Gender Variation in Ewe-English Codeswitching

1

GENDER VARIATION IN EWE-ENGLISH CODESWITCHING

BY

GLADSTONE DEKLU

10340362

THE DISSERTATION IS SUBMITTED TO THE DEPARTMENT OF

LINGUISTICS, UNIVERSITY OF GHANA, LEGON IN PARTIAL

FULFILMENT FOR THE AWARD OF BACHELOR OF ARTS DEGREE

IN LINGUISTICS

JUNE 2014

2

DECLARATION

I, Gladstone Deklu, do hereby declare that with the exception of references which

have been duly acknowledged, this long essay is the result of my personal research

carried out under the supervision of Dr. Evershed Kwasi Amuzu of the Department

of Linguistics, University of Ghana, Legon.

……………………………… Date………………………………

Gladstone Deklu (10340362)

(Candidate)

……………………………… Date………………………………..

Dr. Evershed Kwasi Amuzu

(Supervisor)

3

DEDICATION

I dedicate this work to the almighty God, to my parents Mr. Robert Attah Deklu

and Mad. Felicia Adjei and to my siblings Carl, Nicholas and Eugenia and my

nephew Osmond.

4

ACKNOWLEDGEMENT

I wish to sincerely thank all who played various roles in support of this work.

Firstly, I give thanks to the Almighty God. His grace and mercies have seen me

through my studies in the university and through this research work. I cannot

forget the tremendous help I received from my supervisor, Dr. E. K. Amuzu. Your

insightful questions, critical observations and comments have helped me

throughout the research work. You are truly a mentor.

My deepest acknowledgement also goes to my family. You have nurtured me

never to give up on anything I do. “If you want to do something, do it well once

and for all”: these words from you, Mum, has kept me going throughout my

studies in the University of Ghana.

I am deeply indebted to all members of E. P. Church, Legon. Special thanks go to

Winfred Dotse-Gborgbortsi, Gotah Selikem, Nuviadenu Joana and Bribi Doris

Yayra; your support and encouragement have seen me this far. Words cannot

express how much I am grateful to you, Ms. Thompson Rachel. I can only say

thank you and God continue to shower his blessings upon you.

To all who helped in various ways to make this work a success, God richly bless

you.

“Mawu na yra mi katã.”

5

TABLE OF CONTENT

COVER PAGE…………………………………………………………………….i

DECLARATION…….…..…………………………………….………………… ii

DEDICATION……………….………………………………………………….. iii

ACKNOWLEDGEMENT...….………………………………………..……….. iv

TABLE OF CONTENT……...……………..………………………...…………..v

ABSTRACT.………………….……….…..…………………………….............. ix

CHAPTER ONE: INTRODUCTION…...……………………….……..………. 1

1.1 Background.………………………......…………………………………..…… 1

1.2 Problem Statement …………………...…….………..………………………. ..2

1.3 Research Questions……….. ….……..……………......…………………… …3

1.4 Research Objectives…..…….………..……………...………………….…...... 3

1.5 Definition of Key Terms …………..…………..……………………..………. 3

1.6 Overview of the Study ..…………..…………………..…….………..……. …5

CHAPTER TWO: LITERATURE REVIEW, METHODOLOGY AND

THEORETICAL FRAMEWORK..………………….….…..……….…………. 6

2.0 Introduction………………………..…...………………………………………6

2.1 Literature Review ………..……..…………………………………….……. …6

2.1.1 Codeswitching………………………………................................. …6

2.1.2 Motivations for Codeswitching……...………….……..……...………..........9

2.1.3 Language and Gender………………….…………………………………...11

6 2.1.4 Labov (1991) on Gender and Language choice…………….……………....13

2.1.5 Codeswitching and Gender…………………………………………............16

2.2 Methodology…………………………………………….………………………….....17

2.2.1 Research Design……………………………………………………………17

2.2.2 Research Sites………………..……………………………………………..18

2.2.3 Research Instruments………..…….………………………………………..20

2.2.3.1 Questionnaires…………………………………………………...20

2.2.3.2 Out – Group Interviews…..……………………………………...21

2.2.4 Sources and Methods of Data Collection…………………………...22

2.2.5 Method of Data Analysis……………………………………………23

2.3 Theoretical Framework…………………....…………………………….........23

2.4 Summary of Chapter………………..………..…………………………..……25

CHAPTER THREE: SOCIOLINGUISTIC BACKGROUND OF EWE-

ENGLISH BILINGUALS…………………………………...…………..............26

3.0 Introduction……………………………………..…………………………….26

3.1 Demographic Information………………………………….………………....26

3.2 The Use of Codeswitching at Home…………….…………….……………...27

3.3 Language Usage of Ewe-English Bilinguals...............…..….………………...33

3.4 Summary……………………………………….…………..…………….........39

7

CHAPTER FOUR: GENDER VARIATION IN EWE-ENGLISH

CODESWITCHING…..........................................................................................40

4.0 Introduction……………………………..…………………..............……...…40

4.1 Codeswitching Habits among Male and Female Ewe-English Bilinguals……40

4.2 Gender Variation In Codeswitching Patterns Among Age Groups Of Ewe-

English Bilinguals………………..………………………………………......46

4.3 Codeswitching as a Prestigious Form or as an Adulteration of a Language….47

4.4 Sociolinguistic Behaviour of Ewe Women In Relation To the Principles of

Labov……………………………………………………………………….…49

4.5 Summary………………………….……………………………………..….…51

CHAPTER FIVE: FINDINGS AND RECOMMENDATIONS………………52

5.0 Introduction ……………………………………...……..…………………….52

5.1 Sociolinguistic Background of Ewe-English Bilinguals...................................52

5.2 Gender Variation in the Use of Codeswitching.................................................53

5.3 Attitude towards Codeswitching among Ewe-English Bilinguals……..……..54

5.4 Reaction Of Ewe Women To The Principles Of Labov (1991).……….……..54

5.5 Recommendations……………………………………………………..……..55

APPENDIX I: QUESTIONNAIRE ………………………….…………………...56

APPENDIX II: INTERVIEW GUIDE……………………………………………62

REFERENCES….…………………………………………………………….......64

8

ABSTRACT

There is evidence showing that codeswitching (CS) is pervasive in the speech of

bilinguals. Ewe-English bilinguals are no exception to this trend. This study

investigates gender variation in the use of CS by Ewe-English bilinguals. Using the

variationist sociolinguistic approach by Labov to analyse data, it is found that

females are generally in the lead in engaging in CS. However, women that are

above age 50 tend to engage in it less than their male counterparts. The current

study also investigates the attitude of Ewe-English bilinguals toward CS. It is

discovered that there is an increase in the prestige enjoyed by CS even though

there is still an overt negative attitude toward it. However, based on data collected

and analyzed it is discovered that CS enjoys covert prestige. Thus even though

Ewe-English bilinguals explicitly express negative attitude toward it, there is an

implicit positive attitude toward CS that is seen in the pervasive use of CS by Ewe-

English bilinguals.

9

CHAPTER ONE

INTRODUCTION

1.1 Background

Code switching among bilingual Ghanaians is not a new phenomenon. It can be

defined as “the switching from the linguistic system of one language or dialect to

that of another” (Merriam-Webster dictionary, 2013). For the purpose of this study,

I will define code switching or CS as the switch from one language to another

language in a single conversation. This switch can be intra-sentential or inter-

sentential. Code switching has received a lot of attention over the years. The first

major work on code switching between English and a Ghanaian language was done

by Forson (1979). Forson (1979) claims that, in the 1970s and 1980s, educated

Ghanaians were able to use their Ghanaian language and English fluently when

they interact with other colleagues with the same level of education or fluency in

both languages. They only resort to code switching due to certain factors such as

the degree of formality of the speech situation, the linguistic repertoire of

addressees and the topic under discussion.

Asilevi (1990) however indicates that Ghanaian bilinguals engage in

codeswitching because they have no other choice than to use codeswitching in

their interactions with other bilinguals. This is due to the pervasiveness of code

switching. This assertion was confirmed by Amuzu (2005a). In his study, he held

an out – group interview whereby respondents were made to discuss various topics

with an Ewe monolingual. He discovered that even though respondents were

restricted to the use of Ewe in the interview, they ended up using English words or

phrases.

10

There exist opposing views about codeswitching (CS): one view is that it is

a marker of prestige while the other is that it is an adulteration of a language.

“While CS is viewed as an index of bilingual proficiency among linguists, it is

more commonly perceived by the general public as indicative of language

degeneration” (Bullock and Toribio 2009:1). These views present a paradox as to

whether code switching is a prestigious phenomenon or a phenomenon that

destroys a language.

This study will focus on the pervasiveness of code switching among the

speakers of the Ewe language with much emphasis on gender. It will throw some

light on how much code switching is done by male and female speakers of Ewe.

1.2 Problem statement

Many researchers including Dzameshie (1996), Yevudey (2009), and Amuzu

(2005 and 2012) have worked on Ewe-English codeswitching. These studies

mainly focused on the motivation for Ewe-English codeswitching and the grammar

of codeswitching. However, they were relatively silent on which gender of Ewe-

English bilinguals engage in this practice more. This research seeks to fill this gap

in the literature and also deal with the sociolinguistic background of interactants

who engage in codeswitching.

Labov (1991) claims that two general principles regarding sexual

differentiation in sociolinguistics exist. The first principle states that in a stable

sociolinguistics stratification, men use a higher frequency of non-standard forms

than women. This implies that women are more receptive to standard forms and

that have apparent social prestigious forms. The second principle posited by Labov

(1991) states that in majority of linguistics changes noted, women use higher

11

frequency of incoming forms than men. These principles will be dealt with in

detail in the next chapter of this work. The study will further consider whether

codeswitching is perceived among Ewe women as a prestigious form or an

adulteration of the language using the two principles propounded by Labov (1991).

1.3 Research Questions

1. Which gender engages in Ewe-English code switching more and what

sociolinguistic factors account this?

2. What is the reaction of female Ewe-English bilinguals in light of Labov

(1991)’s principles on language use of males and females?

1.4 Research Objectives

This research seeks to

1. Find out the gender of Ewe-English bilinguals that engages in Ewe-English

code switching more and to determine the sociolinguistic factors that are

responsible for that gender’s pervasive use of code switching.

2. Discover the reaction of female Ewe-English bilinguals to the principles on

language use of males and females by Labov (1991).

1.5 Definition of Key Terms

This section provides the definition of key terms used in the study. These

defnitions will provide a better understanding of concepts discussed in the study.

