University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA, LEGON THE SOCIO-PRAGMATICS OF CODE-SWITCHING IN WHATSAPP GROUP CHATS AMONG GHANAIANS BY OPOKU-FOFIE SYLVIA (10551037) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF MASTER OF PHILOSOPHY DEGREE IN ENGLISH. DEPARTMENT OF ENGLISH UNIVERSITY OF GHANA, LEGON JULY, 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION I declare that this thesis is the result of my original research, except for references to other works which have been duly acknowledged. I also declare that no part has been presented as part of the requirement for any degree in any university. Date……………………………………… ………………………………………….. Sylvia Opoku-Fofie (Candidate) Date………………………………….. Date……………………………………. ……………………………………….. ………………………………………… Prof. Tayo Lamidi Dr Evershed Kwasi Amuzu (Supervisor) (Supervisor) i University of Ghana http://ugspace.ug.edu.gh DEDICATION This thesis is dedicated to the founder of the Moments of Glory Prayer Army (MOGPA), Rev Isaac Osei Bonsu, for investing in my education. ii University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGMENTS I give thanks to the Almighty God for His grace and mercies for my life and how far He has brought me. I also give God the glory for bringing into my life my able supervisors, Professor Tayo Lamidi and Dr E. K. Amuzu, who have been a blessing to me. I am so grateful to them for their encouragement, support and constructive criticisms which shaped this thesis to its present quality. You were always there to help explore every avenue that helped to make this thesis the best. God bless you all. I also thank Rev Isaac Osei Bonsu, the brain behind my going back to school to pursue this degree. The many efforts you put into my education without which this would not have happened, is very much appreciated. May God bless and recompense you for every dime you spent on me. I also acknowledge Michael Owusu Aduomi and Jacob Kusi, my research assistants, for their love, support and contributions to making the completion of this work successful. May you also achieve higher heights in life. God bless you all. Finally, I acknowledge my siblings and friends, especially Francis Mintah Dadzie and Dennis Boamah Boateng, who encouraged me in my pursuit of this work. I have really learned and benefitted greatly from the experiences we shared. iii University of Ghana http://ugspace.ug.edu.gh ABSTRACT Though English is widely considered as the language of the internet, the internet is currently multilingual since the practice of code-switching has found its way into online communication especially, WhatsApp. In this study, the socio-pragmatics of code-switching in WhatsApp chat groups among Ghanaians is examined by (a) scrutinizing the motivations for Ghanaians’ choice of language on WhatsApp (b) investigating the nature of code-switching (c) investigating the relationship between a social variable like gender and particular switching used on WhatsApp. The study focuses its analysis on five homogenous chat groups and one heterogeneous chat group. It also adopts a variationist methodology which includes participant observation and interview surveys in order to identify chatters’ motivations for switching codes during WhatsApp group conversations. Also, extracted data from WhatsApp chat conversations are used as part of the data collected for the analysis. Using the Community of Practice theory, the Markedness Model and Variationist Sociolinguistics Approach as the theoretical backdrop, the study reveals that Ghanaians adopt novel ways of writing in order to switch codes on WhatsApp. They employ code-switching as a mechanism for improving their communication. Further, code-switching used in WhatsApp conversations bear semblance to spoken code-switching. Besides, this study establishes that the use of code-switching in WhatsApp group chats depends on the particular community of practice that chatters may find themselves in. Also, the study proves that the gender of chatters causes disparity in terms of language choice and language use in WhatsApp group chats. The findings of this research set the beginning of the investigation of language contact in WhatsApp conversations in Ghana since there has not been any known study on code- switching in WhatsApp interactions in the Ghanaian community. iv University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION ........................................................................................................................................... i DEDICATION .............................................................................................................................................. ii ACKNOWLEDGMENT .............................................................................................................................. iii ABSTRACT ................................................................................................................................................. iv TABLE OF CONTENTS .............................................................................................................................. v LIST OF FIGURES ................................................................................................................................... viii LIST OF TABLES ....................................................................................................................................... ix ABBREVIATIONS ...................................................................................................................................... x CHAPTER ONE ........................................................................................................................................... 1 INTRODUCTION ........................................................................................................................................ 1 1.0 BACKGROUND TO THE STUDY ................................................................................................... 1 1.0.1 THE LINGUISTIC SITUATION IN GHANA ............................................................................ 4 1.0.2 THE WHATSAPP APPLICATION ............................................................................................ 7 1.0.3 Language use on WhatsApp ........................................................................................................ 8 1.2 PROBLEM STATEMENT ............................................................................................................... 11 1.3 AIMS AND OBJECTIVES .............................................................................................................. 12 1.4 RESEARCH QUESTIONS............................................................................................................... 13 1.5 ORGANIZATION OF THESIS........................................................................................................ 13 CHAPTER TWO ........................................................................................................................................ 14 LITERATURE REVIEW, THEORETICAL FRAMEWORK AND METHODOLOGY ......................... 14 2.0 INTRODUCTION ............................................................................................................................ 14 2.1 LITERATURE REVIEW ................................................................................................................. 14 2.1.1 Bilingualism ............................................................................................................................... 15 2.1.2 Code-switching Research ........................................................................................................... 16 2.1.2.1 Blom and Gumperz (1972) ................................................................................................. 17 2.1.2.2 Myers-Scotton (1993b) ....................................................................................................... 19 2.1.2.3 Auer (1998) ......................................................................................................................... 19 2.1.3 The Socio-pragmatics of CS in Ghana ....................................................................................... 20 2.1.3.1 Forson (1979) ...................................................................................................................... 20 v University of Ghana http://ugspace.ug.edu.gh 2.1.3.2 Albakry & Ofori (2011) ...................................................................................................... 21 2.1.3.3 Amuzu (2012) ..................................................................................................................... 21 2.1.3.4 Quarcoo & Amuzu (2016) .................................................................................................. 22 2.1.4 Language and the Internet .......................................................................................................... 24 2.1.4.1 Computer-mediated Communication .................................................................................. 24 2.1.4.1.1 WhatsApp Messenger .................................................................................................. 26 2.1.5 Written and Spoken Text Genre ................................................................................................. 28 2.1.6 Code-switching research in CMC .............................................................................................. 30 2.1.6.1 Georgakopoulou (1997) ...................................................................................................... 30 2.1.6.2 Siebenhaar (2006) ............................................................................................................... 31 2.1.6.3 Goldbarg (2009) .................................................................................................................. 31 2.1.6.4 Marjie (2010) ...................................................................................................................... 32 2.1.6.5 Halim and Maros (2014) ..................................................................................................... 33 2.1.6.6 Anchimbe (2015) ................................................................................................................ 33 2.1.6.7 Amuzu, Kuwornu & Opoku-Fofie (Forthcoming) .............................................................. 34 2.2 THEORETICAL FRAMEWORK .................................................................................................... 35 2.2.1 Community of Practice .............................................................................................................. 35 2.2.2 Markedness Model ..................................................................................................................... 40 2.2.3 Labovian Sociolinguistics/Variationist Sociolinguistics ............................................................ 46 2.2.4 Justification for the three models ............................................................................................... 48 2.3 METHODOLOGY AND DATA COLLECTION ............................................................................ 49 2.3.1 Selection of WhatsApp Chat groups .......................................................................................... 49 2.3.1.1 WhatsApp Data ................................................................................................................... 50 2.3.1.1.1 WhatsApp Chat groups ................................................................................................ 54 2.3.1.1.1.1 Ahwiaa R/C & L/A Old Students WhatsApp Chat group ..................................... 54 2.3.1.1.1.2 PENSA KNUST Organa Alumni Family WhatsApp Chat group ......................... 55 2.3.1.1.1.3 MyChisco WhatsApp Chat group ......................................................................... 55 2.3.1.1.1.4 Zighiss 2009 WhatsApp Chat group ..................................................................... 56 2.3.1.1.1.5 Fraser Boys WhatsApp Chat group ....................................................................... 56 2.3.1.1.1.6 ND 1-14 WhatsApp Chat group ............................................................................ 56 2.3.1.2 Interviews ............................................................................................................................ 57 2.3.1.3 Other methods ..................................................................................................................... 58 2.3.2 Ethical Issues ............................................................................................................................. 58 vi University of Ghana http://ugspace.ug.edu.gh 2.4 SUMMARY ...................................................................................................................................... 59 CHAPTER THREE .................................................................................................................................... 61 QUALITATIVE ANALYSIS OF LANGUAGE USE ............................................................................... 61 3.0 INTRODUCTION ............................................................................................................................ 61 3.1 ANALYSIS OF CODE-SWITCHING STRUCTURES ................................................................... 61 3.1.1 Posts on WhatsApp .................................................................................................................... 61 3.1.1.1 Forwards ............................................................................................................................. 61 3.1.1.2 Original Posts ...................................................................................................................... 65 3.1.2 Types of Code-switching ........................................................................................................... 65 3.1.2.1 Tag-switching ..................................................................................................................... 65 3.1.2.2 Inter-sentential CS............................................................................................................... 66 3.1.2.3 Intra-sentential CS............................................................................................................... 68 3.1.2.4 Single Word Switches as Discourse Markers ..................................................................... 71 3.1.2.4.1 Functions of Akan Discourse Markers in English Constructions ................................ 74 3.1.2.4.1.1 Saa ......................................................................................................................... 74 3.1.2.4.1.2 Wai ........................................................................................................................ 75 3.1.2.4.1.3 ooo ......................................................................................................................... 76 3.1.2.4.1.4 Koraa ..................................................................................................................... 77 3.1.2.4.1.5 Papa ....................................................................................................................... 78 3.1.2.4.1.6 Paa ......................................................................................................................... 79 3.2 Marked versus Unmarked codes on WhatsApp ................................................................................ 80 3.2.1 Unmarked CS on WhatsApp ...................................................................................................... 81 3.2.1.1 English as the unmarked code choice on WhatsApp .......................................................... 81 3.2.1.2 Pidgin as the unmarked choice on WhatsApp ..................................................................... 82 3.2.1.3 CS as an unmarked code choice on WhatsApp ................................................................... 85 3.2.1.3.1 Akan-English CS as used to display multiple identities in WhatsApp group chats ..... 87 3.2.1.3.2 CS as used to negotiate the development of a conversation ........................................ 88 3.2.2 Marked CS on WhatsApp .......................................................................................................... 91 3.2.2.1.1 Factors motivating chatters’ use of code-switching in the WhatsApp group chat ....... 95 3.2.2.1.1.1 Inability to express thoughts in English ................................................................ 95 3.2.2.1.1.2 Quoting someone else ........................................................................................... 99 3.2.2.1.1.3 Being Emphatic ................................................................................................... 101 3.2.2.1.1.4 Humour ............................................................................................................... 103 vii University of Ghana http://ugspace.ug.edu.gh 3.3 INTERVIEWS ................................................................................................................................ 