithes, he used a word used in Abar to refer to a kind of abnormally formed corn 123 to make his explanation clearer to his congregation. When he was with his

Christians out of the church he only used CPE. Meanwhile, with his wife and children with whom he lives in Lower Fungom, he used his own mother tongue-

Limbum.

4.6.1.3 Njangi (a local co-operative, financial institution) houses

We visited three Njangi houses in Abar. These Njangi houses comprised individuals from multiple and diverse villages of Lower Fungom. Worth mentioning is that every Njangi house we visited was indeed a multilingual milieu; the participants of the Njangi each spoke in their mother tongues. In one of the houses, we had people from Ngun, Abar and Munken. We recorded them for about an hour and discovered that each person spoke in their mother tongue when they had to. It was not very easy to collect information once the Njangi had started, as only the chair person spoke and the rest simply responded when they were called or just presented their own financial contribution. Whenever they desired to say something to their neighbours, they whispered, and so we ended up having recordings with spaces and with inaudible conversations.

4.6.1.4 Market

Lower Fungom has two markets: one in Yemgeh and another one in Abar. The latter is the main market; most people would rather go there. Abar market days are days of festivity. People come together not only to buy and sell but also to 124 meet friends and relatives, drink, eat and dance. On such days, stores and drinking parlours are operational and some even have electricity from generators. We imagined that with this influx of many diverse people, multilingualism would be at its peak. We thus had a trader from Weh (a neighbouring village to Lower Fungom) wear a microphone which was not immediately recognized by most people in the market. We discovered that most traders come from neighbouring villages outside Lower Fungom. Those who sell food stuff such as garri, palm oil and groundnut, come from Lower Fungom but those who sell non-perishable goods generally come from outside Lower

Fungom. Consequently, contrary to our imagination, CPE was the order of the day on market days. It is true that there were still a lot of diverse languages being used by the indigenes of Lower Fungom, especially with the elderly group. This ‘foreign’ trader from Weh could not help us a lot as he spoke in

CPE almost all the time.

4.6.1.5. Homes

We also had plans of studying the linguistic habits of a group in a home setting.

Our intention was to understand linguistic habits in relation to family relations

(such as son-in-law), biography (such as whether a person worked for long in the South West Region, was brought up in a particular area, etc.), power

(Kwifon in the village, Chief’s son, priest) and residence (where he/she comes 125 from). We intended to focus on this group to see how language is actually used.

We had envisaged examining this group in waves with the first wave intended to be the communicative strategies of people from the same village, and the second wave would be about conversation with people from other villages of Lower

Fungom. As a result, we visited a multilingual home in one of the villages of

Lower Fungom, Mundabli (the wife was Mundabli while her late husband was

Koshin) and discovered that in Mundabli, their mother tongue is generally favoured, especially when the participants are in their own village. We had a gentleman from Mundabli talk with one of our major consultants and he (the gentleman) mastered only his mother tongue and CPE. He used CPE with our consultant and later told us that he liked to use another language every now and then. We also visited another home in Buu and there, we met people from Fang,

Abar, Koshin and Buu. Everyone during this encounter spoke their first language. Only one lady from Koshin was using another language mainly because the then-regent of Buu had forbidden her from using Koshin as Koshin and Mundabli were in dispute.

These five groups gave us the chance to see if setting influences language choice. It also gave us a chance to see that in Lower Fungom, multilingualism is very common; the vast majority of people have competence in a good number of languages even if they may lack performance. However, observations of these 126 five groups did not provide us with the most detailed intricacies of multilingualism; it did not provide us with insight into the personal linguistic decisions of participants. Sticking to groups would have produced interesting results with great depth and peculiarities. However, it would not have provided us with the most detailed intricacies of multilingualism; it did not provide us with the personal linguisitic decisions of participants. We concluded that a more productive study would be even more small-scaled, with focus on individuals.

We chose to keep the settings described above as part of this study, but only if one of our consultants was present in it. We could still have a comprehensive study covering markets, homes, schools, Njangi and churches with just a few consultants wearing microphones. This made us switch from a collective approach to an individualistic approach in data collection.

4.7 Sampling strategies

4.7.1. Setting

The preceding discussion sets the groundwork for the discussion of our sampling techniques. As far as choosing the study’s area of focus is concerned, we chose to have an open mind (within the confines of focusing the study on

Lower Fungom). We tagged along with the consultants, accompanying them wherever they went if it benefited the study.

127

The objective of this qualitative study was to describe linguistic practices as they occurred naturally. As such, we focused on the natural setting, without any attempts to manipulate the situation under study. To capture a sufficient level of detail about the natural context, we lived in the community in constant contact as it is the practice in ethnographic studies.

Data was collected from those who interacted with the major consultants in Buu,

Missong, Abar, Fang, Munken, Mashi and Mufu. However, our major consultants lived in Buu, Abar and Missong. We selected consultants in this area because Abar is the center of all thirteen villages of Lower Fungom; it is thus host to many persons from other villages. It is more heterogeneous than every other village in Lower Fungom. We were able to ascertain this fact during our feasibility study when we conducted interviews with individuals in Abar, Buu and Mundabli. We understood from our feasibility study that most of the villages of Lower Fungom were incredibly linguistically diverse; with only a few exceptions. This means that it was possible to select any of these linguistically diverse villages as our setting. We therefore had no specific guiding factor in choosing these villages. Several factors conspired in favour of

Abar as the village of focus in this study. At the time of this research, Abar was the centre of the thirteen villages; it was the region with the only secondary school in the area, the only major market and the only health centre. It was also 128 the gateway through to several villages of the area. The consultants would almost always come to Abar.We collected data from consultants when they were at their homes, at the jobsites (which could be several kilometers away, like in the case of Kulo who also worked in Nkang in the Furu Awa region), in their farms and in diverse kinds of gatherings. Wherever the consultants found themselves, that area became part of the study’s setting. For example, when we went for the third trip, we found out that one of our major consultants, Janet, had travelled out of Lower Fungom to her husband’s village Esu (still in Menchum

Division, North West Region) to give birth. So, we travelled to Esu and worked with her there. In sum, we did not only work in Abar, Missong and Buu, but also in any other village the consultants visited.

4.7.2. Population sample

As far as the sample population is concerned, we did not initially have a specific number we aimed at working with; we kept the study open, as it is characteristic of qualitative investigations to be emergent. We went for participants in a logical and connected sequence to the point where new insights were no longer forthcoming (saturation). After we had selected our major consultants, we began collecting the data. Next, we listened to the initial recordings with them and selected another set of consultants. This selection was not randomly done, but it was based on the distinctiveness of the consultants. We chose the second set of 129 consultants whose linguistic choices were unusual, unexpected and interesting.

For example, we realized that one of our major consultants, Janet (see section

5.2.1) always conversed in Buu with mama Turner (see section 5.2.4) who was from Mufu. Mama Turner was from Mufu and so was Janet. Mama Turner only moved to Buu because she married from Buu. Janet's mother was from Mufu and so we expected both of them to speak in Mufu, at least from time to time.

This was not the case, we found this interesting and worth further investigation.

We found it necessary to follow-up interlocutors such as mama Turner to know their own backgrounds, to understand the reasons behind their linguistic habits.

We interviewed them to know if their linguistic choices were a result of their limited linguistic repertoire, the language ideologies of Lower Fungom or other sociological factors, such as employment, inter-tribal conflict, social mobility and education. We followed them to their homes, in their respective villages, arranged for meetings with them and with the help of our interview guide, sought answers to the questions on their background and linguistic choices.

We kept adding to the number of this second set of consultants as we listened to additional recording from our major consultants and from our observations. This went on until responses from consultants became repetitive. We ended up with three major consultants and twenty minor consultants making a total of 23 consultants for the study. 130

Qualitative studies are directed at describing the aspects that make up idiosyncrasies, the unusual feature of something, rather than determining the most likely, or mean experience, within a group (Polkinghorne, 2005).

Accordingly, this inquiry was not concerned with how representative the respondent sample was or how their experiences were distributed in the population. Instead, our main goal in sampling was to find individuals who could provide rich and varied insights into the language use to maximize our findings. This goal Dornyei (2011) asserts is best achieved by means of some sort of 'purposeful' or 'purposive' sampling.

Of the three strategies of purposive sampling; homogeneous sampling, typical sampling and criterion sampling, we chose homogeneous sampling where we selected participants from a particular subgroup who shared some important common characteristics. In this way, this strategy allowed us to conduct an in- depth analysis of identifying common patterns in a group with similar characteristics. Dornyei (2011) admits that a well-designed qualitative study usually requires a relatively small number of respondents to yield the saturated and rich data that is needed to understand even subtle meanings in the phenomenon under focus.

131

Saturation and sample size were very much related to the selection of participants in this study. Iteration on the other hand was concerned with the data collection and analysis. The participant selection process remained open long enough into the data collection process; after initial accounts were gathered and analyzed, additional participants were added for additional data. It is this cyclical process of moving back and forth between data collection and analysis that Dornyei (2011) refers to as 'iteration'. Although iteration is a key process in qualitative sampling, Dornyei feels that it should go on until saturation is reached. Glaser and Strauss (1967) defined this as the point when additional data do not seem to develop the concepts any further but simply repeat what previous informants have already revealed. In other words, saturation is the point when the researcher becomes 'empirically confident' (p. 61).

4.7.3. Participants

We chose to collect information from three remarkable individuals (our major consultants), and from a few persons they interacted with who displayed unexpected linguistic behaviours. We used purposive homogenous sampling in choosing the major consultants; we chose participants with the same linguistic behaviour. From our feasibility study, we learned about the linguistic abilities of a few individuals. From these individuals, we chose our consultants- those who were indeed multilinguals (see chapter five for details). These consultants 132 deployed at least six of the lects of the thirteen villages of Lower Fungom in actual daily use. This criterion was very pertinent for this study, and the choice of such a group with similar characteristics permitted us to explore similar patterns among them. As a reminder, this is a small-scale ethnographic case study that aims at digging very deep into the linguistic habits of individuals in great detail. We believe that this can be accurately achieved with this limited number of participants. These participants are profiled in chapter five; the information on them is part of the results of the sociolinguistic survey that we conducted.

4.8 Data analysis

This section discusses the range of processes and procedures we took to move from the raw data into some form of explanation, understanding or interpretation of the people and situations we were investigating. We moved back and forth between data collection, data analysis and data interpretation in this study. For example, while we were analyzing some part of the data, we came across some very crucial information portrayed by one of our major consultants. We went back to the field and asked further questions to his interlocutors and some other minor consultants. Dornyei (2011: 243) stresses that it is not at all unusual to decide well into the analytical phase that there is need to collect some more data 133 of a specific sort, or to go back to the original data transcripts to recode the text according to some newly conceived categories.

4.8.1 Grounded theory

We chose grounded theory because it is an effective approach to building new theories. The resulting theory or hypothesis helps to generate future investigation into the phenomenon, which was something we aimed at. We considered investigations into the use of local languages in rural areas to be a relatively new area of study which needed further inquiries. The constant comparison which is part of grounded theory helped reveal the idiosyncratic

(distinctive) nature of the consultants. It helped us generate patterns from which tentative hypothesis where formed. We compared the linguistic habits of the consultants against one another. This revealed the reasons behind the choices each consultant made and the way these consultants use the languages they knew. In addition, grounded theory permitted us to collect data at different rounds and this helped to ensure meaningful and validated results.

4.8.1.1 Coding in grounded theory

For data analysis, we first broke up the data into chunks and assigned conceptual categories to the data segments. For example, we listened to the recordings made by the consultants. We took note of who initiated the conversation because we feared that if the consultants themselves intiated every conversation, then they 134 may be using multiple languages just as a show-off. We identified the languages used, the dynamics involved in their usage (code switching, language choice), and the reasons why each consultant chose a particuar language. We put the languages actually used in one group, the dynamics involved in their usage in another group and the reasons behind the language choices of the consultant in a third group. More particularly, we paid attention to:

1) What people (consultants and interactants) said;

2) The language in which they said it;

3) The topics that were discussed;

4) The setting;

5) Why they used specific languages;

6) How they used these languages (language choice and code switching);

7) What was common and different across the consultants.

Figure 2 provides a summary of the conceptual framework for analyzing our data.

Figure 2: Framework for data analysis

Developed codes (e.g setting, rebuking, to be understood, secutity)

Coded the data (gotten from recordings and observations)

Found patterns and relationships between the codes

Summarized the data and provide 135

hypotheses

These codes enabled us to see the difference and similarities among consultants; they enabled us to categorize our consultants (see section 5.2). We consequently placed them on a sort of continuum of who is the heaviest multiple language user and the lightest. From the codes, we built hypotheses such as "code switching or language choice in Lower Fungom. The codes also informed us of the reasons behind the language choices of the consultants. These reasons include, the relationship between the living and the supposedly dead, mimicking and security.

4.8.1.2 Constant comparison

As already mentioned above, sampling, data collection and analysis for this work took place in an iterative manner, consisting of collection of data and constant comparison between results and new findings in order to guide further data collections. For this reason, the development and identification of themes did not take place prior to data collection, but instead as part of the data collection and analysis process.

4.8.1.3 Theoretical Saturation

We stopped selecting participants and collecting data when we reached theoretical saturation in both processes. Theoretical saturation with minor consultants was reached once we noticed that the consultants' responses stopped 136 generating new information. For example, during our second trip, our last minor consultant was Mbang Silas from Buu. We could not spend as much time with him like we did with the other minor consultants because his responses were very much like those from previous minor consultants. He was thus the last consultant we added to our list. We used early findings to shape the on-going data collection. We conducted more interviews during our third trip, after we had listened to the recordings and after further observation by the researcher.

This was to address issues brought up by preliminary analysis such as unusual linguistic practices by our consultants and their interlocutors. After listening to one of Boniface's (see section 5.2.2) audio recordings, we found one of his interactions particularly fascinating. What Boniface was able to achieve with his choice of language was enthralling (See section 6.4). We decided to interview him further and also his interlocutors. We also interviewed mama TR (already seen above) regarding Boniface's linguistic behaviour on that day.

During our third trip to Lower Fungom, we went even deeper in analyzing our data. We did not only do a free translation of the participants’ conversation, but also included a word-for-word translation. This process was intended to enable us to tell whether any word or fragments of a different language had been inserted into the actual speech of the consultants. It is easy to believe that with the degree of contact between the villages of Lower Fungom, there was bound to 137 be some form of code switching/mixing in their conversations. Worth noting is the fact that this was a first attempt in this direction; our transcriptions were still far from being totally dependable as, to become such, one needed either to be proficient in a number of local languages, or to work closely with a number of trained consultants. Even though our data lacked phonological and morphological precision, our objectives were still achieved and our findings are indeed groundbreaking.

4. 9 Data presentation

ELAN (Eudico Linguistic Annotator) adds annotations to a video or audio clip.

Video and audio data from the field work for this project was analysed and presented with the help of the ELAN software. Basically, the presentation comprised tiers that demonstrate the language use, the main speaker, the interactants, the interlocutor, free translation of words and sentences, and comments about the setting, cultural facts, or participants.

Findings for this study are presented in three different chapters in relation to the research questions. The findings are divided into subsections:

1) Data from interviews (self reports);

2) Data from audio and video recordings (natural occuring conversations); and

3) Data from observation. 138

The findings are part of the chapters that follow, but before moving to these findings, here is a discussion of how ethical issues were handled in this study.

4.10 Ethical consideration

Researchers face ethical challenges in all stages of a study, from designing to reporting. These include anonymity, privacy and informed consent. Our consultants were free from any coercion; they participated on their own free will. Whenever anyone was not willing to participate, we made sure we made them know that their decision was not a problem and then we moved on to another person. Participants were also fully informed regarding the reasons for research; we told them that it was for academic purposes.

We were involved in recording the natural conversation of consultants at their work places, homes and many more places. We understood that these individuals needed their privacy; we made it very clear to our consultants that they were free to pause the recording for a while or better still, to ask us not to include it as part of our data. This was very important especially because part of the recording was done at the health centre of Lower Fungom where one of our major consultants was the nurse. So, the participants understood that they were, without judgement, allowed to ask us not to video or even audio record whenever it was inconvenient for them. We also told the consultants to let us 139 know when they did not feel comfortable with us translating any of their conversations into English and this was what happened.

As far as anonymity was concerned, we used pseudonyms instead of the original names of the consultants. The consultants did not have a problem with their real names being used; the decision was entirely ours. This study celebrates these individuals, their ideologies and their locality. However, maintaining anonymity is paramount in qualitative research, albeit challenging. Consequently, we assigned fictional names which we could easily remember to each consultant.

Also, we always sought the consultants' permission before another person (like a judge) could listen to their recordings.

Conclusion to chapter four

This chapter aimed at providing ample information on how the research was carried out. The chapter has discussed the research design (qualitative), method of data collection (case study), instruments and tools of data collection

(interview guide, observation, audio and video recorders), procedure of data collection, piloting, sample strategies (setting and population), data analysis and presentation. The chapter ends with a discussion of the ethical considerations we took. Chapter five presents a detailed description of the consultants and their linguistic practices. 140

CHAPTER FIVE

PRESENTATION OF PARTICIPANTS AND THEIR DAILY LINGUISTIC ACTIVITIES (SELF REPORTS)

5.1 Introduction

This chapter comprises the first of the data presentation and analysis chapters. It discusses in detail all the participants involved in this study based on data from the self-reported survey mentioned in chapter four. As a reminder, this study gathered data from two sets of individuals: the ‘major’ and the ‘minor consultants' (see chapter four). We met with the latter set once or twice only, and our discussions with them focused on their linguistic interactions with the former. Section 5.2 presents and discusses data on the major consultants while section 5.3 centres on the minor consultants. The Chapter rounds off with a conclusion.

5.2 The sociolinguistic profile of primary consultants as reported in the sociolinguistic questionnaire

5.2.1 Ntamneh Mbang Janet

Mme Mbang Janet was born in Buu in 1981 to a Buu father and a Mufu mother.

This vibrant and extraordinarily multlinigual lady was married to man from Esu who at the time the fieldwork was done was the Chief of the Health Centre in 141

Abar. She reportedly spoke more than thirteen languages: all the languages and dialects of Lower Fungom plus English, French, CPE and Esu to varying degrees. She reportedly had native competence in Buu, Abar, Missong, Fang,

Koshin, Munken, Mundabli, Mufu, and CPE. She was fluent in Mashi, Ajumbu and English. She also spoke Biya, French and a bit of Ngun. Her father was from Buu (Tsa quarter) and her mother from Mufu (Gen quarter). Her husband spoke Esu, English and CPE and they had five children with whom Mme Mbang

Janet spoke only Buu. Her mother, Mama Mbang Frida (discussed below in section 5.4), reportedly spoke Mufu, Mungaka, Abar, Missong, Buu and CPE.

However, Mama Mbang Frida only spoke Buu, her husband’s language with

Janet. Mbang Janet's father reportedly spoke Buu, Fang, Koshin, Missong and

CPE. Notice that her father did not speak Mufu, his wife’s mother tongue.

Below is a more detailed discussion of Janet that will help understand her attitude towards people and their languages.

Mme Mbang Janet went to primary school, classes one and two at Abar center.

For classes three to seven, she was sent to Ekok (Eyumodjock subdivision,

Manyu division, South West Region, Cameroon) at her late uncle’s (her father’s youngest brother, who was a policeman), who was from Buu. She came back home during summer holidays. She did her Forms one to three in Wum, as there was no secondary school in Abar at the time. She spent some of her holidays 142 back home. During one of these holidays, in 1993, as she was to go to Form

Four, she got pregnant. During this period, she lived with the former

Cameroonian Member of Parliament, honorable Nkangkolo. Nkangkolo was first cousins with her father. Six months after she gave birth, she travelled again to Ekok where she began studying tailoring. After a while, her uncle (guardian) had marital squabbles in his polygamous home and also had no money to pay for her studies. Mme Mbang Janet had no choice but to dive into petty trades; she travelled to Onitsha in Nigeria and bought kitchen utensils such as plates which she sold in Abar. Soon afterwards, she returned to Abar where she settled finally. At the time of our research, she had five children from her previous relationship, having just lost the sixth baby she had with her new husband.