12

1.5.1 Multilingualism

Multilingualism can be defined as the act of using two or more languages by an

individual or a speech community. A person is referred to as a multilingual if that

person can speak two or more languages-his mother tongue and an additional

language(s). In Ghana, many educated people are bilinguals because they can

speak English which is the official language and at least one indigenous Ghanaian

language. Multilingualism can be considered as the complete mastery and

proficiency in two or more languages or, the ability to be able to make basic

utterances that enable one to be able to communicate with speakers of other

languages.

1.5.2 Gender

Gender is the series of physical, biological, behavioral, and social features that are

associated with and used in the differentiation between masculine and feminine.

Gender is a social construct. It includes a set of behaviours that a particular sex is

expected by society to possess (see section 2.1.3 for further details).

1.5.3 Out-Group Interview

The out group interview is a kind of interview that constitutes respondents that

have little or no formal education. This group of persons therefore has little or no

communicative competence in English.

13

1.5.4 Collaborator

For the purpose of this study, a collaborator is a person that does not speak and/or

understand English hence cannot communicate in English (cf. Amuzu 2005a). This

person acts as a third party in an interview where participants exclusively

communicate in Ewe.

1.6 Overview of the Study

This study is in five chapters. Chapter one gives a general introduction to the study

and also presents the problem statement and objectives of the study. It also gives

the research questions of the study. Chapter two is the review of literature and the

methodology employed in the study. It also presents the theoretical framework that

the study employs in the collection and analysis of data. A brief background to the

research area is then given in the chapter. Chapter three and four contain analyses

of the data collected and chapter five deals with the conclusion of the study and

highlights the major findings as well as some recommendations.

14

CHAPTER TWO

LITERATURE REVIEW, THEORETICAL FRAMEWORK AND

METHODS

2.0 Introduction

This chapter opens with a review of literature in two sections. The first section

reviews works done on CS and its motivations. Section two focuses on language

and gender, reviewing basic language and gender issues necessary for

understanding the current study. These concepts include, ‘language and gender’,

‘sex and gender’, and ‘gender and language choice’. I will further concentrate on

gender and codeswitching. After the literature review, the methods employed in

the study will be elaborated. The chapter ends with a presentation of the theoretical

framework employed in the study.

2.1 LITERATURE REVIEW

2.1.1 Codeswitching

Codeswitching (CS), as defined by Bullock and Toribio (2009), is the capacity of

bilinguals to alternate between two or more languages in a single speech event.

This alternation can be within a sentence (intra-sentential) or between two

sentences (inter-sentential). This phenomenon is basically a result of language

contact.

Chan (2009) opines that even though earlier works focused on bilingual

communities in the United States of America, works were done later by linguists in

other regions of the world. He further asserts that, these works were initially done

15

on Indo-European languages, but gradually, interest grew in languages that are not

Indo-European. This interest led to the expansion of research into codeswitching.

In Ghana, a lot of work has been done on CS. Asilevi (1990) studied Ewe-

English code switching in conversational discourse. He discovered that, Ewe-

English CS has definable rules with which it operates. As he rightly puts it “Ewe-

English code-mixing like code-mixing in other languages with our type of

language contact also has restraints and proceeds in accordance with some

definable rules” (Asilevi 1990:92). He also ascertained that CS is neither random

nor fully as a result of linguistic deficit.

Amuzu (2005a) also studied the grammar of Ewe-English codeswitching.

This study gathered that CS is a pervasive phenomenon among bilingual Ewes.

Also, in his interviews, many bilingual Ewes one way or the other engaged in CS

even though they were supposed to communicate monolingually in Ewe in the

interviews. This confirms Asilevi’s (1990) position that, Ewe-English bilinguals

have no other choice than to use code switching in their interactions with other

bilinguals.

Forson (1979) focused on Akan-English code switching. In his view, CS

does not occur between Akan and other indigenous Ghanaian languages but

between Akan and only English which is the official language of Ghana.

Furthermore, CS has been studied in churches in Ghana by scholars such as

Asare-Nyarko (2012) and Albakry and Ofori (2011). In the study of Albakry and

Ofori (2011), CS was found to be used in 50% of the catholic churches that were

under study. They also assert that CS was done during certain church activities

(e.g. homilies, prayers of the faithful and announcements).

16

Some researchers have discovered that CS is not a deficiency in a language

but a way through which speakers are able to express their social roles and

identities and/or manage ongoing talk (Chan 2009). It is worth noting that, “CS

comprises a broad range of contact phenomena and is difficult to characterize

definitively” (Bullock and Toribio 2009:2). The linguistic index may vary from the

insertion of single words in a sentence to vacillating two or more languages. Also,

the bilingual mastery of people who engage in CS is not the same. Similarly, the

type of contact situations in which they find themselves vary therefore, CS does

not have a uniformed pattern (Bullock and Toribio 2009).

Early research in multilingualism focused on the macro sociolinguistics

point of view (i.e. analyzing how languages and their varieties are distributed in a

society). This involves the languages that are peculiar to certain domains of

society, for instance, the family, school, workplace and many others. However,

Angermeyer (2006), following studies by Glom and Gumperz (1972), asserts that

language users do not only change their code choice due to changes in the domain

in which they are communicating but they may switch their language at any time

without even being aware of it. This triggered interest in micro-analysis into

speech data by linguists who study bilingualism. As a result of this, CS has become

central to many if not all bilingual studies (Angermeyer 2006).

For CS to occur in any conversation, participants of that conversation or at

least one of the participants must be a bilingual or a multilingual. Without this

condition, there cannot be any form of CS. As Forson (1979) claims, code

switchers are by definition bilinguals. He asserts that, bilingualism is a necessary

condition for CS but CS is not a necessary product of bilingualism.

17

2.1.2 Motivations for Code Switching

Code switching can be employed for many reasons. One of the most basic reasons

why people engage in CS (especially inter sentential CS) is to exclude a party that

is not a speaker of the language from a conversation. This is a deliberate act to

deprive that person of understanding the message conveyed.

Another factor that brings about CS is the lexical gaps in a language. When

a new concept is introduced into a community, there may be no name in the

language of the host community for that entity. It may also be that, the name for

the entity may not be popular in the language of the host community. Speakers

therefore switch into a different language to refer to that entity or borrow that name

entirely. This assertion is confirmed by Bandia (1996), who asserts that, African

writers resort to the use of indigenous words and expressions when they cannot

adequately express certain African socio cultural realities in European languages.

Forson (1979) claims that CS may be employed based on the topic of discussion

and the linguistic repertoire of speakers. Speakers with enough exposure to English

engage in CS when communicating monolingually in Akan. He observed that in

the discussion of certain topics, even bilinguals with enough exposure to English

rarely engage in any form of CS. These topics include funeral rights, farming

arrangements and many others.

According to Ibhawaegbele and Edokpayi (2012), CS is employed as a

stylistic strategy by novelists and poets in Nigeria in their writings. According to

them, CS can be done to indicate the type of relationship that exists between two

people and also indicate the degree of formality in any situation in the novel.

Writers use CS to show the level of seriousness attached to a particular issue that is

discussed by characters in the novel. CS can be used to give a stylistic effect to a

novel and also indicate the cultural difference that exists in the setting of the novel.

18

Scotton (1982) observes that bilinguals engage in CS to affirm their identity

as ethnic brethren by using their native language and to affirm their identity as

educated persons by using English. This study establishes that CS may be

employed to show authority or superiority.

With the use of CS in classrooms, Arthur (1996) reveals that teachers in

Botswana use CS in the classroom to encourage participation of pupils. This is

done by their use of tag questions which elicited answers from pupils and

encouraged their participation. He however indicates that there are instances where

the intended outcome of CS was not realized in the classroom.

Albakry and Ofori (2011) notes that many people choose the code choice

that brings them the greatest reward. A code choice that makes them communicate

most effectively. This confirms Dzameshie’s (2001) concept of economy of code

choice principle. This principle states that “if you are a bilingual, use the language

that communicates your message most clearly and expeditiously whenever you are

talking with other bilinguals with approximately the same linguistic repertoire.”

Dzameshie (2001:15), argues that this principle is “tacit knowledge that is

possessed by all bilinguals and it forms part of their communicative competence”

which guides them in their day-to-day interactions. He explained further that for a

bilingual or monolingual, the basic and universal motivation behind the use of

one’s communicative competence is the desire to maximize the full benefits of

one’s competence in social encounters. Hence, in communicating, a bilingual will

employ CS if it is the code that provides the greatest reward to his communication

need.

19

2.1.3 Language and Gender

Language and gender is an area of sociolinguistics and linguistic anthropology

which is concerned with the investigation of varieties of speech associated with

particular gender or social norms for gendered language use. According to

Sunderland (2006), “popular understandings of gender and language existed

probably for centuries before ‘gender and language’ was considered worthy of

study”. This includes both prescriptive ideas of how men and women are expected

to speak, and linguistic ideas about how they actually do.

Language and gender, in the view of Eckert and McConnell-Ginet

(2005:79) cited in Shitemi (2009), involves interpreting the use of linguistic

resources to accomplish social ends. Leading authorities in the study of language

and gender agree that gender makes a difference in the use of language. This

difference however, is not universal hence, some men may exhibit feminine

conversational qualities and some women may exhibit masculine conversational

styles. In traditional studies of language and gender, two paradigms are often

employed; these are ‘dominance’ and ‘difference’.

In the dominance model which is associated with Zimmerman and West

(1975), there is a notion that in a mixed sex conversation, men are more likely to

interrupt than women. They put it this way:

…males assert an asymmetrical right to control topics and do so

without evident repercussions. We are led to the conclusion that…

men deny equal status to women as conversational partners with

respect to rights to the full utilization of their turn and support for

the development of topics. Thus we speculate that just as male

dominance is exhibited through male control of macro – institutions

in society, it is also exhibited through control of at least a part of

one micro institution. (Zimmerman and West 1975:125)

20

This is confirmed by Marie Shimomura (2003) as cited by Stanhope-Essamuah

(2005). She states that men are most probably interruptive and assertive than

women in any speech event. Also, Stanhope-Essamuah’s (2005) study of

University of Ghana students indicates that 73.4% of men as compared to 63.4% of

women interrupt during conversations. She states that, “women wait their turn to

speak in conversations than men do”.

The difference model is concerned with striking the difference between the

various ways males and females express themselves in communication and the

motives behind them engaging in any form of conversation. Tannen (1990) claims

that, men live as “individuals in a hierarchical social order in which they are either

one-up or one-down”. This suggests that they strive to gain the upper hand at

whatever they do. In other words, men are competitive in both act and speech.