105 3.3.1 Inability to find an appropriate expression in the unmarked code ........................................... 106 3.3.2 Switching to create humour ..................................................................................................... 107 3.3.3 Switching for convenience sake ............................................................................................... 109 3.3.4 Switching to include and exclude others in a conversation ..................................................... 110 3.3.5 Switching in order to accommodate others .............................................................................. 111 3.4 SUMMARY .................................................................................................................................... 112 CHAPTER FOUR ..................................................................................................................................... 114 VARIATIONIST ACCOUNT OF LANGUAGE USAGE....................................................................... 114 4.0 INTRODUCTION .......................................................................................................................... 114 4.1 VARIATIONS IN CS STRUCTURES ACROSS CHAT GROUPS.............................................. 114 4.1.1 Distribution of the Types of CS across the WhatsApp chat groups ......................................... 114 4.1.2 Distribution of Discourse Markers found of the data ............................................................... 120 4.1.3 Code choice by Gender ............................................................................................................ 124 4.2 SUMMARY .................................................................................................................................... 127 CHAPTER FIVE ...................................................................................................................................... 129 SUMMARY AND CONCLUSION.......................................................................................................... 129 5.0 INTRODUCTION .......................................................................................................................... 129 5.1 SUMMARY .................................................................................................................................... 130 5.2 SUMMARY OF FINDINGS .......................................................................................................... 133 5.3 FUTURE RESEARCH ................................................................................................................... 136 REFERENCES ......................................................................................................................................... 138 APPENDICES .......................................................................................................................................... 153 6.1 APPENDIX 1 .................................................................................................................................. 153 6.1.1 SNAPSHOTS OF THE STATEMENT OF INFORMED CONSENT .................................... 153 6.2 APPENDIX 2: Interview Responses ............................................................................................... 159 6.3 Appendix 3: WhatsApp Data .......................................................................................................... 170 LIST OF FIGURES Figure 1:Emoji Collection Fig. 1. Emoji Collection ................................................................................... 10 Figure 2: WhatsApp Extracted data ............................................................................................................ 48 viii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1: Distribution of CS in WhatsApp chat ......................................................................................... 115 Table 2: frequency of types of CS across chat groups .............................................................................. 118 Table 3: Distribution of DMs across chat groups ..................................................................................... 121 Table 4:Frequency of Akan Discourse markers found in English constructions ...................................... 122 Table 5: code level among males and females in mixed WhatsApp chat groups ..................................... 124 Table 6: code level between Fraser Boys and Zighiss .............................................................................. 126 ix University of Ghana http://ugspace.ug.edu.gh ABBREVIATIONS CS = code-switching CMC = Computer-mediated Communication DM = Discourse Markers MM = Markedness Model CoP = Community of Practice x University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.0 BACKGROUND TO THE STUDY Communication via digital media arrived in the lives of many people during the 1990s. The internet is currently the world’s leading computer network which provides a variety of services to its users. It has been explained by Crystal (2004:1) as “an association of computer networks with common standards which enable messages to be sent from any central computer (or host) on one network to any host on any other”. Since the internet has become a global medium of communication due to the intensity of its use, it has had an effect on language. Moreover, it has been a major field of research in the study of language since it offers easy access to data which is already transcribed. Another reason could be that though the English language is globally considered as the language of the internet because people from different nationalities use it as a common code of communication through the Web (Crystal, 2004), the internet is currently multilingual. It is this multilingual characteristic of the internet, for example, the use of code- switching that has aroused the attention of bilingualism and language contact experts. The phenomenon, code-switching, has been variously described as the alternation of words and phrases between two languages or dialects. For instance, Weinreich (1953:73) defines code- switching as “the practice of alternately using two languages”. Myers Scotton (1993b) describes code-switching as “alternations of linguistic varieties within the same conversation”. Despite the several definitions that have been given by various scholars, code-switching would be described as a change from one language to another within a single utterance or sometimes within a sentence in this study. 1 University of Ghana http://ugspace.ug.edu.gh Code-switching (CS) has aroused significant interest among scholars for the past several decades. However, most studies have analysed code-switching on the basis of its oral production (Gumperz, 1982; Auer, 1984; Myers-Scotton, 1993; Li Wei & Milroy, 1995; Amuzu, 2012), with few on its written production (Paolillo, 1996; Georgakopoulou 1997). The use of verbal code-switching has become a usual activity for bilinguals to the extent that it is not only used in informal domains but also in formal domains. Currently, code-switching is used in many domains. For instance, it has become a common code in classroom settings (Asilevi, 1990; Anyidoho & Dakubu, 2008; Olusegun, 2012) and church settings (Andoh, 1997; Albakry & Ofori, 2011; Asare-Nyarko, 2012). It has also been pervasively used in students’ academic discussions (Obiri-Yeboah, 2008; Quarcoo, 2013, and Quarcoo & Amuzu, 2016), in radio discussions (Yevudey, 2009; Brobbey, 2015) and radio and television advertisements (Anderson & Wiredu, 2007). Although CS in Computer-mediated Communication (CMC) evidently qualifies as written in terms of the written representation of linguistic signs, it also bears resemblance to spoken conversational CS, most obviously in terms of its dialogic context and its discourse functions. A social media network like the WhatsApp application has enabled Ghanaians to alternate codes during private and group chat sessions. This application is quasi-synchronous in that it may allow for participants of a group to be online at the same time in order to produce a discussion while there are times that not all participants may be online during a discussion but may give their views on issues when they come online. In view of this, this study focuses on the interactions of members of WhatsApp chat groups. Usually, members of WhatsApp chat groups see themselves as constituting communities which make them establish their own group culture. 2 University of Ghana http://ugspace.ug.edu.gh This is because many of the people in the chat groups are church members, past students of a school reminiscing their old school days, youth in a church, work colleagues, etc. Much of the literature on CS occurring in CMC shows that many scholars usually study traditional pre-web modes of interpersonal CMC (such as Internet Relay Chat (IRC), email, mailing lists, and Usenet groups) complemented by quantitative analysis of language choice while only a few have focused on CS occurring in chats which are quasi-synchronous. Moreover, these studies which have been conducted on CS occurring in CMC have been done on different languages in other countries and it will be beneficial for such studies to be done in the Ghanaian community. Yet, CS usually occurs on many computer-mediated platforms although the keys on the keyboard of computers and phones do not represent many of the letters of the alphabet of the local languages in Ghana. Nevertheless, Ghanaians go to the extent of using numbers in conjunction with some letters just to write in their local languages. Also, technology has advanced to the extent that some Play Store Applications (apps) have been developed to help Ghanaians to get access to keyboards that would enable them chat in their local languages on their phones with ease. These apps include Kasahorow keyboard, Ka Twi keyboard, Ghana keyboard and Chale keyboard. What they have to do is to download these apps onto their phones, tablets or other communication gadgets so they can conveniently write in their local languages. However, not many Ghanaians are aware of the existence of these keyboard apps. As a result, this current study examines the socio-pragmatics of code-switching in the WhatsApp group chats among Ghanaians. 3 University of Ghana http://ugspace.ug.edu.gh 1.0.1 THE LINGUISTIC SITUATION IN GHANA Africa is noted for its diversity of languages. With the approximately 6700 languages that exist in the world, about 30% of these are found in Africa (Bamgbose, 2011). The existence of many languages has resulted in bilingualism and multilingualism. The bilingualism and multilingualism in the Ghanaian context is not different from that of other African countries. The 16th edition of Ethnologue lists about seventy-nine (79) living indigenous languages (belonging to Proto-Niger Congo language groups) spoken by a population of about 29 million (as at 2013) in the country. Nine of these languages are government-sponsored and are supported by the Bureau of Ghana Languages, which was established in 1951, and publishes materials in them. These languages are Akan, Dagaare, Dangbe, Dagbane, Ewe, Ga, Gonja, Kasem and Nzema. According to the Ghana Statistical Service on the 2010 population Census, Akan is the dominant language in the country with about 47.5% of the population speaking it as a mother tongue and with about 40% of Ghana’s population speaking it as a second language/Lingua Franca (Ghana Statistical Service, 2017). Akan was used as a broad term in the 1950’s to represent a cluster of dialects: Akuapem, Asante, Fante, Wasa, Agona, Akyem, Bono (Abron), Kwawu, Gomua and Bono (Obeng, 1997: 64). Moreover, Akan is the most extensively documented and has the largest amount of written texts of all the Ghanaian languages. Ga and Ewe are other vehicular languages in the country. In addition to the local languages that co-exist in Ghana, there is also Pidgin which “has been considered an alien phenomenon” in Ghana (Dako, 2002: 54). Pidgin has a low status in Ghana and usually associated with illiteracy (Dako, 2002: 54). However, there is a second variety of pidgin which is known as Student Pidgin which is spoken by males in secondary and tertiary 4 University of Ghana http://ugspace.ug.edu.gh schools. This kind of Pidgin is used by the males in those educational settings for expressing their interactional needs and their in-group solidarity. Although these languages co-exist with English in Ghana, the English language is the de facto lingua franca of the country. English was introduced to Ghana with the arrival of the Europeans as slave traders and colonizers. After independence, there was the problem of choosing one local language over the others to be the official language of the country. Therefore since the English language had played a prominent role in the history of the country, it was preferred as the official language of the country in order to forestall the possibility of uprising sentiments among Ghanaians when one indigenous language was chosen as official language of the country over the others (Sey, 1973). Therefore, the English language was seen as a neutral language. In Morris’ (1998) interview with Prof J. A. Sackey, a lecturer at the University of Ghana, he (Prof J. A. Sackey) said that “English brought (Ghanaians) all together under one umbrella”1. Thus, English aided communication between the ethnic groups in the country. The English Language is the language of government, administration, legal procedures, and official communication. It is also the language used in Ghanaian politics, education, entertainment, and the media as well as the medium of instruction in schools. Besides the English language, the teaching and learning of French have been promoted by the Ministry of Education in conjunction with the French Embassy since the 1980s (Associates for Change, 2010). It is the country’s second foreign language. However, only few Ghanaians can communicate in this international language. Because of the language contact situation in Ghana where the English language is associated with formal education and seen as a prestigious code, code-switching between English and the 1 Interview of J.A Sackey, Lecturer of University of Ghana, by Morris on 27 November 1998 in Cape Coast. 5 University of Ghana http://ugspace.ug.edu.gh Ghanaian local languages has become a normal practice among Ghanaian bilinguals. In Forson’s (1979) study, he observed that code-switching was restricted to informal speech domains since any formal gathering required an unmixed code. Current studies (Georgakopoulou, 1997; Siebenhaar, 2006; Mvan Gass, 2008; Goldbarg, 2009; Halim & Maros, 2014) have, however, revealed that this habit of alternating between languages is not restricted to only verbal communication but is also found in informal written communication. With the advancement in science and technology, many gadgets have been invented to aid humans in their everyday life. The WhatsApp application is one of the many inventions created to help users in communication. Tagoe (2013) claims that WhatsApp has developed as the de facto mode of communication via the digital media in the Ghanaian community. This application works on specific android phones (smart phones) and on some computers. Since the arrival of this mobile messaging application in Ghana, it has become very popular in the country and has also reduced traditional phone calls and text messaging among Ghanaians. Due to its popularity, many Ghanaian consumers consider the availability of this app on mobile phones before they buy them. Currently, it is the most commonly used app in Ghana (Tagoe, 2013). Therefore, the general assumption is that Ghanaians would stick to the English language for chats on the WhatsApp platform since it is associated with formal education. In addition, WhatsApp chat is also cheaper than text messages and calls. However, this is not the case since it is not uncommon to find some Ghanaians alternating codes especially, between English and Akan during chat sessions. As a result, the present study examines the use of CS on WhatsApp group chats among Ghanaians. It also explores the social motivations behind the use of CS on this platform and how members of chat groups make particular code choices in order to show their solidarity and familiarity with one another. 