Language acquisition path

Mme Mbang Janet learned Koshin from the uncle with whom she lived in Ekok, as one of his wives was from Koshin. As far as Fang is concerned, Janet’s father was related to Fang. This prompted many visits to this area. Janet reported that people from Koshin and Fang did not understand Abar, consequently, she (who is able to speak both Fang and Koshin) alone could switch from Abar to any of these languages. It is worth mentioning that Kulo Rene, one of our primary consultants, (as we shall see later in section 5.2.3) makes a similar claim.

143

While Janet stayed in Wum as a student, she joined a Njangi group of Biya and

Ngun indigenes. During these meetings, Janet would pick up a few words of

Biya and Ngun. People from Biya and Ngun could understand Abar, so when she met them, she would speak in Abar, but if they knew her personally, they would speak to her in Ngun or Biya and she would respond accordingly. She also asserted that some people in Biya were from Buu because they had fled historical inter-tribal wars to settle in Biya. They could thus understand Buu. So, with the Biya, she sometimes spoke Buu.

Mme Mbang Janet picked up Kung from her classmates in Wum. At that time, there was no secondary school in Lower Fungom. Most of its indigenes went to school in Wum, hence the school was linguistically mixed. Janet reported that she had not lived with anyone from Biya, Kung or Ngun; she had picked up vocabulary items from these languages during her time in Wum.

Mme Mbang Janet’s mother was a granddaughter of Mundabli, with her mother coming from Mufu. The similarity between these languages made her understand Mundabli. Janet also learned Mashi here and there; she was not related to anyone from Mashi. She used this language whenever she met someone from Mashi who did not understand Abar.

144

Janet's grandmother was from Munken; Janet said that in Buu, maternal families were more cherished, so her grandmother used to speak to them in Munken. She also used Munken with someone from that area who did not understand Abar.

The table below summarizes all what has been discussed above of Janet’s linguistic pattern.

Table 2: A summary of Mbang Janet's linguistic pattern Language Reported Acquisition paths Present in my recordings name competence (see section 4.2) Buu 5 Born in Buu and Yes (one of them is Buu is her father’s KPAAM_CAM_RO_multilingualis primary language m_Janet_ABARHOME_2015_12_0 7) Abar 5 She attended the Yes (one of them is only primary KPAAM_CAM_RO_MULTILING school at the time UALISM_JANET_BUUFANG_20 located in Abar 15_11_22 Fang 5 Father is related to Yes (one of them is Fang and they KPAAM_CAM_RO_multilingualis have a very good m_Janet_ABARHOME_2015_12_0 relationship, so 7) there is constant contact Mundabli 5 Mother is a No granddaughter of Mundabli, this means constant contact and it is similar to mother’s language Missong 5 Just around in Yes one of them is primary school in KPAAM_CAM_RO_multilingualis Abar/Missong m_Janet_ABARHOSPITAL_2015_ 11_26) 145

Biya 2 She lived around No people from Biya who also encouraged her to join the LF meetings in Wum and also mingle with other people from Biya Mashi 4 Here and there as No she went about LF Munken 5 Grandmother’s Yes (one of them mother was from KPAAM_CAM_RO_MULTILING Munken, so she UALISM_JANET_ABARHOSPIT regularly used AL_2015_11_25 Munken with them Mufu 5 Mother is from Yes (one of them is Mufu KPAAM_CAM_RO_MULTILING UALISM_JANET_BUUFANG_20 15_11_22 Ngun 2 As a student in Yes (one of them is Wum, as she KPAAM_CAM_RO_multilingualis attended LF m_Janet_ABARNJANGI_2015_11 meetings _25) Ajumbu 3 In Buu from a No friend who got married in Buu and could not speak Buu Koshin 5 Uncle with whom Yes one of them is she lived growing KPAAM_CAM_RO_multilingualis up married from m_Janet_ABARHOSPITAL_2015_ Koshin 11_28) Kung 4 Had a close friend No from Kung while 146

in secondary school in Wum. As they were both from LF, they bonded. English 4 In school as it was Yes one of them is the language of KPAAM_CAM_RO_multilingualis instruction m_Janet_ABARHOSPITAL_2015_ 11_28) French 3 Her uncle who Yes (one of them is was a policeman KPAAM_CAM_RO_MULTILING with whom she UALISM_JANET_BUUFANG_20 lived, in Yaounde, 15_11_22 Kribi and Bafia CPE 5 From classmates Yes (one of them is and around LF KPAAM_CAM_RO_MULTILING UALISM_JANET_ABARHOSPIT AL_2015_11_26

5.2.2Nto Boniface

Mr Nto Boniface was a calm and hardworking man born in Missong, in 1980 to a Koshin father and a Missong mother. He was the son of the then Fon of

Missong who was reported to be a hundred and twenty-two years old in 2016.

His father was from Koshin (Sawe), while his mother was from Missong.

Boniface’s father reportedly spoke Koshin, Missong, Mashi and Mufu, while his mother was reported to speak Buu, Fang, Ajumbu and Mundabli. Husband to two wives and father to five children and six grand children, Mr Nto Boniface was a wonderful hunter and farmer. At the time of this study, he lived in 147

Missong with his family. His first wife was from Missong and they had been divorced for some time when we met Boniface. His second wife was from Buu.

She reportedly spoke Buu, Missong, Munken, Biya, CPE and French. Boniface himself reportedly spoke Missong, Mashi, Buu, Munken, Mundabli, Mufu and

CPE as a native speaker; he was fluent in Menkaf and English; he spoke a bit of

Abar and could only understand some French. Boniface attended primary school at Abar-Missong. He went to Wum for his secondary school, but he only completed form four because his father had fallen sick and could not pay his registration fee for the Cameroon General Certificate of Education (GCE) examination. He left and started to work at the CDC (Cameroon Development

Corporation) and other plantations in the South West Region of Cameroon. Even though he had lived outside the village for a long time, Boniface still used his mother tongue most of the time. Some people who had lived out of Lower

Fungom rather used a lot of CPE. We observed that Boniface was one of those who used their mother tongues most of the time. When Boniface switched from

Missong to another language, he had pertinent reasons. Mr Nto Boniface enjoyed using his mother tongue; he used this language most often, especially when he was in his village. Nevertheless, it is also worth mentioning that he inserted fragments of CPE into his Missong. This is because, as mentioned earlier, he had spent time away from his village and Lower Fungom, he had been in the South West Region of the country working in plantations, and also in 148

Nkongsamba where he used Cameroonian Pidgin English frequently. The table below summarizes the sociolinguistic profile of Boniface’s linguistic pattern.

Table 3: A summary of Boniface's linguistic patterns Language Reported Acquisition paths Present in my name competence recordings (see section 4.3) Buu 5 Maternal grandfather’s Yes (one of them primary language KPAAM_CAM_RO_M ULTILINGUALISM_N TO_ABARMARKETD AY_2015-12_02) Abar 5 Similar to Missong Yes (See KPAAM_CAM_RO_M ULTILINGUALISM_N TO_ABARMARKETD AY_2015-12_02 Mekaf 5 Similar tradition with No Missong and his father is related to Mekaf so there is constant contact Mundabli 5 Similar to Mufu Yes (see KPAAM_CAM_RO_M ULTILINGUALISM_N TO_TRADITIONALCE REMONYMISSONG_2 015_12_03

Missong 5 Mother’s first language Yes (See KPAAM_CAM_RO_M ULTILINGUALISM_N TO_ABARMARKETD 149

AY_2015-12_02 Biya 2 Yes (See KPAAM_CAM_RO_M ULTILINGUALISM_N TO_ABARMARKETD AY_2015-12_02 Mashi 4 Wife’s grandparents Yes (see were from Mashi, she KPAAM_CAM_RO_M uses Mashi with their ULTILINGUALISM_N children. He also had a TO_TRADITIONALCE friend from Mashi in REMONYMISSONG_2 primary school. 015_12_03 Munken 5 Similar to Missong so Yes he did not need to learn it Mufu Similar cultures with No Missong, he partakes in their events and so do they. They also share boundaries and farms English 3 In school as a language Yes (see of instruction KPAAM_CAM_RO_M ULTILINGUALISM_N TO_ABARMARKETD AY_2015-12_02) French 2 In the plantations in No Nkongsamba and in those in the South West Region CPE 5 From classmates and Yes (see around LF KPAAM_CAM_RO_M ULTILINGUALISM__N TO_NJANGI_2015_12_ 03 150

5.2.3 Kulo Rene Domo

Mr Kulo Rene, born in 1975 in Buu was phenomenon. His father was from Abar while his mother was from Mundabli. He and two of his brothers were born in

Buu before his father’s brothers brought him back to Abar. Mr Kulo Rene was the third child of his father. His father first married a woman from Missong; she gave birth to a girl who passed away soon afterwards. He later married Kulo’s mother, a woman from Mundabli (Banleh quarter) and yet again had a son who passed away soon afterwards; then he had Kulo. At that time, it was Kulo’s uncle (his father’s brother) who was the Fon of Abar. After this man died in

1980, Kulo’s father was made chief. Then he kept marrying. He married a third wife from Abar (Mban quarter), a fourth wife from Abar (Mbu quarter), a fifth from Fang, a sixth from Mundabli, a seventh from Missong, an eighth from Esu, and a ninth from Mashi. All these women spoke their languages to their children and when he was around, he learned them.

The remarkable and humourous Kulo Rene first married when he was twenty- five. He went to primary school at Abar-Missong, from class one to class seven.

In those days, this was the only school in Lower Fungom. He attended Forms one and two in CPTS Ndu. His father had a friend who was a pastor in Ndu.

This man invited Kulo to Ndu for studies; Kulo went there by helicopter. After two years there, he was home sick and asked to be taken to Abar for the summer 151 holidays. After a difficult trip by car, he got to Abar. As Kulo got there, his father’s relatives advised his father not to let Kulo return to Ndu because it was too far away. They preferred Bamenda which was nearer. Kulo thus went to continue school at Progressive Comprehensive Secondary School (PCSS)

Bamenda (now known as PCHS) in Forms three and four. He became too stubborn in form five and paid neither his school fees nor his registration fee for the Cameroon General Certificate Education examination- ordinary level.

Consequently, he was dismissed from the school. He then developed interest in earning money and left Bamenda to Douala (Cameroon’s economic capital) where he engaged in several odd jobs for livelihood. He later returned to Wum where he trained as a motor mechanic and then driver. After four years in the driving school, Kulo met his first wife who had made it to form three in Wum, and started a family. With his first wife from Abar (Ugako quarter), Kulo had three children. After many years and due to family squabbles, Kulo separated from his wife and took a second one. She too was from Abar (Utong quarter).

Mr Kulo Rene, as discussed above, unfortunately stopped formal schooling in

Form four, but back in the village, he was noted for being smart and was also a friend to the nurses. One of his friends, Robert (from the West Region), who was the chief of the health centre in Lower Fungom at the time, decided to pay for his training in a medical school in Kumba where he became an EPI 152

(Expanded Program on Immunization) vaccinator after nine months. He then went to Ekondo Titi (in the South West Region of the country) for practical lessons, and was later transferred to Wum. The same Robert intervened and had him transferred to Abar as the EPI vaccinator in Lower Fungom. Kulo’s work extended beyond Lower Fungom towards Furu Awa. He was trained as an EPI vaccinator but at the time this research was conducted, due to lack of staff, his job description extended beyond that of a vaccinator. For example, he weighed pregnant women, examined them, and gave pregnant women and children from ages zero to nine the ATT vaccine. He also worked as a midwife, as he had participated in training workshops in Esu on midwifery.

Language acquisition path

Mr Kulo Rene reported to speak Koshin because his aunt (father’s sister) had married from Koshin and this aunt used Koshin with her children. She lived just around his job site, and when he visited them, they would all speak Koshin. He also attended classes with some boys from Koshin from whom he learned their language as they interacted. In addition, Kulo had a girlfriend from Koshin, who helped him perfect his knowledge of Koshin.

Kulo says that Mufu and Mundabli are very similar, so he did not need to learn

Mufu. This is also the case with Biya, Munken and Missong which are like

Abar. 153

As for Kung, Kulo’s grandfather’s mother was from Kung. He and his father used to go to Kung for a Njangi. There, he said, the people from Kung would force him to learn the language, saying that he was part of them. They always spoke to him and his father in Kung.

Mr Kulo Rene had also been a businessman; he traded in coffee for five years when this was not common in Lower Fungom. He would buy from Lower

Fungom and sell in Foumban. This is how he learned French, although he had picked up a few words and phrases from the streets with friends in Douala.

Kulo used Abar with everyone from Missong because they understood Abar. He asserted that he typically code switched when he met someone who did not understand his language: for example, people from Koshin did not understand

Abar. He said that even with someone from Abar he (Kulo) could use another language, provided the person understood it. He remembered an instance where he was walking with a friend of his and they remembered how they would speak

Mundabli, after spending a good time there. “Sometimes, a language expresses a description better than another especially for fun," Kulo affirmed. In such a case, they would use this language to discuss the matter.

Kulo’s job site was another avenue for multilingualism. He said that when it came to the very elderly who could not understand CPE, he needed to explain to 154 them what was wrong with their health and how they should take their drugs in a language they understood best. His job made him practice and become better at all these languages. He was, at the time of this research, struggling to improve on the language of the Aku (Hausa), as he worked in Nkang and even in Abar.

He said he was well loved in Nkang because he always made a deliberate effort to use patients' language. They had even given him a name in their language: Dr.

Hang’Ori (meaning Dr Mad man). Kulo reported a linguistic repertoire of at least fifteen languages; he spoke Abar, Mundabli, Buu, Missong and CPE like a native speaker. He was fluent in, Koshin, Fang, Munken, English and Ngun. He spoke a bit of Mufu, Kung, Biya, Mashi and French. He learned French from a

Bamileke (Western Region of Cameroon) man in whose shop he was the sales person. Although he was the son of the late Fon of Abar, he could not be crowned Fon after the death of his father because the indigenes of Abar claimed he was not completely from Abar, since his mother was from Mundabli and

Kulo was born in Buu. Below is a table that summarizes the sociolinguistic profile of Kulo.

155

Table 4: A summary of Kulo's linguistic patterns Language Reported Acquisition paths Present in my recordings name competence (see section 4.3) Buu 5 Born and raised Yes (see there KPAAM_CAM_RO_MULTILI NGUALISM_KULO_2015_11_ 22 Abar 5 The only primary Yes (See KPAAM-CAM- school at the time RO_MULTILINGUALISM_KU was in Abar. LO_ABARMARKET_2015_11_ Father’s primary 24 language Fang 4 Father married from Yes (see Fang and the KPAAM_CAM_RO_MULTILI women spoke Fang NGUALISM_KULO_2015_11_ constantly to their 22 children Mundabli 5 Mother was from Yes (see Mundabli KPAAM_CAM_RO_MULTILI NGUALISM_MUNDABLI_AB AR_KULO_ABAR_2016_05_3 1 Missong 5 Father married a Yes (see woman from KPAAM_CAM_RO_MULTILI Missong NGUALISM_KULO_2015_11_ 22) Biya 3 Similar to Abar No Mashi 3 Father married a Yes (See KPAAM-CAM- woman from Mashi RO_MULTILINGUALISM_KU LO_ABARMARKET_2015_11_ 24 Munken 4 Similar to Abar Yes (see KPAAM_CAM_RO_MULTILI NGUALISM_KULO_ABARSH AAHOUSE__2015_12_04 Mufu 3 Similar to mother’s No primary language 156

Ngun 4 Yes (see KPAAM_CAM_RO_MULTILI NGUALISM_KULO_ABARBA SE-2016-05-18 Ajumbu 2 No Koshin 4 His aunt who lived Yes (see close by married KPAAM_CAM_RO_MULTILI from there and used NGUALISM_KULO_2015_11_ this language with 22) her children constantly. Also had a girlfriend from Koshin Kung 3 Grandfather’s No mother is from Kung, he went there for meetings English 4 In school as Yes (See KPAAM-CAM- language of RO_MULTILINGUALISM_KU instruction LO_ABARMARKET_2015_11_ 24 French 3 From one of his Yes (See KPAAM-CAM- bosses whose first RO_MULTILINGUALISM_KU official language LO_ABARMARKET_2015_11_ was French and 24 from trading in coffee in Fumban CPE 5 From classmates Yes (See and around LF KPAAM_CAM_RO_MULTILI NGUALISM_KULO_ABARBU U_2015_12_02

157

5.3 Minor consultants

We held interviews with seventeen individuals from Lower Fungom who interacted with the major consultants presented in the previous section. The present section summarizes the findings as gleaned from the the self reported sociolinguistic questionnaire.

1) Yamuo Francis from Mundabli reportedly spoke only CPE and Mundabli. He was conversing in these languages with Kulo when we took interest in him.

2) Mr Jum Denis is from Buu and was best friends with Kulo. He reportedly spoke five languages: Buu, Fang, CPE, English and Abar. We took interest with him as he interacted with both Mr Rene and Mme Janet. Mr Jum had a fascinating mother who spoke many languages, many from Lower Fungom but she did not speak Buu at all. Buu is her late husband’s primary language and the language spoken in the village in which she had lived for over fifty years. This mother lived in Kung in her youth and then moved to Ajumbu. She was forcefully brought back to Buu in blindfolds, where she was expected to bear children to expand the family lineage which was fast fading away. We speculate that this act infuriated her and caused her to shut her mind to Buu.

3) Mr Nkole Lucas was fifty-nine and he was the-then regent of Buu. (A regent in Lower Fungom is temporary leader who acts in the absence of a Fon). His 158 wife was from Buu and he reported speaking Buu, Fang, CPE, English, Mufu and Missong. He spoke only Buu with native speaker fluency; the others he claimed to only understand and to speak a bit. He interacted with Mr Kulo and

Mme Janet.

3) Pa Buh Mbobeh Joseph was sixty-one and he is also from Buu. He was good friends with Kulo and enjoyed making jokes with him. He was married to a woman from Mundabli and they both lived in Buu. He reportedly spoke Buu,

Mundabli, Abar and CPE. He interacted with Janet and Kulo.

4) Mr Ugeh Ntung Blasius was forty-one and was a herbalist from Muken. He reported to speaking Munken, Missong, CPE, Abar, English and French. He interacted with Boniface in Missong.

5) Mr Yelo Paul was fifty-five and he is from Abar. He interacted with Janet and

Kulo using Abar as he only spoke Abar, CPE and English.

6) Mama Tuna Regina, forty-two, is from Mufu and was married in Buu. She reportedly spoke Buu, Mufu and CPE. She lived in Buu with her husband. Her husband spoke only Buu and CPE. She interacted with both Kulo and Janet.

7) Mr Ebioh Nkam Elvis was a thirty-two-year-old farmer from Missong. He reportedly spoke Missong, CPE and French. He only just returned to the village; 159 he was in the South West Region and in Mbanga where he learned English and

CPE.

8) Pa Mbang Diah Shadrach was a farmer when he was much younger. He is from Buu and he did not know his age although we believe that he could be about seventy years old. He is the father of one of our major consultants Janet.

He reportedly spoke Mufu, Buu, Missong, Munken, Abar, Fang and CPE like a native speaker.

9) Mr Akwa Eugene Meh is a thirty-seven-year-old businessman from Abar. His parents were from Weh, a village about twenty-five kilometers from Abar. His wife was from Abar. He claimed to speak Weh, CPE and English. He interacted with Kulo in Abar.

10) Mama Mbang Frida is the mother of one of our major consultants, Janet. She is from Mumfu, she was married to Pa Shadrack (seen above) from Buu and she lived in Buu. She said that she was 50 years old and she was a farmer. She interacted with Janet and Kulo.

11) Mama Mbande Frida was a farmer from Mundabli who married in Abar and lived in Abar. She did not know her age. She said that Kulo was her brother just because she was neighbours with Kulo’s mother in Mundabli before they both got married to men from Abar. 160

12) Pa Domo Gha Linus was born in 1949 to Abar parents. He was a retired driver and mechanic who lived in Abar. He had two wives: one from Buu and another from Abar.

13) Nkangkolo Isaac Nkolo was born in 1976 to Buu parents. His wife was from

Missong. At the time of this research, he was a teacher in Kung. He interacted with Kulo and Janet.