Women, on the other hand, are more likely to approach speech as “a

network of connections”. Thus, “conversations are negotiations for closeness and

people try to seek and give confirmation and support, and to reach consensus”

(Tannen 1990). Women’s speech is said to be less aggressive and assertive than

men’s speech. It is rather seen to be on the emotional and sensitive side hence,

their ability to be more polite than men. On the other side, men engage in talks

about physical strength and competition. To be able to understand the phenomenon

of language and gender better, a distinction must be made between sex and gender.

According to Talbot (2010), “the earliest work on men, women and

language attended to ‘sex differentiation’. Studies of such differences were carried

out by Europeans and other westerners with an interest in anthropology”. These

studies, as asserted by Talbot (2010), have focused on the existence of affixes and

pronouns that are specific to men and women, whether being spoken about or as a

speaker. The question then arises as to whether gender is the same as sex. Sex is a

21

biological categorization based primarily on reproductive potential whereas gender

is the social elaboration of biological sex (Eckert and McConnell-Ginet 2003). Sex

is determined by the biological make up of an individual while gender is a social

construct which is determined by norms of society. Hence, gender patterns in one

community may be different from another community. However, the notion of sex

is universal. It is the same from one community to another and there is no

difference.

2.1.4 Labov (1991) on Gender and Language Choice

Language use has characteristics that are peculiar to the respective genders.

According to Labov (1991), “sexual differences are institutionalized in most

languages as the grammatical category of gender”. There are two general

principles regarding sexual differentiation according to Labov.

Principle I: In stable sociolinguistics stratification, men use a higher frequency of

nonstandard forms than women.

Principle II: In the majority of linguistics changes, women use a higher frequency

of the incoming forms than men.

These are explained below:

Principle I

This principle gives an idea about what happens in a stable situation. Given any

stable situation, women seem to be more conservative and approve of variants of

language that have apparent social prestige. Men, on the other hand, use forms that

22

are perceived to be archaic and non-standard. Women are more likely to use

grammatically correct linguistic forms and also articulate more clearly than men

do. They are more cognizant and intentional in their speech utterances (Hass 1979

as cited by Stanhope-Essamuah 2005). It is believed by many researchers (Fischer

1958, Labov 1966:6, Anshen 1969, Milroy and Milroy 1978) that, women use

standard linguistic forms more than men to attain the status and power that has

been denied them by society.

In a study of the alternation between [n] and [ŋ] in the unstressed English

variable /ing/, it was discovered that, male speakers tend to use the colloquial form

[in] more than females in England (Fischer 1958). The same study was carried out

in New York City by Labov (1966) and in Australia by Bradley and Bradley

(1979). There were other such studies in many other English speaking areas.

Also, Principle I was found in operation in the study of Canadian French

for a number of variables (Thibault 1983 in Labov 1991). Mougeon and Beniak

(1987) in their study in Ontario showed that, men more often than women, borrow

core terms such as English “so” (59% vs. 41%) and also use archaic conjunctions

such as ça fait que instead of alors (68% vs. 32%). In these studies, women are

found to use standard forms associated with the highest social class and formal

speech more than men.

Labov (1991) asserts that, women are more communicatively competent

than men. They tend to use more communicative symbols to express their position.

This is the case because women possess less material power as compared to men.

As Labov (1991:214) rightly says, “in disadvantaged communities, sensitivity to

exterior standards of correctness in language is associated with upward social

23

mobility” hence women speak more standard language than men in order to rise in

the social ranking.

Principle II

This principle states that in a linguistic change, women are more likely to use the

incoming forms than men. This has been proven to be true in many cases. Gauchat

(1905) as cited by Labov (1991) showed that, women were in the lead of men in a

number of variables in the Swiss French village of Charmey. In the palatalization

of /l/, the aspiration of /ɵ/, the monophthongization of /aw/ and the

diphthongization of /o/ and /e/, women were found to be in the lead. Also, in a

study led by Guy Bailey at Texas A&M as cited by Labov (1991), 25% of female

respondents engaged in the unrounding of long open o to [a] as opposed to only

16% of male respondents. This again confirms the fact that women are the

innovators in any sociolinguistic situation of language change.

From the studies conducted by various scholars, anytime there is a new

form introduced in a language system, women are more likely than men to adopt

these new forms. It can thus be said that, in stable sociolinguistics stratification,

forms that are mostly used by women are of high prestige in that community. Also,

it can be predicted that, in the introduction of any new form or innovation in a

language system, women are more probable to adopt and use these new systems

than men.

This study will focus on finding out if these principles are true in the case

of Ewe-English code switching. The study will reveal the extent to which Ewe-

24

English bilinguals respond to these principles, and whether CS is seen as a

prestigious form or an adulteration of the language by bilinguals of Ewe.

2.1.5 Code Switching and Gender

It is believed that men and women use language in distinct gendered manner. Men

are expected to use language forms that will differ from that of women. Based on

the principles discussed earlier in this work, that women use more standard, new

and prestigious forms than men, it is expected that women will engage in CS more

frequently than men if CS is considered a prestigious form or a new form of

language use. This has been the case in many studies.

Poplack’s (1980) cited in Cheshire and Gardner-Chloros (1998) shows that

women engage in intra-sentential CS more often than men in New York Puerto

Rican community. Also, in a study conducted in Nairobi Kenya by Jarego and

Odongo (2011), it was uncovered that, women code switched more than men do.

Out of 373 instances of CS, 193 were done by females and 180 by males.

However, any time both sexes engage in a single conversation, males were found

to engage in CS more (Jarego and Odongo 2011). They further assert that, male

speakers tend to use more nonstandard forms when they are communicating with

females. This is a style they use to indicate their confidence and dominant social

position while females use standard forms in their conversations with men. The

same study indicates that, out of 333 instances of unswitched code, males and

females used 172 and 161 respectively, meaning men tend to use more archaic

forms than females who tend to use new forms.

25

In a similar study by Gulzar et al (2013), it was realized that female EFL

teachers engage in intrasentential CS more often than their male counterparts. Out

of 288 instances of intrasentential CS, males accounted for only 85 instances while

females engaged in 203 instances of CS. However, Awan and Sheeraz (2011) in

their study of English-Urdu CS, establish that male lecturers engage more often in

CS than female ones. Out of 1208 instances of CS, 805 representing 67% of the

instances were made by male respondents. They further assert that, English-Urdu

CS was seen as a less prestigious form because Urdu holds a less prestigious

position in the society in which it is used. This explains why women are found not

to engage in CS more frequently than men.

This study will focus on finding out which gender of Ewe-English

bilinguals engage in CS more often. This will help ascertain the applicability of

Labov’s two principles earlier discussed.

2.2 Methodology

This section deals with the research design and the research site. It also talks about

the research instrument, the procedures used in data collection, the sampling

method and the data analysis method used in the study.

2.2.1 Research Design

Both qualitative and quantitative research methods were employed in this study.

Qualitative research methods focus on the different interpretations of certain

phenomena in community by individual researchers. This method is used to

unearth the motives and reasons behind the behavior of people. The qualitative

26

research method is relevant to this study because the study seeks to investigate and

comprehend how pervasive CS is among Ewe-English bilinguals.

Quantitative research method on the other hand is a numerical method that

is used to analyse any phenomenon that involves precise measurement of variables.

This research method is employed to enumerate or express in numerical terms the

relationship that exists between variables in a study. This method is relevant to this

study because it helps me to quantify the occurrence of CS in the speech of

respondents. This will help establish the relationship between CS in males and

females.

2.2.2 Research Sites

The study was conducted in Ho Municipality, Ho-West District, Keta Municipality

and Ketu South District in the Volta Region of Ghana. All four locations speak

Ewe.

Ewe [eβe] belongs to the cluster of languages known as GBE. GBE is a sub–group

of Kwa languages and some of its members are spoken in Ghana, others in Togo,

Benin and Nigeria. The members of the cluster include Fon, Aja, Xwla-Xweda and

Gen. This study however focuses on Ewe spoken in Ghana, specifically the Aŋlɔ

dialect spoken in Keta Municipality and Ketu South Districts and Ho dialect

spoken in Ho Municipality and Ho-West District.

Aŋlɔ Dialect speaking area

The Aŋlɔ dialect of Ewe is spoken in about 41 towns and villages located at the

southern part of the Volta Region of Ghana (Akyeapong 2001). Most of these

towns and villages are located within the Keta municipality and others in the Ketu

27

south district of Ghana. According to the 2010 population and housing census, the

population of Keta municipality and Ketu south district is 147,618 and 160,756

respectively.

Majority of the members of these districts are native speakers of Aŋlɔ with

a minority speaking other dialects of Ewe or other Ghanaian languages. Adult

literacy for male and female in both districts stand at 72.1% and 44.0%

respectively for Keta municipal and 56.4% and 36.9% respectively for Ketu district

as at 2005 (National Population Council Fact Sheet II, 2005).

However, the absence of current data makes it difficult to predict the

current literacy rate. Based on the data available, one can conclude that many of

the residents and speakers of Aŋlɔ have had some formal education and are

exposed to varying degrees of English language. The map below shows the towns

and villages that form modern day Aŋlɔ.

FIGURE 1

MAJOR TOWNS AND VILLAGES IN CONTEMPORARY AŊLƆ

28

SOURCE: Greene (1996) as cited by E. Akyeapong (2001) History, Memory,

Slave Trade and Slavery in Aŋlɔ (Ghana) Slavery & Abolition: Journal of Slave &

Post - Slave Studies. 22: 3 1- 40 DOI: 10.1080/714005205

Ho Dialect speaking area

The Ho dialect is spoken in towns and villages in the Ho municipality and some

towns and villages in the newly created Ho West district (e.g. Dzolo, Kpedze,

Dzolokpuita, and Anyirawase). According to the 2010 population and housing

census, the Ho municipality has a population of 271,886 persons. This was the

population before the Ho West district was carved out of it. The main language

spoken in this area is the Ho dialect of Ewe. However, speakers of other dialects of

Ewe, Akan, Ga and other Ghanaian languages can be found in the Ho municipality

and Ho West district. According to the national population council facts sheet II,

literacy rate is high in the Ho municipality with 84.9% and 70.7% for male and

female respectively as at 2005. This rate includes the Ho West district since as at

the time, it was considered a part of the Ho municipality.

2.2.3 Research Instruments

2.2.3.1 Questionnaires

Questionnaires were used to collect data on the sociolinguistic background of

subjects of the research. 180 questionnaires were administered in all.

Questionnaires were filled by members of community that are native speakers of

Ewe. Key questions asked in the questionnaire include how often expressions are

mixed with English when communicating with members of the immediate family

as well as friends, the languages used when communicating with family and

friends, attitude towards CS and the educational background of respondents. These

29

provided information on the sociolinguistic and educational background of the

respondents as well as the attitude of Ewes towards CS. The questionnaire also

served as a means of locating the right candidates for the interview. Due to the

unwillingness of people to fill the questionnaires in Keta especially, I had to resort

to the snow ball method of data collection in order to get the right respondents to

fill the questionnaire. This challenge was however peculiar to respondents above

50 years in age. The questionnaire provided a means of measuring the

representativeness of the data collected from the interview. A sample of the

questionnaire can be found in the appendix.