6 University of Ghana http://ugspace.ug.edu.gh 1.0.2 THE WHATSAPP APPLICATION WhatsApp is a social media application that is usually used on specific phones and computers and operates with internet connectivity. With the introduction of smart phones into the system, WhatsApp has become a popular messaging service. This smartphone application functions on almost all recent kinds of devices and operating systems. This application was produced by Brian Acton and Jan Koum with the sole aim to making communication easier and faster. With its inception, it was tagged “Simple. Personal. Real time messaging” (Jisha & Jebakumar, 2014: 2). It has been on the market since 2009 (Sánchez-Moya & Cruz-Moya, 2015). However, this application receives periodic updates which help to improve its functionality since its release date. WhatsApp is one of the most commonly used individual messaging phone applications that allow individuals to relate and connect with one another in all parts of the globe as well as share their feelings and thoughts in a fraction of a second. Though it is used as a means of sending and receiving messages to and from individuals or groups, it contains a range of purposes such as text messages, attached images, audio files, video files, GIF (Graphics Interchange Format) files, contacts, locations, documents and links to web addresses. This mobile application boasts of over 800 million users throughout the world as of 2014 and was sold to Facebook for $19 Billion (Karpisek, Baggili, & Breitinger, 2015). In the year 2015, a calling feature was introduced in the version 2.11.552 (Arce, 2015). Recently in 2016, a video calling feature was introduced in the version 2.16.318. This feature is available to all users on Android, iOS and Windows devices (MacGoogan, 2016). Recently, WhatsApp has added a ‘quote and reply’ feature in the version 2.16.7.1 that allows users to select specific messages and respond to them (Brown, 2016). However, this app is very addictive and can influence its users. 7 University of Ghana http://ugspace.ug.edu.gh This application allows its users to provide personal information and create their own digital profile (Sánchez-Moya & Cruz-Moya, 2015:53). In creating their digital profile, users are prompted to include a picture, a nickname and a status. Also, WhatsApp offers social information to its users; for example, contacts can see when their friends are online, when they are typing and when they last accessed the app. Moreover, it offers delivery notifications, and highlights when a message is sent and when it is delivered to the recipients’ phone (Church & de Oliveira, 2013). Green ticks are used to convey delivery information to the sender. One tick indicates that the message has been sent while the two ticks show that the message has been conveyed to the receiver’s phone. However, two green ticks imply that the message has been read by the recipient. Though persons on WhatsApp can write messages to their individual contacts, they can also create groups to share information. As already stated, the main research site for this thesis is the WhatsApp group chat. The group chat feature was developed to enable communities of users with a common interest to interact and stay in touch with family and friends. This feature allows at most 256 people in a chat group. Therefore, it gives the user the opportunity to create unlimited number of groups. 1.0.3 Language use on WhatsApp Devic (2008:32) argues that the language use on the internet displays the principle of economy of words which aids and speeds up the typing process; thus serving as time savers. In view of this, the language used on WhatsApp chat displays some features that are characteristic of verbal communication (Sánchez-Moya & Cruz-Moya, 2015). There is the use of colloquial and sociolectal vocabulary. Usually, the kind of language used during chat sessions is familiar and common to the parties who engage in the chatting. Therefore, it is very common for members of 8 University of Ghana http://ugspace.ug.edu.gh a particular age to conventionalize their own slang expressions and jargons that would be meaningful to them. Also, discourse particles are very characteristic of WhatsApp chats (Devic, 2008:32). These discourse particles are characteristic items of verbal language discourse. Examples of such discourse particles are well, yes, yeah, oh, ah, okay, uh, ooo and um. These have various functions. They can signal turn-taking, organise the conversation and show interactive closeness of those communicating. Moreover, discourse particles can indicate new topics for discussion, mark important information and show speaker attitude (Schiffrin, 1987:24). Moreover, speech-like constructions such as assimilations and elisions are found within messages sent on WhatsApp (Devic, 2008:32). Though elision and assimilation are noted for occurring in spoken discourse, they are usually present in chats. Examples include: gerrout (get out), bfdae (birthday), getogeda (get together), c’mon (come on), dassor (that’s all) etc. Also, numerous techniques are applied which are supposed to compensate for a person’s inability to depict pitch of voice, level of inarticulateness and the nonverbal cues (facial expressions, gestures) which are key constituents of human communication in spoken discourse (ibid). Among them are the emoticons (emoji) that typically portray stylised facial expressions, which in their simplest forms express either delight or displeasure. Emoticons have been a vital part of this chat application to convey feelings and several other thoughts which would otherwise need many words to express. Note that these features mentioned above are not peculiar to WhatsApp but are widely used in CMC. A collection of these is shown in Figure 1 below: 9 University of Ghana http://ugspace.ug.edu.gh Figure 1: Emoji Collection (image source: http://shutterstock.com/) The bare infinitive of the verb and a variety of acronyms are also used to express nonverbal actions (Dürscheid & Stark, 2011). Examples of some acronyms that are commonly used are Lol (laughing out loud), imho (in my humble opinion), ATM (at this moment), B4 (before), BBL (be back later), B/C (because), BF (boyfriend), smh (shaking my head), WTG (We thank God), Idk (I don’t know), Ukr (you know right), OMG (Oh my God) etc. Another technique is the means of emulated prosody which is about the repetition of letters of the alphabet (pleeeeease), punctuation (help!!!!!) and rendering of entire words in capitals (WILL YOU COME) just to vividly realise the means of emphasis and to stress what is being communicated. Sometimes, however, punctuation marks may be omitted. All these techniques are sole properties of verbal communication. 10 University of Ghana http://ugspace.ug.edu.gh 1.2 PROBLEM STATEMENT Even though the English language is considered the de facto lingua franca of the internet, many users of the Web are not native speakers of English. Nevertheless, the research works on Computer-mediated Communication (CMC) do not reveal this multiplicity. Since code-switching is a common linguistic practice among bilinguals, it has easily found its way into communication via digital media. Code-switching in this field is considered an area for linguistic research. Consequently, a social media network such as WhatsApp has given room for some Ghanaians to switch codes during private and group chat sessions depending on their linguistic abilities and their intended communicative needs; thus, Ghanaians innovatively use languages in varying degrees to which there is the need for such switches to be investigated. Notwithstanding its global popularity, Church and de Oliveira (2013) have observed that interactions on WhatsApp have not received much attention from scholars. Also, code-switching in computer-mediated communication has not been previously investigated in the Ghanaian context; there has not been a known research on CS in Ghanaian WhatsApp messages. Therefore, this study investigates the socio-pragmatics of code-switching in WhatsApp group chats among Ghanaians. An investigation into the socio-pragmatics of CS occurring on WhatsApp would enable the study to explain the factors that account for members of chat groups to switch codes. Also, it is through this that the functions of CS on WhatsApp can be identified. The WhatsApp platform was chosen for this study because it is a popular and the fastest growing global social networking tool with over 1.2 billion monthly active users as at January 2017 (Statista, 2017). This study investigates CS in CMC because CMC as a discourse field contests the supposition that only verbal face-to-face communication is the essential site for code- switching (Androutsopoulos, 2013: 668). Moreover, CMC offers opportunities for written CS at 11 University of Ghana http://ugspace.ug.edu.gh an unprecedented scale. The reason for studying CS in WhatsApp group chats is due to its use on the platform even though English is the expected norm. CS on this platform ranges from the use of English to the indigenous languages and pidgin. The English language is the expected language choice on WhatsApp and also, affords chatters ease in its usage besides being the language of formal education. Anyidoho and Dakubu (2008) have observed that English and Akan are possibly the two prominent languages in Ghana. Akan is the most dominant local language in Ghana, spoken as a first language by almost 47.5% of the population and also as a second language or as a lingua franca by at least 40% of the remaining Ghanaian population (Ghana Statistical Service, 2017). Akan is also used as a lingua franca across ethnic groups (Adika, 2012). Apenteng and Amfo (2014: 219) further claim that “the Akan language has had a long contact with the English language from the time of colonization up to date”. Therefore, it is very common for Ghanaians to switch between Akan and the English Language. Pidgin is a language typically used as a means of expressing solidarity and therefore since members in these chat groups see themselves as constituting communities that have certain commonness; they may use language patterns that would bring out their in-group loyalty. 1.3 AIMS AND OBJECTIVES This study aims to study code-switching in some selected Ghanaian WhatsApp chat groups and primarily seeks to achieve the following objectives: 1. To find out the motivation for Ghanaians’ choice of language on WhatsApp 2. To investigate the nature and types of switches being used on WhatsApp 3. To examine the relationship between social variables (e.g. gender) and particular linguistic variables like discourse markers and specific switching patterns on WhatsApp 12 University of Ghana http://ugspace.ug.edu.gh 1.4 RESEARCH QUESTIONS In examining code-switching employed by Ghanaians in their group chats on the WhatsApp messenger, the study will attempt to answer the following central questions: 1. What is the motivation for code-switching on Ghanaian WhatsApp platform? 2. What is the nature of code-switching on Ghanaian WhatsApp platform? 3. What are the specific switch patterns on Ghanaian WhatsApp platform? 4. What are the social and pragmatic functions of CS on the Ghanaian WhatsApp platform? 5. What is the relationship between social variables (e.g. gender), and particular linguistic variables like discourse markers and specific switching patterns on the Ghanaian WhatsApp platform? 1.5 ORGANIZATION OF THESIS This thesis is made up of five chapters. Chapter one presents the study by setting the background through which the discussion of the linguistic situation in Ghana as well as the research rationale and questions are grounded. Subsequently, chapter two provides a review of relevant and related literature. It discusses the research done on spoken and written code-switching occurring in computer-mediated communication. Further, it gives a detailed discussion of the theoretical framework namely, Community of Practice theory, the Markedness Model and Variationist Sociolinguistics and the methodology used in this thesis. Chapter three discusses the data in relation to the occurrence of the code-switches. It also discusses the factors, reasons and social motivations of code-switching on the WhatsApp platform with the Markedness Model as theoretical background. Chapter four discusses the data in relation to linguistic variations associated with social categories like gender. This analysis is done with Variationist Sociolinguistics as theoretical backdrop. Chapter five concludes the study and presents the summary, major findings and recommendations. 13 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW, THEORETICAL FRAMEWORK AND METHODOLOGY 2.0 INTRODUCTION This chapter is made up of three sections. The first section reviews some literature on code- switching. It begins with a thorough discussion and explanation of some relevant terms and concepts. The second section discusses the theoretical frameworks on which the study is grounded. Finally, the third section of this chapter provides the methodology used in selecting the chat groups and collecting the data. It also discusses the ethical issues that needed to be addressed before and during the data collection. 2.1 LITERATURE REVIEW Code-switching (CS) as an area of linguistic investigation has attracted the attention of many scholars across the globe. Many researchers have endeavoured to determine the social motivations and reasons behind this phenomenon while others have looked at it from a grammatical perspective. In this section, some related works on CS are presented and examined. Also, several aspects of language and the internet and most importantly, the focus of this study, WhatsApp and CS in Computer-mediated Communication (CMC) are discussed. 14 University of Ghana http://ugspace.ug.edu.gh 2.1.1 Bilingualism Bi/multilingualism is usually the product of language contact. Language contact results when speakers of different languages come into contact with one another (Li Wei, 2000). Over the years, this phenomenon has attracted the attention of many language experts. The prevalence of bilingualism across the globe shows that many people usually employ two or more languages in their daily activities. The increase in bilingualism in the world can be attributed to globalization since people have recognized the benefits of adding another language to their language (Bhatia & Ritchie, 2004). Mackey (2000: 22) defines bilingualism as “the alternate use of two or more languages by the same individual”. This definition buttresses his argument that bilingualism is an asset of an individual and not the community. This is because a person’s use of two languages presumes the existence of two different language societies; however it does not presume the reality of a bilingual community (Mackey, 2000). Conversely, Mackey’s (2000) argument is at variance with Myers-Scotton (1993a) and Bhatia & Ritchie’s (2004) view that a speech community can be considered bilingual because languages develop alongside the growth of human societies. Hence, a community can be easily identified by a particular language, norms, beliefs and practices peculiar to it. In Africa, specifically West Africa, bilingualism is an asset because of the many languages that co-exist. The co-existence of these languages provides speakers with several linguistic choices. Hence, many people have adapted themselves to social settings where more than one language is spoken. This has generated communicative strategies such as code-switching and borrowing which have become the fundamental topics of bilingualism study. 15 University of Ghana http://ugspace.ug.edu.gh 2.1.2 Code-switching Research Code-switching, a phenomenon in sociolinguistics, has attracted the attention of linguists for several decades. Yet the spelling of the term is not consistent and not agreed upon by linguists. It is so easy to find the forms codeswitching, code switching as well as code-switching all referring to the same phenomenon with no difference in meaning. Milroy and Muysken (1995: 12) point out that the Research Network of the European Science Foundation (ESF) has made deliberate attempts to standardise the terminology in order to avoid the confusion that these terms convey. However, these attempts have failed because different authors may choose to deal with these terms in different ways. In order to maintain consistency in the spelling of the phenomenon in this study, the term code-switching would be used. The issue of terminological confusion is not only restricted to the spelling of the label for the phenomenon. This is because not all linguists agree on the areas covered by the terms code- switching, code-mixing, borrowing or code-alternation. The major issue is the perceived difference between code-switching and borrowing (Poplack, 1980; Myers-Scotton, 1992) on one hand, and code-switching and code-mixing on the other hand (Belly, 1976; Sridhar & Sridhar, 1980; Kachru, 1983; Bokamba, 1988). Despite these confusions, some scholars such as Poplack (1980), Eastman (1992), Myers-Scotton (1993b), Milroy and Muysken (1995) use code-switching as a cover term under which they differentiate between switches that occur between sentences and those that occur within sentences. For instance, Poplack (1980) has distinguished between three types of code-switching: tag-switching, intra-sentential code-switching and inter-sentential code-switching. Also, Milroy and Muysken (1995:7) explain CS as “the alternative use by bilinguals of two or more languages in the same conversation”. Auer (1995), in contrast, employs the term code-alternation as the 16 University of Ghana http://ugspace.ug.edu.gh superordinate term for code-switching and transfer. To him, code-alternation refers to “a relationship of contiguous juxtaposition of semiotic systems, such that the appropriate recipients of the resulting complex sign are in a position to interpret the juxtaposition as such” (Auer, 1995:116). However, alternation refers to the occurrence of one code being changed to another code halfway through a discourse. For the scholars (Belly, 1976; Sridhar & Sridhar, 1980; Kachru, 1983; Bokamba, 1988) who maintain a distinction between code-switching and code-mixing, code-switching is reserved for switches at the clausal boundary while code-mixing is reserved for switches within clauses. This is because, to them, the switches that occur