14) Pa Nkole Simon of Buu did not remember his age but he could be about 60 years old. He had two wives. His first wife was from Missong (Boniface’s sister) and his second wife was from Buu. He interacted with Janet and Kulo.

15) Mr Botan Bodwin was 30 years old, he was a farmer and a fisherman from

Mufu. He was married to a woman from Mufu; he was friends with Kulo and he also knew Janet.

16) Mama Chefesoh Chu Margaret was born to Abar parents and she married from Abar. She did not remember her age although we estimated it to be 55. She was a cleaner at the health centre and had recently been trained as a midwife.

She was also a farmer. She claimed that Kulo was her younger brother and that she was friends with Janet.

17) Marcus is Boniface’s brother. He lived in the South West Region of the country and he was a mechanic. He interacted with all three consultants. 161

The table below summarizes what has been discussed above regarding the linguistic patterns of the minor consultants.

Table 5: A summary of the linguistic patterns of the minor consultants

Year Place Current Known Language present in Minor of of residence languages recording consultant birth birth name Maggy 1962 Abar Abar Abar (5), Abar Buu (5), KPAAM_CAM_RO_MUL Mundabli TILINGUALISM_JANET (4), CPE _ABARHOSPITAL_2015_ (5), Ngun 11_25B (3) Mr Botan 1986 Mufu Mufu Mufu, Mufu Bodwin Mundabli, KPAAM_CAM_RO_MUL Mashi, TILINGUALISM_KULO_ French, 2015_11_22 CPE, English Pa Nkole 1940 Buu Buu Buu 5, Buu Simon Fang 5, KPAAM_CAM_RO_MUL Missong TILINGUALISM_NTO_A 3, CPE 5 BARMARKETDAY_2015 -12_02 Nkangkolo 1976 Buu Buu Buu 5, No Fang 5, Kung 3, Abar 5, CPE 5, English 3 Pa Linus 1949 Abar Abar Abar 5, CPE Mashi 5, KPAAM_CAM_RO_MUL Missong TILINGUALISM_JANET 5, English _ABARHOSPITAL_2015_ 162

3, CPE 5 11_25B Mbang 1967 Mufu Buu Buu 5, No Frida Mufu 5, Mundabli, 5 CPE 5, Fang 5, Missong 5 Pa Shadrack 1967 Buu Buu Buu 5, No Mufu 5, Mundabli 5, Fang 5, Missong 5, Abar 5, CPE 5, Munken Nkam Elvis 1985 Miss Missong Missong Missong ong 5, CPE 5, KPAAM_CAM_RO_MUL French 3, TILINGUALISM_NTO_M English 3, ISSONGPALACE_2015_1 2_04 Mama Tuna 1975 Mufu Buu Buu 5, Buu Mufu 5, KPAAM_CAM_RO_MUL CPE 5 TILINGUALISM_JANET _BUUFANG_2015_11_22 Mr Yello 1965 Abar Abar Abar 5, CPE and Abar Paul CPE 5, KPAAM_CAM_RO_MUL English 3 TILINGUALISM_JANET _ABARHOSPITAL_2015_ 11_25B 1976 Munk Munken Munken 5, Missong and Munken en KPAAM_CAM_RO_MUL Missong TILINGUALISM_NTO_M 5, CPE 5, ISSONGPALACE_2015_1 English 3, 2_04 French 3, Abar 5 163

Joshua Buu Buu Buu 5, Buu KPAAM_CAM_RO_MUL Mundabli TILINGUALISM_KULO_ 5, Abar 5, 2015_11_22 CPE 5 Luke Buu Buu Buu 5, Buu KPAAM_CAM_RO_MUL Fang 5, TILINGUALISM_KULO_ Mufu 5, 2015_11_22 Missong, English 4 Jumia Buu Buu Buu 5, Buu and CPE KPAAM_CAM_RO_MUL Abar 5, TILINGUALISM_KULO_ CPE 5, 2015_11_22 English 4, Frankline Mund Mundabli Mundabli abli 5, CPE 5 KPAAM_CAM_RO_MUL TILINGUALISM_KULO_ ABARBASE-2016-05-18 Mbang Mund Abar Abar 5, Mundabli and CPE Frida abli Mundabli KPAAM_CAM_RO_MUL 5, TILINGUALISM_MUND ABLI_ABAR_KULO_AB AR_2016_05_31 Marcus

Conclusion to chapter five

This chapter has presented information on the linguistic lifestyle of some extraordinary individuals who allowed us to tag along with them and ask questions whenever necessary. More information on the linguistic lifestyle of 164 these individuals including the number of languages they speak can be found in the online archives, along with the rest of the data.

165

CHAPTER SIX

PARTICIPANTS’ DAILY ACTIVITIES (NATURAL INTERACTIONS)

6.1 Introduction

In this chapter we present the first part of the findings revealed by our data; we move from self-reported information to the natural interactions of consultants as we examine further the multilingual situation of Lower Fungom. Given the highly multilingual state of individuals in Lower Fungom and given that language choice is hardly ever haphazard, this part of the study discusses how rural multilinguals in Lower Fungom use their linguistic repertoire daily. The reasons for their language choices are rather discussed in chapter eight. Here, our preoccupation is providing an elaborate summary of our entire data without the analysis of it. This part presents data to determine the veracity of these multilingual individuals' who claim the ability to speak between six to eleven languages, plus five dialects, and whether they indeed use multiple languages or one language in their daily interactions.

6.2 Presentation of findings

6.2.1 Kulo Rene

In this section we will examine the different sources of our data and then present the data that we collected from these sources. The analysis of the data is done in 166 chapters seven and eight. This section discusses data obtained from one of our main language consultants. He is Kulo. Recall from chapter five, section 5.2.3 that Kulo is a 43 year old male from Abar. He speaks all languages of Lower

Fungom and works as a nurse at the lone health centre in Lower Fungom. For details see chapter five section 5.2.3. Section 6.2.1.1 below presents a transcript of Kulo’s daily activities as he puts his repertoire to use. It is based on our observations while section 6.2.1.2 presents data from recordings.

6.2.1.1 Data obtained through observation: languages used by Kulo Rene

During our second trip, we started working with Kulo the day after we arrived in

Lower Fungom. He wore our microphone and audio recorder as we travelled from Abar for the ‘Pig for Pikin’ meeting (see chapter four) with the entire

Lower Fungom in Fang, on 22 November 2015, passing through Buu. We felt that it would be absolutely important for Kulo to wear the recorder that day, as he had the possibility of meeting many more people on such a day than on an ordinary day. He had a better chance to meet and interact with several more people from the different villages of Lower Fungom. On this day, Kulo wore the recorder for about three hours. In Fang, we observed him after retrieving the recorder from him and found him discussing in CPE with Yamuo Francis, whose parents and grandparents are all from Mundabli. They were discussing in

CPE in the midst of a mixed group of people who had gathered for the meeting, 167 but the pair later switched to Mundabli, probably to discuss something more personal.

A similar incident happened between Kulo and one of his very close friends,

Jumia from Buu. Jumia’s father was Buu while his mother was Kung. Jumia reportedly spoke Abar, Buu, Kung and Ajumbu. However, whenever he met

Kulo, he almost always used Buu. We interviewed him and found out why he preferred using Buu. These discussions are presented in detail in chapter eight.

Later in our trip, we found out that Jumia was paired with Kulo and sent to

Nkang to administer vaccines against Measles and Rubella. While the pair was still in Abar and while we were charging our laptops at a barber’s shop, they kept us company and conversed in CPE the whole time.

We can make the following remarks from our observations and interviews with

Kulo. Kulo was in the habit of using Buu when he was sad or very happy; he said that Buu was his first language. He also kept a habit of soliloquizing in

Buu. Kulo told us that he enjoyed using Buu; that Buu was the first language he learned, and it was the language with which he grew up. His mother used

Mundabli with him, but his father and friends used Buu. Even many years after their father was made chief in Abar, and after he had passed away, Kulo and his siblings continued to use Buu whenever they met. However, Kulo used Abar 168 with his children. He told us that although his children could understand some of the languages of Lower Fungom, they could not speak them fluently. They were well-versed only in Abar, English and CPE. He did not use CPE with them very often because he preferred for them to perfect their knowledge of English since they were all students. However, he used CPE with the children in the presence of strangers from outside Lower Fungom.

During our third trip, we worked with Kulo for twenty days, mostly observing him without recording, usually from 7:30 am to midday; we tagged along as he went about the village. We would talk together when we were alone and he would speak to others if we went to join them or if they joined us. On one of such days, at about 10am, a corpse was brought to Abar from the hospital in

Wum (the capital of Menchum division in the North West Region of

Cameroon). We went to the burial service as it was one of his relatives (from his wife’s family). We spent about an hour there and then left back for our abode.

As we left, we asked Kulo the languages he used there and he said he only used

Abar. We pointed to a few persons with whom he spoke and asked him the language he used with them just for confirmation and he said Abar. It was a mixed crowd and he spoke to people from Abar, Missong, Buu and Ngun who were already present at that time.

169

On our way back, he even met one father from Munken who was going to the same burial, Pa Obong George. Kulo spoke Abar to him although he (Kulo) is fluent in Munken, and Pa Obang responded in Munken. Both men conversed each in his mother tongue until they separated.

We went to Buu and met Nkangkolo Isaac from Buu; Isaac was using Buu as he was working with Doriane (one of the other researchers in Lower Fungom).

Kulo spoke with Isaac in Buu and Isaac responded in Buu. Later that day, as we were just reentering Abar, we stopped at a drinking parlour where Kulo greeted people and spoke in Abar, although Isaac Nkangkolo was present. Isaac responded in Buu. The vendor was a lady from Mekaf (Mekaf is a village close to Lower Fungom, it is still actually in Fungom); she did not understand the languages of Lower Fungom as she had only been in Lower Fungom for about five months. She spoke in Cameroonian Pidgin English to the men.

The following day, we observed Kulo from 3pm till 7pm. He kept using only

Abar with Abar indigenes who came to offer their sympathies to our recently bereaved landlord. We continued observing Kulo; we even went together to his farm. On our way there, we ran into some young boys from Koshin with whom

Kulo spoke Koshin and he explained that the boys do not understand Abar. We were joined by a gentleman from Mundabli; he joined our discussion in CPE but

Kulo later switched to Mundabli. 170

In the farm, Kulo heard some noise from a neighbouring farm. He called out and greeted in Abar as he was sure that it was an Abar indigene. Later that day we went to observe Kulo again, when he was at his job site. He had gone to help a lady from Nkang (Nkang is still in Menchum division but it is not part of Lower

Fungom) in Furu Awa, who had given birth. While we were there, he kept using only CPE. In Nkang it was Mashi that was reportedly used as a mother tongue, although CPE iwas used most of the time because of the presence of a lot of strangers such as the Aku (a subgroup under mainstream Mbororo, a grazing community in the North West Region of Cameroon. They are a small subgroup of the Fulani ethnic group and are predominantly found in Menchum division) in the region. Still on that day, he met some students from Koshin travelling to

Zhoa (also in the Menchum division, and the headquarters of the Fungom subdivision) for the 20th May march past. He spoke to them in CPE, and so we asked him why he spoke Koshin in the morning to Koshin boys but he was that evening using CPE.

He also ran into an Abar indigene carrying wood. They started a discussion in

CPE but Kulo switched again to Abar because he needed a favour once more: he was requesting palm wine. We ran into Mama Chu, who invited us to her house to share a meal. There, Kulo attended to a lady (Lovelyn from Abar) whose baby was sick there and they conversed in CPE. Kulo told us that the girl was born in 171

Lower Fungom but had travelled to Bafoussam, in the Western part of the country, for a good number of years. As we left Mama Chu’s house, we ran into a lady, Mami Uji from Fang. She initiated a conversation in Fang and Kulo responded in Fang. They were talking about Njangi. Shortly after this, he met one young girl from Koshin, he asked the girl to go and wait for him in his house. The girl is a secondary school student and Kulo typically uses English with students but his time he used Koshin.

As we stood in front of our house a very interesting thing happened. A gentleman was passing by, Pa Akong Julius from Abar. Kulo perceived Pa

Akong while he was still at a distance and while he approached us; Kulo told him that he was walking exactly like one of the prominent persons in Lower

Fungom who had been a parliamentarian (1973-1978) in Cameroon – the late

Hon. Joseph Che Nkangkolo from Buu. Kulo initiated the conversation in Abar but as he was mimicking Pa Nkangkolo, Pa Akong switched to Buu and explained that the reason for his walking like the late honourable is because he was already getting as old as Pa Nkangkolo himself. Kulo joined him in Buu and they kept discussing the subject and mimicking the former Member of

Parliament. The conversation finally ended in Buu. Worthy of note is that Pa

Akong’s parents are all from Abar, and his father’s mother is from Munken. He 172 has no relations with Buu at all. Kulo says that Buu and Abar are very similar, so most people who speak Buu understand Abar and vice versa, even kids.

Later that day, Kulo spoke to a young boy from Missong whose name he could not remember at that time. He spoke in Abar and the boy responded in Missong.

Kulo also made some sentences in Aku (the language of the grazing community discussed above). He was speaking to one of them who was a shop owner in

Abar. Kulo reported using Aku with this person because of their friendship; he was trying to learn the language of one of his friends.

We also observed that Kulo kept using Abar with his own children who lived just behind our house. He also used Abar with Mami Maggy from Abar, who stopped by to see Mr. Abang Gabriel, the then chief of the health centre, who lived temporarily in a house just behind our house.

Still in the morning hours of this day, Mami Mbang Frida from Mundabli passed through our house while we were standing outside. Kulo started speaking to her in Mundabli and she responded in Mundabli. We were inspired to make a video of this event.

Just like in the recording on the 24th (see chapter seven) in the market, Kulo met

Madam N Janet from Ngun. They met in Abar as Madam N Janet was returning 173 from her Njangi meeting. Kulo conversed with her in Abar and Madam N Janet responded in Ngun.

The following day we continued observing Kulo. We went to Missong with him and he spoke in Abar to everyone he met. He also spoke to Boniface in Abar and

Boniface kept responding in Missong. The next day we observed Kulo using

Abar almost throughout as he was in Abar. He spoke in Abar to his colleague

Emmanuel, and to the guy in whose drinking parlour we were charging our computers and phones, Mathiew from Abar. Back in Abar, he spoke in Abar to

Mami Asangbe from Mufu and she responded in Mufu because she knew that

Kulo understood Mufu.

May 26, 2015 was market day in Abar. On such a day we would rather observe the consultants than make them sit down to listen to recordings. Market days were festive days; there was a lot of drinking and meeting of friends from other villages. Everyone’s ultimate aim on such a day was to eat, drink and be merry, not to work. If you attempted to work with them early in the morning, they would be very apprehensive, and in the afternoon they would be drunk or at least tipsy. We met Boah Stephen in the market. Kulo started speaking to him in

CPE then switched to Koshin.

174

Kulo spoke CPE to some youngsters and English to students. Then he also began a conversation with Mami Mbande Ruphina from Fang who lived in

Abar, she was married to someone from Meta (a distant village from Lower

Fungom) but was widowed at the time. Kulo discussed with her in Fang but ended the discussion in Fang. Kulo also used Kung with Pa Muluh Patrick from

Kung. Pa Muluh lived in Kung and Kulo did not know his wife. That same evening, we accompanied Kulo to a drinking spot, at Mami Mbang Frida’s house where she sold spirits. As she saw Kulo, she started using Mundabli and

Kulo joined her in Mundabli.

The next day we still observed Kulo as we were with him in Missong. He still spoke only in Abar, even when he was spoken to in Missong. In the house where we charged our electronics, the inhabitants were from Missong (both parents);

Kulo still spoke to them in Abar and they responded in Missong. We went to

Mashi with Kulo, and accompanied him there as he had to administer EPI

(Expanded Programme on Immunization) to pregnant women and children up to nine years old. Those concerned from Mundabli, Mufu and Mashi assembled in

Mashi. Kulo asserted that he didn’t use Abar as they did not understand Abar.

He mostly used Mundabli as the bulk of the populace was from Mufu and

Mundabli. With those from Mashi he used CPE. Kulo reported that he did not like using Mashi; he explained that one of his step mothers came from Mashi 175 and they did not get along well. When he was spoken to in Mashi by an elderly person, he typically would say a few phrases in Mashi then change to CPE. On this day only young girls came by “as these days”, he said, “only young girls are pregnant”.

Kulo always spoke in Abar to his children. We asked him why he did not use

Buu with them. Kulo said that he was truly not related to Buu. It was Mundabli which he preferred to teach them. However, during recordings and observations,

Kulo and his children only used Abar. When we asked Kulo what language he would use if he was speaking Kung (which he did not master) and forgot a word, he said that he would use CPE. He would rather use CPE and be told the word in

Kung. He said that if he had to mimic or if he had to say a joke which somehow had its origin from Munken or Biya, he would use these languages. He would not, however, leave Abar to use these languages because understanding was not impeded. We also observed that Kulo spoke in English to students whenever he met them, especially when they came to his jobsite. As we would have realised,

Kulo made use of eight languages: English, Kung, Fang, Mashi, CPE, Mundabli, koshin and Buu during our observations.

6.2.1.2 Data obtained through recordings

We made two audio recordings of 3 hrs 30 minutes each and three videos with

Kulo. One of the recordings was on a market day in Abar and the second was in 176

Fang from Abar through Buu. In the ‘market day’ recording, Kulo used a total of nine languages: Abar, CPE, Buu, Mundabli, Mashi, French, English, Missong and Ngun. This is the same number of languages he used in the second audio recording; he used CPE, Mundabli, Buu, Fang, Missong, Mufu, Abar, English and Koshin.

A video recording was made by one of our guides, Tsongham Nelson, at our base in Abar, wherein Kulo was discussing with some individuals from Ngun and Mundabli who stopped by to greet us. With the people from Ngun, Kulo used Abar, while with Francis from Mundabli, Kulo used Mundabli. The second video was taken at a drinking spot in Abar on a ‘Country Sunday’ (a traditional holiday). In this video, Kulo is conversing in Abar while the lady selling the corn beer is talking in her mother tongue, Munken. They are later joined by a small number of Abar men who all joined the discussion in Abar and Kulo converses in Abar with them.

In sum, from our intense observation and from the recordings, it is safe to conclude that Kulo is indeed very multilingual. We have been witnesses to his using: Abar, Buu, Ngun, Missong, Mashi, Kung, Fang, English, French, CPE,

Koshin, Mundabli and Mufu. Based on our research, Kulo did not use Ajumbu,

Kung and Munken.

177

6.2.2 Mbang Janet

Our second language consultant was Mme Mbang Janet, a 37 year old lady from Buu. As we already spelled out in chapter five section 5.2.1, she lay claim to a linguistic repertoire of at least ten languages and five dialects. As we did with Kulo above, we present an insider’s evidence of Janet’s claimed linguistic repertoire. Section 6.2.2.1 presents data based on observation and section 6.2.2.2 presents data based on recordings.

6.2.2.1 Data based on observation

We kept record of her daily linguistic habit through audio and video recordings and observation. We had two audio recordings of 3 hrs 30 minutes and eight videos of 10, 17, 4, 4, 10, 10, 4 and 20 minutes. The recordings had to be as short as possible because we only video recorded when she was conversing and sometimes, 4 minutes were all it took. We observed her for seven hours (9am -

1pm and 3pm-6:30pm) on our 6th day of work in Lower Fungom; five hours

(9am-2pm) on our 7th day, and five hours (8:30am-1:30) on our 11th day of work. During our observations, we listened and asked questions such as “With whom were you talking”, “what language did you just use” (in case we could not identify it), and “why did you use or not use this other language”. We listened to her speak Fang, Koshin, Missong, Buu, Abar, Mundabli, Mufu, Esu,

English, and CPE more or less effortlessly. We also listened to her struggle with 178

French. We played a recording in Ngun in which the other consultant Kulo was involved, and she was able to interpret the conversation faultlessly (one of our judges confirmed this).

We noticed that Janet mostly approached people in their languages, even more than they did to her. She switched a lot to other people’s languages, saying that she had tried most of them, but they did not understand Abar or Buu perfectly and she would end up repeating herself too many times for them to understand, or will have to translate what she said in their own language. However, we noticed that even when it involved intelligible languages, she would still switch.