2.2.3.2 Out - Group Interviews

Prior to my journey to the research site, I adopted the interview style used by

Amuzu (2005b). This made use of out–group interviews instead of observing and

recording subjects as they engage in in–group conversations. Out–group interviews

were conducted to ascertain the frequency of code switching among the various

genders. But upon my arrival at the research site, it was difficult getting a

monolingual Ewe speaker who will act as a collaborator. The collaborator is

expected to be a person with no formal education and could not communicate in

English. Due to this difficulty, I changed my interview style to a one on one

interview. However, persons interviewed were asked to communicate

monolingually in Ewe. The out–group interviews provided the necessary

conditions to restrict the use of code switched items.

In all, eighteen persons were interviewed, nine from each gender under study.

Persons were interviewed based on the age group in which they belonged. Each

age group had 3 representatives for each gender. Each interview lasted for 25

30

minutes due to time constraints on the collection and analysis of data. In

conducting the interview, there was an interviewer, and the subject who engaged in

series of conversations in Ewe. Also, persons interviewed were chosen on the basis

of their level of education. Amuzu (2013), mentions that the level of education of a

Ghanaian determines the degree of exposure to the English language. However,

proficiency can be influenced by other social factors. Hence only persons with a

form of tertiary education were interviewed.

Topics discussed ranged from technical or foreign concepts (such as the

election petition in Ghana, the current state of the Ghanaian economy and the use

of technology) to local or Traditional concepts (e.g. funeral rights, festivals and

chieftaincy rites). The interviewer coordinated the interview and asked the

necessary questions. The respondents responded to these questions and explained

issues raised by the interviewer as well as present their opinions on issues raised.

The medium of communication during the interview was strictly Ewe. The

interview guide used in conducting the interview is placed in the appendix.

2.2.4 Sources and Methods of Data Collection

The study draws data primarily from the questionnaire administered in the

communities under study and through the interviews that were conducted. In

administering the questionnaire, native speakers of Ewe were selected at random.

This was done to get an adequate representation of Ewe speakers for the study.

However, the population was first stratified based on sex and age before random

selection was done. Only Ewe-English bilinguals with tertiary education were

selected for the interview section. This exercise satisfied the Labovian model used

in the collection and analysis of data.

31

2.2.5 Method of Data Analysis

Purposive sampling was employed in the analysis of recordings from the interview.

This method made it possible to select only documents or variables that will best

answer the research question posed (Cresswell 1994). Hence, I purposively

sampled major questions contained in the questionnaire and only the instances of

CS in the interview.

2.3 Theoretical Framework (Labovian Sociolinguistics)

In an attempt to provide a theoretical explanation of the pervasiveness of Ewe-

English CS, William Labov’s theoretical model called “Labovian sociolinguistics”

or “Labovian paradigm” will be employed in the collection and analysis of data for

the study. The main objective of the Labovian paradigm is to “compare the texts or

people with one another” (Hudson 2001:146). This theory states that, data should

be collected from native speakers of a language. Also, in the selection of speakers,

social variables such as age, sex and social class should be considered so as to give

a clear representation of the data collected. These variables are analytical tools that

help the researcher to detect quantitative patterns. In this study, speakers were

selected based on age, sex and the dialect of Ewe spoken.

In the collection of data, tape recordings are often used (see Labov 1972:2,

Trudgill 1974, and Milroy 1980). However most of the instances in which the

Labovian paradigm is used, it is used to analyze phonetic variables. The Labovian

method is relevant to this study because this study focuses on the various ways by

32

which Ewe is used and tape recordings will play a crucial role in the analysis of

texts and identifying the instances of CS in the speech of Ewe-English bilinguals.

The Labovian approach allows for a simple method of apportioning score

to texts as a method of indicating the similarities and differences between the uses

of linguistics variables in speakers’ speech. According to Hudson (2001), a score is

derived for each variable in each section of data collected. This makes it possible

to compare texts with respect to a variable at a time. Also, scoring can be done

based on groups that are found in the study. This is to reduce the burden of large

scores that may occur from a large set of variables to be studied.

The approach also stipulates that, the location of a speaker which is referred

to as place and the race of a person can influence the variables that are used. This

was studied by Trudgill (1975/1983) as cited in Hudson (2001). In his study, it was

discovered that, the location of a person influences the use of a linguistic variable.

“The factor of race has been shown relevant by Labov and his associates in the

study of New York, working on the distinctive features of the speech of black

adolescents” (Labov 1972b:7 as cited in Hudson 2001). Therefore in the selection

of interviewees, these factors must be put into consideration.

In terms of the degree of belongingness of a person to a group, it is reported

that, the degree to which an individual belongs to a group can also influence the

pervasiveness of a linguistic variable in this study CS. According Milroy (1980) as

cited in Hudson (2001) people with extremely closed networks are more likely to

display a pervasive use of the linguistic variable than those that are in looser

relationships. Also, different sections of a community recognize different ranges of

linguistic variable that serve as a means of identification of that section.

33

In addressing the issue of sex and prestige pattern, Hudson (2001) posits

that certain factors must be taken into consideration. For instance, in some

countries, men have the upper hand in receiving formal education than women. If

the country in question is a diglossic one, then men will be more exposed to

prestige and standard forms than their female counter parts. Hence, they may use

more prestige forms than women. Also, he also claims that the variable under study

must be genuinely stratified. In selecting speakers for a comparative analysis, both

sexes must be well represented in terms of level of education, age etc. for instance,

when a female with a university education is interviewed for a comparative study,

her male counterpart must also have the same level of education and at the same

age. This makes way for a genuine comparison to be made.

Finally, the style of speech employed by anyone is dependent on the

situation in which that speech is made. As Labov (1994:157) rightly puts it, “the

speech of a person changes depending on the degree of attention that person pays

to speech forms used”. Therefore in a formal situation, a person pays more

attention to speech forms than in a casual conversation. A typical Labovian

interview must consist of sections that cater for each of these situations of speech.

This study caters for this difference in style by designing the interview in a way

that makes room for different speech styles.

2.4 Summary of Chapter

This chapter opened with a review of literature on the works that have been done

with regards to the study. Areas covered in the review include language and

gender, a differentiation between gender and sex as well as gender and language

choice. Also, literature on code switching and its motivations were reviewed. This

34

was followed by a review of literature on the gender and code switching. The

chapter then proceeded with a discussion of the methodology of the data collection

and theoretical framework to be used in data analysis.

CHAPTER THREE

SOCIOLINGUISTIC BACKGROUND OF EWE – ENGLISH BILINGUALS

3.0 Introduction

This chapter discusses the sociolinguistic background of Ewe-English bilinguals

based on their responses to items in the questionnaire administered. A detailed

analysis is first made of questions about how often Ewe-English bilinguals engage

in CS when conversing with various people ranging from brothers and sisters,

spouses, parents and children at home. This is then followed by an analysis of how

Ewe-English bilinguals use language. This section describes the languages used by

Ewe-English bilinguals in their daily interactions with friends and family and how

the various languages are employed in their daily interactions. This will help in

finding out the tendency of Ewe-English bilinguals to code switch. With all the

analysis, a gender differentiation is made between the respondents, indicating the

sociolinguistics behavior of both males and females. The chapter then closes with a

summary of the issues and findings made.

35

3.1 Demographic Information

The study was conducted among 180 respondents in the Keta and Ho

municipalities of the Volta Region of Ghana. 90 respondents were drawn from

each of the municipalities. 85 respondents were speakers of the Ho dialect of Ewe

out of this, 53 were females and 32 were males. This was closely followed by the

speakers of Aŋlͻ dialect with 71 respondents. Of the respondents who speak the

Aŋlɔ dialect of Ewe, 45 were males with 26 females. The Tͻŋu dialect speakers

were represented by 24 respondents out of which 13 were males and 11 were

females. Respondents were taken to represent each age group hence there was

equal number (i.e. 60) of respondent across each of the age groups, thus 20 – 35,

36-49 and 50+. This is shown in the table 1 below.

Table 1 Which dialect of Ewe do you speak?

Which dialect of Ewe do you speak?

Total Ho Tͻŋu Aŋlͻ

Sex of

respondent

Male 32 13 45 90

Female 53 11 26 90

Total 85 24 71 180

3.2 The use of Code Switching in the Home

The study reveals that 26 males which represent 29% of male respondents very

often mix expressions from their mother tongue (Ewe) and English when

conversing with their brothers and sisters. 42% often do same with 21% and 8%

saying they rarely and do not mix expressions when conversing with their brothers

and sisters respectively. In the case of the female respondents, 25 of them

representing 28% very often mix expressions when conversing with their brothers

and sisters. 37% of them also often mix expressions when conversing with their

36

brothers and sisters. However, 21% and 14% of the female respondents rarely and

don’t mix expressions respectively. Comparatively, males mix expressions more

often than females when conversing with their brothers and sisters.

Table 2 How often do you mix expressions from English and your mother tongue

when conversing with your brothers and sisters?

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL

MALE Very often 12 9 5 26 (29%)

Often

9 14 15 38 (42%)

Rarely 7 5 7 19 (21%)

I don’t 2 2 3 7 (8%)

TOTAL 30 30 30 90 (100%)

FEMALE Very often 12 6 7 25 (28%)

Often 11 10 12 33 (37%)

Rarely 6 5 8 19 (21%)

I don’t 1 9 3 13 (14%)

TOTAL 30 30 30 90 (100%)

Also, results from the study showed that majority of the male respondents (38%

and 46%) do not mix expression from their Ewe and English when conversing with

their mother. Only the minority do so. They represent 3% and 13% respectively.

Similarly, majority of the female respondents (30% and 39%) do not mix

expressions when communicating with their mother but this percentage is lower

37

compared to the males. 10% and 21% of the female respondents very often and

often respectively mix expressions when conversing with their mother.


when conversing with your mother?

The situation is not any different when it comes to mixing expressions when

conversing with their fathers. Of the male respondents, 13% and 16% claim to very

often and often respectively mix expressions when conversing with their fathers

with a majority of 44% saying they rarely engage in this phenomenon. Finally,

27% of the male respondents do not mix expressions when conversing with their

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL


Often 3 9 0 12 (13%)

Rarely 13 12 9 34 (38%)

I don’t 12 8 21 41 (46%)

TOTAL 30 30 30 90 (100%)


Often 7 9 3 19 (21%)

Rarely 7 9 11 27 (30%)

I don’t 7 12 16 35 (39%)

TOTAL 30 30 30 90 (100%)

38

fathers. Just as before, the majority of the female respondents also claim not to

(28%) and rarely (37%) mix expressions when conversing with their fathers.