within clauses involve the incorporation of the rules of the languages involved. In this vein, many theories have been developed to explain the sociolinguistics of code-switching phenomenon. Below is a discussion of some major studies done on the sociolinguistics of CS: 2.1.2.1 Blom and Gumperz (1972) The linguistic anthropologist Gumperz championed the beginning of a new approach to code- switching research which sought to look at why bilinguals switch codes and what exactly they mean by this linguistic behaviour. He stressed the necessity to study the social function of language usage. In view of this, Blom and Gumperz (1972) explored code-switching between standard and vernacular varieties of Ranamål (a northern Norwegian dialect) and Bokmål (the standard language of official transactions, religion, and the mass media) in a naturally occurring conversation in the small town of Hemnesberget, Northern Norway. They concluded that although there are linguistic similarities between the two languages (Ranamål and Bokmål), the local people thought of them as distinct codes, which was mainly because of the different social 17 University of Ghana http://ugspace.ug.edu.gh roles they performed (local cultural identity versus formal education/official transactions, religion, mass media). They then coined the terms situational and metaphorical switching. Situational switching arises when “within the same setting the participants’ definition of the social event changes” (Blom and Gumperz, 1972: 424). In their study, situational switching was marked by a move from Ranamål to Bokmål and vice versa, and concurrently by a change in channel cues such as speed, rhythm, amount of pauses and so forth. Metaphorical switching, alternatively, “allows for the enactment of two or more different relationships among the same set of individuals” (Blom & Gumperz, 1972: 425). This implies that metaphorical switching does not include an essential change of the speech event. Blom and Gumperz (1972) contributed to the research into code-switching since it was they who first expounded on the insight that code-switching is nothing less than skilful communicative action which can convey social meaning and can help achieve communicative goals. However, their theory has been criticized as being out-dated since it emphasizes the lack of volitional action on the part of the speaker during an interaction. Also, Pride (1979) argues that situational switching and metaphorical switching are inconsistent in relation to their definitions. To him, there seem not to be a difference between the situational and the metaphorical switching as Blom and Gumperz (1972) described. Further, Myers-Scotton (1983b) contends that Blom and Gumperz did not discover the similarities between these two types of CS. Maehlum (1996) also disputes that in Norway, there seems to be nearly no distinction between Bokmål and Ranamål as being suggested by Blom and Gumperz. 18 University of Ghana http://ugspace.ug.edu.gh 2.1.2.2 Myers-Scotton (1993b) Carol Myers-Scotton (1993b) also brings another dimension into the study of the social motivation of CS. By analysing bilingual conversations taken from Kenya and Zimbabwe, she offers a theoretical viewpoint to the study of the socio-pragmatics of CS behaviour. In this theory, Myers-Scotton argues that there is an innate ‘markedness evaluator’ in speakers that includes a cognitive ability to evaluate markedness. Unlike the early attitudes towards CS behaviour, she proves that CS is a skilled behaviour that could be used to express intended meanings rather than linguistic insufficiencies. Moreover, this model by Myers-Scotton can be applied to major occurrences of CS behaviour. Though Myers-Scotton recommends the Markedness Model (MM) which argues that bilinguals are competent in the languages they speak but their choice for a particular code is determined by the need to index to a set of Rights and Obligations required of that particular code, the model has been criticised. Auer (1998:9) criticizes the MM model for obviously denying the applicability of a Conversational/Sequential Approach to the study of the social motivations of CS. 2.1.2.3 Auer (1998) Auer (1998) emphasizes an interactional perspective of language alternation. Thus in his study, he attempts to apply a sequential analysis approach to language alternation as a means for interactional meaning. This shows that the sequence of a bilingual conversation is independent “both from grammar (syntax) and from the larger societal and ideological structures to which the languages in question and their choice for a given interactional episode are related” (Auer, 1998:5). In Constance, West Germany, he researched into the conversations of younger children of Italian immigrant workers with Italian background. The conversations produced by the 19 University of Ghana http://ugspace.ug.edu.gh children were characterised by alternations between German and Italian (dialect). The analysis led to the conclusion that the recognition of a linguistic choice as marked is not automatically predetermined by conventionalized societal norms and values. Alternatively, such messages are ‘locally produced’ in the discourse structure of the specific interaction. Hence depending on the specific discourse, the same linguistic choice may as well express different meanings for speakers and addressees. Finally to him, the sequence of a discourse is ‘sufficiently autonomous’ and offers the basis for ascertaining the exact shared meaning intended in every bilingual conversation. 2.1.3 The Socio-pragmatics of CS in Ghana 2.1.3.1 Forson (1979) In Ghana, the scholarly works on the socio-pragmatics of CS began in the 1970’s. Forson’s (1979) work is the first major research done in this area. The outcomes of his study showed that CS was absolutely an unexpected choice in any formal setting comprising even educated bilinguals. The reason was that in any formal setting the expected code choice was purely an indigenous language or the English language. Similarly in informal conversations, CS was employed deliberately by the bilinguals to build their social identities or express ideal meanings. Because of this, he referred to CS as the ‘third tongue’ of bilinguals aside their indigenous language and English (cited in Amuzu, 2012). Forson’s findings are inconsistent because language is subject to change. Moreover, current studies (Amuzu, 2012 and Asare-Nyarko 2012) on the socio-pragmatics of CS have shown that there has been a change in relation to the domains in which Ghanaians use CS. CS has also been gradually incorporated into the Ghanaian system that it has become a default language of many 20 University of Ghana http://ugspace.ug.edu.gh Ghanaians to the extent that it is used in formal settings like the classroom (Asilevi, 1990; Anyidoho & Dakubu, 2008) and church settings (Andoh, 1997; Albakry & Ofori, 2011, Asare- Nyarko 2012). 2.1.3.2 Albakry & Ofori (2011) Albakry and Ofori (2011) examined the language use in 10 selected Catholic Church masses in the capital city of Ghana, Accra. Using Myers-Scotton’s (1997; 2003) Markedness theory as their theoretical framework, they engaged in a participant observation which enabled them to observe the codes used in the sermons and also between the believers. They concluded that the English language dominated though the local languages were also employed. One noticeable thing observed in the data presented is that anytime English was used as the default code in the masses or informal interactions between the church members, there was no/minimal code- switching. Thus, they pointed to “a trend towards English monolingualism” in urban educated youth, which has been observed by other researchers. A line of similarity could be drawn to Quarcoo and Amuzu’s (2016) study of language use in out-of-classroom academic discussions and the use of language on WhatsApp platforms, where English is the default language of communication. 2.1.3.3 Amuzu (2012) Amuzu (2012) develops insights into the social motivations of conversational CS in Ghana. His work shows that the perception of Ghanaians on CS behaviour has changed with time. Before the 1990’s, CS was considered as a marked code which had its clear social functions. However in the 1990’s, research showed that CS is pervasively used in both formal and informal domains in the Ghanaian community. 21 University of Ghana http://ugspace.ug.edu.gh As he presents in-depth stylistic analysis of Ewe-English and Akan-English code-switching employed by some Ghanaians in informal and semi-formal contexts using Myers Scotton’s Markedness Model, he concludes that CS has become a norm in the Ghanaian context to the extent that it looks primarily unmarked since it satisfies little or no pragmatic function farther than assessing speakers’ solidarity though both unexpected CS and expected CS are usually used by the bilinguals. Amuzu’s claims that CS primarily looks unmarked among Ghanaians sound convincing since CS itself has become a variety of language in Ghana. This argument is supported by Guerini (2012) who shows that CS has become a new variety of language in Ghana. However, Amuzu’s (2012) study was only based on spoken conversations; therefore, this current study fills the gap by investigating the socio-pragmatics of code-switching in written conversations. 2.1.3.4 Quarcoo & Amuzu (2016) Quarcoo and Amuzu (2016) examined the use of CS in out-of-classroom academic discussions of students in a teacher training university in Ghana. Using an ethnographic approach for data collection and the Community of Practice concept and the Markedness Model, they concluded that monolingual English was used for core academic discussions while students departed to Akan-English CS/Akan when passing side comments. The use of Akan-English CS/Akan created solidarity among the students since they were all Akan speakers. Their study confirms Obiri- Yeboah’s (2008) findings from her study of CS in academic discussions among undergraduate students at the University of Ghana. Additionally, their study reveals that CS is used by the students to complement English anytime they need to discuss concepts which look abstract in English. 22 University of Ghana http://ugspace.ug.edu.gh Moreover, their study points out that within a larger CoP, other CoPs may be identified. Thus, their study indicates that students constituting an ethnically-based sub-community of practice within the larger study group would often switch to either their ethnic language or to CS involving their ethnic language and English (e.g. Ga/English CS, Ewe/English CS) which restricted the communication to their in-group; thus, excluding those from other ethnic groups within the community. A similar situation is presented in Chapter 3 section 3.2.2 of this study where two participants in a WhatsApp chat group try to exclude the other participants from a discussion by suddenly switching to French. This proves that there could be in-groups within a larger CoP. Therefore, it is only the use of monolingual English, Akan or Akan-English CS that serve as the neutral languages that integrate all the students in the study group as a CoP. From the different domains in which CS has been studied, it is important to note that anytime English is used as the default code of communication, there is minimal CS and this is evident in Quarcoo and Amuzu (2016) and Albakry & Ofori (2011). This could be attributed to the fact that English as a prestigious variety is self-sufficient and does not function as the language of grammar into which insertions could be made from the local languages. However, research works like Forson (1979), Asare-Nyarko (2012) and Amuzu (2012) prove the pervasive use of CS in situations where a local language is used as the default language. The use of language in out-of-classroom academic discussions is similar to that of WhatsApp since English is the language of discourse on this CMC mode. Therefore, this will better help in the discussion of the motivations for CS on WhatsApp. 23 University of Ghana http://ugspace.ug.edu.gh 2.1.4 Language and the Internet 2.1.4.1 Computer-mediated Communication Computer-mediated Communication (CMC) is an innovative approach to communication which is more rapid than written interactions but much slower than verbal interactions (Herring, 2001: 614). It arose with the advent of technology. It can simply be defined as communication through networked digital media. December (1996) defines it as “the process by which people create, exchange, and perceive information using networked telecommunications systems that facilitate encoding, transmitting, and decoding messages”. CMC makes it possible for several people to connect to one another at the same point in time which would have been difficult to be realized through other means (Herring, 1999a). Herring (2001) observes that computer-mediated communication is a broader interdisciplinary field which contains computer-mediated discourse (CMD). To her, CMD emphasizes the “language and language use in computer networked” settings and also relies on discourse analysis approach to study language use (2001: 613). Herring (2001: 613) describes CMD as: the communication produced when human beings interact with one another by transmitting messages via networked computers. Broadly, there are two types of CMC: Synchronous and Asynchronous. The synchronous CMC demands participants to be online at the same time and be active in producing a conversation. Hård af Segerstad (2002:4) explains that in asynchronous CMC, “the sender and the receiver of a message do not need to be online at the same time, neither do they need to respond immediately to a received message”. Hård af Segerstad (2002) observes that asynchronous CMC has characteristics of written language since it does not require immediate response from receivers 24 University of Ghana http://ugspace.ug.edu.gh which enables the receiver to plan and edit the message before replying. Examples include email, SMS, news boards and discussion forums. Synchronous CMC, on the other hand, permits interactive written communication and a typical example is Internet Relay Chat (IRC). With actual Synchronous CMC, it enables the recipient to witness the formation of a message letter by letter. Though it might be assumed that all kinds of chats are synchronous, there are times they take on the feature of asynchronous CMC. This is because the recipient may decide to read the message instantly or wait for some hours, days or weeks before attending to the sent messages. That is why Devic (2008) incorporates all kinds of chats into quasi-synchronous CMC. Quasi synchronous CMC may partly require participants to be online at the same time and be active in a chat; thus, “the recipient can read the message as soon as the author finishes and sends it” (Devic, 2008: 30). Language is very crucial in CMC since it is usually influenced by the Internet. In the Preface to Language and Internet, Crystal (2004) discusses ‘‘the role of language in the Internet and the effect of the Internet on language’’. In his discussion, he identifies four different situations in which the internet is contained. These situations are Electronic mail (e-mail), Chat groups (synchronous and asynchronous), Virtual worlds and World Wide Web (www). According to Crystal (2004: 48), the internet language (what he refers to as ‘Netspeak’) is a hybrid of ‘‘speech + writing + electronically mediated properties’’. As pointed by Herring (2001: 616), it is usually perceived that the language of CMC is “less correct, complex and coherent than standard written language”. Though the language of the internet has some features of speech, the actual measure is very limited (Crystal, 2004). In all the internet situations discussed by Crystal (2004), it is the World Wide Web that has the least characteristics of speech. Chat group and Virtual world 25 University of Ghana http://ugspace.ug.edu.gh exchanges have greater amount of spoken features while e-mails occupy the middle of the continuum. Crystal (2004: 29) notes that though it might be argued that interactions in e-mails, chat groups and virtual worlds are conveyed in writing, they show many features of verbal discourse. Like verbal discourse, these three internet situations are time bound demanding abrupt reaction; they are also transitory, “in the sense that messages may be immediately deleted (as in e-mails) or be lost to attention as they scroll off the screen (as in chat groups); and their utterances display much of the urgency and energetic force which is characteristic of face-to-face conversation” (Crystal, 2004: 29). With these properties shared in common with spoken discourse, “the situations are not all equally ‘spoken’ in character”. This is because e-mails are written but not spoken; however, members of chat groups and virtual worlds actually speak to one another since they were meant for people to ‘chat’. Crystal (2004: 11) defines a chat group as the “continuous discussions on a particular topic, organized in ‘rooms’ at particular Internet sites, in which computer users interested in the topic can participate”. 