Of the three consultants, she was the only one who acted in this manner. Kulo would not switch to Missong or Munken; he would simply carry on with his preferred Abar because he assumed that they understood him. Boniface would not switch to Abar either for the same reason. As we can see, Janet was the only female among our consultants. One wonders if this behaviour was gender based.

Our research did not gather any data to verify this. We hope that some future quantitative investigation will examine this as a variable. Janet conversed in

Missong at length with Etung Prince and his brother Etung Divine who joined in later. Janet spoke Buu to all her kin and sometimes to people who were not from

Buu but who could understand Buu; for example, she spoke Buu to the night watch of the health centre in Abar, who responded in Abar. 179

6.2.2.2 Data obtained through recordings

In the recording done on November 22, during the ‘Pig for Pikin’ meeting in which Janet wore our microphone from Buu to Fang, she used nine languages:

Missong, Buu, Fang, Koshin, Mufu, Abar, CPE, English and French, within a period of three hours thirty minutes.

On November 25, Janet wore our microphone as she was working at that period as a medical assistant in a vaccination campaign. For about three hours thirty minutes, Janet used ten languages: Abar, CPE, Buu, Fang, Missong, Munken,

Mufu, Wey, Koshin and Esu.

In the video at her Njangi group she used Ngun, Mufu, CPE and Abar. We also recorded her at the hospital on several occasions, during which she used Buu,

Abar, Missong, Koshin, English, CPE, Fang and Munken.

Mme Mbang Janet as we mentioned earlier, seemed to enjoy talking to people in their own language. We made a video of Janet selling clothes to an Aku customer. This Aku lady spoke a variety of CPE spoken by the Mbororos, which was not too distant from the regular variety. This is the variety that her customer spoke and understood very well. Janet, without delay modified her speech to resemble that of the Aku lady.

180

During our third trip to Lower Fungom, Janet travelled to her husband’s village

Esu to give birth. We naturally had to follow her there to observe her linguistic habits. There, away from Lower Fungom, Janet still chose her tongues conscientiously and carefully. She always began a conversation in Esu which she did not master at all. She would say a few words, usually greetings then continue in CPE.

In sum, Janet used Buu, Abar, Fang, Mufu, CPE, English, French, Koshin,

Missong, Esu, Ngun, Mundabli and Munken to our hearing. We did not hear her use Kung, Ajumbu or Biya. She told us at one point that she did not speak

Ajumbu, and that she would talk in CPE when she met someone from Ajumbu.

6.2.3 Nto Boniface

Our third consultant was Mr Nto Boniface, a 38 year old male from Missong. He was a hunter by profession and claimed a multilingual repertoire of at least seven languages and three dialects: Biya, Buu, Missong, Abar, Mashi,

Mundabli, CPE, English, French and Munken. For his detail profile, see chapter five, section 5.2.2. As we have done for the preceding two consultants, we will present data from our observation of the activities of Boniface and from our video and audio recordings.

181

6.2.3.1 Data gathered through observation

Like with the other two consultants, we also spent time observing Nto. Nto, from our observation, almost always spoke Missong. He more often than not conversed in Missong with people from Abar who could understand him. We noticed that he even spoke in Missong with the sub-chief of Mufu who lived in

Abar. The sub-chief, Mr. Nfua, spoke to him in Mufu but he replied in Missong.

During one of the days we were observing him, while Boniface was on his way from Abar to his home in Missong, he met someone from Mufu commonly referred to as ‘Ça Passe’, they walked together towards Missong and conversed in Missong and Mufu, each one maintaining his language most of the time. It was ‘Ça Passe’ who joined him in Missong at the initial part of the conversation.

Boniface also met Pa Gabriel from Ngun, who lived in Missong with his relatives; Boniface spoke to him in Missong. We observed him for more than one hour, walking with him from Abar to his village Missong. We noticed that he used only Missong and CPE.

6.2.3.2 Data obtained through recordings

In a recording done later the same day, December 3rd, he was found speaking

Mashi a whole lot. This could be because his wife’s grandparents were from

Mashi. Boniface, in this one hour thirty-five-minutes recording in his village, used five languages: Missong, Buu, Mundabli, CPE and Mashi.

182

In a recording done on December 2nd, on an Abar market day, Boniface used seven languages: Missong, Buu, Mundabli, Abar, Biya, Mashi and CPE. In another recording where Boniface was in his Njangi group, comprising people from Abar and Munken, Boniface used CPE and Missong throughout. This was the same thing that happened in a recording done at the palace in the presence of his father, the Fon of Missong. In this video, while Boniface was with his tribesmen and his father, the Fon at the palace, he used only his mother tongue, but while he walked around the village, he also spoke Biya to a certain Pa Jo from Biya and Mundabli to a few indigenes from Mundabli who had come for a visit. Boniface used CPE with the youths in mixed gatherings and with strangers. From our observations and recordings, Boniface used Biya, Missong,

CPE, Buu, Mundabli, English, Mashi and Munken.

All in all, be it through recordings or direct observation, our data indicates that all three consultants were indeed multilingual and that multilingualism in Lower

Fungom was a daily reality. All three consultants selectively drew on the languages in their linguistic repertoires as dictated by their intentions, their needs, participants and several other factors. Each one of them actively used several languages every day, depending on those they came across on a particular day.

183

However, as a matter of fact, all three consultants made it a priority to use their

L1; the use of their L1 was truly dominant in their daily conversations.

6.3 Characterization of the three consultants

This study being qualitative in nature used research methods that uphold depth and nuances. Instead of editing these elements out in search of a general picture, we made them part of the analysis. Here is a characterization of our three major consultants to reflect their varied nature. Although these individuals use their linguistic repertoires differently, there is a pattern.

Mme Mbang Janet is at one extreme: she does the most movement between languages, whether the interlocutor understood hers or not. This is true if we consider the fact that she would leave Abar to use Missong, which the two other consultants would rarely do since people from Missong understood Abar. She would even go as far as modifying her CPE to resemble that of her Aku interlocutor. In addition, even when she did not master a language, such as her husband’s language, she would start her conversation in Esu rather than just resort to CPE.

Mr Kulo Rene left his language quite often as well, but not as often as Janet. He would rarely leave his language for another if the interlocutor understood his; he 184 would most often do so in cases where family ties obliged him. However, he did leave his language just to mimic another, for humour and just for interjections.

These are some of the things Boniface would hardly do. He was at the other extreme of the continuum. In a recording that lasted an hour, he used only

Missong and CPE, although he met people from different villages like Mufu.

What is clear here is that, these consultants all left their languages (their L1) from time to time albeit in varying degrees.

Conclusion to chapter six

This chapter has discussed one of the research questions of this thesis: it has examined how typical rural multilinguals use their linguistic repertoires daily.

Data gathered through observations and recordings of natural interactions have been used to illustrate how the consultants lead a multilingual lifestyle indeed.

The recordings and observations revealed accurately when one code was preferred over another and the competences of these multilinguals. In the chapter that follows, we discuss the kinds of linguistic activities attested in

Lower Fungom.

185

CHAPTER SEVEN

MULTILINGUALISM WITHOUT CODE SWITCHING?

7.1 Introduction

Wolff’s (1991:4) declaration that code switching (CS) is a characteristic concomitant to stable multilingualism is indeed quite popular. This characteristic is not an exceptional one but a regular form of language use in linguistically diverse settings. As we examine how multilingualism operates in rural Africa, we explore the possibility that Lower Fungom could be an area of linguistic diversity devoid of code switching. In this chapter, we examine instances of language use in Lower Fungom in comparison to instances of language use elsewhere in Africa and beyond. We begin with a recap of what we consider in this study as code switching. A more detailed discussion of codeswitching is found in chapter two.

7.2 Definitional issues

Auer’s (1999) discusses a continuum of language alternation phenomena. This continuum ranges from code switching, through code mixing to fused lects, with code switching and fused lects marking the polar extremes of the continuum and code mixing the point in between.

186

According to Myers-Scotton (1993: 47), code switching refers to “the use of two or more languages in the same conversation, usually within the same conversational turn, or even the same sentence of that turn.” It is worth emphasizing that she treats code switching and mixing as a continuum, with the two concepts only distinguished by the degree or position (inter-sentential or intra-sentential) of the alternating varieties.

One term to be remembered here in code switching is “alternating”, or a back and forth movement in language use. In addition, Poplack (1980) says that for

“another language” to be considered code switched, there must be an asymmetrical relationship in which one language dominates (matrix language) and items from the other language (embedded language) are inserted. Again, she stresses that the languages are said to alternate.

Franceschni (1998) asserts that CS is a language universal behaviour of multilingual speakers. However, this may not be the case with the very multilingual individuals of Lower Fungom. Consequently, in Lower Fungom, we talk about language choices instead of CS. The difference here lies in words that characterize CS such as “back and forth” movement from one code to another or alternating use. With language choice, there is no alternation; not inter- nor intra-sentential. As a matter of fact, language A gets completely dumped for language B. 187

Before we discuss further the kind of communicative practices attested in Lower

Fungom, the following section provides a review of the practices found in other parts of the world, and also in rural parts of Africa such as the Lower

Casamance in Senegal.

7.3 Code switching around the world

Outside of Africa, Poplack (1980) studies the speech of some Puerto Rican residents of a stable bilingual community in the United States of America, exhibiting varying degrees of bilingual ability. These bilinguals were able to switch from their L1, Spanish, to English. Details of her work will not be discussed here but emphasis is laid on the ability of balanced bilingual (those who know two languages equally well) to use two languages in communication.

Our attention is also drawn to (and remain in our minds throughout this section) the title of her work: “Sometimes I’ll start a sentence in Spanish y termino en español’: towards a typology of code-switching.”

In Africa, Myers-Scotton (1993) examines the social motivations for code switching in the urban cities of Nairobi, Kenya and in Harare in Zimbabwe. She notes that in urban Africa, knowing and using more than one language every day is commonplace. In one of her examples, two teenage boys from different ethnic groups (Kalenjin and Kikuyu) are chatting after school in Nairobi, Kenya. She 188 says that English and Swahili are the official languages of multilingual Kenya.

Myers-Scotton insists that at home, Nairobi schoolboys largely speak their own ethnic group languages to parents and elderly relatives. At school, English is the medium of instruction. However, when these students are with peers, their interaction is most otfen a combination of Swahili and English.

Myers-Scotton (1993) further presents an example from Zimbabwe, where the interaction is in Shona and English. Here, a female nurse is talking with a male teacher in Harare, Zimbabwe, about people who buy cars out of the country and sell them soon afterwards. Both are in their twenties. Myers-Scotton says that while English, Ndebele and Shona are official languages of Zimbabwe, Shona is the main language heard in Harare. She holds that such educated young people typically do not use Shona alone when with peers, even though this is their first language. They instead speak a variety consisting of Shona and English even in informal conversations.

In Senegal, as part of a broader discussion of multilingualism in Lower

Casamance, Cobbinah, Hantgan, Lupke and Watson (2017) briefly present conversational data collected as part of documentary work on the languages of the region. They argue that a key point that emerges from their data analysis is that the ways different languages are used do not allow for any straightforward categorization. In the Lower Casamance region, there are instances of extended 189 single-language use (no code switching), there are multiple language use where the languages are nevertheless used relatively “discretely” (inter-sentential but not intra-sentential code switching) and then there is intense code-switching, within sentences and even words (see Cobbinah et al (2017) for details).

Extended single-language use, in this part of Senegal, might take place among individuals from the same household, for instance, while more extreme instances of codeswitching might take place among speakers who are very familiar with each other and have clear knowledge of each other’s linguistic repertoires. They provide an example that show intense code switching, and the four languages involved are the Gubëeher variety of Bainounk, Wolof, Kujireray and French.

7.4 Code Choice in Lower Fungom

The linguistic situation of Lower Fungom superficially resembles Lower

Casamance; a large number of languages are used in a relatively small area, and residents tend to be highly multilingual. However, the use of local languages in conversation is quite different from what is found in Lower Casamance. In particular, the presence of individuals with large language repertoires does not seem to correspond with the lack of extensive code switching in interaction. In

Lower Fungom, there is a strong preference towards single-language interactions. This is reflected not only in ideological orientations, as revealed through speakers’ reports on how languages should be used, but also in 190 observational data. Intra-sentential codeswitching in Lower Fungom appears to be largely limited to the use of words from Cameroonian Pidgin English.

Stretches of discourses are clearly identifiable as belonging to a single local language. Deviations from this tendency appear to be rare but can be used to achieve specific pragmatic effects, as seen in the extract below.

N (Nkole) is a sixty-year-old man from the village of Buu who is married to B’s older sister. B (Boniface) is thirty-eight years old and is from Missong, where he is the son of the chief. Worthy of note too is that B’s mother is from N’s village.

They each speak Buu, Missong, and Cameroonian Pidgin English, among other languages. On a market day in the centrally-located village of Abar, B enters a drinking establishment and comes across N and they have the following conversation.

(Buu, Missong, Cameroonian Pidgin English)

B: ndɛ...a ye ne...be de bɛ Uncle... How are you? Isn’t there kola?

N: nfo question wa tumɛ You had asked me before

B: a fɛ ŋkwo mi tumɛ be? a fe so hɛnɛ. n du we kwe fa mi ɛmu be… What is it about? You remembered. I asked you to buy Kola for me… [noise]

B: aI sa n sɛ keke wu! 191

No! Don’t flatter me!

N: a kɛ ya lɛ dzɛŋ? ŋ wu yɛ bu ka follow wa ton Did you come up to Fang? I heard that you were chased there

B: ŋ ka follow be mi? ŋge du ye a ka de mi. e bɛ kɛhɛ manto Chased away? It was not me, it was Manto

N: a ke wou ye kem djo uwa de? Are you all listening to what I am saying?

B: ben wou gin ta? What should we listen to?

N : a gɛ kɛ kɛ ta ? So, where did you go?

B: offlicense wone mi wo me ma bahɛ ti ma I reached here and saw you in this off license.

[... 30 secs pause ]

N: bi kie lahe You are still a child.

[After some grumbling, N stops speaking to B who then leaves.]

Worthy of note is that consistent with broader patterns of the local culture, it is expected that a junior person should accommodate (Giles, 1982) a senior person’s primary linguistic identity to the extent possible especially if they constitute members of a family of some sort as is the case of Nkole and

Boniface seen above. In choosing to use Buu, this is exactly what B (junior) 192 initially does in his interactions with N (senior). However, when N insists on reprimanding B, the latter first did his best to argue against N’s insinuation and contain himself using the Buu language, but then shifted to his own primary language Missong, when he had had enough. This had the effect of putting an end to the interaction. What is crucial to note here is that the change in code did not go unnoticed. By-standers and both interlocutors understood it in the same way as an act of social distancing. Converging use of language (as seen in chapter two) creates confidence and togetherness whereas divergence in language use tends to be ambivalent and dissociative. Moreover, B’s act was effective only because the interlocutors share an understanding of the significance of the pool of communicative resources from which language choice is drawn. Heller (1988:7) asserts that conventions must be shared for their violation to have meaning.

The violation here does not lie in N's intonation as that did not change. It neither lay in the literal act of language choice itself. Judging from one of the language ideologies of Lower Fungom, it is most probable that the violation came from his choice of Missong.

Language choices are meant to index a certain population and, by association, some set of distinctive features. This would be an instance of a second order index following Silverstein’s (2003) terminology seen in chapter two. The 193 perception of a linguistic feature distinguishing a given population is taken to index a speaker’s membership in that population. (See also Johnstone &

Kiesling (2008).

By choosing to use Missong, B chooses to index or represent himself as a member of the village of Missong, a community in which he is the chief’s son and where N, being from the village of Buu and having only in-laws in Missong, does not have the right to criticize his actions. Thus, choices made by the two interactants rely on an ideology where the local lects of Lower Fungom index not “identity” as “categorical identification” but, rather, as “relational identification” and whose interpretation depends on the specific position that an individual occupies within the relational “web” indexed by a lect (Di Carlo et al,

2016: 31).

In Lower Fungom, the beliefs and principles of maintaining one language for one village fosters the use of a stretch of single languages during conversations than the use of several of these local languages within the sentence. Research in the area (see Agwara 2013) reports very high rates of self-reported multilingualism. In her study, typically, men and women reported to speaking 6

'languages'. As a matter of fact, nobody is monolingual in Lower Fungom. At the very least, everyone is competent in a local language plus CPE.

194

Remarkably, although in both Lower Fungom and Lower Casamance local languages are social strategies used to maximize alliances, Lower Fungom practices a stricter language usage dynamics in marking alliances. As a matter of fact, with the presence of individuals with large language repertoires, which they use habitually, one would presuppose instances of CS in the conversations of inhabitants of Lower Fungom. It would be typically expected that when multilinguals hold a conversation, multiple languages could be used. This is what obtains in Lower Casamance but not in Lower Fungom.

Deviations from this tendency is vehemently frowned upon; it actually plainly marks a crack in alliance (albeit temporarily). Consider this example below where Kulo (Abar) was on his way to another village (Fang) but stopped mid- way (in Buu) to visit his friend Pa Joseph (from another village of Lower

Fungom- Buu) and they engaged in a discussion. Kulo chose to use Buu,

Mundabli, CPE and Abar depending on who his interlocutor was. He changed these languages almost immediately, in the same discussion, effortlessly, without mixing them. Indeed, if Kulo had inserted words from his father's language -Abar in this discussion, communication with these participants would not have been impeded. If he had used only Abar or his mother's language-

Mundabli, he would have communicated with all three successfully as these participants all shared Abar, Buu, CPE and Mundabli in the linguistic repertoire. 195

Kulo was mobile throughout this recording, as such, he greeted a few passers-by on his way to Pa Jo's house and to other neighbouring locations.

The languages used here are Buu, Mundabli, Abar and CPE. Participants include K (Kulo), PJ (Pa Jo), PJW (Pa Jo's Wife). PB1 (Passer-by 1), PB2

(Passer-by 2), PB 3 (Passer-by 3), J (Jumia ), PB4 (Passer-by 4), PB5 (Passer- by 5)

Scene one

(Pa Jo's house, the conversation begins as Kulo spots Pa Jo standing outside his house. He calls him and begins talking while still a few metres away from him.

While he is approaching Pa Jo, Kulo also spots a friend- Kwembe. Kulo calls him and speaks to him, greets some mothers sited near Pa Jo's house then he finally reaches Pa Jo's house).

1. k-PJ Pa Jo! ayeneh eyikehe ayikebenesanto yebene Free Trans Pa Jo, how are you, you are eating. what are you eating there? what?

2. K- PB1 kwembe, I di come, make I salot pa here Free Trans kwembe, I will come there, let me say hello to pa here

3. K- PB2 mami bin tʃi ta ton Free Trans good morning mothers

4. PS2-K a tʃi ta kwa Free Trans good morning 196

5. K-PJ ah! ben diə bɛn ʃʷom mena Free Trans ah! you people have cooked palm nut?

6. PJ- K a ʃʷom be amerika ŋ yi ke Free Trans It is American palm nut that I am eating (footnote, what is American palm nut)

7. K- PJ la ka nə de, ʃʷom bə amerika (….) amerika wa Free Trans Do I lack American palm nuts?

8. PJ- K mʊ kə a mɔːŋ wu Free Trans take and get the taste yourself

9. K- PJ ay! dzoŋ yen Free Trans (sarcastic) No! thank you

10.PJW-K bɛn tsɨ li a tʃɪn? Free Trans Did you guys sleep well there? (Kulo does not respond to this as he was distracted)

11. K- PJ bishie bibiami dzonkufa ideya meeting Free Trans let me follow the others to fang for a meeting

12. PJW-K ben nia nu a tʃi tʃia Free Trans Will you guys remain standing?

13. K-PJW ben feziebe nemsijo ben tchi ne ŋgoko noh Free Trans You guys said there was something, so bring it nah

14. PJW-K ay! mbu tʃi mbu Free Trans no! I just asked

15. PJ- K nam fedzuo ayena kemu Free Trans sit down and eat child a tem eke ndie eyikə nemfe zohfoh Free Trans you are boasting with whom that you can feed yourself?