However, these figures are lower than the male figures.


when conversing with your father?

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL


Often 5 6 3 14 (16%)

Rarely 12 12 16 40 (44%)

I don’t 8 5 11 24 (27%)

TOTAL 30 30 30 90 (100%)


Often 6 6 6 18 (20%)

Rarely 10 8 15 33 (37%)

I don’t 5 13 7 25 (28%)

TOTAL 30 30 30 90 (100%)

When conversing with their spouses, 17% of male respondents very often mix

expressions from their Ewe and English compared to 23% of the female

respondents. Also, 48% and 53% of male and female respondents respectively

often mix expressions from their Ewe and English when conversing with their

spouse. On the other hand, 13% of males say they do not mix expressions when

conversing with their spouses with only 7% of females saying so. This indicates

that females are more likely to mix expressions from their mother tongue and

39

English when conversing with their spouses. This trend is clearly elaborated in

table 5.


when conversing with your spouse?

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL


Often 5 11 13 29 (48%)

Rarely 2 2 9 13 (22%)

I don’t 1 2 5 8 (13%)

TOTAL 9 21 30 60 (100%)


Often 1 14 18 33 (53%)

Rarely 1 4 6 11 (18%)

I don’t 0 2 2 4 (7%)

TOTAL 7 25 30 62 (100%)

The study also revealed that majority of males (41% - very often and 36% - often)

mix expressions from their Ewe and English when conversing with their children

with just 21% and 2% indicating that they rarely and do not mix expressions

respectively. Similarly, majority of the female respondents (45% - very often and

33% often) mix expressions when conversing with their children. Like the male

respondents, only 21% of the female respondents rarely mix expression when

conversing with their children and 2% said they do not. The implication is that,

40

even though both males and females mix expressions from Ewe and English when

communicating with their children, females are more likely than males to engage

in the phenomenon.


when conversing with your children?

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL


Often 3 5 12 20 (36%)

Rarely 2 2 8 12 (21%)

I don’t 1 0 0 1 (2%)

TOTAL 8 18 30 56 (100%)


Often 0 8 11 19 (33%)

Rarely 0 4 8 12 (21%)

I don’t 0 1 0 1 (2%)

TOTAL 4 24 30 58 (100%)

Finally, 42% of males very often mix expressions from Ewe and English when

conversing with their friends with 29% often engaging in this phenomenon. Only

11% of the males in this study say they do not mix expressions when conversing

with their friends with 18% rarely doing this. On the other hand, more females than

males (25% - rarely and 9% I do not) do not mix expression from Ewe and

English when conversing with their friends. This implies that, males engage in CS

more often when conversing with their peers than females do.

41


when conversing with your friends?

SEX OF

RESPONDENTS

AGE OF RESPONDENTS

20 – 35 36-49 50 + TOTAL


Often 10 11 5 26 (29%)

Rarely 3 3 10 16 (18%)

I don’t 1 0 9 10 (11%)

TOTAL 30 30 30 90 (100%)


Often 9 11 8 28 (32%)

Rarely 6 6 10 22 (25%)

I don’t 1 2 5 8 (9%)

TOTAL 30 28 30 88 (100%)

3.3 Language usage of Ewe – English Bilinguals

The study revealed that, 126 out of 180 respondents speak Ewe more frequently

when conversing with their friends. This consists of 64 males and 62 females. Of

90 males that were sampled for the study, 9 speak English most frequently when

conversing with their friends with 16 using Pidgin English. Only 1 uses Hausa

most frequently when interacting with friends at home. The second language most

frequently used by males is as follows. 17 use Ewe as the second most frequent

42

language employed in conversing with friends. Also, 44 males use English as the

second most frequent language that they speak with their friends with 9 and 3

using Pidgin English and Akan respectively. For the female respondents, 22 use

English as the most frequently used language of communication when interacting

with their friends. 3 use pidgin, 2 use Akan and 1 uses Hausa. Also, 42 females use

English as the second most frequent language when conversing with their friends.

21 use Ewe, 2 use Akan and Pidgin English each. This implies that, in an

interaction that involves a male and his friends, he is most likely to use Ewe

followed by English. This is elaborated more in tables 8 and 9

Table 8 What Language do you speak to your friends? (the most frequent language

used)

Count

Language you speak to your friends

Total

Ewe English

Pidgin

English Akan Hausa

Sex of

respondent

Male 64 9 16 0 1 90

Female 62 22 3 2 1 90

Total 126 31 19 2 2 180

Table 9 What language do you speak to your friends? (second most frequent

language used)

Count

Language you speak to your friends

Total

Ewe English

Pidgin

English Akan

Sex of

respondent

Male 17 44 9 3 73

Female 21 42 2 2 67

Total 38 86 11 5 140

43

Also, majority of both males (76) and females (79) use Ewe as the most frequent

language when conversing with their brothers and sisters. 6 males use English and

6 also use Pidgin English when interacting with their brothers and sisters. Only 2

of the male respondents use Akan most frequently when interacting with their

brothers and sisters. However, 37 males use English as the second most frequent

language that they speak to their brothers and sisters, followed by 14 who use Ewe

and 8 and 1 using Pidgin English and Akan respectively. For the female

counterparts, 4 use English the most frequent language of interaction between their

brothers and sisters, 3 use Pidgin English and 4 use Akan. 44 of female

respondents use English as the second most frequent language when conversing

with their brothers and sisters, 10 use Ewe, and 3 use Pidgin English.

Table 10 What language do you speak to your brothers and sisters? (Most frequent

language used)

Language you speak to your brothers and

sisters

Total

Ewe English

Pidgin

English Akan

Sex of

respondent

Male 76 6 6 2 90

Female 79 4 3 4 90

Total 155 10 9 6 180

Table 11 What language do you speak to your brothers and sisters? (second most

frequent language used)

Language you speak to your brothers and sisters

Total Ewe English Pidgin English Akan

Sex of

respondent

Male 14 37 8 1 60

Female 10 44 3 0 57

Total 24 81 11 1 117

Furthermore, in conversing with their father, 89 male respondents use Ewe most

frequently and only one said he uses English. However, 31 males use English as

44

the second most frequent language used in interaction with their fathers. 1

respondent uses Pidgin English and another one uses Ewe as the second most

frequent language of interaction with their fathers. On the contrary, 82 of the

female respondents use Ewe as the most frequent language of interaction with their

fathers and 8 use English. For the second most frequent language of interaction, 28

females use English as opposed to 5 who use Ewe and one person who uses Pidgin

English.

Table 12 What language do you speak to your father? (most frequent)

Count

Language you speak to your

father

Total Ewe English

Sex of respondent Male 89 1 90

Female 82 8 90

Total 171 9 180

Table 13 What language do you speak to your father? (second most

frequent)

Count

Language you speak to your father

Total

Ewe English

Pidgin

English

Sex of

respondent

Male 1 31 1 33

Female 5 28 1 34

Total 6 59 2 67

The situation is even more interesting when it comes to the language they use in

conversing with their mothers. 88 females use Ewe as the most frequent language

if interaction with their mothers. Only 2 use Akan. Also, 18 use English as the

second most frequent language of interaction with their mothers, 2 and 1 use Ewe

45

and Pidgin English respectively. Of the male respondents, 89 use Ewe as the most

frequent language of conversing with their mothers and 1 uses Akan. For the

second most frequent language of interaction, 19 use English, 1 uses Ewe and

another uses Pidgin English. This implies that, more male respondents than

females use English as a language of interaction with their mothers.

Table 14 What language do you speak to your mother? (Most frequent).

Count

Language you speak to your

mother

Total Ewe Akan

Sex of

respondent

Male 89 1 90

Female 88 2 90

Total 177 3 180

Table 15 What language do you speak to your mother? (second most

frequent)

Count

Language you speak to your mother

Total

Ewe English

Pidgin

English

Sex of

respondent

Male 1 19 1 21

Female 2 18 1 21

Total 3 37 2 42

In addition, 49 out of 57 male respondents speak Ewe to their spouses more

frequently followed by 6 who use English. Also 1 male respondent uses Ga as the

most frequent language of interaction and another uses Akan. Furthermore, 37

males use English as the second most frequent language of conversation and 2 use

Ewe. This shows that, 18 males communicate monolingually with their spouses. Of

46

60 female respondents, 51 use Ewe as the most frequent language of conversation

with their spouses. 8 use English and 1 uses Ga. Also, 34 females use English as

the second most frequently used language when conversing with their spouses and

4 uses Ewe. 22 of the female respondents communicate monolingually in Ewe with

their spouses.

Table 16 What language do you speak to your spouse? (most frequent)

Count

Language you speak to your spouse

Total Ga Ewe English Akan

Sex of

respondent

Male 1 49 6 1 57

Female 1 51 8 0 60

Total 2 100 14 1 117

Table 17 What language do you speak to your spouse? (second most frequent)

Count

Language you speak to

your spouse

Total Ewe English

Sex of

respondent

Male 2 37 39

Female 4 34 38

Total 6 71 77

Finally, the study showed that 42 out of male 56 males use Ewe when conversing

with their children and 14 use English. Also, 29 use English as the second most

frequent language when conversing with their children while 14 use Ewe. On the

other hand, out of 55 female respondents, 37 speak Ewe more frequently when

conversing with their children and 18 speak English. Also, 27 females use English

as the second most frequently used language when chatting with their children

while 19 use Ewe in doing so.

47

Table 18 What language do you speak to your children? (most frequent)

Count

Language you speak to your children

Total Ewe English

Sex of

respondent

Male 42 14 56

Female 37 18 55

Total 79 32 111

Table 19 What language do you speak to your children? (second most

frequent)

Count

Language you speak to your children

Total Ewe English

Sex of

respondent

Male 14 29 43

Female 19 27 46

Total 33 56 89

3.4 Summary

This chapter analyzed the sociolinguistic background of Ewe – English bilinguals.

The chapter opened with an introduction which is followed by an analysis of the

code switching attitudes of male and female Ewe – English bilinguals. It was

revealed in the analysis that females are in the lead in mixing expressions when

conversing with their parents, spouses and their children. However, males mix

expressions more often than women when communicating with friends they

frequently interact with at home and with their siblings. An analysis of the

language usage patterns of Ewe – English bilinguals then followed. When it comes

to the language usage of Ewe-English bilinguals, males were found to use Ewe

more frequently when interacting with their friends, fathers, mothers and children

than their female counterparts. Similarly, females are more prone to the usage of

48

Ewe when communicating with their brothers and sisters, and spouses than males.