2.1.4.1.1 WhatsApp Messenger The advancements in science and technology have led to the creation of many gadgets, software and internet-based applications. Kietzmann, Hermkens, McCarthy, & Silvestre (2011) have noted that these gadgets and applications help people to get things done quickly in their day-to- day routines. The creation of the social media is one of the developments in technology which has helped to improve the lives of people. As years go by, there has been an intense growth in the usage of the social media. However, as observed by Kietzmann, Hermkens, McCarthy, & Silvestre (2011), anytime the term social media is mentioned, people tend to consider Facebook, Twitter, WhatsApp, YouTube etc. as making up the social media without really knowing what 26 University of Ghana http://ugspace.ug.edu.gh the term means. The social media employs mobile and web-based technologies to produce extremely communicating platforms through which persons and communities share, co-create, discuss, and modify user-generated content (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011). This makes the social media a very powerful tool for disseminating information. That is why BBC Business Editor, Tim Weber (2010), expounds that, ‘‘These days, one witty tweet, one clever blog post, one devastating video–—forwarded to hundreds of friends at the click of a mouse–—can snowball and kill a product or damage a company’s share price’’. This shows that individuals should be educated on both the good and the bad sides in the use of the social media. Kaplan & Haenlein (2010: 61) describe social media as “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content”. The term Web 2.0 was first employed in 2004 to “to describe a new way in which software developers and end-users started to utilize the World Wide Web” (Kaplan & Haenlein, 2010:61). WhatsApp is different from other social media platforms since it is a cross-platform that allows an individual to connect with his contacts. Moreover, it does not network its users like how other social media platforms such as Facebook and Instagram allow users to connect with both known and unknown friends, have an idea of their friends’ friends and their activities in news feed unless they have some privacy settings. However with WhatsApp, an individual’s contacts are his alone and nobody else can connect with those contacts through that individual. 27 University of Ghana http://ugspace.ug.edu.gh 2.1.5 Written and Spoken Text Genre In discourse, a text may denote any verbal or written passage of whatever length. It is described as “a verbal record of communicative act” (Brown and Yule, 1983:12). Though both verbal and written language can be used to create and preserve social relationships, it is so evident that spoken and written language makes different demands on language users. With all the different characteristics of spoken and written language, Tannen (1982) argues that both spoken and written texts are two ends of the same continuum since some written genres (chats and text messages) contain typical features of spoken genres while some spoken genres (e.g. scripted news reports, public speech) also contain features of written discourse (Chan, 2009: 108). Scholars like Du Bartell (1995) and Hἁrd ἁf Segerstad (2002) argue that communication via digital media has been “treated as a sort of hybrid between spoken and written communication” since it exhibits the characteristics of both spoken and written communication. A major technique to code-switching studies is to consider the nature of the text genre (spoken/written). Chan (2009) argues that CS in written discourse may be less frequent since the writer has enough time to produce what he intends to say and even explain terms while a speaker in a spoken discourse may be in a rush to express himself quickly to his interlocutors. Furthermore, some scholars argue that verbal code-switching and written code-switching may express different meanings. McClure (2000) and Montes-Alcalá (2000), for instance, claim that code-switching in written discourse usually serves as a stylistic device and an identity marker whereas code-switching in a spoken discourse may perform many functions. In the Ghanaian community, spoken code-switching has become a norm in both informal and formal settings. Amuzu (2012) rightly asserts that spoken code-switching is pervasively used in the Ghanaian community to the extent that it does not perform much or no pragmatic function than to show 28 University of Ghana http://ugspace.ug.edu.gh speakers’ solidarity because spoken CS has become an expected choice for many Ghanaians. Therefore, this study hypothesizes that CS employed in written discourse may perform specific pragmatic functions since in the Ghanaian context, spoken code-switching is the expected norm which performs little pragmatic function. In written discourse, Ghanaians usually employ the English Language when communicating their thoughts. Marjie (2010) examined English items in Swahili-based texts and her analysis of the chats, emails and phone text messages revealed that writers adopt bilingual attitude in these electronic modes. Her study showed that speakers of Swahili transfer their spoken code-switching into written code-switching in electronic writing (emails, chats and phone text messages). Unlike Ghana, Tanzania and Kenya use Swahili as national language. However in Tanzania, Swahili enjoys the prestige of the official language of the state while Kenya uses English as the official language (Myers-Scotton, 1993b). Since Swahili dominates all aspects of life in Tanzania, it has become the preferred code over the English language; therefore, it is used in both written and spoken communication. In Ghana, the English language enjoys much prestige over the local languages. Opoku-Amankwa & Brew-Hammond (2011) explain that the English language enjoys prestige in Ghana because many people consider literacy as the capability to read and write English or another dominant transnational language like French, and hardly perceive mother tongue literacy as important. As a result, they have observed that there is little enthusiasm in the study of the indigenous languages in terms of reading and writing. 29 University of Ghana http://ugspace.ug.edu.gh 2.1.6 Code-switching research in CMC Having discussed the researches that have been undertaken in code-switching based on its spoken form and also on Computer-mediated Communication (CMC) in sequence, this part of the section brings the two phenomena together by presenting some studies that have been done on code-switching in CMC. 2.1.6.1 Georgakopoulou (1997) Georgakopoulou (1997) investigated the self-presentation and the interactional alliance of Greek- English code-switching in email communication. The data for the study was made up of Greek messages sent among friends and colleagues in an academic environment, who were all native speakers of Greek, living in London. Using interactional sociolinguistics and ethnography of communication as the theoretical frameworks, she made the following observations: Firstly, she noticed that the opening and closing sections such as greetings routines hardly occurred in the corpus. However, when they occurred, they followed a particular pattern. Georgakopoulou (1997) argues that the unusual occurrence of opening and closing sections in emails brings out their dialogic nature which is the feature of spoken discourse. Moreover, she noticed the practice of pasting copied parts from a received email into a responding email which is “arguably a stimulation of the conversational sequential mechanism of turn taking” (Georgakopoulou, 1997:146). The conclusions drawn after the study were that participants build their self-presentation and alignments with their recipients through code-switches and style-shifts. Those code-switches and style-shifts in the corpus functioned as hedges which possessed the quality to lessen possible face-threatening acts. Georgakopoulou (1997) argues that the hedges in the corpus improve familiarity and the participants’ in-group loyalty. The findings of this study show the need for a 30 University of Ghana http://ugspace.ug.edu.gh researcher to look into the contextual approaches to any study in Computer-mediated communication. 2.1.6.2 Siebenhaar (2006) Using both quantitative and qualitative approaches to language disparity in Swiss-German Internet Relay Chat (IRC), Siebenhaar (2006) studied code choice and code-switching between the varieties of German spoken in a diglossic situation in Switzerland. The data was analysed using Auer’s (1984) Conversation Analysis Model. He concluded that the variety of German adopted in IRC was determined by the variety already in use in a particular session. Thus, chatters usually use the linguistic code already in use when joining discussions. This usually causes a situational code-switching since the chatters may not have any choice than to use the variety already used in a chat room. The writer also argues that the preference for a particular variety is also age related. This is because participants of a middle-aged group frequently adopt standard German while younger and older participants use more Swiss German which is a dialect of Standard German. Siebenhaar notes a change from the use of only Standard German to the use of both Standard German and its dialects in writing. He points out that Computer-mediated Communication influences language and brings out novel ways of writing. 2.1.6.3 Goldbarg (2009) Goldbarg (2009) investigated written CS, which is specifically naturally-occurring code- switching in the emails of five Spanish-English bilingual Latinos. Using multi-dimensional scaling and tree diagrams in the discussion of the data, the researcher concluded that the use of the English language in emails is mostly associated with formality while the use of the participants’ native language is associated with familiarity, informality and group solidarity. 31 University of Ghana http://ugspace.ug.edu.gh Moreover, the participants only switch to Spanish in order to personalise transactional or work- related English-dominant emails. This study provides a template for the current study. However, this current research looks at code-switching in WhatsApp group chats. 2.1.6.4 Marjie (2010) Marjie (2010) examined how English words are treated in Kiswahili-based texts by focusing on how such English words are spelled and the factors that cause the use of those words. The corpus for the study was made up of two types of texts: print (newspapers) and electronic texts (chats, emails and phone text messages). Using Myers-Scotton’s Markedness Model and the Matrix Language Frame as well as Winford’s Approach to borrowing, she observed that writers of newspapers and electronic texts (chats, emails and phone text messages) have different orientations towards the English language when writing. This is because English words undergo borrowing when they are used in newspapers. However, the findings revealed that the English words used in electronic writing are code- switched, i.e. the English words were spelled as they would be in English. This observation led to her argument that it is writers who draw the line between borrowing and code-switching but not linguists. Her argument seems logical since writers are in the best position to decide when to treat a word as code-switched or borrowed. The findings of this study are very relevant to the current research in that it demonstrates how language alternation occurs in both print and electronic writing. Also, the study is useful to the current thesis because of its focus on language alternation in a computer-mediated communication environment; thus serving as an example to such studies in an African context. 32 University of Ghana http://ugspace.ug.edu.gh 2.1.6.5 Halim and Maros (2014) Halim and Maros (2014) examined the amount of code-switching in the status updates sent by five Malay-English bilingual users in their Facebook interactions. The corpus for this research was gathered within one year. Using Gumperz’s (1982) Conversational code-switching model augmented by Zentella (1997), San (2009) and Montes-Alcalá’s (2007), the results of the study showed that code-switching occurs in virtual communication to serve as quotations, addressee specification, reiteration, message qualification, clarification, emphasis, checking, indicating emotions, availability, principle of economy and free switching purposes. This research asserts that bilinguals are required to be competent in both languages so that they can use code- switching as an effective instrument to improve communication in an electronic platform. As a result, this study buttresses Blom and Gumperz’s (1972) assertion that code-switching obliges bilinguals to be totally proficient in both languages. However, the question that crops up is how can a bilingual be fully competent in both languages. How can equal competence in two languages be measured? However, Halim and Maros’ (2004) study provides a template for the current study since it also looks at code-switching in a different CMC mode. 2.1.6.6 Anchimbe (2015) Anchimbe (2015) examined the illocutionary motivations behind Cameroonians’ CS from English to French and to Cameroonian Pidgin English (CPE) in blogs. He applied postcolonial pragmatics to the data taking into account the bilingual country setting where English and French compete for dominance. The findings of the study show that anytime members of the forum switch to French or CPE, there is a motivation for it. Thus, they would switch to French when they desire to denigrate, abuse, exclude and challenge others while they switch to CPE to show solidarity towards their in-group. This research points to the fact that colonialism has brought 33 University of Ghana http://ugspace.ug.edu.gh into contact foreign and local languages and cultures. Thus, the analysis of CS is done vis-a-vis colonialism as a factor. The relevance of this study lies in the fact that it is able to use the data from the blogs to point out the division that exists in Cameroon along Francophone and Anglophone lines, which in turn places emphasis on the identity constructed by various bloggers. Thus, within a larger community, the Anglophone bloggers are able to distance themselves from the Francophone through their use of English, which creates their in-group identity. This is similar to the situation presented in Quarcoo and Amuzu (2016) and what occurs on the WhatsApp platform where there exist sub-communities of practice within the larger community. Therefore, the smaller communities may use language which is understood only by their in-group in order to exclude others from a particular discussion. 2.1.6.7 Amuzu, Kuwornu & Opoku-Fofie (Forthcoming) Using the Markedness model (1993, 1999, 2006), Amuzu, Kuwornu & Opoku-Fofie (Forthcoming) investigated the pragmatic use of Ghanaian language DMs in English-based conversations on WhatsApp. The findings revealed that chatters make conscious effort to use the DMs to re-enact their in-group solidarity and the familiarity they have developed in verbal conversational contexts. Moreover, the DMs function as overt markers of Ghanaians’ in-group identity and solidarity on social media. The study also confirms Landone’s (2012) observation that the use of DMs in written conversations means more than a mere transfer of verbal conversational norms by participants. The study points out to the fact that Ghanaian chatters’ use of DMs is informed by social factors which include the level of interactivity participants have established among themselves. 34 University of Ghana http://ugspace.ug.edu.gh 2.2 THEORETICAL FRAMEWORK Since code-switching (CS) became an intriguing field of study, linguistic investigation into it has been done in two distinct but related directions: Structural (e.g. Poplack, 1980; Forson, 1979; Myers-Scotton, 1993a; Asilevi, 1990; Amuzu, 1998 and 2005; Quarcoo, 2009) and Sociolinguistic (e.g. Gumperz, 1982; Auer, 1984; Swigart, 1992; Myers Scotton, 1993b; Li Wei & Milroy, 1995; and Amuzu, 2012). Whereas the structural approach to CS principally looks at its grammatical aspects, the sociolinguistic approach perceives CS mainly as a linguistic phenomenon which focuses and concentrates on questions such as how social meaning is produced in CS and what particular discourse purposes it serves. Myers-Scotton’s (1993b) study represents one of the most prominent investigations done in the area. Since this study seeks to investigate the socio-pragmatics of CS on WhatsApp platforms, it uses Lave & Wenger’s (1991) Community of Practice theory, Myers-Scotton’s (1993b) Markedness Model and Labovian (1966) Sociolinguistics as theoretical frameworks. The section below presents a discussion of each framework. 2.2.1 Community of Practice The theory of Community of Practice (CoP) was propounded by Jean Lave and Etienne Wenger as a source of a social theory of learning (Lave & Wenger, 1991; Wenger, 1998). This theory has been extended by Wenger (1998) and applied across many learning situations. As a critique to traditional methods of learning, the theory argues that learning is a social process and learners acquire knowledge through the process of sharing information and experiences with a group, which is similar to apprenticeship. Therefore, becoming a participant of a community of practice includes learning in order to fit into it either as a peripheral or core member. 