16.PJ-K nam fə dzo-fə ke a bɛn yi wan Free Trans sit down and eat 197

(Pa Jo is actually screaming at Kulo, asking him to sit down and eat)

17. K-PJ ke lewou lɛla malaya Free Trans stop troubling me eh

18. PJ-K andu ya yikə (in response to Kulo’s statement in line 11) Free Trans who is going up?

19. K_PJ nobo etʃi ewu no tawula Free Trans it is the whiteman over there

[For a couple of seconds later, Kulo tries to speak to Pa Joseph still in Abar but he does not respond and Kulo leaves for the palace where Pa Joseph would join him in a while. On his way, Kulo meets some people seated in front of their houses and he greets them before discussing with his friend Jumia ]

20. K- PB3: atʃita, te banu Free Trans Good morning, you people are out?

21. K-J: Denis Free Trans Denis

22. J- K: patron in charge Free Trans The most important person

23. K- J: I no see you ... ok, you be dey? Free Trans I did not see you ... OK were you there?

24. K-J: be buoh tabaŋ meh Free Trans Friend (Boh in CPE) give me cigarette

25. J-K: masa, tabaŋ bə sa ta Free Trans Man! Where is cigarette?

198

[Throughout this section which is left out in this excerpt, Kulo is discussing with

Denis in Buu and then he returns to discuss with Pa Jo who has now joined them at the palace still using Abar while Pa Joseph is using Buu]

Scene two (Palace in Buu)

Kulo and Pa Jo entered the chief’s house in Buu. They were not alone in the room as other people kept going in and out to prepare it for a meeting. Kulo sat down and Pa Jo remained standing.

26 K-PJ: Pa Jo! iyə! Free Trans Pa Jo! How are you?

27 K-PJ aboh koh daŋ koh oti corner mohngən noh Free Trans You want to sit beside me, right?

(Pa Joseph did not respond, he simply sat down next to Kulo)

28 K-J: mboh koh mbaŋ Pa Jo but ŋgaŋ Free Trans I want to greet pa Jo but no

29 J-K: nobi ye Free Trans I'm listening to you

30 K-J: notefiə ŋwo iloŋ yiə Free Trans If you feel like listening you will, that's your trouble

[Notice that below, Kulo switched from Abar to Buu to talk to some boys who were helping to bring chairs into the palace to contain the guests. The addressees did not include Pa Jo but were limited to some Buu people who were in the same room]. 199

31 K-PB4 ke be m-kpoko biə bə du lebin tu ke ta bebi noh

Free Trans If there are more chairs, just keep bringing them

32 K-PB5 dzahɛ wantou noh Free Trans You too, go that way [to a chicken]

33 K-PB5 adena tioŋ tizakɛ Free Trans So, you say people should walk around with hunger?

[Kulo ends up this scene in Buu with a chicken as his addressee; there was a

chicken chirping around Kulo, while food was being served. Kulo interpreted

the chirping as the chicken's refusal for people to sit around and eat. He did

return to Pa Joseph’s house for a short while because his (pa Jo) wife had asked

him (Kulo) to return afterwards to eat the food she was cooking when he first

visited. He talked to pa Jo’s wife in Mundabli and to Pa Jo for the rest of the

recording in Abar].

The languages Kulo used were not his first languages per se. His father spoke

Abar and his mother Mundabli, but he was born and bred in Buu. As such Pa Jo asserted (from our interview with him) that he (Kulo) was a son of Buu; that they were a family. Kulo was thus expected to linguistically accommodate Pa

Jo. According to the culture shared by Kulo, Pa Joseph and his wife, it was expected that a junior family member should accommodate a senior person’s primary linguistic identity. That is why Kulo, in the initial part of the interaction 200 chose conscientiously the languages for both his interlocutors. He did not use

Mundabli with Pa Joseph or Buu with Pa Joseph’s wife. He only switched to his primary language after his conversation with Pa Joseph became aggressive.

Worth noting, he did not cease using Pa Joseph’s wife’s primary language; only

Pa Joseph's. This act of Kulo’s disheartened Pa Joseph, communication between them was strained, with Pa Joseph refusing to be friendly with Kulo anymore.

In choosing to use Pa Joseph’s language and that of his wife, Kulo indexed himself as a member of both villages; he preserved both alliances. This is a phenomenon which is greatly upheld in Lower Fungom. This membership permits Kulo to obtain material and spiritual security from both Buu and

Mundabli and this was of utmost importance to Kulo. However, he felt that temporarily leaving one of these memberships was necessary; it expressed his emotions at the moment.

Here is another excerpt showing language use in Lower Fungom. It is a conversation carried out by one of our major consultants Janet from Buu, a trader in bush meat (from Fang) and a friend of Janet- Maureen (from Buu) who had come visiting her. Worth noting is that, with sociolinguistic interviews, their linguistic repertoires were explored. All three participants in this conversation shared similar linguistic repertoire. They all spoke or at least understand Buu, 201

Abar, Fang and CPE. However, the main speaker, Janet carefully selects a different local language for a different interlocutor.

Here, M (Mbang Janet) is having a conversation with Tr (Trader) and V

(Visitor). The languages used are: Fang (Normal), Buu (Bold), and CPE

(underlined and Italics)

1. M-T atsie na ya yan di wo Free Trans To whom do you want to give it?

2. Tr-M Pa Anthony

3. M-Tr Pa Anthony?

4. M- Tr a negbo neh mahen ne tiele Free Trans Is he your father or your in-law?

5. Tr- M a negbo Free Trans my in-law

6. M-Tr gene dzon gie me noko niam yen, ma keyen be mi no baleiyo ma sefe keyan bami Free Trans say something better about this meat, I don't have any food at home, don't do that to me.

7. Tr-M metsi anyen fien nuŋgho fo fiane shife Free Trans I want to go and sell it over there

8. M-Tr nayen fiene feku weh aŋka bene begezie niam Free Trans you want to go and sell it over there why? Are they the only people who eat meat?

9. M-Tr a ŋkike na kale kedu a ta kegie bime ya sista yami badzie tse ta nwoho ngwo gon owe Free Trans Don't you know that one day you can come here and say "Sister, give me 500 to buy catridge"

202

10. Tr-M andoua aya gimie ba ŋkantefe badjie tsa yage mi bə ŋkamtefeh be di tsa Free Trans give me with 2000 and one hundred five times (Give me 2500)

11. M-Tr nayi! nta ya tekem tefe badziefe me ne tshite gwo yagi ŋkamtefe bə di Ifeh... ma nə tʃetə gwo Free Trans I will give you 2000 and hundred two, I have bought so well (I don over buy)

12. Tr-M ememe fie bene mane tʃie ntome yohoyen wum Free Trans Someone had to even buy it at home it is just that I told them that I wanted to go sell it in Wum

13. M- Tr a ntem bagu Free Trans Did you shoot it with a gun?

14. M-Tr an tsete sefele yin abale beli yene Free Trans You scattered this part, removed it and ate with your fufu, didn't you?

15. M-V Aunty Mau a wule la kpuan (you hear the money) Free Trans Aunty Mau do you hear the amount he is asking?

16. V-M ehe wu Free Trans I heard

17. M-V ye ntə kefie be kpante kpin layane bwoeh Free Trans He says it is 2500, is that alright?

18. M- Tr sometime e denai luh? Free Trans maybe it will be bitter

19. Tr- M yi denai ilu eh Free Trans it cannot be bitter

20. Tr-M a nemen tangme be na be kezie keli eneke ta sikpin keliaŋke elu denai kpin Free Trans the time for bitter not reached... there are some small seeds in the forest which these animals eat that makes them bitter, but it is not yet time for those seeds to be produced.

203

21. M-V bimbe kabedge afake beben tintsan ke ndiemi segun bugo tugo ndin neyəŋə dey be bringam for me yesterday Free Trans Won't you give me egusi so that I can cook that cabbage of mine? It was given to me yesterday.

22. V-M die be be tsongbe noh Free Trans cook it with groundnuts, will you?

23. M-V la koŋghode yennda bə tsongbə Free Trans I do not want to see groundnuts

24. V-M dzong moadi ndjockə ayinda se niam na mi? kedzo ngamne dzeya ahe tinse Free Trans Good for you. You have even seen meat, what about me? Have I seen any?

25. M-Tr a ne tsəte fə bami Free Trans you have really dealt with me

26. M-Tr Kpo niam e ntagə (money meat e strong) Free Trans you are really expensive

27. Tr-M even tsiete nfanbe duta begwoyen bekenbe teh Free Trans Even in Fang, they could have bought it for 3000

28. M-Tr te ever gwo niam wokah, andona niam mien ne fain tinə intʃətə siasə mien way Ngianə nane 3500 a niam me tchətə zdong me ne gom kpo mo ever gwosə Free Trans If it wasn't scattered, that it still had all its part, completely, I could even buy it for 3000 or 3500, like I used to do, without a problem

29. Tr- M eni kalele na ŋkegi babale kegi tʃesi gieyan Free Trans whenever I have meat I will come and show you

30. M-Tr a gi tʃesi mien Free Trans Yes, come and show me

31. Tr- M yeŋke fu pa me Free Trans Let me go and see that Pa 204

32. M- Tr ok, ke yoŋ ne kə Free Trans Ok, thank you

The above discussion suggests language use which indexes a language ideology devoid of personal prestige but more oriented towards the indexing of affiliation with a given group. Language choices in Lower Fungom are meant to index a certain population and, by association, some set of distinctive features. This would be an instance of a second order index following Silverstein’s (2003) terminology seen in chapter two. The perception of a linguistic feature distinguishing a given population is taken to index a speaker’s membership in that population. (See more in Johnstone and Kiesling (2008). That is why mixing these languages as typical of multilingual settings would mean different things in Lower Fungom, none of which is positive.

A neutral person whom was called a judge (the regent of Buu) in this thesis, was asked to translate this recording word after word into English after Janet herself had given a free translation of the same video recording. With this exercise, the fact that there is no code mixing between languages of Lower Fungom was ascertained. Janet did not insert a word from Buu or Abar while she was discussing in Fang to the trader. She also did not insert a word in Abar while she was discussing in Buu with her friend Maureen who is half Abar, half Buu.

After this exercise with the recording, the regent of Buu was interrogated on the 205 same subject, his facial expressions and words made it clear that such linguistics activity is not done nor is it appreciated in Lower Fungom.

Crucial to our understanding of code switching in Lower Fungom is Epps

(2012). Epps examined multilingualism in the Amazon basin, one of the most linguistically diverse regions on earth, comprising some 300 indigenous languages corresponding to over 50 distinct lineages (Epps 2012:1). She asserts that the diversity of language families in Amazonia is some ten times higher than that of Eurasia and Africa, and is rivalled only by Papua New Guinea. Her study revealed that one’s identity as belonging to a particular group is not a given, it must be actively constructed and maintained by a number of ways most important of which is by how one speaks.

Kulo in the excerpt discussed above spoke in Buu throughout his stay in Buu except when Pa Joseph insisted on making him eat palmnuts. His last utterance is actually directed to a chicken that was around his seat and this utterance is still in Buu. Once more, it is clear from that excerpt that a shift in codes in

Lower Fungom is very systematic and it is not meant to demonstrate the interlocutor’s mastery or accommodation of several languages but to signal togetherness or separateness. A switch in codes detaches or disassociates a user from one group and unites them with another.

206

Epps (2012: 12) goes ahead to cite Chernela (2013: 227) who believes that for the Wanano, “since speaking-and-being … are inseparable, to speak a language not your own is to ‘become’ another... By speaking a language other than his or her ‘own’ –an alterlect –a Wanano person ‘stops being Wanano’ and ‘becomes something else’.” In Epps’ study, speaking another’s language came with a negative connotation as it meant relinguishing one's true identity and in favour of another's. This is actually the reverse in Lower Fungom. All along, when

Kulo used Buu with Pa Joseph, he did not relinquish his identity in favour of Pa

Joseph's. He did not stop being Abar only to be Buu; he was both Abar and Buu.

Kulo, like many others in Lower Fungom, through his language practices portrayed the multiple identities he embodied. When he felt offended by his uncle, he managed the situation by assuming another identity of his throughout his stay with Pa Joseph. Kulo was member of the Abar community (with a specific position given by his being a son of the former Fon of Abar) but also of the Buu comunity (being son to Pa Joseph by virtue of his place of birth) and also of the Mundabli community (being a son to a woman from Mundabli).

Whenever he used one or the other of these languages as seen in the excerpt,

Kulo was representing himself as a member of one or the other community. As a matter of fact, this is not just accommodation; it is the actualization of relational identification as Di Carlo (2016) puts it.

207

Further evidence attesting to the fact that mixing two local languages is an aberration in Lower Fungom is the situation described in chapter five where two indigenes of Abar (Kulo and Pa Goerge) talking about Pa Nkangkolo of Buu, ditched their mother tongue, Abar, completely for Buu, just in a bid to joke around and to mimick.

In addition, in an interview with Pa Pa Linus, one of the pillars of Lower

Fungom based in Abar, we were informed that individuals from Abar who were performing burial rites, and who had resorted to traditional methods for avenging deaths, used the language of the village from where the “medicine” was taken. If it were possible to start a sentence in one local language and end it in another like in bi/multilingual contexts elsewhere, reference would have been made to this “medicine” in the language from where it originated and the rest of the interaction would have remained in Abar. This was not the caseː when it got to that part, Abar was abandoned completely.

All those who were interviewed had the same opinion on this matter. They declared assuredly that code switching, in the sense discussed in this work, was seriously disliked and not likely to happen in people’s speech (at least as far as the languages of Lower Fungom are concerned). Everyone almost always used

CPE if they had to code switch. Janet asserted that they were not known to use 208 one local language and mix with another in Lower Fungom. She said that if she were struggling with a language like Biya and she forgot a word or did not know the word at all, she would substitute it only with CPE and then she would be corrected or told the word in Biya.

We also note from the studies reviewed in this thesis that in rural regions only ex-colonial languages are endowed with prestige; communities with no minority languages naturally would not rush to code switching. They would not

“sometimes … start their sentence in English and finish in Spanish” (Sometimes

I’ll start my Sentence in Spanish y termino en español) like Poplack (1980) so eloquently puts it. Egalitarian and reciprocal multilingualism as practiced in

Lower Fungom, therefore, would not breed intra-sentential or even inter- sentential code switching. Some of the pioneer scholars of code switching,

Gumperz & Hernandez (1971), wrote that code switching would typically occur whenever minority language groups come into close contact with majority language groups.

Di Carlo (2016) discusses in detail the fact that prestige, except for the colonial languages (such as English and French) is not among the main symbolic assets negotiated in the local linguistic market. He argues that this reality has tremendous consequences for the understanding of the local language ideology of this community: “Instead of the indexing of a social identity implying 209 personal prestige, what we seem to be uncovering here is suggestive of a language ideology more oriented towards the indexing of affiliation with a given group, devoid of any behavioral or moral reflexes.” (Di Carlo 2016: 83).

Conclusion

This chapter has analyzed part of the data collected in this study. We have also examined the kind of code switching found in Lower Fungom, in comparison to other multilingual rural parts of Africa like Zimbabwe, Kenya and Senegal, and beyond Africa like the bilingual Puerto Rican community in the United States of

America. We contrasted our analysis with existing data which upheld the idea that intra-sentential and inter-sentential code switching occurs in contact situations where there is a minority or a subordinate language group; it is typically absent in societies with egalitarian and reciprocal multilingualism.

Single-language use is very important in Lower Fungom; multilinguals here have not been seen to switch local languages. They use one language and when they leave it, they begin speaking exclusively in another, and this has been shown to send a strong message to the interlocutor. This is attributed to the language ideology guiding linguistic activities in the region. Indexicality as the language ideology (see section 2.3, chapter 2) handicaps code switching in this very complex multilingual community. It instead fosters instances of extended single-language use. This is what the next chapter will discuss further. The next 210 chapter looks into the reasons why these multilinguals choose one language over the many others in their repertoire in a given conversation. 211

CHAPTER EIGHT

LANGUAGE CHOICE IN LOWER FUNGOM

8.1 Introduction

The previous chapter looked at use patterns exhibited by our three main consultants. It established that the three consultants were indeed multilingual.

The chapter presented scenarios where the consultants in some context chose one language over another without mixing any of the local languages. When it was necessary for them to use another code, instead of mixing them, they would switch to an entire code. This chapter, like the preceding one, is dedicated to data analysis.

It discusses additional notions concomittant to multilingualism in general but which are seemingly very absent in Lower Fungom. The choice of a language in any speech act is made irrespective of such traditional notions like prestige that are associated only with colonial languages. This indicates the ‘linguistic market’ of Lower Fungom (Boudieu 1991) is structured in a way that largely escapes the model of polyglossia scales, dominant in mainstream sociolinguistics, where each language/variety is found at a given ‘rank’ reflecting the degree of prestige attributed to its speakers. This and more will be 212 discussed in the following paragraphs, highlighting the reasons why participants in Lower Fungom prefer one language to another.

8.2 The notion of di/polyglossia in Lower Fungom

In studying how multilinguals manipulate their linguistic repertoire, it is traditional to immediately think of diglossia. Diglossia (Ferguson 1959),

“extended” by Fishman (1967), has been discussed in the literature review section of this study. It has been considered the conceptual foundation of most studies on African multilingualism. As a reminder, this notion describes situations where the use of different languages was understood to depend on the idea that “one set of behaviors, attitudes and values supported—and was expressed in—one language”, while “another set of behaviors, attitudes and values supported and was expressed in the other” (Fishman 1967: 29).

This notion compartmentalizes the linguistic space of a speech community: different languages can be associated with different social spheres, and these spheres are typically ranked in a prestige hierarchy. In African settings there are usually three recognized “ranks” along this scale: European languages are highest, African lingua francas occupy a middle position and local languages are lowest (Di Carlo 2016). We therefore would speak of polyglossia rather than 213 diglossia in African contexts as seen in Mkilifi (1972) for Tanzania, Johnson

(1975) for Ghana, and Whiteley (1973) for Kenya.

Connell (2009) holds that in many parts of Africa, it is common to find people who, in addition to their home language, would often have command of another local language, a regional or national lingua franca and have some command of a former colonial language. He affirms that “multilingualism of this sort often leads to a diglossic or polyglossic situation; with each language typically having a reasonably well-defined role in terms of the functional domains it serves” (p.

131).

This study, as it discusses how individuals in Lower Fungom manage their numerous lects, examines this well-known concept in a very complex linguistic community. In the section that follows, we examine some of the key criteria for di/polyglossia as championed by Ferguson (1959) in relation to Lower Fungom.

v Compartmentalization

Diglossia, as a recap, is based on the idea that a society's linguistic space can be straightforwardly subdivided into externally-defined "compartments". These compartments according to Romaine (1994: 47) are ascribed clearly defined functions. Function, as Fasold (1984: 53) puts it, is the heart and soul of the diglossia concept and it is clearly linked to prestige. With compartmentalization 214 naturally comes restricted access to some of the domains for some people who may lack mastery of the language/dialect. In addition, compartmentalization and its notion of a division of labour, always assumes the existence of a High and

Low variety, and this is important to note as far as Lower Fungom is concerned.

In Lower Fungom there is no “labour”, as Di Carlo (2016) puts it, that needs to be carried out by a generally agreed-upon specific code. What’s more, and in accordance with Dakubu (1997) and Jaspers (2016), it is rather typical in Lower

Fungom to find individuals who use sets of languages, rather than just one, in what the classical diglossia framework would refer to as one and the same compartment.

In fact, we find Janet using Mufu, Abar and Ngun in a particular Njangi meeting, on the same day, in

KPAAM_CAM_RO_MULTILINGUALISM_ABAR_NJANGI_2015_11_25.

She used Mufu with one of the Njangi members (Mme Nfua) who was from

Wum and was married to a man from Mufu but they lived with their family in

Abar. Janet spoke Abar with other members of the meeting who originated from

Abar and then she also spoke Ngun to Madam N Janet from Ngun.

At work, at the health centre in Lower Fungom, Janet does the same thing; she uses Abar, Fang, Buu, CPE and English as seen in 215

KPAAM_CAM_RO_MULTILINGUALISM_ABAR-FANG-BUU_ENGLISH-

CPE_JANET_ABARHOSPITAL_2015_11_28.

It is important to note that compartmentalization comes with separateness.