The chapter closes with an overview indicating the various issues discussed in the

chapter.

CHAPTER FOUR

GENDER VARIATION IN EWE – ENGLISH CODE SWITCHING

4.0 Introduction

This chapter compares the various variations in the CS patterns of Ewe males and

females. First a general comparison between the CS patterns of males and females

is done. This is followed by an analysis of the CS patterns of male and female

Ewe-English bilinguals among the various age groups under the study. The study

came up with data comprising both male and female conversations in an interview.

All interviews were on a one-on-one basis where an individual is interviewed at a

time. As stated in the methodology, the interview was conducted in Ewe and

respondents were expected to answer all questions in Ewe without mixing their

expressions. Despite the need to communicate monolingually in Ewe, there were

instances among both male and female respondents in which expressions were

mixed with English. There is therefore the need to analyze the degree of CS among

male and female respondents as well as among the various age stratifications of the

study. A discussion based on the results from the analysis is then done to indicate

whether CS is regarded a prestigious form or an adulteration of a language and

then the response of Ewe women with regards to the principles of Labov is

determined. The chapter concludes with a summary of topics discussed.

49

4.1 Code Switching Habits among Male and Female Ewe-English

Bilinguals.

It is generally believed that females tend to use more innovative forms than men

any time these forms are introduced. The attitude towards CS among the Ewes has

been noted to be negative. According to Yevudey (2009), majority of Ewe- English

bilinguals have a negative attitude towards CS. This negative attitude is also seen

in the survey made by this study on the prestige being enjoyed by CS among Ewe

speakers. However, the phenomenon is still pervasive among the various age ranks

and gender of Ewe-English bilinguals.

According to this study, there were 717 instances of CS recorded in about

487 minutes of recorded data. Out of the 717 instances of CS, males recorded 316

which is about 44% of CS instances recorded. Female Ewe-English bilinguals

accounted for 401 instances of CS. This represents 56% of all instances of CS

recorded by the study. This implies that, there exists a higher level of

pervasiveness of code switching among females than males. This confirms Jarego

and Odongo (2011)’s claim that, females tend to mix expressions more than males.

I would like to indicate that most of the instances of CS used by males were

inter-sentential with a few instances of tag switching and intra-sentential CS. Some

instances of inter sentential CS of males are illustrated below. Codeswitched items

are in bold. ‘I’ represents the interviewer and ‘R’ the respondent. Also, the Ewe

transcription is presented at the left with the English translation to the right.

50

I: Vovo toto kae le ku kɔnu wo

wɔwɔme le egbe ŋkekewo me kple

blema me?

R: Vovo toto geɖe le ku kɔnuwo

wɔwɔ me. Tsã ne ame kua, ɖeko woa

ƒoƒu ko wo wͻ ku… they don’t do

all these things… fifia ya la, ele too

larvish.

I: What differences exist in how

funeral rites are conducted in these

days and in the past?

R: There are many differences in

funeral rites. In the past if a person

dies, people gather and funeral rites

are performed… they don’t do these

things… now it is too larvish.

I: Ame aɖe wo gblͻ be tohehe si

wona amemawo me le eteƒe o, wohã

nesusu nenemaa?

R: Nya maa, it doesn’t hold any air.

Mélé ya aɖeke o

I: Nukae nye woƒe nukpɔkpɔ le

akɔdanya hehe le υɔnudrɔƒe gã le ƒe

eve si vayi me? Ɖee nyahehe sia gblȇ

nu le dukɔa ŋutia?

R: Le miaƒe democracy nu sinye

government for the people a, enyo

be woyi υɔnudrɔƒe but egagblȇ nu le

mia ŋu le ganyawo gome. Ga geɖe yi

gbe.

I: Ame aɖewo be vovo toto siawo na

be deviwo le gbegblem. Nukae nekpɔ

tso nya sia ŋuti?

R: …ke woadi be woa find the cause

of death…fifia ɖeviwo le nusiawo

katã teƒe kpɔm ena be dzo sasa le

ɖeviwo me vam.

I: Some people say that the

punishment met out to those people

are undeserving, do you share in this?

R: That issue, it doesn’t hold any

air. It doesn’t hold any air.

I: what is your view on the election

petition at the Supreme Court last

two years? Did this cause any

damage to us as a nation?

R: According to our democracy

which is government for the people,

it is a good thing that they went to

court but it caused a lot of damage to

51

us in terms of money. So much

money went to waste.

I: Some people are of the view that,

these changes spoil children. What is

your take on this?

R: …and they will like to find the

cause of death…these days children

witness all these things and it is

causing them to go into spiritualism.

There are many other instances where this pattern is realized among male

respondents in the study. In both male and female CS, 142 nouns were

codeswitched followed by verbs that accounted for 87 instances of CS. 79

conjunctions and 58 adverbs were codeswitches. An analysis of the recorded data

indicate that 47 instances of adjectives, 35 instances of interjections and 27

instances of adjuncts were codeswitched. Finally, 21 instances of adverbial phrases

and 18 instances of adjectival phrases were codeswitched by Ewe-English

bilinguals.

In the case of female respondents, most instances of their CS were intra

sentential with a few instances of inter sentential CS and tag switching. Instances

of intra sentential CS showing the different grammatical categories codeswitched

is indicated below. Just as before, codeswitched items are in bold. ‘I’ represents the

interviewer and ‘R’ the respondent. The Ewe transcription is presented at the left

with the English translation to the right.

I: Vovo toto kae le ku kɔnu wo

wɔwɔme le egbe ŋkekewo me kple

blema?

R: Nye mekpɔ difference aɖeke le

kunu wɔwɔ…

52

I: Nukae hea hunyahunya dea amewo

dome le fiaɖu nyawo me?

I: What differences exist in how

funeral rites are conducted in these

days and in the past?”

R: I haven’t seen any difference in

funerals…”

I: What brings disagreements

between people when it comes to

issues of chieftaincy?

R: Enye mekpɔ be title-a ta wo wɔa

dzre le fiadunyawo me.

I: Nuka nesrɔ tso akɔda nya hehea

me le ʋɔnudrɔƒe gã?

R: Mesrɔ nu geɖee. Gba mesrɔ be wo

dumevi wo mateŋu atso ko aƒonu o

because ese kple ɖoɖo le dukɔame…

ne eƒonu anyhow ko woa punish

wo…

I: Le keke sia wo me, miese be gana

ga le gbɔɖim ne etsɔe sɔ kple dukɔ

bubuwo ƒe ga eyata, gadzraoƒe gã da

afɔtoƒe aɖewo ɖe anyi be gadzraoƒe

wo katã na zɔ edzi be woateŋu atsi

wɔna sia nu. Nukae nekpɔ tso afɔtoƒe

sia wo ŋuti?

R: I think it is because of the title

that people disagree over chieftaincy

issues.

I: What did you learn from the

election petition at the Supreme

Court?

R: I learnt a lot. Firstly, I learnt that

as a citizen you cannot talk anyhow

because there are rules and

regulation in the country…if you talk

anyhow you will be punished.

I: These days, we are told that the

Ghanaian currency is depreciating

when compared with the currencies

of other countries so the central bank

has given some directives for all

other banks to follow in order to end

53

this event. What is your take on these directives?

I: Aleke wo wɔna fiaɖokɔnuwoe le

nutosiame?

R: ...there and then ko wógba ɣé ɖe

ta nawo

I: How are chieftaincy rites

performed in this area?

R: …there and then they will pour

powder on your head.

R: …the way ega nyawo va le yiyim

la, ehia vevie be wóawɔ numawo.

…tsã la minimum amount ke anɔ

account me nawo enye five cedis

gake fifia ezu fifty cedis… and nusi

mekpɔ be enyo enye be…

R: …the way issues about money are

going, there is the urgent need for

them to do those things.…in the past,

the minimum amount required in

your account is five cedis but now it

is fifty cedis… and what I think is

good is that…

54

4.2 Gender Variation in Code Switching Patterns among Age Groups of

Ewe-English Bilinguals.

Another important way CS patterns can be looked at is through the various age

ranks as stratified for the study. There are variations in CS patterns among the

various age groups also. Out of the 717 instances of CS recorded by the study,

respondents between the ages of 20 – 35 accounted for 294 instances. This

represents 41% of all instances of CS recorded. By implication, younger Ewe-

English bilinguals engage in CS more often than the older ones. Also, within this

age group females accounted for 172 (59%) instance of CS and males accounted

for 122 (42%) instances of CS. This implies that in conversations, females are

more likely to switch code than males indicating the degree of pervasiveness of CS

among female Ewe-English bilinguals.

There were 278 instances of CS recorded for respondents between the ages

of 36-50. This figure represented 39% of all instances of CS recorded in the study.

Among this age stratification too, males recorded 112 instances of code switching

which represents 40% and females recorded 166 representing 60% of CS instances

recorded by the age group. This indicates that, among this age stratification also,

females are more likely to engage in CS than males and CS is more pervasive

among females of the age group than males.

For the respondents who are 50+, they recorded 145 instances of CS which

represents 20% of CS recorded by the study. Females of this age group accounted

for 63 instances of CS which represents 44% and males accounted for 82 instances

which is 56% of CS recorded by the age group. The implication of this is that

contrary to the claims of Jarego & Odongo (2011), males of this age group are

more susceptible to engage in CS than females. This result also confirms the stance

55

of Awan & Sheeraz (2011) that males engage in CS more than females. This trend

may be because more males than females are educated and speak English.

4.3 Code Switching as a Prestigious Form or an Adulteration of a

Language

The question of whether code switching is a prestigious form or an adulteration of

one’s language has been a major argument among scholars. The study revealed

that, there still exists a negative attitude towards CS among Ewe-English

bilinguals. However, the degree of negative attitude has declined. This implies that,

CS is gradually gaining prestige among Ewe-English bilinguals. This assertion is

based on the survey made on the attitude of Ewe-English bilinguals towards the

use of CS in their interactions compared with an earlier survey made by Yevudey

(2009). Even though the same questions were not asked, questions from both

surveys seek to determine the attitude of Ewe speakers towards CS.

According to Yevudey (2009), majority (65%) of Ewe-English bilinguals

have a negative attitude towards CS. This figure confirms Forson’s (1979) claim of

attitude towards Akan-English CS. In my study however, only 48% of respondents

have a negative attitude towards CS with 42% having a positive attitude. Also,

11% of respondents are indifferent. Across the two genders, CS has not enjoyed a

lot of positive attitude. Both majority of males and females have a negative attitude

towards CS.

The current trend of attitude towards CS is figuratively presented in table 20.