35 University of Ghana http://ugspace.ug.edu.gh After this theory was propounded as a learning theory, it was never used in linguistic studies till it was presented into language and gender studies by Eckert & McConnell-Ginet (1992) in a very significant research work. They adopted Lave and Wenger’s community of practice theory because to them, it emphasises a community “defined by social engagement - after all, it is this engagement that language serves, not the place and not the people as a bunch of individuals” (Eckert & McConnell-Ginet 1992: 463). Moreover, they adopted the community of practice theory as a substitute to the speech community theory for language and gender studies. Bucholtz (1999) has observed that the inconsistencies of the speech community theory for language and gender studies have not been thoroughly addressed. To her, the community of practice theory avoids the inconsistencies that are linked to the speech community framework. Bucholtz (1999) argues that the CoP theory is widely applicable to linguistic studies, and also allows language analysts to completely theorize social explanations and avoid the problems that were associated with the speech community theory. Also, Bucholtz (1999: 210) have observed that the community of practice model “considers language as one of many social practices in which participants engage”. In Homles and Meyerhoff’s (1999:173) Introduction to Language in Society’s Special Issue on Communities of Practice, they observed that a CoP bears semblance to a speech community although members of a speech community may not necessarily speak the same language on similar situations. What is important is that members of a speech community need to share in common at least one language and the rules governing its use so that they can make sense of the social meanings of its use (Gumperz, 1972:25). Social network analysis is also similar to the CoP theory since it also explores the relationships between people and how these relationships pull them into or away from a group identity 36 University of Ghana http://ugspace.ug.edu.gh (Davies, 2005:559). What differentiates these two concepts is that social network analysis is only grounded on linguistic disparity instead of viewing linguistic disparity as a part of the network practice (Davies, 2005:560). To Davies (2005:558), a community is not just explained based on linguistic behaviour. However, other social practices (norms and values, dress code, food, greeting, etc.) also help to define a community. Therefore, linguistic behaviour is an aspect of what defines a community. Eckert and McConnell-Ginet define a community of practice as: an aggregate of people who come together around mutual engagement in some common endeavor. Ways of doing things, ways of talking, beliefs, values, power relations - in short, practices - emerge in the course of their joint activity around that endeavour. (Eckert & McConnell-Ginet 1992: 464) The definition implies that a CoP is not similar to the traditional concept of community in that a CoP is primarily described based on the participation of the members. Therefore, it is the practices of the community and the distinct roles played by every member that build it socially (Eckert &McConnell-Ginet, 1992: 464). These practices include what the members of the community do, why they do them and what they aim to achieve. As already stated, members of a CoP may be found at a particular location; however members of a community of practice need not necessarily be situated at one place. Members make up a virtual community of practice “when its members use ICT as their primary mode of interaction” (Dubé, Bourhis & Jacob, 2005:147); thus, they work in partnership online such as within discussion boards and newsgroups or a mobile community of practice (Kietzmann et al., 2013) when they interact through mobile phones just to work and contribute to their community work. To Kietzmann et al. (2013:4), a mobile community of practice (MCOP) is “computer mediated, marked by advances in means of mobile communication (e.g., from pager to smartphone), supporting infrastructure 37 University of Ghana http://ugspace.ug.edu.gh (e.g., faster networks, better signal coverage) and of communicative acts (e.g., from speaking and mobile email to texting and tweeting)”. For example, participants of a WhatsApp chat group form a community of practice because they are defined by their membership and practices which grow out of their mutual engagement (Eckert & McConnell-Ginet, 1992: 464). Members of a WhatsApp chat group can be recognised as (1) a social group although they are not located at a particular place, (2) they interact on the platform for a special purpose, (3) they are part of the mobile community and thus, their practices affect the individual participants as well as language use on the platform. Since a CoP is developed around practice, Wenger (1998:73) categorizes three characteristics that define this practice: a. Mutual engagement b. A joint enterprise c. A shared repertoire It is the mutual engagement that defines every community of practice. It refers to the kind of interaction that exists between members of the community. Regular interaction between members of a community of practice is very useful to develop the practices of the group. Therefore when these practices are developed, their meanings are negotiated by the participants of the community. One feature of participation in a community of practice is reciprocity (Wenger, McDermott & Snyder, 2002). Participants of a community believe in building a more effective community that would profit all the members. Because of this, they all contribute their quota just to make the 38 University of Ghana http://ugspace.ug.edu.gh community functional. In Wenger, McDermott & Snyder’s (2002:37) words, reciprocity means “a pool of goodwill—of “social capital,” to use the technical term—that allows people to contribute to the community while trusting that at some point, in some form, they too will benefit”. Wenger (1998:81) explains that the negotiation of a joint enterprise refers to a process that “gives rise to relations of mutual accountability of those involved”. It is this process which involves the relations of joint responsibility that develops into an aspect of the practice of the community. The negotiation of this joint enterprise reveals participants’ understanding of their individual roles in the chat group. Davies (2005: 560) describes the shared repertoire as “the way of doing – which can be instantiated through linguistic features, personal common ground, a shared perspective on the world, and many other behaviours”. It may also include specialized gestures, stories, symbols, pictures and concepts which the group has created in the course of its existence. Further, participants are familiar with these and their continual use become part of the practices of the community. The CoP theory was used to get access to the data since for some of the chat groups, the researcher had to become a member of those groups. Also, there was the need to keep abreast of the in-group practices of the various groups. The data was generated from people who have been participants of those chat groups for a while and have formed for themselves communities but the researcher had to become a member in order to get the information of who they are and what they talk about. Therefore, this theory was mainly used in the data collection methodology. 39 University of Ghana http://ugspace.ug.edu.gh 2.2.2 Markedness Model Myers-Scotton’s (1993b) Markedness model is a theory that offers an explanation for the social motivations of code-switching. It, therefore, investigates speakers’ socio-psychological motive for engaging in code-switching (Myers-Scotton, 1993b: 75). It is centred on the principle that both speakers and linguists can simply recognize and distinguish between marked and unmarked codes. As a result, she proves with sufficient examples from the bilingual conversations taken from Kenya and Zimbabwe settings that her model can bring out the social meaning of code- switching. Myers-Scotton (1998: 22) indicates that all speakers have a ‘markedness evaluator’ that contains a cognitive facility to measure markedness. According to this model, speakers need to cultivate two abilities in order to identify markedness. These are: (1) The ability to identify that linguistic alternatives or choices fall along a multidimensional continuum from more unmarked to more marked and that their ordering will be different, according to the particular discourse type; (2) The ability to recognize those marked choices will receive various receptions from unmarked choices (Myers-Scotton, 1998: 22). Thus, speakers attain these types of capabilities when they come into contact with both marked and unmarked codes (Myers-Scotton, 1998: 22). Therefore, speakers are mandated to be exposed to the use of marked and unmarked codes in a community or group discourse so that they can acquire the particular languages that are needed to be used under which circumstances. This implies that participants in a conversation are expected to choose a code that will index the context in which they are operating. 40 University of Ghana http://ugspace.ug.edu.gh It is the ‘markedness evaluator’ that helps to determine the particular code that is marked or unmarked. Unmarked choices act according to the social expectations attached to the different codes while marked choices violate these kinds of social expectations and can, accordingly, be used purposefully by speakers (Milroy and Gordon, 2003: 213). With the Markedness Model, the rate of occurrence of the codes involved in the CS can be used to describe the less frequently- used language as the marked code and the more-frequently used language as the unmarked code or choice (Myers-Scotton, 2002: 206). It must be noted that the speaker’s use of the marked or unmarked code rests on his rational decision to use a particular code so that it fulfils a specific function in a particular communication (Myers-Scotton, 2002:218). This implies that the researcher needs to clearly describe both the marked and the unmarked codes in a given interaction so as to examine the functions of the several codes used in the interaction. Therefore in WhatsApp interaction, English is the unmarked code while the use of the local languages in such an interaction is seen as marked. Myers-Scotton (1993b) postulates that every linguistic code in a bilingual community is linked with particular social roles (Rights and Obligations sets). Rights and Obligations (RO) is “a theoretical construct for referring to what participants can expect in any given interaction type in their community” (Myers-Scotton, 1998: 23). Consequently, the RO sets are negotiated by speakers to create a social position based on the norms established by the society. The RO sets also offer speakers the linguistic choices available in any situation. So in an interaction, speakers know the particular codes available and select them to reflect their intentions bearing in mind the implications that those choices will convey. Since this model was developed after Grice’s (1975) Co-operative Principle, it operates with a central claim referred to as the Negotiation Principle. The principle states, “Choose the form of 41 University of Ghana http://ugspace.ug.edu.gh your conversational contribution such that it indexes the set of rights-and-obligations (RO) which you wish to be in force between the speaker and addressee(s) for the current exchange” (Myers- Scotton 1993b:113). The Negotiation Principle is explained in three maxims which offer guidelines for the several cases of code-switching choices: The unmarked choice maxim: “Make your code choice the unmarked index of the unmarked RO set in talk exchanges when you wish to establish or affirm that RO set” (Myers Scotton, 1993b:114). The unmarked choice is the language expected in a given context. The unmarked choice maxim involves certain conditions that need to be met. Primarily, the speakers involved in the conversation must be bilinguals who view their bilingualism as a sign of their solidarity. Also, the conversation must be casually shared by members of an in-group. Lastly, the bilinguals involved must be quite competent in the languages employed in the CS and must be able to assess the social values associated with those languages. Under this maxim, the model distinguishes between CS as a sequence of unmarked codes and CS itself as the unmarked code choice. CS as a sequence of unmarked code choice involves the change in code choice (expected for the given interaction) when participants or topics change during an interaction. This example (1) below is extracted from Myers Scotton’s (1993:116) unpublished data, which helps to explain sequential CS. Example 1: John has been visited by his brother Edward, and he switches from one language to another as the addressee changes. He speaks to his subordinate (Sub) in English and switches to Swahili when he addressed the receptionist (Rec): 42 University of Ghana http://ugspace.ug.edu.gh Sub: Where has this guy gone to? Where has this guy gone to? Edward: He’s just gone out. He will soon be back. He’s just gone out. He will soon be back. John: (To subordinate when he returns). Why Why did you change the plan of our did you change the plan of our stand at the stand at the showground? Who showground? Who recommended the recommended the change? change? Sub: (Looking guilty). Nobody told me. Nobody told me. John: Go and change it according to our Go and change it according to our previous plan. Also make sure the previous plan. Also make sure the painting is done properly. painting is done properly. (To Edward when subordinate has left). (To Edward when subordinate has left). I’ve told this man how to build our stand, I’ve told this man how to build our but he went and did a different thing. stand, but he went and did a different thing. He’s a stubborn person. Nti mtu mjeuri sana. I’ll make him pay for the paint he I’ll make him pay for the paint he spoilt. spoilt. (Calling to receptionist) (Calling to receptionist) Bring the guest a soda so that he may Letea mgeni soda anywe drink Rec: (to Edward) Nikuletee soda gani? What kind of soda should I bring you? Edward: Nipe Pepsi. Give me a Pepsi In the next type under the unmarked choice maxim, the CS itself serves as the unmarked choice and occurs when two languages are used simultaneously in the same conversation by bilinguals (Myers-Scotton, 1993b:117). This implies that before speakers engage in such conversation, they are already conscious of their bilingual personality. Thus, CS becomes the default means of 43 University of Ghana http://ugspace.ug.edu.gh communication among bilinguals. For example, many Ghanaians switch between their local language and English in many conversations with peers. Below is example (2) which is a conversation in Akan-English CS between friends on a WhatsApp platform. One of them was celebrating her birthday and as usual, they were to send messages to congratulate that person. English is in normal font while Akan is in bold font. Example (2) CHATTERS MESSAGES TRANSLATION ROCK: Happy birthday. Nyame nhyira wo. Happy birthday. God bless you IZZY: Ah nnipa yi se nnԑ ne bday monso Ah the person said it is not her moayere moho na happy bday birthday and you are bothering happy bday her with your birthday wishes LORD: Luvly…….Harry toaso Lovely…….Harry continue RODER: Happy birthday madam Happy birthday madam ROCK: Campaign start for Harry….adԐn Have you started campaigning for Harry why? IZZY: Nana twaso koraa. Nana, stop it In the extract, the WhatsApp group members congratulate a member of the group on her birthday although that person has told them it is not her birthday. However, another member of the group questions the others on their birthday wishes since the person perceived to be celebrating her birthday says it is not her birthday. The group members seem unperturbed and just send their 44 University of Ghana http://ugspace.ug.edu.gh wishes. Since they are conscious of their bilingual personality, they switch codes between English and Akan. The marked choice maxim: “Make a marked code choice which is not the unmarked index of the unmarked RO set in an interaction when you wish to establish a new RO set as unmarked for the current exchange” (Myers-Scotton, 1993b:131). When speakers deviate from the established RO set and introduce a marked choice to achieve a particular communicative intent, this type of CS results. This implies that bilinguals switch to a marked code so as to convey different rights and obligations than the one indexed by the unmarked choice. According to Myers-Scotton (2000:139), the marked choice is a violation though it can have a positive or negative connotation. Thus, it is positive when it decreases social distance and negative when it widens social distance. Moreover, a brief marked switching can be for emphasis. Exploratory choice maxim: “When an unmarked choice is not clear, use CS to make alternation exploratory choices as candidates for an unmarked choice and thereby as an index of an RO set which you favour” (Myers Scotton, 1993b:142). This maxim is employed when a speaker is not able to recognize the expected code in his interaction with others. Myers-Scotton (1993b) explains that this type of CS is rare and would occur under certain conditions where the unmarked choice is not evident. This is because in some situations, it is very difficult for both a speaker and addressee to