Kulo's linguistic practices are clearly in conflict with this definition. With his relatives in the same home (Pa Jo from Buu and his wife from Mundabli), he used two distinct languages with the two interlocutors (as seen in the previous chapter). He used Buu with Pa Jo who is from Buu and Mundabli with Pa Jo's wife who is from Mundabli. Kulo used both Abar and Buu with the same person

(Pa Joseph from Buu). He used Buu with him when they just met but switched to Abar when their conversation was no longer pleasing. So, within the same compartment, Kulo used more than one language, a complete violation of the principle of compartmentalization.

v Prestige

One of the most important criteria for di/polyglossia is prestige. With prestige,

the H variety enjoys superiority to the L variety, as the latter is believed to be

inferior in a number of respects. Ferguson (1959: 237) explains that “even

where the feeling of the reality and superiority of H is not so strong there is

usually a belief that H is somehow more beautiful, more logical and better able

to express important thoughts and the like.” The situation in Lower Fungom is 216

very interesting because in the midst of the eight or more distinct languages (Di

Carlo 2016) and five other dialects, there is no H and L labeling.

Agwara (2013) as observed in chapter two corroborates this theory as she

explores factors accounting for multilingualism in Lower Fungom. Her analysis

revealed that significant rates of multilingualism in the area are explained in

terms of blood relations, marriage, perceived proximity and similarity, religion,

education, kindred and movements. In addition, her data suggests evidence of

pre-colonial multilingualism explained in terms of trade, dependability and the

search of security. We see that one of the reasons people learn and use several

languages in her study is not prestige or superiority. It is true that there are

languages spoken by only about 200 people in a village like Buu, and other

languages spoken by about 3000 in a village like Fang. Even this drastic

difference in population does not render Fang superior to Buu. There is no

village considered more economically viable or more powerful than another.

There is a complete absence of prestige as far as the languages of Lower

Fungom are concerned, and this we will continue to see even as we further

examine the pillars of this phenomenon.

Patterns like these are described in another rural area of Cameroon. Connell

(2009) examined language use in a market setting in the village of Somié. Aside the predominant Mambila dialect of Ba, other languages present in the local 217 ecology, in roughly decreasing order of prominence include Fulfulde as a lingua franca, French, and various other varieties of Mambila, including Mbar, as well as some close relatives of Mambila in the Mambiloid group, more distant relatives within the Bantoid subgroup of Benue-Congo, and Cameroonian Pidgin

English (Connell 2009: 133). Except for French, Connell (p. 134) found that none of these languages “appears to have greater prestige than the others...the society is essentially unstratified, and language knowledge or use does not serve as a social class marker”. Notions of prestige are associated with languages that have entered the region recently along with an exogenous ideology, while they seem to be absent with respect to relationships among local languages.

Data gathered for this study indicates that the desire to create and maintain as many relationships as possible is the basis for code choice in Lower Fungom, not domains or the functions the languages play.

As such, we realize that we cannot completely rely on the notion of polyglossia to explain how multilingualism actually works in Lower Fungom, because there is a complete absence of the H and L dichotomy, and especially because of the fact that languages in Lower Fungom are not assigned specific tasks. For trade, religion, politics and social purposes, in ordinary conversations in homes and in formal conversations, a single language can cut across or play these “labour” functions or, as seen above, several languages can be used conveniently in a 218 single function. Polyglossia will be appropriate in explaining the functionality of multilingualism in an urban area because in such areas there obviously are H and L varieties and all the other pillars on which the concept rests. This suggests that the diglossia framework cannot capture all of lower fungom's sociolinguistic realities connected with multilingualism.

If these conventional notions fail to hold grounds in Lower Fungom, an ethnographic study was needed to investigate and shine the light on the factors that lead multilinguals in Lower Fungom to pick and choose languages in interactions. This is what section 8.3 handles.

8.3 What motivates language choice among Lower Fungom multilinguals?

8.3.1 The co-existence of humans and the supernatural

According to the belief system, life does not end with death. Most

Cameroonians like typical Africans believe that the dead are not really dead; they are just invisible members who are by far superior to human beings. They refer to these ancestors (or their dead family members) as the ‘living dead’ and their reality and presence in the community are duly acknowledged and honoured by community members. Hence, they do actually communicate with them. In speaking with the ‘living dead’ languages choice is not haphazard in

Lower Fungom. The languages the individuals use to speak with the ‘living 219 dead’ in Lower Fungom are quite significant especially because they are certain of a response.

During our last trip in Lower Fungom, our landlord lost one of his relatives.

Several relatives of his had been dying after very short illnesses and the people believed that someone was causing these deaths and so they decided to perform some traditional ritual. We went to the ‘die house’ to offer our condolences and discovered this very interesting phenomenon. The dead child was from Abar, the individual performing the ritual was from Abar but the spiritual package was taken from Missong. This individual spoke to the people while performing the burial in Abar but switched to Missong as soon as and he picked up the package and performed the ritual with it. In other parts of the world, the name of the person who prepared the spiritual package and maybe some other relatives of his will be mentioned, but in Lower Fungom bits and pieces of Missong cannot be used with Abar. Consequently, this performer stopped using Abar for a while and picked up Missong (which is mutually intelligible) throughout the performance with the package. When that ritual was done, he ordered some men waiting nearby to pour earth and cover the coffin. We did not know about the switch in language until later that day when we were discussing with one of the custodians of Abar- Pa Pa Linus. He explained the languages used to perform such a rite in Lower Fungom. 220

In addition, several of the minor consultants told us during out interviews that they were very specific with the languages they use with their ancestors. This is particularly perceptible with those involved in intermarriages or those who have moved away from their villages. Mami Ngonge was from Mundabli but she was married to a man from Abar and she lived in Abar. She told us that she used

Abar with her husband, her children and people around her. She could sing and do every other thing in Abar including talks to God but not to her ancestors.

Mami Ngonje told us that she could only use the language she was brought up with by her parents who had passed away. Pa Pa Linus used her exact same words too to talk about this issue.

Mami Mbang Frida is from Mufu but was married in Buu. She did everything using Buu which she did not even know prior to getting married (that is, she did not know Buu-the place before she got married). She talked to her children mostly in Buu. She was even a member of a secret society of women in Buu.

However, she said that she could not make a mistake of using this language with her ancestors. She told us that her ancestors would ask her if she was from Buu.

To her, it was inacceptable for her to use Buu while talking to her ancestors.

221

8.3.2 Accommodation

A speaker's attitude towards another speaker in conversation can influence his/her language choice. According to Giles’ (1973) accommodation theory as discussed in chapter two, an individual tends to shift his/her speech styles to resemble those they would like to associate with. Giles calls this convergence while he uses the term divergence for a process in which a speaker makes his/her speech sound more unlike that of his/her interlocutor. In Lower Fungom, family relations are sacred, even extending several generations to distant ancestors. One is expected to speak the language of his great-grandparents or at least make an effort to accommodate linguistically to their kin, whether distant or immediate. We proceed in this discussion with a view from Di Carlo’s (2016) study which included the naming system in Lower Fungom and the importance of having multiple identities. He reports that throughout Lower Fungom, children are named as per their maternal and paternal affiliation; they have names from their maternal kin and also from their paternal kin. This is especially obvious if the child’s parents come from two different villages and hence are speakers of two different languages/varieties. In such a case, the child is expected to learn both and to use them in the appropriate circumstances. The father’s language is the exclusive code to be used for communication with paternal kin, while the mother’s language must be used with the maternal kin. In 222 essence, both personal names and linguistic competence seem to require that the child acquires distinct identities with respect to each kin group.

We saw from the literature review of this thesis that linguistic neutrality (Myers-

Scotton 1976) is a strategy which multilinguals use for accommodation. The language of neutrality in Lower Fungom was CPE and it was usually employed in a mixed crowd. Kulo and his friend Jumia from Buu ditched Buu for CPE because they entered a barber’s shop where people from other tribes were present. He told us that he used this language so that everyone would understand what was being discussed. By doing so, he avoids a situation where those present feel that he and his friend were saying unpleasant things about them.

Their behaviour expressed their conscious desire for social inclusion and intergration.

Kulo met Pa Joseph's wife on his way to Fang for the ‘Pig for Pikin’ meeting and he only spoke to her in Mundabli. He walked around the village, returned to her house and continued using this same language. Pa Joseph's wife is from

Mundabli and so is Kulo’s late mother. It is inadmissible for Kulo to meet this mother and use his father’s first language, Abar. Pa Joseph is Buu; Kulo and Pa

Joseph’s wife meet as can be seen in the recording

KPAAM_CAM_RO_MULTILINGUALISM_KULO_2015_11_22, discussed in the previous chapter. Kulo spoke in Buu to Pa Joseph and almost at the same 223 time he speaks in Mundabli to Pa Joseph’s wife. He did not use Pa Jo’s language with this man’s wife, which is what he was raised with and which he is currently using with Pa Joseph. Kulo switched from Abar, which he was using on his way to Buu, and uses Buu with indigenes of Buu like Pa Joseph and his very good friend Jumia. He further switched to Mundabli to speak with Pa Joseph’s wife.

Nkangkolo Isaac, from Buu, is one of the participants we interviewed with whom Kulo often spoke Buu. He corroborated this information when he told us that whenever he met people whom he knew were from Fang, he would immediately switch to Fang, not because some of his relatives were from Fang, but because he had taught in Fang for a long time. He said that this showed his affection for them and they were very happy when he did that. Nkangkolo

Isaac’s switches to Fang are very important because he would not even switch to use his wife’s language. His wife was from Missong but he spoke to his wife and children only in Buu. He told us that whenever he met with someone from

Abar he would still use Buu because the person from Abar would understand

Buu. Buu and Fang are neighbours, consequently people from Fang understand

Buu, but Nkangkolo chose to switch to Fang to show those from Fang that they were a family and that he was happy to have spent some time with them.

In one of our recordings as we were walking with Boniface from Abar to

Missong, he used CPE with us but whenever he met anyone from Missong, he 224 spoke to them in Missong and then returned to CPE with us. We asked him why he could not converse in CPE since the people talked to us in CPE. Boniface said that they would take offence. They would ask him if he did not know them.

For solidarity's sake, Boniface had to move back and forth, from CPE to

Missong to accommodate the people from Missong.

Janet was the leading practitioner of accommodation or solidarity. In her audio of November 22nd, she used Fang in speaking to all natives of Fang, as she had relatives from Fang. Janet also used her husband’s language when she met someone from his village. While Janet was heavily pregnant, she left Lower

Fungom and went to give birth in her husband’s village and enjoy the company of her in-laws. We went to Esu to continue our observation of how Janet used her very rich linguistic repertoire. Janet was not even fluent in Esu, but whenever she met someone from Esu, she began by uttering a few sentences

(greetings especially) in Esu before switching to CPE.

Worthy of note is the fact that these consultants were performing one of their duties, as Myers-Scotton spells out. People really do take offence when these obligations are not practiced. This holds true if we consider what happened between our consultant Kulo and the wife of a renowned politician in Lower

Fungom, Madam N. Janet from Ngun. On May 20th 2016, Kulo met Madam N

Janet from Ngun. They met in Abar as Madam N Janet was returning from a 225

Njangi meeting called ‘work hard’, which is always held in Abar. Kulo conversed with her in Abar and Madam N. Janet responded in Ngun, just like in the recording on May 24th 2016, in the market. Kulo reported that she was the one related to Abar; her great-grandmother was from Abar. Consequently, the switch is expected to come from her, not Kulo. He said that whenever he wanted to speak to her in Ngun, his tongue would be heavy, but with her husband Mr

Njah Vincent who was not related to Abar, he would use Ngun comfortably.

With Kulo, Nkangkolo Isaac spoke only Buu. He told us that it was Kulo who had to drop Abar and talk to him in Buu because he had a relationship with Buu, while Nkangkolo did not have one with Abar.

May 26th 2016 was a market day in Lower Fungom. Market days bring together people from all areas of Lower Fungom. Kulo ran into Pa Muluh from Kung and talked in Kung with him. Pa Muluh lived in Kung; although Kulo did not know his wife, he knew that this man was not related to Abar. Kulo told us that whenever they met, Pa Muluh would use Kung with him, as he knew that Kulo’s great-grandfather (grandfather’s father) was Kung. Kulo explained that he could not respond in Abar because Pa Muluh did not understand Abar. He could not also use CPE because Pa Muluh was older and they were not friends. The same evening, we went with Kulo to a drinking spot in Abar, Mami Mbang Frida’s house where she sold spirits. As she saw Kulo, she started using Mundabli. Kulo 226 told us that she typically used Mundabli with him because she knew that his mother was from Mundabli. He respected her like his aunt and replied in

Mundabli. Mami Mbang Frida explained to us in an interview that she used

Mundabli most of the time with Kulo because she and Kulo were from the same area; not only that, but they were also relatives. (Kulo’s mother was Mami

Ngonje’s mother’s sister). She called Kulo her brother. She however admitted that, she sometimes used Abar with Kulo because at that moment, they both lived in the same place, and it was Kulo’s main language.

Pa Pa Linus from Abar said that if one had a relative from another village, if they met, it would be better to use the relative's language to let them know that they have not been forgotten. If one did not use the relative’s language, it would indicate that their relationship was not strong.

Affinity which could be because of marriage or ethnicity could push for accommodation in Lower Fungom. Mama Regina's case resulted from the former and curiously, it clashed with the latter. We interviewed Mama Regina as a secondary consultant; she was friends with both Janet and Kulo. She did not know her precise age when we worked with her but she gave a rough estimate of about 45 years old. Her parents were from Mufu and her husband is from Buu.

He speaks Buu and CPE, and he does not speak Mufu. Her father spoke Mufu and CPE. Her mother spoke Mufu, CPE and Mashi. Mama Regina herself spoke 227

Buu, Mufu and CPE. She did not speak Abar; consequently, she could not converse in Abar with Kulo or Janet. She was from Mufu and it was absolutely important that she knew this language: she was born in Mufu and her parents spoke to her in Mufu. She used Mufu with her parents and those from Mufu. She admitted that she prayed and sang in Mufu. Curiously, whenever we met her in the field, she would speak in Buu, even when we met her in Mufu, she spoke to us in Buu and was very happy when we replied in Buu. She learned Buu when she got married at the age of 16. She overtly claimed to be from Buu. She also said that even when she was in Mufu, she would speak in Buu to thoses she met in Buu to let them know that she was from Buu and also because she would like to welcome them. If they met and she spoke Mufu, these visitors would feel that she desired to challenge them. Mama Regina also told us that she would not speak Buu to someone from Mufu, no matter the setting in which they found themselves. Janet’s mother was Mufu, as was Mama Regina. One would expect

Janet and Mama Regina to speak Mufu too, but this was largely not the case.

Although they were both from Mufu, they both did not speak Mufu to each other. Mama Regina rather felt that they were both from Buu and it was natural for them to use Buu and then, if they wanted to communicate in secret, they would use Mufu. She said her children typically used Buu, and if they said something in Mufu it would not be something serious. She said Buu was their language (her children). Kulo and she spoke Buu, as she said, his father was 228 from Buu. She explained to us that when she met Kulo and spoke to him in Buu, she was reminding him of his ‘foundation’; she was bringing him close to her. If she were angry with Kulo she would not use Buu, she would use CPE to send him away from her. She asserted that Kulo would most certainly know that there was something bothering her. She also said that when she spoke Buu with Kulo, he was happy “e heart di sweet”. Her husband had passed away 10 years before but she remained in Buu. She was even the chief of Mufu’s younger sister, yet she remained in Buu although a widow, using Buu almost always.

Mbang Frida is another case we studied. She was one of the rare women we found who would rather use Buu (her husband’s language) with her children and not her own language. This was also evidence of the fact that people generally did not use language haphazardly. Mama Mbang Frida was a farmer; she was 50 years old and lived in Buu. Worthy of note is the fact that her parents were both from Mufu. She told us that she and her children were almost always in Buu, and they only went to Mufu maybe twice in a year, so that using that language was not circumstantially pressing. She did not even pray in Mufu. Worth mentioning is the fact that she did not even know Buu (the village), let alone the language, before her marriage; at the time we were there, she spoke with Janet and her other children almost always in Buu. She used Mufu with her ancestors, 229 as she said that she was from Mufu and if she used another language, her ancestors would ask her whether they were from that village.

8.3.3 The desire to communicate in secret

Everyone we worked with in Lower Fungom desired to be secretive at some point in their lives. Not every conversation is meant for the ears of the public or even the person seated next to the consultants. Pa Joseph from Buu admitted during an interview that he would use his wife’s language, Mundabli, with her if someone from Buu was present and he just would like to ask her to bring part of the palm wine he had kept in the house. He felt that the visitor did not need to know the quantity of palm wine that was available.

Jumia admitted to occasionally leaving Buu and joining Kulo with Abar or using any other language they both master whenever he wanted to keep information between him and Kulo. This would happen if the person (s) close to them understood Buu but not Abar.

When Boniface met his fellow hunter and wanted to ask how he was coping with hunting in his own area, he did so in Mundabli. Meanwhile, a few seconds before, both men had been using CPE with each other. He explained to us that sometimes hunting was frowned upon and so CPE would not be a safe choice for this conversation as everone would understand them. 230

Kulo met a secondary school student from Koshin but who lived in Abar, he wanted to ask the girl to go and wait for him in his house at the palace but his children were around, and he did not want them to understand him. Out of respect for his children, he hid his message in a language they did not understand. We realized that the girl was a student and Kulo normally spoke to students in English, so we asked him why he had used a mother tongue this time. He told us that he did not want his children to understand what he had just said.

Blackwell from Munken and Jumia from Buu told us that mother tongues were shadowy; if one intended to conceal information, they were the best languages to use.

8.3.4 Mimicking

We stood outside our home in Lower Fungom one day during our last visit when a gentleman, Pa Akong Julius from Abar, was passing by. Pa Akongnwi, a politician, just like Pa Nkangkolo was a CPDM councilor in Lower Fungom. As he walked towards our home, Kulo told the man that he was walking exactly like Pa Nkangkolo from Buu. Kulo initiated the conversation in Abar but to respond, Pa Akongnwi used Buu; he said that he was getting already as old as Pa

Nkangkolo and it was normal for him to walk like him. Kulo joined him in Buu 231 and they kept discussing the subject and mimicking the former Member of

Parliament. Worthy of note is that Pa Akongnwi was not truly related to Buu; none of his parents or grandparents was from Buu, only his aunt was married in

Buu.

Kulo also told us that this happened very often; he remembered a recent occasion when he was walking with his friend Jumia, from Buu on their way to

Nkang to administer a vaccine. They remembered a funny confrontation they had experienced in Fang and the pair began using this language as they mimicked the event. We asked him why the speech event was not done in Buu, since the pair almost always used Buu. Kulo told us that they had to immitate and quote certain people from the confrontation. This could not be done in Buu.

8.3.5 The need to obtain a favour

This was greatly practiced by Kulo, although everyone would tend to do this to obtain friendship, information, or just to have some kindness, especially from a relative. All the consultants went out of their way to ingratiate themselves to others. On one of the market days in Abar (May 26, 2016), we were with Kulo in the market when he met Boah Stephen. He and Kulo started discussing in

CPE and later switched to Koshin. We felt that it was an attempt to conceal something from us but we still asked to confirm. It was not a case of secrecy:

Kulo said that he was begging Stephen to help carry his vaccination kit on a bike 232 to Koshin the following day so that he would leave very early in the morning on foot to join them there.

Later that day, Kulo was using CPE with some friends in the market square when he spotted Mami Mbande Ruphina from Fang, with whom he began a conversation. Mami Mbande was married to someone from Meta but was a widow then. She was in the market on that day to sell 'Shaa' (locally made corn beer). Kulo wanted to drink corn beer on credit. Kulo’s father married from

Fang, in Lower Fungom, they would say, he married Mami Mbande’s sister.

Kulo was discussing in Abar but realized that he needed to persuade even more to obtain the favour. So he switched at once to Fang, calling her ‘my in-law’.

Mami Mbande just continued the conversation in Fang and so did Kulo. The conversation ended in Fang, and she ended up giving Kulo some corn beer to drink. Kulo’s strategy was thus successful.