56

Table 20 Which of the following best describes your perception of code switching as a

style of speech?

Which of the following best describes your perception of

code switching as a style of speech? Total

It

indicates

high level

of

education

It shows

one’s

level of

modernity

It is just

another

way of

language

use

It shows

incompetence

in the use of

Ewe

It

adulterates

the Ewe

language

Sex of

responde

nt

Male

24

13

11

24

18

90

Female 18 20 8 29 15 90

Total 42 33 19 53 33 180

Similarly, majority of Ewe-English bilinguals (i.e. 83 out 180 respondents) say that

CS should be stopped while 75 out of 180 respondents say CS should not be

stopped. 22 respondents are indifferent to whether CS should be stopped or not.

This goes further to confirm the earlier assertion that CS is enjoying an increase in

the degree of prestige attached to it. CS is gradually becoming a prestigious form

among Ewe-English bilinguals.

Table 21 Should people stop mixing expressions from Ewe and English when

conversing?

Should people stop mixing expressions from Ewe and

English when conversing?

Total Yes No No Opinion

Sex of respondent Male 38 36 9 90

Female 45 39 13 90

Total 83 75 22 180

It can be deduced that even though Ewe-English bilinguals have a negative attitude

towards CS, it is gaining some prestige over the years considering the number of

57

people that have a negative attitude towards it five years ago and currently. More

people are developing positive attitudes towards CS.

The question about why majority of people have a negative attitude towards

CS but the phenomenon is still pervasive in their language usage patterns comes up

at this point. This can be rightly answered based on what Swigart (1992) called

practice and perception. According to Swigart (1992), there exists contradictions

between what people say they do and what they actually do. As Labov (1972)

rightly observed, speakers may be unconscious of their usage of low prestige

phonological trends and only come to be appreciative of their phonic intentions

instead. “Speakers ‘codify’ their ideas about of how language should ideally be

used by perceiving real language use in an (unconsciously) categorical manner”

(Swigart 1992:48). Speakers have an idea of how a particular language should be

used but they fail in using language in this ideal way. Ewe-English CS is enjoying

some form of covert prestige among Ewe-English bilinguals. This is seen in the

pervasive use of CS among the various genders and age stratification of Ewe-

English bilinguals. CS is used in the day-to-day interactions of Ewe-English

bilinguals yet they explicitly deny their engagement in the phenomenon.

4.4 Sociolinguistic Behaviour of Ewe Women in Relation to the Principles

of Labov

According to the first principle, men are more likely to use nonstandard forms

when communicating than women. This implies that if CS is regarded as a

nonstandard form of Ewe, men should be found using more of it than women.

“Women use more standard forms, responding to the overt prestige associated with

them” (Labov 1991:210). A lot of evidence to this effect has been provided in

58

chapter two. However, in this study, females were found generally to engage in CS

more often than males even though CS does not enjoy any overt prestige. The

implication of this is that Ewe women in general do not conform to the first

principle postulated by Labov (1991). They rather act contrary to that effect. This

can however be attributed to the practice and perception analogy made by Swigart

(1992). She asserts that there is a contradiction between what people say and what

they actually do. Thus even though there is an overt expression of negative attitude

towards CS, there exists a covert positive attitude towards it. This explains why

females are in the lead in the usage of CS. It can therefore be concluded that in

practice Ewe women do not conform to the first principle of Labov but based on

perception they rightly conforms to this principle. Despite the general implication,

females in the age group of 50+ were found to engage in CS less often than males.

This implies that, among the older female Ewe-English bilinguals, there is

conformity to the first principle of Labov.

Furthermore, the second principle postulated by Labov (1991) states that,

whenever there is a linguistic change, women are found to be in the lead in the use

of these new forms introduced in the language. This implies that, women tend to

use more of the incoming prestigious forms than men. Before indicating the level

of conformity to principle two, it is worth noting that, CS is a new form introduced

into the Ewe language. This can be proven first from the level of pervasiveness of

CS among the age stratifications of the study. It was observed earlier that, CS is

most pervasive among respondents between the ages of 20-35 followed by

respondents in the age ranges of 36-50 and then those in the range 50+. This

indicates that CS is more common among the younger generation.

59

Secondly, speakers of Ewe have been in existence and communicating

monolingually before colonial days. It was until colonial times that Ewe-English

bilinguals emerged. It can therefore be concluded that, Ewe-English code

switching is a current trend among speakers. Based on the second principle

postulated by Labov (1991) and the earlier assertion that Ewe-English CS is a new

form, it can be predicted that, Ewe women will be in the lead in the use CS. This

was confirmed by the study as females generally use CS more than males.

4.5 Summary

This chapter captured a quatitative analysis of the findings of the study with

regards to gender variations in Ewe-English code switching. It opened with an

introduction, which was followed by an analysis of the code switching patterns

among male and female Ewe-English bilinguals. It then touched on gender

variations in Ewe-English code switching among the age groups as stratified for

the study followed by an analysis of the attitude of Ewe-English bilinguals towards

CS. The chapter ended by indicating the response of Ewe women to the principles

of Labov.

60

CHAPTER FIVE

FINDINGS, CONCLUSION AND RECOMMENDATIONS

5.0 Introduction

The use of CS has become pervasive in the speeches of Ghanaian bilinguals. Ewe-

English bilinguals have not been left out of this growing trend. Although, this

linguistic phenomenon has led to many varying opinions and attitudes, it continues

to gain prominence in all fields of interpersonal communication. The study of CS

has received a lot of attention from scholars including Dzameshie (1996), Yevudey

(2009), and Amuzu (2005 and 2012).

This present study has aimed at studying gender variation in Ewe-English

CS. The study employed the Labovian sociolinguistics in the collection and

analysis of data. The main research questions sought to find out the gender of Ewe-

English bilinguals that engage in the phenomenon more. The study examined the

response of Ewe women to the paradox of CS in relation to the two principles on

language use among males and females by Labov (1991).

The following sections provide a summary and an outline of key findings

of the research. We conclude by providing some recommendations for future

related studies are provided.

5.1 Sociolinguistic Background of Ewe-English Bilinguals

This study came up with an analysis of the sociolinguistics background of Ewe-

English bilinguals. From the data, it was revealed that females are in the lead in

mixing expressions when conversing with their parents, spouses and their children.

61

Males on the other hand mix expressions more often than women when

communicating with friends they frequently interact with at home and their

brothers and sisters. An analysis of the language usage patterns of Ewe – English

bilinguals showed that when it comes to the language usage of Ewe-English

bilinguals, males were found to use Ewe more frequently when interacting with

their friends, fathers, mothers and children than their female counterparts.

Similarly, females are more susceptible to the usage of Ewe when communicating

with their brothers and sisters, and spouses than males.

5.2 Gender Variation in the Use of Codeswitching.

It is an agreed fact among leading authorities in the study of language and gender

that gender makes a difference in the use of language. These variations can also

occur between different age stratifications. This present study revealed that,

women more often than males use CS in their day-to-day interactions. Of the 717

instances of CS, females accounted for 56% instances of CS compared to 44%

instances of CS recorded by males. It was also revealed that majority of males CS

is inter-sentential with the rest being instances of intra-sentential CS and tag

switching. Females on contrary tend to engage more in intra-sentential CS than

inter-sentential and tag switching.

In terms of age stratifications, respondents were categorized into various

age groups (i.e. 20-35, 36-49 and 50+). Respondents between the ages of 20-35

accounted for the highest instance of CS (41% of all recorded instances of CS).

They were followed by respondents between the ages of 36-49, who recorded 39%

of the instances of CS. Respondents of the 50+ age stratification recorded the

62

lowest instances of CS (20%). Female respondents among the various age groups

except those in the group 50+ recorded the highest instances of CS. This goes

further to confirm the assertion that females codeswitch more than males.

5.3 Attitude towards Codeswitching among Ewe-English Bilinguals

Many Ewe-English bilinguals have a negative attitude towards the phenomenon

even though they pervasively engage in it. This study gathered that majority of

both male and female Ewe-English bilinguals have negative attitudes towards CS

and want the practice to be stopped. Even though attitude towards CS is negative,

the phenomenon is gradually gaining prestige among Ewe-English bilinguals. This

was discovered when a comparison was made between the present data and an

earlier data used by Yevudey (2009) to investigate attitude towards CS.

Also, it was realized that even though speakers openly indicate negative

attitude towards CS, they tend to engage in the phenomenon. This trend of

behaviour is explained with the practice and perception analogy postulated by

Swigart (1992) which indicates that there exists contradictions between what

people say they do and what they actually do. It can therefore be said that CS is

enjoying covert prestige among Ewe-English bilinguals while overtly it not

regarded as a prestigious form.

5.4 Reaction of Ewe Women to the Principles of Labov (1991).

It was also discovered that females generally engage in CS more often than males

even though CS does not enjoy any overt prestige. This implies that Ewe women in

general do not conform to the first principle postulated by Labov (1991). But based

on the practice and perception analogy by Swigart (1992), it can be said that Ewe

63

women in practice do not conform to the first principle of Labov but in perception

they conform to the first principle of Labov. This is becuse CS seem to enjoy

covert prestige among Ewe-English bilinguals. Despite the earlier discovery,

females that belong to the age stratification 50+ conform to the first principle

postulated by Labov (1991). Based on the data gathered by the study, Ewe women

were found to be in the lead in the use of CS. This confirmed the second principle

postulated by Labov (1991) that whenever there is a linguistic change, women are

found to be in the lead in the use of these new forms introduced into the language.

5.5 Recommendations

I would like to recommend that a deeper study of the attitude of Ewe-English

bilinguals towards CS be made. That study should not be based only on the

response given by respondents but also by participant observation. Such multiple

approaches will minimize the margin of error in the study.

Also, a comparison between the various dialects of Ewe spoken in Ghana should

be made. This will throw more light on the dialectal variations in CS (i.e. if any

exists) and the forms they take.

In conducting future studies into CS using interviews, these interviews should

range between 45 minutes to 2 hours. This will help collect a more natural data

than shorter interviews since respondents tend to relax averagely about fifteen

minutes into the interview.

64

APPENDIX I: QUESTIONNAIRE

I am a level 400 student of the University of Ghana, Legon. I am reading Bachelor

of Arts Linguistics and working on a research with the title “Gender variation in

Ewe – English Code switching”. This questionnaire asks you a number of

questions designed to ascertain your bilingual experience especially at home. The

questions range from your language of communication with your immediate family

to how other people use languages they know to communicate.

The information you provide will be kept highly confidential. Please complete the

questionnaire for your answers are very important to me. If you are interested, the

results of the research will be discussed with you at the end of the study.