choose the set of Rights and Obligations that applies. Therefore when code-switching is employed, it verifies the linguistic code that seems more unmarked. Here, it is the speaker that determines the code that 45 University of Ghana http://ugspace.ug.edu.gh would be unmarked in a particular conversation. So if the first code is not common to the addressee, the speaker proposes another language. Therefore, this type demonstrates CS as a ‘true negotiation’ (Myers-Scotton, 1993b:143). This model is used in chapter three in the qualitative analysis of the data. 2.2.3 Labovian Sociolinguistics/Variationist Sociolinguistics This theory was necessitated when Labov and Weinrich developed a theory of language change which employed ethnographic dialectological, probabilistic approach to the study of linguistic variation (Weinreich, Labov & Herzog, 1968). William Labov introduced the quantitative study of language variation and change in sociolinguistics in the 1960’s. This theory is termed the quantitative approach to linguistics because it adopts methods that are rather distinct from other sociolinguistic approaches (Paoletti, 2011). Hudson (1996) notes that quantitative approach to language studies is relevant to theoretical linguistics since it encompasses aspects of language which theoretical linguists consider fundamental to linguistic studies. Variationist sociolinguistics hypothesizes language as “an object possessing orderly heterogeneity” (Weinreich, Labov & Herzog, 1968: 100). This suggests that this theory accounts for all linguistic variations of speakers explained by the involvement of social factors. Labov’s early influential works on Martha’s Vineyard (1963) and in New York City (1966) demonstrated the significance of social determinants of linguistic variations and their relationships with the social structure. As a result, he suggested a social approach to language study through this sociolinguistic framework in which the linguistic theorization was linked with the society. Variationist studies are explicitly comparative and usually involve the identification of a variable which is the focus of investigation. Hudson (1996: 169) defines a variable as “a collection of alternatives which have something in common”. He distinguishes between different types of 46 University of Ghana http://ugspace.ug.edu.gh sociolinguistic variables according to the core of linguistics (phonetics, phonology, morphology, lexicon, syntax). The theory establishes that there is a broad relationship between linguistic forms and social factors like age, gender, ethnic background and social status. This is because in a community, speakers who belong to different age groups, genders, ethnic groupings and social classes may display systematic variations in their use of language. In view of this, the model requires a variationist methodology. These methods and procedures were first developed by Labov (1966) and later became the standard for other research works. Before these methods could be used, there is the need to define the sample size of the community the researcher is interested in investigating and also construct stratification for the sample in terms of ethnicity, gender, education or social status. The next stage of the methodology is the processing of the figures involved. At this stage, the researcher counts the occurrences of each variant in each text and compares the figures from different texts. Finally, there is the interpretation of the results. In the light of this, this approach to language studies adopts various data collection techniques such as written questionnaires, interviews, participant observation and surreptitious recording. According to Eckert (2012), social meaning in variationist sociolinguistics has come in three movements. The first movement was associated with Labov’s (1966) study of linguistic variation in the New York City. Fundamental to this wave is the concept of the vernacular. It also finds age, social class, ethnicity and gender stratification in variation. The second wave of variationist studies was introduced by Milroy (1980) with the study of phonological variation in social networks in Belfast. In this Belfast study, Milroy (1980) adopted ethnographic methods which led to the conclusion that high network strength was connected to the use of vernacular forms. This implies that variations in speakers’ use of vernacular can be influenced by their integration 47 University of Ghana http://ugspace.ug.edu.gh in a social network (Eckert, 2012). As observed by Eckert (2012), variation in both movements was seen as marking social categories. The third wave views the meaning of variation as an important characteristic of language. This wave is championed by Eckert (2008, 2012: 98) and it uncovers “ideology in language itself, in the construction of meaning, with potentially important consequences for linguistic theory more generally”. This theory is used in chapter four for the analysis of variations in language usage. 2.2.4 Justification for the three models This work selected the Communities of Practice (CoP) model, the Markedness Model (MM) and Variationist approach for a few reasons. Firstly, the Markedness Model assumes that speakers make rational choices when selecting a right code in terms of reasons and motivations behind those choices. Thus, bilinguals would switch between languages when they want to communicate certain meanings or when they want to show group solidarity. Moreover, the MM model describes the nuances in the selection of codes. The CoP model also explains the fact that chatters conventionalize a particular way of talking; thus, there is solidarity in relation to the language they use on WhatsApp irrespective of their social status. The Community of Practice (CoP) theory explains human behaviour in relation to the language patterns which members of chat groups use. Moreover, these language patterns indicate the in-group familiarity they have with one another. The Variationist theory also helps to illustrate the effects of social constraints on the patterns of variations across the groups in terms of their language use. Therefore, these theories complement one another. Whereas the Community of Practice (CoP) framework views members of chat groups as constituting communities and helps in data collection, the Markedness model provides the sociolinguistic and pragmatic explanations of 48 University of Ghana http://ugspace.ug.edu.gh code-switching which help to bring out the social motivations behind the act. The Variationist model also shows how the chat groups differ in terms of their use of language. These models help to ascertain the socio-pragmatics of the code choices used on the WhatsApp. 2.3 METHODOLOGY AND DATA COLLECTION This section involves the gathering of relevant data from available WhatsApp chat groups and the compilation of the databases just to analyse the occurrence of code-switching in the data. The first set of data comprises text-based chats of six WhatsApp groups. The researcher used both old and recent conversations of the chat groups since language is dynamic and is bound to change. Thus, chatters still contribute to the group discussions. Also, other data collection methods (interviews and participant observation) informed by the Labovian Sociolinguistics were used. 2.3.1 Selection of WhatsApp Chat groups Six WhatsApp chat group data were gathered and used for this study. These six groups consented to allow their conversations to be used for this study. Moreover, age and gender were considered. Age was a factor for selecting the groups because chatters who were relatively young were required. Moreover, it was the young people who easily made their chats available for the research work. Moreover, these six groups were chosen because three are of mixed sex groups while three groups are single sex groups. Participants of all the chosen chat groups are Ghanaians and belong to different ethnic groups. These variables were needed in order to bring out the different perceptions people have on what motivates them to switch codes on WhatsApp. 49 University of Ghana http://ugspace.ug.edu.gh 2.3.1.1 WhatsApp Data The WhatsApp chat groups used for this study have been in existence for at least two years. Therefore, naturally-occurring chat conversations that had existed for years were captured. In forming the corpus for the study, the data were obtained with the help of research assistants who are members of other WhatsApp chat groups while the researcher also joined some of the chat groups in order to obtain data. Usually, WhatsApp chat conversations are stored in the users’ mobile phones and some steps must be undertaken so as to export the data for the investigation. The researcher got the data by transferring the text from the groups she belongs to on her phone into her email inbox and finally onto the laptop. The research assistants also sent theirs into her mail. After receiving the corpus in an email, it is downloaded and automatically becomes a txt file. After it has been saved as a word document, it looks like the image below: 50 University of Ghana http://ugspace.ug.edu.gh Figure 2: WhatsApp Extracted data 51 University of Ghana http://ugspace.ug.edu.gh Once the corpus has been formed, the next step is the corpus preparation. This preparation is required to remove the names/contacts, time and dates from the text for further processing. This was done in order to protect the identity of the participants. However, the date, time and the contacts of the participants were replaced with names manufactured by the researcher so that those names acted as exhibits to those contacts. So that any time those contacts were found in the data, they were represented by those names. Also, since the researcher was conscious of the fact that some of the participants reside outside Ghana, she attached the status US, UK, Italy, Aus., Russia etc. depending on the country where the participant resides. Example (3) illustrates the explanation on the preparation of the corpus: Example (3) SAM: Saaaa na afe aso.....if like u bi team aa like u b Really and also… if you were like a Black Terror...full of hope bt always 4th team, you would have been Black place... Terror…full of hope but always 4th 🏃52🏃52🏃52🏃52🏃52🏃52🏃 place…🏃52🏃52🏃52🏃52🏃52🏃 awaaayyyy 52🏃 away EA: Ka na ka biom😏😏😏😏mouth lazt Say it and say it again😏😏😏😏 lazy mouth EA: Lazy# Lazy DAVE-RUSSIA: Oh Nana Abotr3😂😂😂 Oh Nana patience 😂😂😂 DAVE-RUSSIA: PhD, I believe, is not at any point considered PhD, I believe it is not at any point to understanding a human being by studying considered to understanding a human her (in that context or relationship/marriage) being by studying her (in that context of relationship/marriage) DAVE-RUSSIA: *of (not or) Jude: Mills Mills Jude: Is it true or it's a Joke Is it true or it's a joke? EA: ���� ���� Jude: BEY: But April has 30 days. Lol But April has 30 days. Laugh out loud Jude: Mteeeewww (chuckling) BEY: 😂😂😂😂😂 😂😂😂😂😂 Jude: Thanx errhhh Thanks BEY: Welcome dear Welcome dear 52 University of Ghana http://ugspace.ug.edu.gh FRANK: I'm sure the writers point is "claiming to study I'm sure the writer’s point is "claiming her" after 5 yrs. I will ve a lot of difficulty to study her" after 5 yrs. I will have a understanding any man who dates a lady for 5 lot of difficulty understanding any man yrs and tells me he is studying. It sounds not who dates a lady for 5 years and tells been serious. A more better response such as me he is studying. It sounds that he’s "we are preparing to marry" sounds more not being serious. A much better mature. response such as "we are preparing to By the way, congrats Rockson. I ve booked ur marry" sounds more mature. wedding, I'm praying nothing hinders my By the way, congrats Rockson. I have coming booked a seat at your wedding, I'm praying nothing hinders my coming EA: 😂😂😂😂😂😂 😂😂😂😂😂😂 DAVE-RUSSIA: DAVE-RUSSIA: #Trumpanosomiasis #Trumpanosomiasis K O: Borga u no go slp??? Been-to won’t you sleep? QUOFI-US: As again oooo Ask again DAVE-RUSSIA: Lol Laugh out loud ANNA: Lol sen Why laugh out loud? In the example (3) above, the contacts of the participants have been replaced with manufactured names. Moreover, contacts that live outside Ghana have been given the status Russia and US. Furthermore, the researcher was conscious of the gender of the contacts so what she did was to attach the exhibit F (female) and M (male) for the females and males in the mixed groups during the analysis. In getting the gender of the participants of the groups where the researcher was not a member, she inquired from the research assistants. For the groups the researcher was a participant, she inquired from the group administrators and sometimes, contacted those individuals and asked them. Since the data for the study is already written, all texts in the study are presented as they appear in the corpus. Because other languages besides the English language are found in the corpus, the data is presented as follows: I. There is an orthographic representation of the data rather than phonological transcription. II. All English items are in normal font. III. All Akan items are bold. IV. Other languages are italicised. 53 University of Ghana http://ugspace.ug.edu.gh The extract in example (3) above illustrates the presentation of data. English is in normal font, Akan items are bold while pidgin is in italics. 2.3.1.1.1 WhatsApp Chat groups The chat groups used for the study comprise chatters who are Ghanaians although some of them have travelled outside the country. The topics that are usually discussed in these chat groups range from social life, birthdays, football and other issues relating to the activities of the members of the groups. The chat groups that were used for the study include Ahwiaa R/C & L/A Old Students WhatsApp chat group, PENSA KNUST Organa Alumni Family WhatsApp chat group, MyChisco WhatsApp chat group, Zighiss 2009 WhatsApp chat group, Fraser Boys WhatsApp chat group and ND 1-14 WhatsApp chat group. 2.3.1.1.1.1 Ahwiaa R/C & L/A Old Students WhatsApp Chat group This group was created by a former school prefect of Ahwiaa R/C & L/A School. It was created on January 5, 2016. It is a mixed group made up of twenty (20) people who are past school mates. Members of this group started their early days in school together. So the group was formed to bring these past school mates together so that they interact with one another. The age range of members is 32-36years. Besides, the lowest educational qualification among the group members is at least a Junior High School certificate while the highest is post-graduate degree. Of the participants of the group, Akan speakers dominate. There are other participants from Ewe and Gonja ethnic groups. Although members of the group were not in the same class during their school days, they are able to share the same platform and interact with one another. 54 University of Ghana http://ugspace.ug.edu.gh 2.3.1.1.1.2 PENSA KNUST Organa Alumni Family WhatsApp Chat group This chat group was created on the 25th of December, 2014. It is made up of about one hundred and sixty-nine (169) members; however, only few of the participants actively contribute to group discussions. Participants were members of the Organising Committee of PENSA KNUST who completed KNUST between the years 2011 to 2016. The group administrators continually add PENSA KNUST organizing committee Alumni members to the group when they get their contacts. For the educational background of the members of this group, the minimum educational qualification is a degree while the maximum is post-graduate degree. The ages of the participants range from 24–34years. Aside their consent that their chat conversations should be studied, other factors that informed the selection of this group are that they are a mixed group comprising both sexes and the members are from different ethnic backgrounds. Though the Akans dominate the group, other members are Ewes, Gas, Gonjas and Dangmes. 2.3.1.1.1.3 MyChisco WhatsApp Chat group MyChisco chat group is a mixed group comprising forty (40) participants who went to the same Junior High School. It was created on 22nd of March, 2014. The participants are between the ages 26-30years. Members who have the lowest educational qualification have attained a Senior High School Certificate while others have attained diploma and degree. Moreover, some of the group members are pursing post-graduate degrees outside Ghana. In this chat group, there are Akans, Ewes, Hausas and Krobos. However, the Akans and Ewes dominate. 55 University of Ghana http://ugspace.ug.edu.gh 2.3.1.1.1.4 Zighiss 2009 WhatsApp Chat group Zighiss 2009 WhatsApp Group consists of twenty-eight (28) members. It was created on the 5th of June, 2014. This group is an all-female group. Members were General Arts classmates of Zion Girls’ Senior High School in Winneba, who finished school in 2009. The age range of the participants is 26-30. About 80% of the participants are nurses with diploma in nursing while the highest educational qualification is a degree. There are nine Akans, seven Ewes, six Gas, two Gonjas and two Krobos. 