Kulo did something similar another day while we were on our way to his farm in

Abar. We were discussing in CPE and we were joined by a gentleman from

Mundabli. He joined our discussion in CPE but Kulo later switched to

Mundabli. When this gentleman had left us, we asked him why he had switched.

He told us that he wanted to beg the man for palm wine, as he always harvested good palm wine. This man happened to be his uncle (Kulo’s mother was from

Mundabli, so all the men there were said to be his uncles). So, Kulo wanted to 233 say “uncle, please send some palm wine for me.” His target was to use the word

'uncle' in Fang to remind him of their relationship. Kulo said that the word

“uncle” in CPE would not have had the same effect.

Although Kulo practiced this a lot, he was not alone in it. Janet met Mami

Beatrice from Koshin at the health centre. She joined a discussion in which CPE was the language in use, so she continued with CPE until she wanted to discuss money matters with Mami Beatrice. The ladies were in the same Njangi group, known as ‘Work Hard’, and Janet wanted to borrow using Mami Beatrice’s name. She abandoned CPE and discussed this in Koshin. We did not even get to ask her why she had switched because soon afterwards, she on her own accord, urged us not to feel bad that she stopped using CPE. She said that she was only discussing money matters. We asked her why the conversation could not be done in CPE, and she said it would feel awkward ("No, e go dey somekind" were the words she used).

Other participants of this study (different from our three major consultants) also had similar comportment. Mr. Pa Linus, one of the cultural custodians of Abar whom we interviewed, shared the same opinion. Pa Linus said that people from

Mashi did not speak Abar, and that they did not interact with them a lot. So he mostly used CPE with them, but if he were trying to get some information from the person or if he needed a service, he would use Mashi because he would want 234 them to be soft-hearted. He said when someone knew those languages, his life was much easier. Mr Botan Bodwin from Mufu also thought in this light. He said that to know all the languages of Lower Fungom was a blessing because nobody would refuse to do you any favours.

We also interviewed Luke, the then regent of Buu; born in 1953 and a farmer.

He added his leading voice to those above in affirming that when one spoke some people’s languages, they feel at home and they tend to be warm-hearted.

He said that, if they were in the market, they might even bring down the prices.

8.3.6 Avoidance

Avoidance is a divergent communicative act. Here, a speaker's language choice can emphasize social difference or disapproval of the addressee. When we asked

Pa Nkole Simon of Buu in Lower Fungom to list the languages he understood and spoke, he mentioned Missong among others. When we asked him which language he liked people from Missong to speak to him, he said that any language he understood would be fine. He added that the languages were one, so using Missong or CPE was just fine, and so we mentioned Boniface his brother- in- law from Missong. He said that when Boniface met him, he (Boniface) always spoke Buu because his mother was from Buu and his (Boniface's) wife was his (Pa Nkole Simon) sister. When Boniface had a problem, he would go to 235

Missong for him and Boniface would do the same should he have a problem. In the recording of Boniface seen before in this study, on a market day labeled

KPAAM_CAM_RO_MULTILINGUALISM_NTO_ABARMARKETDAY_201

5, he ran into the said Pa Pa Nkole Simon. They began speaking in Buu. We remember that Boniface was the son of the Fon of Missong, a potential heir apparent. In actual fact, he was supposed to use his language most of the time to uphold and assert his background. Consequently, when he switched, it was a major cause for concern; it was highly marked. Boniface began using Buu with

Pa Nkolo and Pa Nkolo had a series of unpleasant questions for Nto. He told

Boniface that he had heard that he had been chased away from Fang recently.

Boniface explained to him that it was some other person called Manto, but Pa

Nkolo persisted with this line of inquiry. Pa Nkolo was his uncle, much older than he was. Consequently, Boniface could not outrightly ask him to stop talking to him or change the topic. Boniface resorted to his linguistic arsenal; he replied in Missong. This worked perfectly as Pa Nkolo immediately left, but not without expressing his displeasure at Nto’s behaviour. Boniface could not savour his victory as one bystander called Kinka (who was also older than Nto) walked up to him and told him not to ever repeat such an act. He actually found the act alarming and he openly begged Boniface to always treat his relatives kindly, no matter what they did.

236

The interaction above is in line with what Franceschini (1998) says, and which we discussed more elaborately in chapter two. She asserts that “language serves to differentiate the speaker from others and by marking the speaker’s belonging to a certain group by means of a similar use of language.” The converging use of language, to her, creates confidence and cohesion, whereas divergent use of language tends to be ambivalent and dissociative and is easily associated with either ‘personal’ or ‘foreign’. She maintains that speakers know how to use these codes and they are well aware of which values are transported by means of these codes.

For further illustration of this fact, consider this other interaction from Kulo.

Kulo, on his way to Fang, stopped at Buu with everyone else headed to Fang for the ‘Pig for Pikin’ meeting. As mentioned earlier, on November 22nd, he talked with others on his way to Buu in CPE and Abar, but as soon as he reached Buu and met its indigenes, he began using Buu. He went to his very good friend Pa

Joseph’s house, he greeted him in Buu and his wife in Mundabli. He conversed with Pa Joseph in Buu until he took offence at one of Pa Joseph’s statements.

From then on, Kulo spoke in Abar with Pa Joseph untill he left Buu. Pa Joseph did not like it and so he left his company, but not after throwing some unpleasant words at him; Pa Joseph told him that he was an animal in the bush.

This is obviously different from a domesticated animal at home. Pa Joseph 237 found this act barbaric; however, he kept speaking in Buu to Kulo whenever they spoke. Kulo persisted with his use of Abar in Buu, and switched to Buu to talk to his best friend Jumia and others from Buu.

Pa Joseph reportedly spoke Buu, Fang, CPE, Mundabli and Abar. Worthy of note is the fact that he spoke Abar like a native speaker. He said that he learned it from Abar girls who got married in Buu. He also said that when he met someone from Abar, he spoke Abar, but did speak Buu with Kulo because although his mother was from Mundabli, his father, according to Pa Joseph, was from Buu. Consequently, to him, Kulo’s first language was Buu. He said that although Kulo's father came from Abar parents, and was also born in Abar, he grew up in Buu. Kulo himself was born in Buu; he grew up in Buu before he was taken to Abar. This was the same as the other consultant, Janet. He told us that one could not meet their child and start using another man’s language with him/her. Pa Joseph learned Mundabli in Mundabli as he used to go there for a dance, but it was his wife who had perfected his knowledge of this language.

He, however, did not use Mundabli with Kulo, although Kulo’s mother was from Mundabli and he also knew that Kulo had native fluency in this language.

So, when Kulo decided to use any other language with Pa Joseph, bearing all this in his mind, it was a consious decision that marked sourness and distancing. 238

8.3.7 The need to convey a feeling within oneself (soluloquy)

There are some languages that better convey a thought; they express extreme emotions better. It is not strange to find Cameroonians of English expression exclaim in CPE or a language like Duala while using English. It is common in

Buea- South West Region of the country, where this thesis is being written, to hear people say ‘bana loba ni kasoh’ approximately translates as 'Oh my God in heaven' in Duala, even if they don’t speak that language. This also happens in

Lower Fungom. Kulo manifested this behaviour quite often. This can be seen in his recording of November 22nd. He typically exclaimed in Fang and Koshin.

His famous exclamations include, “I have lost” in Fang and “This world is not good” in Koshin. Although he could not really explain why he chose to exclaim in these languages, it is clear from our experience with similar situations that it is because these languages to him better expressed his emotions at a particular point in time. These languages to him conveyed the intensity of his emotions at their actual state: mild, moderate, or intense.

Sometimes, it is not only related to the strength of the emotion, but also the nuances of it. Some cultures just have different or more precise words for specific emotions. Sometimes, it’s not just because the word itself has a more precise meaning, it is also because it indexes a cultural idea of a word which the interlocutors would recognize, and would be able to call up images in their 239 minds; they would see the whole idea and emotion a lot better than if the word were said in another language. In sum, it is also about very fine nuances of emotion that these languages carry.

8.3.8 Communication

Sometimes, consultants code switch for basic communication. It is true that

Lower Fungom is very linguistically diverse. However, not everyone could speak all its languages and lects like Janet and Kulo. Consequently, when our consultants met people who simply did not understand any of their languages, they were forced to plunge into theirs. When we asked Janet why she typically left Buu or Abar for Fang or Koshin, for example, she told us that she had tried the people several times and that she knew for sure that they did not understand her. She said that she did not want to repeat herself several times in an attempt to communicate, so she switched to their language whenever she wanted to communicate with them. This is especially true as CPE was not highly esteemed in Lower Fungom, especially when it was used by someone who could speak the language of the interlocutor in question. Boniface said that if he used CPE with an indigene of Lower Fungom who knew he could speak their language, they would ask him if he did not know them.

The regent of Buu told us that CPE helped one to communicate with those whose language one did not know. He gave an example of a day when he went 240 to the market to buy ‘manyanga’ (oil from palm kernels); he met a woman from

Mundabli and he spoke Buu as he asked for the price of the oil. The woman did not understand and was saying something different from what he had asked, so he changed languages and spoke to her in CPE, as he was unable to speak

Mundabli. Then the lady understood him.

Nkangkolo Isaac explained that CPE was a language one used when there was no other means of communication. Consequently, he could not use CPE with someone from Buu or Abar. It felt degrading to him.

As we walked around the village with Kulo during our second trip, we visited a drinking spot, and we realized that one person conditioned the discussion in the area. People spoke to one another privately in their mother tongues but whenever they did a joint and audible discussion, it was in CPE. We asked about the languages each spoke and discovered that there were some people present who were from Missong, and then Akwa Eugene who was from Weh. As a matter of fact, the discussion was happening in front of his house. It was his wife who made and sold the corn beer that was drunk there. His parents were from Weh and his wife was from Abar (Mban). She spoke Abar and CPE. Akwa

Eugene spoke Weh, CPE and English. He had learned Weh from his parents and in his neighbourhood and learned CPE in school. He used CPE to interact with people with whom he did not have a language in common. He told us that he 241 would not use CPE with people from Weh. He felt that CPE helped him to do his business, and his business was always out of his village. In order to communicate with him, the rest of the crowd had to use the language common to them all and that was CPE.

Blackwell from Munken, during our discussion with him, declared that people from Munken understood Missong and vice-versa. As a result, he did not need to switch languages, but there were some people with whom he could not use

Munken: for example, the Fon of Missong, even though the Fon understood this language. He explained that whenever he was in Missong, he would use

Missong with the Fon. He added that he did not use CPE with indigenes of

Lower Fungom unless he did not know their language and they could not understand his. He too felt that if he spoke CPE with them, they would feel insulted.

8.3.9 Protection

Life in general could be terribly hazarduous in rural settings and the need for security crucial. Many people in Lower Fungom also use language for protection, as seen in Di Carlo (2016). Safety is just one of the many advantages of knowing and using several languages. The regent of Buu, for instance, explained that knowing several languages helps one to be safe, as one would always hear what another person was saying against them. 242

Mr. Pa Linus also felt that knowing another person’s language could keep one safe. He said that if he were kidnapped, he should be able to ask help even from a little child around and save himself. In his opinion, if someone wanted to harm you, just let them know you understand their language, and they would hold back.

Madam Mbang Frida (Janet's mother) told us that she learned several languages, including her husband’s language-Buu, so that not even her husband would cheat her. It is even true that the need for security even deepens the impulse for a sense of communal life among the people of Lower Fungom. In addition, Mr

Botan Bodwin said that he sometimes used a language for his protection; if he went to a village and realized that they were doing suspicious things or talking suspiciously, he would start speaking their language to let them know that he understood what they were saying. This may cause them to change or, at the very least, he would be able to plead with them and make them see reason. He said that to know all the languages of Lower Fungom was a blessing because nobody would refuse to do someone any favours. Kopytoff (1987) strengthens these views as he says that;

Each person was attached to several groups of solidarity. Depending on the context, one expected support from each and offered it to each of them. In times of conflict, one tried to mobilize the maximum contextually relevant group. Since traditional African societies were structured in terms of 243

corporate groups, individual survival was possible only by being under the protective umbrella of one or another such group, and the larger and more powerful it was, the safer one was. Kopytoff (1987: 24)

Thus, the construction of multiple identities is a response to individuals’s fear of the unknown; it is a response to the feelings of insecurity.

8.3.10 Topic

Throughout the recordings, especially with the use of CPE, we realized that the topic of discussion also affected the language to be used. Most of our consultants, Kulo, Nto, Nkangkolo Isaac from Buu, and Nkam Elvis from

Missong, all declared that CPE was for general conversations. When Kulo wanted to talk about politics where he would appreciate the input of several others, or when he wanted to inform the general public about a vaccination campaign, he used CPE. When he was using CPE and decided to say something unpleasant like an insult, he typically used one of the languages of Lower

Fungom.

We also realized that Boniface was conversing in CPE in his market recording, but switched to Mashi to ask another hunter how hunting was going on in his area. He felt the need to conceal this information. Whether people choose to conceal information or expose it, this would be demonstrated by the language they choose to use. 244

In Lower Fungom, as we may have already realized, it is normal and very common to find people who actively use more than one language, with the same participant, in the same conversation and same setting. People slip into different languages or refuse to leave their languages for several reasons, some of which have been examined above. Another reason is because of the gender of the interlocutor, as seen below.

8.3.11 The gender of the interlocutor

Nkangkolo Isaac declared that he did not use languages without conscious control. He was born out of wedlock and at the time we worked with him, he did not know his father. His wife was from Missong and they both lived in Buu.

Nkangkolo Isaac admitted to speaking in Buu to his wife all the time, even if he met her using Missong with their children and he needed to add his voice to hers. He affirmed that he could not use another man’s language with his own children. Nkangkolo Isaac told us that his wife insisted on using Missong with their children, so, he too insisted on using Buu with her and their children. He was of the opinion that when a woman married a man from another village, she became a woman from that village too. Consequently, it was his wife who ought to speak Buu to him; he ought not to speak her language. He asserted that if he even dared to speak Missong, even his children would consider him weak; they would think that their father felt inferior about himself and about his language. 245

We asked him if he would not even use his wife's language when he was in a very good mood, and he said that he still would not. He went further to say that sometimes he spoke to her in Buu but she responded in Missong, and he would rather have that than to replace the use of his own language with hers.

The above paragraphs have shown why both the minor and major consultants pick and choose languages. These are: for protection, to convey a thought, as an avoidance strategy, to obtain a favour, for mimicking, to be able to communicate in secret, for accommodation, the co-existence of humans and the supernatural, the topic and the gender of the interlocutor. The paragraphs have reported how rural multilinguals steer their way through life by manipulating their rich linguistic repertoire. They have as such made known how multilingualism functions in this part of rural Africa.

246

8.4 Conclusion to the thesis

With all of the above in mind, it is safe to conclude that Kulo, Janet and

Boniface do not live strange lives; multilingualism is pretty normal today.

However, these individuals are unusually multilingual. They did indeed use several languages on a daily basis depending (among other things) on the people with whom they were talking, the people around them, what the discussion was all about, the environment where they found themselves and the intentions of the participants.

8.4.1 Recap of the problem

The problem addressed in this thesis was the focus that has been placed on urban areas, on colonial languages and on the interaction between colonial languages and local languages while studying multilingualism to the detriment of a more

'pure' rural multilingualism. With this, very little is known about rural multilingualism that involves the active use of local languages on a daily basis.

We felt that for multilingualism and linguistic diversity to be fully understood, data and perspectives from rural areas, too, should be examined and as a matter of priority.

8.4.2 Aims of the research

The purpose of this research was to examine and describe how multilingualism functions in a rural setting. This meant investigating the language usage patterns 247 of multilinguals in Lower Fungom, the kinds of linguistic activities attested in

Lower Fungom and exploring the language ideologies governing language use in Lower Fungom. It also meant investigating the social motivations and impact of the sociolinguistic dynamics of multilingualism in Lower Fungom.

8.4.3 Summary of main parts

The study used a qualitative paradigm to achieve its aims and objectives. We chose the qualitative design based on the kind of research questions we had. The qualitative paradigm gave us the leeway to dig deep into the lives of the subjects of this study and to provide in-depth answers for the research questions of the study. We were able to describe in details the participants and their linguistic habits with this paradigm.

In addition, we chose to do the research as a case study; we were convinced that it would allow for a deep exploration and understanding of the complexity and subtle distinctions contained in language use.

For this small-scale study, we selected a sociolinguistic interview guide, audio and video recorders, non-participant observations as instruments and tools of data collection. We made use of the purposive sampling strategy in selecting three major consultants and 20 minor ones.

248

With these consultants, we examined how rural multilinguals would use their linguistic repertoires daily. These consultants wore a microphone as they went about their daily activities. With the help of their recordings and of the video recordings we made of them and our regular observations (both participant and non participant), we were able to gather data on how linguistic repertoires are used in rural Lower Fungom. With the grounded theory, we were able to analyze our data and this resulted in very interesting findings which we presented in chapters five, six, seven and eight of the thesis.

In chapter four, we made a very detailed presentation of the participants of the study. In chapter five we presented our overall data; what we obtained from the participants as self reports of their linguistic habits (that is, the languages they have competence in, which they use and how they use them), and what we got from the recordings and observations. In chapter six, we examined instances of language use in Lower Fungom in comparison with instances of language use elsewhere in Africa and beyond; our aim was to find out if Lower Fungom is a multilingual area devoid of code switching. This was geared towards understanding the kind of sociolinguistic dynamics that exist in Lower Fungom.

The data obtained suggests that language choice with these extraordinary individuals is indeed far from being arbitrarily done. It is not necessarily even 249 subjective: more often than not, language choice here is based on the norms and ideologies of the society. Overall our data suggests that:

1) Mulitlingualism is a daily reality even in rural areas. Our consultants were able to use up to seven lects from their repertoire daily.

2) The appropriate construct for the linguistic habits of inhabitants of Lower

Fungom is not code switching but language choice. In Lower Fungom, code switching does not involve two local languages. The switch is between CPE and one local language. Here, one language or dialect is almost always dumped in favour of another depending on several reasons which are generally understood by the participants involved.

3) The ideologies guiding linguistic practices in Lower Fungom are different from those attested in urban areas and even some rural areas in Africa. In Lower

Fungom, the language ideology tilts more towards indexicality than essentialism.

4)That Lower Fungom’s ‘linguistic market’ (Bourdieu 1991) is structured in a way that largely escapes the model of polyglossia scales, dominant in mainstream sociolinguistics, where each language/variety is found at a given

‘rank’ reflecting the degree of prestige attributed to its speakers.

5) The languages the consultants choose to use are designated by their position in the social space or field, just like Bourdieu (1991) holds. Their choices suggest that Lower Fungom is a sort of “linguistic market” wherein linguistic 250 practices are business ventures; they appear to be some sort of investments. The consultants’ choices are clearly made based on the rewards they would accrue.

These rewards include the desire for accommodation, the search for security, solidarity, distancing, mimicking and gender.

8.5 Limitations of the study

Although this study achieved its aim, there were some unavoidable limitations.

First, the time apportioned for this study did not permit us to stay too long in the field. We could only live in Lower Fungom for three months. However, these months were sufficient; we were able to collect data to the point of theoretical saturation.

In analyzing the data, we avoided transcribing the data in any phonetic detail.

We only did a rough transcription, without tones and preliminary word segmentation. While higher levels of accuracy would clearly be desirable, this was difficult to achieve when researching code switching in a region with seven to nine local languages, none of which are associated with any kind of written standard.We are convinced that this does not affect our findings in any way, as we are not concerned with the grammatical aspects of the languages. We simply did a free translation of what the consultant said. Also, we do not speak any of the languages of Lower Fungom. As such, the translations we got from the recordings were provided by the consultants. Although we chose a few 251 recordings which we had already worked on and asked a 'judge' to translate it a second time.

We used ELAN (for transcription and annotations) as it allows both the researcher and other scholars to listen and/or watch the consultants’ actual use of language. This allowed the consultants to follow the analysis with sustained interest, and permitted them to check the data quality and our interpretations more or less directly, depending on their linguistic competence.

8.6 Significance of the study

The intent of this study is to contribute to the overall knowledge of multilingualism. This study has not only drawn attention to one of the richest linguistic areas of Cameroon; it has also focused on collecting a rich, natural corpus from multilingual individuals thereby showcasing the different languages used by these individuals, the persons they interacted with, the proportion of language usage per these consultants, their contexts and the motivation of usage.