Section 1

Please tick the appropriate answer.

1. Age 20 – 35 36 – 49 50 and above

2. Sex MALE FEMALE

3. Level of education.

JHS SHS University

Other (specify)……………………...

4. Do you live with your parents, brothers and sisters? YES NO

Do you live with your spouse and/or child (ren)? YES NO

5. How many languages do you speak?

………………………………………………

65

Please list them

i. ………………………………………………………………

ii. ……………………………………………………………....

iii. ………………………………………………………………

iv. ………………………………………………………………

6. What is your mother tongue?……………………………………………….

7. How many years in the last ten (10) years will you say you spent at the

following places.

a. Your Hometown……………… b. Outside your Hometown …………

c.Your dialects speaking area…………………………………..…..………..

d. outside your dialect speaking area……………………….………………..

8. a. Name one (1) town in which you spend most of this time………………..

b.Was this time spent typically at school?.......................................................

If yes what school……….……...…………………………........…………....

9. Which dialect of Ewe do you speak?

Ho Tɔŋu Aŋlɔ other(s) ………..……….…….

10. Which of the following best describes your perception of codeswitching as

a style of speech?

a. It indicates high level of education

b. It shows one’s level of modernity

c. It is just another way of language use

66

d. It shows incompetence in the use of Ewe

e. It adulterates the Ewe language.

11. Should people stop mixing expressions from Ewe and English when

conversing?

Yes No No Opinion

12. What language(s) do you speak to the following individuals at home?

Please read the following guideline carefully

i. If you speak one language with any of the under listed individuals

write 1 in the box against the language.

ii. If you speak more than one language with the person, write [1] in

the box against the language you speak more frequently to him/her,

[2] for the second most frequent language you speak with him/her

etc.

a. To your brothers and sisters.

Ga Ewe English Pidgin English

Daŋme Akan Other please specify…………….…….

b. To your father.


Daŋme Akan Other please specify…………….……

c. To your mother.



67

d. (if applicable) To your spouse.



e. (if applicable) To your child(ren)



f. To friends you frequently interact with at home.



13. What language(s) does the following people use more frequently to address

you at home?

Please tick the box against the appropriate answer.

a. Your mother.



b. Your father.



c. Your brothers and sisters.



d. (If applicable) your spouse.



68

e. (If applicable) your child (ren).



f. Friends you frequently interact with at home.



14. From your point of view, which one language is used most frequently by

majority of people

a. In your home……………………….… b. In the market ……………..…...

c. At your work place………………………………………………….….......

d. In school (outside the classroom)……..……………………………………..

Section 2

Some people mix expression from more than one language when they

converse. More often English and a Ghanaian language. This section asks

questions about your personal experience with this phenomenon.

15. How often do you mix expressions from English and your mother tongue

(or the Ghanaian language you mostly use) when conversing with the

following people?

a. Your brothers and sisters at home.

Very often Often Rarely I don’t

b. Your mother.


69

c. Your father.


d. (If applicable) your spouse.


e. (If applicable) your child (ren).


f. Friends who speak English and your mother tongue.


Thank you for completing this questionnaire.

70

APPENDIX II: INTERVIEW GUIDE

1. Vovo toto kae le ku kɔnu wo wɔwɔme le egbe ŋkekewo me kple blema?

i. Nuka vovo toto sia wo ateŋu agblȇ le miaƒe dekɔnuwo ŋuti?

ii. Ame aɖewo xɔese be kuwɔwɔ le ŋkeke siawo me le gbegble he va

ɖeviwo dome nyatefe enye nya sia? Nukata?

2. Aleke wo wɔna fiaɖokɔnuwoe le nutosiame?

i. Vovo kawoe le fiadonkɔnuwo le nutosiame kple nuto bubuwo dome?

ii. Nukae hea hunyahunya dea amewo dome le fiaɖu nyawo me?

iii. Nuka miateŋu awɔ be hunyahunya siawo na dzudzɔ.

3. Nukae nye woƒe nukpɔkpɔ le akɔdanya hehe le υɔnudrɔƒe gã le ƒe eve si vayi

me?

i. Ɖee nyahehe sia gblȇ nu le dukɔa ŋutia?

ii. Nunyui kawoe wo he ve name abe dukɔ ene?

iii. Le nyahehea me wo he ame υe aɖewo ɖe υɔnudrɔƒea ŋgɔ be woda le

υɔnudrɔea dzi eye wohe to nawo, nukae nekpɔ tso nenem tohehe ŋuti?

Ame mawo dze na tohehea? Nukae nesrɔ tso woƒe tohehe me?

4.Le keke sia wo me, miese be gana ga le gbɔɖim ne etsɔe sɔ kple dukɔ bubuwo ƒe

ga eyata, gadzraoƒe gã da afɔtoƒe aɖewo ɖe anyi be gadzraoƒe wo katã na zɔ edzi

be woateŋu atsi wɔna sia nu. Nukae nekpɔ tso afɔtoƒe sia wo ŋuti?

i. Afɔtoƒe siawo mele nugblem le dukɔa ŋuti oa? Nukata?

71

5. kaƒomɔ zãzã ɖee, nunyui wonyo vatu mia alo ele nugblem le miaŋu? Nukata?

i. enyo be ɖeviwo alo sukuviwo le miaƒe titiname sukuwo me na zã

kaƒomɔa? Nukata?

72

References

Amuzu Evershed Kwasi. 2013. Producing composite code switching: The role of

the modularity of language production. International Journal of

Bilingualism. Published online 22nd April, 2013.

Amuzu, Evershed Kwasi. 2012. Socio-pragmatics of conversational code switching

in Ghana. Ghana Journal of Linguistics 1(2), 1-22.

Amuzu, Evershed Kwasi. 2005a. Ewe-English Code switching: A Case of Composite

rather than Classic Code switching. Ph.D. thesis. Australian National

University, Canberra.

Amuzu, Evershed Kwasi. 2005b. Code switching in Ghana: Still a ‘third tongue’ of the

educated? Legon Journal of the Humanities XVI, 27-53.

Asilevi, Francis. 1990. English-Ewe Code-mixing in Conversational Discourse: A

case of English as a second language in Ghana. MA thesis. University of

Ghana, Legon.

Awan, Safer and Sheeraz, Muhammad. 2011. Gender Oriented Code Switching: A

Case Study of English Language Teachers in Pakistani Universities.

International Journal of Academic Research 3 (4),

Bullock, Barbara E. and Toribio, Almeida Jacqueline. 2009. Themes in the study

of code switching. In Bullock Barbara E. and Toribio Almeida Jacqueline

(eds.), The Cambridge Handbook of Linguistic Code-switching. 1-17

Cambridge University Press .

73

Chan, Brian Hok-Shing. 2009. Code switching between typologically distinct

languages. In Bullock Barbara E. and Toribio Almeida Jacqueline (eds.),

The Cambridge Handbook of Linguistic Code-switching, pp. 182-198.

Cambridge University Press

Cheshire, Jenny and Gardner-Chloros, Penelope. 1998. Code switching and the

sociolinguistics gender pattern. International Journal of the Sociology of

Language 12 (9), 5-34.

Clopper, Cynthia G. 2004. Linguistic Experience and Perceptual Classification of

Dialect Variation. Ph.D. thesis. Indiana University.

Dzameshie, Alex K. 2001. Toward a Global Explanation of Unmarked

Codeswitching: Evidence from Ewe-English Bilingual Codeswitching.

Reading Working Papers in Linguistics 5, 1-21.

Eckert, Penelope and McConnell–Ginet, Sally. 2003. Language and gender.

Cambridge: Cambridge University press.

Forson, Barnabas. 1979. A description of language situations classified in a corpus

representing a bilingual register range (Akan varieties and English). M.A.

thesis. University of Ghana, Legon.

Gulzar, Malik Ajmal, Farooq, Muhammad Umar and Umer, Muhammad. 2013. Inter-

sentential patterns of code switching: a gender-based investigation of male and

female EFL teachers. International Education Studies 6 (11), 144-159

74

Ibhawaegbele, Faith O. and Edokpayi, Justina N. 2012. Codes switching and code

mixing as stylistic device in Nigerian prose fiction: A study of three

Nigerian novels. Research of Humanity and Social Sciences 2 (6), 12-18

Jarego, Nelson and Odongo, Eston Kwach. 2011. Patterns and motivations of code

switching among male and female in different ranks and age groups in

Nairobi, Kenya. International Journal of Linguistics 3 (1), 1-13

Kroch, Anthony S. 1978. Towards a theory of social dialect variation. Language

and Society 7 (1), 17-36.

Labov, William. 1990. The intersection of sex and social class in the course of

linguistic change. Language Variation and Change 2, 205-254.

Labov, William. 1966. The social stratification of English in New York City,

Washington. DC: Center for applied linguistics.

Mougeon, Raymond and Beniak, Edouard.1987. The extralinguistic correlates of

core lexical borrowing. In K. M. Denning (ed) variation in language.

NWAV-XV at Stanford, pp 337-347. Stanford, CA: Stanford University

Press.

Myers Scoton, Carol. 1982. Possibility of code switching: motivation for

maintaining multilingualism. Anthropological linguistics 24 (4), 432-444.

Namba, Kahuziko. 2004. An overview of Myers Scoton’s Matric language frame

model in Educational research Bullitin Senri International School. http://

www.sisf.minoh.osaka.jp/research/re9/namba.pdf>. February 11, 2014.

http://www.sisf.minoh.osaka.jp/research/re9/namba.pdf

75

Nartey, Jonas N. A. 1982. Code-Switching, interference or faddism? Language use

among educated Ghanaians. Anthropological Linguistics 24 (2), 83-192.

Shitemi, Nora L. 2009. Language and gender. A lecture given at IUPW Fullbright hays

group projects abroad program. Moi University.

Stanhope–Essamuah, Nana Tweikuwa. 2005. The language of women and its

characteristic features. How different is it from men? B.A. long essay.

University of Ghana, Legon.

Sunderland, Jane. 2006. Language and gender an advanced resource book. Milton:

Routlege.

Swigart, Leigh. 1992. Practice and Perception: Language Use and Attitudes in

Dakar. Ph.D Thesis. Seattle: University of Washington.

Talbot, Mary. 2010. Language and gender second edition. Cambridge: Polity

press.

Tannen, Deborah. 1990. You Just Don’t Understand: Women and Men in

Conversation. New York: Ballantine Books.

Zimmerman Don H. and West, Candace. 1975. Sex roles, interruptions and silences in

conversation. In Barrie Thorne and Nancy Henley (eds.) Language and Sex:

Difference and Dominance. pp 105-129 MA: Newbury house.

Gender Variation in Ewe-English Codeswitching

Documents