This in-depth analysis has shined light on individual rural multilingualism in

Lower Fungom; it has most certainly enriched our knowledge of how multilingualism operates in a rural setting.

The findings of this study have not only provided insights to the oversimplified narrative on rural multilingualism, but they has also thrown light on certain notions typical of rural setting. Our data brings to light some notions 252 misconstrued by both African and especially Western countries such as 'tribe'. In

Lower Fungom, as revealed by our data, it is common to find individuals who have two names from two different political units (villages) in mixed marriages.

It is also common to find individuals who use more than seven languages in their daily lives. This is a mess in a tribalistic perpective whose stronghold is oneness; a tribe speaks one language and tribes's men have one affiliation. This is the norm in urban settings but in a rural one like Lower Fungom, everything is much more fluid and permeable.

In addition, the result of this study will contribute in development corporations for rural areas. Through this study, researchers would become aware of the implications of language use in rural areas as far as activities such as bible translation are concerned. This study reveals that language choice is not a trivial matter and that although, as in the case of Lower Fungom where multilingualism is at its peak, it is still inadmissible to simply select one language or dialect to translate the bible into.

The language ideologies that this study showcases broadens our knowledge of culture; it could help in interaction with migrants be it at the local level or international level. This thesis gives value to membership; it exposes the desire to be multiply affiliated, which is something which was not in the minds of many in western societies. 253

We strongly believe that this study has provided a fuller picture of multilingualism, especially multilinualism in Cameroon which is still unfamous and unsung. It has also served as a reminder to the numerous advantages of the method of investigation used in this study.

Significant Original Contribution to Knowledge

With the significance of the study in mind, emphasis in this paragraph is drawn to the significant original contribution to knowledge this thesis has made.

Its findings ilucidate the thesis of this study: rural multilingualism is fascinating and profoundly different from urban multilingualism.

Its methodology (Qualitative/Ethnographic) which was capable of eliciting nuances, perculiarities and complexities in language use and in context is original and quite impactful.

8.7. Suggestion for further research

This study proposes further ethnographic studies in the field of Applied

Linguistics. As this study has demonstrated, ethnographic studies reveal and explain the complexity of the human relations in a specific environment. More studies that would produce situated knowledge instead of universals, that would capture the details of real-life situations are definitely highly needed. 254

REFERENCES

Aikhenvald, A. & Dixton, R. (Eds). (2007). Grammars in contact: A cross linguistic typology. Oxford. Oxford University Press.

Agwara, A. (2013). Rural multilingualism in the north west region of Cameroon. (Unpublished MA thesis). University of Buea, Cameroon.

Auer, P. (1999). From codeswitching via language mixing to fused lects: towards a dynamic typology of bilingual speech. International Journal of bilingualism. 3 (4), 115-158.

Blommaert, J. & Dong, J. (2010). Ethnographic fieldwork: A Beginner’s Guide. Bristol • Buffalo • Toronto: Multilingual Matters.

Bourdieu, P. (1991). Language and symbolic power. J. Thompson. (Ed.). Cambridge: Polity Press, 37-42.

Bryman, A. (2004). Social research methods. (2nd Ed.). Oxford University Press.

Buda, J. (1991). Language Choice. Otsume Review. No 24, 29-35

Charmaz, K. (2002). Grounded theory analysis: In J. F. Gubrium & J. A. Holstein (Eds.). Handbook of interview research. No 25. 675-694. Thousand Oaks, CA: Sage.

Chernela, J. (2013). Toward an East Tukano ethnolinguistics: Metadiscursive practices, identity, and sustained linguistic diversity in the Vaupés basin of Brazil and Colombia. In P. Epps and K. Stenzel (Eds.). Cultural and linguistic interaction in the Upper Rio Negro Region (Amazonia). Rio de Jeneiro: Museu do Indio- FUNAI, 22. 197-244.

Cobbinah, A. (2010). The Casamance as an area of intense language contact: The case of Baïnouk Gubaher. In Lupke and Chambers (Eds) 175-201.

Cobbinah, A., Hantgan, A., Lupke, F. & Watson, R. (2017). Carrefour des langues, carrefour des paradigmes. In M. Auzeanneau (Ed.). Pratiques 255

plurilingues, mobilités et éducation. Paris: Édition des archives contemporaines. pp 79-97

Connell, B. (2009). Language diversity and language choice: a view from a Cameroon market. Anthropological Linguistics, 51, 2, 130-150.

Dakubu, K. E. (1997). Korle meets the sea. A sociolinguistic history of Accra. New York.Oxford University Press. Di Carlo, P. (in press). Towards an understanding of African endogenous multilingualism. Ethnography, language ideologies, and the supernatural. International Journal of Sociology of Language. Di Carlo, P. (2016). Multilingualism, affiliation and spiritual insecurity. From phenomena to processes in language documentation. In M. Seyfeddinipur (Ed.). African language documentation new data, methods and approaches. Language documentation & conservation special publication 10:71–104. Di Carlo, P. (2011). Lower Fungom linguistic diversity and its historical development; Proposals from a multidisciplinary perspective. Africana Linguistica. 17, 53-100. Di Carlo, P & Good, J. (2014). What are we trying to preserve? Diversity, change, and ideology at the edge of the Cameroonian Grassfields. In P. Austin & J. Sallabank (Eds.). Endangered languages: Beliefs and ideologies in language documentation. Oxford and New york: Oxford University Press. 229-262 Di Carlo, P., Good, J., Ojong Diba, R. (2019). Multilingualism in Rural Africa. Oxford Research Encyclopedia of Linguistics. Oxford University Press. Di Carlo, P. & Pizziolo, G. (2012). Spatial reasoning and GIS in linguistic prehistory. Two case studies from Lower Fungom (North West Cameroon). Language dynamics and change, 2 (2), 150-183. Dorian, C. N. (2002). Diglossia and the simplification of linguistic space. International Journal of Sociology and Language. 63–69. Dornyei, Z. (2011). Research methods in Applied Linguistics; Quantitative, qualitative and mixed methodologies. New York: Oxford University Press. 256

Eckert, P. (2012). Three waves of variation study: The emergence of meaning in the study of sociolinguistic variation. Annual Rreview of Anthropology 41, 87-100. Eckert, P. (2000). Linguistic variation as social practice. Oxford: Blackwell. Eckert, P. (1989 b). Jocks and Burnouts: Social categories and identity in the high school. New York: Teachers College. Eckert, P. (1980). Diglossia: Separate and unequal. Linguistics, 1053-1064. Ennaji, M. & Sadiqi, F. (1994). Applications of modern linguistics. Casablanca. Afrique Orient. Epps, P. (2012). Amazonian linguistic diversity and its sociocultural correlates. In M. Crevels & P. Muysken, (Eds.). Language dispersal, diversification, and contact: A global perspective. Oxford University Press. 220-239. Evans, N. (2010). Dying words: endangered languages and what they have to tell us (The language library). Hoboken, NJ: Wiley-Blackwell: NJ. Fasold, R. (1995). The sociolinguistics of society. Oxford: Blackwell. Ferguson, C. (1959). Diglossia. Word 15, 325-340. FINCA International. (2015, April 05). Financing Africa's rural youth. Washinton. Retrieved online on September, 2017. Fishman, J. (1967). Bilingualism with and without diglossia; diglossia with and without bilingualism. Journal of Social Issues 23(2), 29-38. Franceschini, R. (1998). Code-switching and the notion of code in linguistics: Proposals for a dual focus model. In P. Auer (Ed.). Code switching in conversation; Language, Interaction and Identity. London and New York: Routledge. 49-72 François, A. (2012). The dynamics of linguistic diversity: egalitarian multilingualism and power imbalance among northern Vanuatu languages. International Journal of the Sociology of Language. 214, 85 - 110. Giles, H. (1982). Communication accommodation theory. Encyclopedia of health communication. Baltimore: University Park Press 257

Glaser, B. G. & Strauss, A. L. (1967, 1995). The discovery of grounded theory: strategies for qualitative research. New Brunswick and London: A division of transaction publishers. Good, J., Jesse, L., Mve, J. P., Nganguep, C., Voll, R. & Di Carlo, P. (2011). The languages of Lower Fungom region of Cameroon: Grammatical overview. Africana Linguistica, 101-164. Gorra, A. (2007). An analysis of the relationship between individuals' perceptions of privacy and mobile phone location data- a grounded theory study (Ph.D thesis). Retrieved from Research Gate, 2016. Leeds Beckett University. Groom, N. & Littlemore, J. (2011). Doing Applied Linguistics: A guide for students. Routledge. Gumperz, J. (1982). Discourse strategies. Melbourne, Australia: Cambridge University Press. Gumperz, J. (1972).The sociolinguistic significance of conversational codeswitching. RELC Journal 8, 2 : 1–34. Gumperz, J. & Hernandez-Chavez, E. (1971). Bilingualism, bidilectualism and classroom interaction. In C. Cazden, V. John and D. Hymes (Eds.). The functions of language in the classroom. New York: Teachers College Press. Heller, H. (1988). Codeswitching: Anthroplogical and sociolinguistic perspectives. Berlin: Mouton de Gruyter. Himmelmann, N. (1998). Documentary and descriptive Linguistics. Linguistics. 36, 161–195. Honig, H. (1999). Translation studies. In David Katan (Ed.). Translating cultures. An introduction for translators, interpreters and mediators. Manchester: St Jerome Publishing. Irvine, J. (1989). When talk isn't cheap: Language and political economy. American Ethnologist. 16, (2), 248-267. Jaspers, J. (2016). Diglossia and beyond. In O. García, N. Flores & Massimiliano (Eds.). The oxford handbook of language and society. Oxford University Press. 258

Johnson, B. (1975). Stable triglossia at Larteh, Ghana. Patterns in language, culture and society: Sub-Saharan Africa. Proceedings of the symposium on African language, culture and society held at Ohio State University, Columbus, April 11, 1975. 93-102. Columbus, OH: Ohio State University. Johnstone, S. & Kiesling, B. (2008). Indexicality and experience: Variation and identity in Pittsburgh. Journal of Sociolinguistics. 5-33.

Katan, D. (1999). Translating cultures. An introduction for translators, interpreters and mediators. Manchester: St Jerome Publishing.

Kopytoff, I. (1987). Frontiers and frontiers societies. In Encyclopedia of Africa south of the Sahara. (Ed.). J. Middleton. New York. Charles Scribner's sons. 170-171

Labov, W. (1966). The social stratification of English in New York city. Washinton, DC: Centre for Applied Linguistics.

Labov, W. (1963). The social motivation of a sound change. Word, 19: 273- 309.

Lincoln, Y. & Guba, E. (1985). Naturalistic inquiry. Newbury Park, CA: Sage Publications.

Lüpke, F. (2016). Uncovering small-Scale multilingualism. Critical multilingualism studies 4.2, 35-74.

Lüpke, F. (2010). Language and identity in flux: In search of Baïnouk. Journal of Language Contact THEMA 3: 155–174.

Lüpke, F. & Storch, A. (2013). Repertoires and choices in African languages. Berlin/New York, NY: Mouton de Gruyter.

Mason, J. (2002). Qualitative researching. London. Thousand Oaks. New Delhi: SAGE.

Mkilifi, A. M. (1972). Triglossia and Swahili-English bilingualism in Tanzania. Language in society 1.2, 197-213.

259

Myers-Scotton, C. (1982). The possibility of code switching: Motivation for maintaining multilingualism. Anthropological Linguistics, 24. 4, 432-444.

Myers-Scotton, C. (1976). Strategies of neutrality: Language choice in uncertain situations. Language, 52. 4 , 919-941.

Myers-Scotton, C. (1993). Social motivations for code switching, Evidence from Africa. Oxford: Clarendon Press. Neubert, A. & Gregory, M. (1992). Translation as text. Kent: The Kent State University Press. Nida, E. & Wonderly, L. (1971). Communication roles of language in multilingual. The Bible Translator. Pierce, C. (1995). Logic as semiotic: Theory of sign. In Collected papers of Pierce Sanders. Cambridge: Harvard University Press. Polkinghorne, D. (2005). Language and meaning; Data collection in qualitative research. Journal of counseling psychology, 52.12, 137-145. Polome, E. C. (1982). Rural versus urban multilingualism in Tanzania: An Outline. International journal of the sociology of language, 167-181. Poplack, S. (2015). Code switching: Linguistics. International encyclopedia of social and behavioural Sciences, 3, 918-925. Poplack, S. (1980). Sometimes I'll start a sentence in spanish Y TERMINO EN ESPANOL: Toward a typology of code-switching. New York: Center for Puerto Rican studies. Psichari, J. (1929). Un pays qui ne veut pas sa langue. Mercure de France, 207, 63-121. Richards, L. (2005). Handling qualitative data: A practical guide. London: Sage Publications. Rickford, J. R. (1986). The need for new approaches to class analysis in sociolinguistics. Language and Communication, 6 (3) 215-21. Rumsey, A. (1990). Wording, meaning, and linguistic ideology. Retrieved online. 4/14/2017. 260

Schieffelin, B., Kathryn, W. & Kroskrity, P. (1998). Language ideologies: practice and theory. New York: Oxford University Press. Schiffman, H. (2005). Problems with language and ideology. Handout for LING 540. Language policy. Accessed online on 4/14/2018. Silverman, D. (2001). Interpreting qualitative data: Methods for analysing talk and interaction. London/Thousand Oaks/New Delhi: Sage. Silverstein, M. (2003). Indexical order and the dialectics of sociolinguistic life. Language & communication 23, 193–229. Stallcup, K. (1980). La géogaphie linguistique des grassfields. In L. Hyman & Jan Voorhoeve (Eds.). L'expansion Bantoue: Actes du colloque international du CNRS, Viviers 1. 43-57. Strauss, A. & Corbin, J. (1998). Basics of qualitative research: Grounded theory procedures and techniques (1st Ed.). Newbury Partk: CA: Sage. Stroud, C. (2007). Multilingualism in ex-colonial countries. In P. Auer and W. Li (Eds.). Handbook of multilingualism and multilingual communication. Berlin: Mouton de Gruyter. 509- 38. Tabouret-keller, A. (1997). Language and identity: The handbook of Sociolinguistics. Oxford: Blackwell Publishers. Trochim, K. (2006). Research methods knowledge base. Web center for social research methods. Whiteley, W. (1971). Introduction. In W.H. Whiteley (Ed.). Language use and social change. Problems of multilingualism with special reference to Eastern Africa. Oxford, Oxford University Press. William, R. (1977). Marxism and Literature. Oxford: Oxford University Press. Wolff, E. (1991). Spachstandardisierung in Afrika: begriffe, Probleme, Perspektiven. In C, Schubert & E. Wolf (Eds.). Language Standardisation in Africa. Hamburg: 1-19. Yin, R. (1984). Case study research: Design and methods. Bervely Hills: Sage Publications.

261

The Sociolinguistic Dynamics of Rural Multilingualism in Africa. the Case (2024)

FAQs

What is the sociolinguistic approach to multilingualism? ›

The sociolinguistics and multilingualism research cluster focuses on the issues relating to linguistic features and associated links to social meaning, social values, identity and the wider regional and socio-cultural contexts.

What is the official multilingualism in South Africa? ›

South Africa's constitution recognises 11 official languages: Sepedi (also known as Sesotho sa Leboa), Sesotho, Setswana, siSwati, Tshivenda, Xitsonga, Afrikaans, English, isiNdebele, isiXhosa and isiZulu.

What factors have contributed to multilingualism in Nigeria? ›

Multilingualism can be as a result of historical or political movements such as imperialism or colonial- ism; economic movements in the case of migration; increasing communications between different parts of the world and the need to be competent in languages for wider communication; social and cultural identity and ...

What is the meaning of multilingualism? ›

Multilingualism can be defined as “the ability of societies, institutions, groups and individuals to engage, on a regular basis, with more than one language in their day-to-day lives” (European Commission, 2007: 6).

What is the main focus of sociolinguistics? ›

Sociolinguistics is the study of the relationship between language and society. Sociolinguists are interested in how we speak differently in varying social contexts, and how we may also use specific functions of language to convey social meaning or aspects of our identity.

What are the types of multilingualism in sociolinguistics? ›

Types of multilingualism include: Bilingualism (2 languages) Trilingualism (3 languages) Multilingualism (4 or more languages)

Why is Africa multilingual? ›

Multilingualism is part of the experience of growing up in Africa where the interconnected histories of ethnic groups, dynamic geographic landscapes and European colonialism left an enduring legacy. There are more than 1,500 distinct languages spoken in Africa (compare that to about 300 in Europe).

Is Africa the most multilingual continent in the world? ›

The continent has more than 2,000 distinct languages, equating to a third of all languages in global terms. Europe, by comparison, has just 300 languages. Based on the Ethnologue index of linguistic diversity, 20 of the top 25 most linguistically diverse countries are in Africa.

Are you aware that South Africa is a multilingual society? ›

In essence, South Africa has been a multilingual country since the enactment of the Constitutional Act 108 of 1996 of the Republic of South Africa, granting official language status to 11 of these languages and sign language in 2012.

What is the biggest benefits of multilingualism? ›

Studies have found that multilingual individuals have better memory, problem-solving skills, and cognitive flexibility. This may be due to the fact that learning and using multiple languages requires a higher level of cognitive functioning.

What problems does multilingualism bring for individuals? ›

However, multilingualism comes with its set of challenges. Firstly, there is a considerable communication barrier that exists in a multilingual society. Each language has its unique syntax, grammar, and vocabulary, making it difficult for individuals to communicate effectively.

Why is Nigeria a multilingual nation? ›

The language scenario in Nigeria is complex, intricate and dense, leading to an extreme degree of multilingualism due to an array of aboriginal, exogenous and Pidgin languages. Nigeria is one of Africa's centers for language variety, which has metamorphosed into linguistic diversity.

Which language dominates the world? ›

Widely recognised as the most spoken language in the world, Mandarin Chinese holds a significant position in global communication.

What is meant by sociolinguistics? ›

Sociolinguistics is the descriptive study of the effect of any or all aspects of society, including cultural norms, expectations, and context, on language and the ways it is used. It can overlap with the sociology of language, which focuses on the effect of language on society.

What are the arguments against multilingualism? ›

Although this is a typical communication technique, it might cause miscommunication, especially when working with monolingual individuals. Confused Identity: Multilingual individuals often have complex cultural identities, which can lead to a sense of displacement or confusion about where they truly belong.

What is sociolinguistic approach? ›

The idea behind sociolinguistics is that language is highly variable and that it is impacted by complex social factors. Sociolinguistics approaches language with the idea that all kinds of speech are equally correct and valuable, even though some are stigmatized.

What is a sociolinguistic approach to SLA? ›

When a person is learning a second language, they will ideally be exposed to both the common themes of the language as well as variations in the language. Sociolinguistics studies how social factors influence this diversity in language, such as geography, age, and socioeconomic status.

What is the sociolinguistic approach to translation? ›

The principal focus of a sociolinguistic theory of translation is "translation as an act ofcommunication." This focus necessarily implies the analysis ofthe source discourse as an act ofcommunication: who said what, to whom, under what circumstances, for what reasons, and for what purpose.

What is sociolinguistic theory of language learning? ›

Sociolinguistics mainly studies the relationship between social context and language use, that is, how language changes with factors such as social class, education level, age, gender, race, and region. Fishman[1] (1972) pointed out that sociolinguistics examines "who speaks what language to whom, when, and why."

Top Articles
Latest Posts
Recommended Articles
Article information

Author: Nathanial Hackett

Last Updated:

Views: 5729

Rating: 4.1 / 5 (52 voted)

Reviews: 91% of readers found this page helpful

Author information

Name: Nathanial Hackett

Birthday: 1997-10-09

Address: Apt. 935 264 Abshire Canyon, South Nerissachester, NM 01800

Phone: +9752624861224

Job: Forward Technology Assistant

Hobby: Listening to music, Shopping, Vacation, Baton twirling, Flower arranging, Blacksmithing, Do it yourself

Introduction: My name is Nathanial Hackett, I am a lovely, curious, smiling, lively, thoughtful, courageous, lively person who loves writing and wants to share my knowledge and understanding with you.