2.3.1.1.1.5 Fraser Boys WhatsApp Chat group The name Fraser is one of the houses in Achimota Senior High school. As the name implies, these old boys were in the same house during their Senior High School days. It was created on the 24th of May, 2013. Fraser Boys WhatsApp Group is made up of only males who are about thirty (30) in number. The age range of the participants is 27–30 years. Members with the lowest educational background have a Senior High School certificate while those with the highest educational background have post-graduate degree. Of the 30 participants in the group, there are four Gas, one Ewe while the rest are Akans. Since it is an all-male group, they usually chat in pidgin and switch codes between English, Akan, Ga and Ewe. In this group, members insult and tease one another for no reason but no one gets annoyed. 2.3.1.1.1.6 ND 1-14 WhatsApp Chat group This group is an all-female group made up of twenty six (26) participants. This group was created on 16th September, 2015. The age range of the participants is 20–50 years. This is a heterogeneous group which comprises students and floor tutors of Volta hall2. The heterogeneity 2 Volta hall is one of the traditional halls at the University of Ghana campus. 56 University of Ghana http://ugspace.ug.edu.gh of this group is such that among the students, there are PhD students, MPhil students as well as undergraduates. This group was formed purposely to create an interactional platform between JCR (Junior Common Room) members and SCR (Senior Common Room) members. In this group, there are Akans, Ewes, Gas and Gonjas. The main factor that informed the selection of this group is because the group is made up of both senior and junior members, which implies that the members are of different social status. Similarly, the different social statuses of members could help bring out variations in terms of language use. 2.3.1.2 Interviews Interviews were used in eliciting information from the chatters. The purpose of the interviews was to find out from the chatters the rationale behind their code-switching during conversations. Since the researcher and her assistants were observers of the chat groups used for the study, they interviewed participants of the WhatsApp chat group based on when they switched to a language that is not expected on the platform. Therefore, private messages were sent to those participants based on the particular code choice used on the group page. These participants were then asked why they decided to choose a particular code over another. Respondents of the interview were from all the six groups used for this study. (See Appendix 2 for examples of the questions asked during the interviews). The questions usually asked the participants were unstructured and the interviewer sometimes ended up also switching to an unexpected code. The interviews were in question and answer format since the researcher was only interested in eliciting naturally occurring responses from the respondents. 57 University of Ghana http://ugspace.ug.edu.gh 2.3.1.3 Other methods The researcher employed other methods such as participant observation and background checks. The researcher and the assistants participated in and observed the conversations and the practices of the WhatsApp groups. These observations produced insightful information into the participants’ relationships with other members of the chat group. It enabled us to observe the code choice and also, the patterns of CS usually employed by the chatters. Moreover, snapshots of some conversational patterns were taken after the data had already been exported for analysis since they looked useful to be included in the analysis. The participant-observation work also involved background checks where the researcher and the assistants had to chat participants of the various chat groups privately in order to obtain and verify useful information from them. 2.3.2 Ethical Issues This study was based on voluntary participation. Since many participants did not understand the essence of this research, they asked so many questions to clarify why their conversations were being collected as data. Some of the ethical issues that were applied include the following: 1. Participants’ approval was sought although they were not asked formally to fill and sign consent forms. Since the researcher communicated with the participants via chat, consent forms could not have been sent to them. However, there are snapshots of the approval of participants which are presented at the Appendix 1 section of the study. 2. Participants were informed about the research though the particular language practice expected from them was not revealed. This was to avoid a situation where participants would have intentionally switched codes in their expressions and thus, make the data unnatural. 58 University of Ghana http://ugspace.ug.edu.gh 3. The researcher also assured the participants of confidentiality of the information that might be found in the chats. They were assured that the data would only be used for academic purposes. 4. Moreover, it must be noted that chatters on WhatsApp usually make use of several ‘nicknames’ instead of their real names. Besides, they have the ability to change their names any time they want. Consequently since the chatters usually sign in with their ‘nicknames’, their identities are concealed anyway. Aside this, it is the phone numbers of the chatters that usually appear if the phone numbers have not already been saved on the research assistants’ phone. 5. Likewise, there were some groups that did not comment when the permission was being sought. They only ignored it and moved on with their discussions. Therefore, their silence on this meant to the researcher that they did not care if their chats were studied or not. Also, since many of the chatters had used their ‘nicknames’ instead of their real names, they knew that their identities were concealed. 2.4 SUMMARY This chapter was divided into three parts. The first part started with a discussion of the related and relevant works done on the code-switching phenomena. This included the studies on the sociolinguistics aspect of CS. As a result, there was a discussion on the research works of Gumperz and Blom (1972), Myers-Scotton (1993) and Auer (1998). This part also discussed the phenomenon of Computer-mediated communication (CMC) by focusing on the language of the internet. It then reviewed some studies done on the language choice and code-switching in Computer-mediated environment. The discussion on the researches carried out on code- switching occurring in CMC proved that many of the works have mostly concentrated on CMC environments where English is not a native language or functions as a major lingua franca for majority of the participants (except Siebenhaar (2006) who emphasizes on German-dominated 59 University of Ghana http://ugspace.ug.edu.gh CMC) though the current study attempts to do same. Also, those studies have considered traditional pre-web modes of interpersonal CMC (IRC, email, mailing lists, Usenet groups) with very few studies done on the trending social media (WhatsApp, Facebook, twitter etc.). Though CS in CMC is more planned than in speech, the findings of these studies establish resemblance to works done on spoken code-switching where code-switching shows in-group loyalty and solidarity. Nevertheless as pointed by Mvan Gass (2008: 434), researchers interested in the study of code-switching in CMC must remember that the patterns of language use in CMC may not bear semblance to the patterns usually found in spoken discourse; in that there is the possibility for researchers to notice the occurrence of novel patterns of code-switching. The second part discussed the Community of Practice theory, the Markedness Model and the Variationist Approach as the theoretical frameworks adopted for this study. It was explained that the Community of Practice theory is a social theory of learning that describes how individuals acquire knowledge when they join a community. For the Markedness Model, it is a model that expounds on the social motivations for the use of a particular code and its implication in a given society. The Variationist Approach also helps to illustrate the variations found in the various communities. Therefore, there could not have been a more applicable approach for the analysis of CS in WhatsApp conversations than combining these three complementing theories. Finally, the third part presented the methodology which was used for data collection. The data collection procedures included exporting WhatsApp data to an email, interviews and participant observation. 60 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE QUALITATIVE ANALYSIS OF LANGUAGE USE 3.0 INTRODUCTION This chapter deals with language choice and use in WhatsApp group chats. It is divided into three sections. Section 3.1 discusses the code-switching structures found in the data. Section 3.2 investigates the choice of languages employed by chatters in WhatsApp group chats. The last section, 3.3, looks at interview responses from chatters concerning their motivations for the use of code-switching during group chats. Myers-Scotton’s (1993b) Markedness Model is used in the interpretation of the data. 3.1 ANALYSIS OF CODE-SWITCHING STRUCTURES 3.1.1 Posts on WhatsApp Simply, posts refer to the messages that are shared through the social media. The Cambridge English Dictionary defines a post as “something such as a message or picture that you publish on a website or using social media”. On the WhatsApp platform, there is the practice of sharing posts. Posts that are sent through WhatsApp chats can be grouped into forwards and original posts. 3.1.1.1 Forwards The forwarding of posts is a practice in all individual WhatsApp chat groups. Forwards are usually sent from one contact to the other without knowing the real author of the posts. For forwarded posts shared on WhatsApp platforms, the unmarked code is the English language. However, some forwards can be entirely written in Akan or Pidgin while others may contain instances of code-switching between the English language and the local languages. 61 University of Ghana http://ugspace.ug.edu.gh Example (4a): This is a forward that is entirely written in Akan. It was sent across WhatsApp group chats when residents in the Ashanti regional capital, Kumasi were told to stay indoors on Thursday, January 19 from 8pm to Friday, January 20, 4am to enable the traditional authorities to perform the necessary rites in giving the late queen3 a befitting burial. This example was posted in KNUST Organa Alumni Family chat group: IZZY: *Asantefoɔ nhoma 1:1-2* Ashanti Book 1:1-2 1. Immediately he proclaimed in a loud *1*. Na ntɛm ara, wɔ tea mu bisaa sɛ hena ne voice, ‘who is a man!’ and there was Ɔbɛrima! Na ɛhɔ yɛɛ dinn silence all over the place. *2.* Na abɔfra ketoa bi kaa sɛ!, Ɔbɛrima ne nea 2. Then a little boy said! ‘a man is ɔde nensa bɛsi ne bo so apue 3n3 anadwo nanka someone who can boldly go out this ɔbɛ te ne kaka nka. ��😂😂😂😂 evening; like he would see where power 🏃🏃🏃🏃🏃🏃🏃 lies’. There is usually cross-posting and members of the chat groups just forward all manner of messages to their groups without knowing the original authors of those posts. Forwards are not actual conversations of participants of a group. They only trigger varied views from the members of a group. Below is a forward that was posted on Ahwiaa R/C WhatsApp group page. In this example, the forward initiates a discussion in the chat group and the participants make comments in relation to the forward. 3 Nana Afia Kobi Serwaa Ampem II was the Queen of the Ashanti Kingdom and mother of the current Asantehene, Otumfuo Nana Osei Tutu II, who is the youngest son of her five children. She was the 13th Queen mother of the Ashanti Kingdom. 62 University of Ghana http://ugspace.ug.edu.gh Example (4b): RICH: I was in church some time ago and the pastor said, "DO SOMETHING CRAZY FOR THE LORD!" I could see ladies removing their wigs. The pastor said again, "DO SOMETHING CRAZIER FOR THE LORD!" I saw people carrying plastic chairs. Now he was like, "can't you do something?...the CRAZIEST?" Someone from the back came and took the offering and was walking outside the church. The pastor shouted, "Hey! hey! gyimii! [Fool] this is not craziness; it’s MADNESS!!! OC.ITALY: Really hahahaha RICH: Cantona this is pure craziness or madness OC.ITALY: Stupidity OC.ITALY: How can you take offering away as a result of craziness From the above example, it can be observed that the first post by RICH is a forward. RICH is not the original author of the message though he sent the post. However, what the forward did was to trigger the views of the members of the group. Consider the example (4c) below. This forward was sent to many chat groups on 14th February, 2017. Example (4c) SLY: Somewhere, some place, right now, my future wife Somewhere, some place, right now, my is planning valentine wid some idiot. Father I pray future wife is planning valentine with some u cancel their plans..... idiot. Father I pray you cancel their S3 me b) me s3m 👏� 👏� 👏� 👏� na me plans..... b) mpayiea ogya oo If I clap my hands and pray, fire ooo fire 🔥🔥👏� 🔥🔥👏� 🔥🔥🔥🔥🔥🔥🔥👏� 👏� 👏� 🔥🔥🔥🔥👏� 🔥🔥🔥🔥👏� 👏� 👏� 🔥🔥😁😁😁 In this forward, there is the use of the English language and Akan. The Akan portion of the post follows a sentence written in English. Although this is clearly code-switching, it cannot be attributed to the members of the group since it is a mere forward whose author is unknown. 63 University of Ghana http://ugspace.ug.edu.gh Also in some forwards, many types of switches (inter-sentential and intra-sentential) may be found. Example (4d) illustrates this: Example (4d): Below is a forward that contains many types of switches: BENE: Teacher: If I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven Sir. Teacher: Naaaaaa [No], listen carefully. If I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven. Teacher: oh...wo nso... [oh you too]Let me put it to you differently. If I gave you 2 apples, and another 2 apples and another 2, how many will you have? Abusua: Six. Teacher: Good. Now if I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven! Teacher: oh...na wo nso wagyimi anaaa?[you too, are you that foolish?] seven no nso firi he?[where does the seven come from?] Abusua: aaah...Sir, Me wɔ rabbit baako wɔ fie dada oo...[Sir, I already have one rabbit at home] Teacher: ofui...� [Fool] crack some ribs�� The above is a single forward but within it, different instances of code-switching could be identified. There are both inter-sentential and intra-sentential CS found in this single forward. The switches that are intra-sentential are underlined while those that are inter-sentential are not underlined. These forwards initiate the varied opinions of the participants of the group. Sometimes, they become the basis for discussions on the platform. 64 University of Ghana http://ugspace.ug.edu.gh 3.1.1.2 Original Posts Original posts, on the other hand, are “utterances” or messages whose real authors can be identified. They can also be seen as the real conversations of the members of the WhatsApp chat groups. In the data, some of the original posts are in one language. Others are made up of code- switched words or expressions (either intra-sentential/ inter-sentential switching) of one language into another language which are discussed in the next section below. 3.1.2 Types of Code-switching This part of the study concentrates on the form code-switching takes in the posts. Milroy and Muysken (1995:8) distinguish between three types of switches. These are inter-sentential CS, intra-sentential CS and tag-switching. However in this study, the types of CS are grouped under four major categories: inter-sentential, intra-sentential, tag-switching and discourse markers. 3.1.2.1 Tag-switching Ammon (2004:54) maintains that tag switching “involves inserting a tag in one language into an utterance which is otherwise entirely in another language”. The reason for this kind of switching is that usually tags contain the least grammatical constraints and as a result “may easily be inserted at a number of points in an otherwise monolingual utterance without violating syntactic rules” (Ammon, 2004:54). Tags that are usually used include in fact, you know, I mean, right, etc. 65 University of Ghana http://ugspace.ug.edu.gh In the extracts below, in fact is used as a tag. Example(5a)PHAT: I lost two laptop bags bcoz of I lost two laptop bags because of kojapharma[escorts] and still de ladies didn't escorts and still the ladies didn't even look at my face than to even consider even look at my face than to even me.......I tot I was de only one, not knowing consider me.......I thought I was I was only continuing a trend...... the only one, not knowing I was In fact, Nyame ne hene ampa only continuing a trend...... In fact, God is king indeed Example (5b) KEN: In fact wo hia Yesu In fact you need Jesus Example (5c) KEN: Happy birthday Esme..... in fact wo ho y3 Happy birthday Esme..... in fact me f3 papapapa. you are very beautiful. In the above, in fact is a tag that is inserted into Akan constructions. It is the only tag that was found in the data. It emphasizes what has been said. For instance in example (5b), in fact emphasizes the speaker’s feeling that the person in question needs Jesus. 3.1.2.2 Inter-sentential CS Inter-sentential code-switching refers to switching that occurs between clauses or sentences. In the data, many of the switches occurred at the clausal boundary. The following are examples to illustrate this: Example (6): This dialogue ensues in Ahwiaa R/C primary WhatsApp chat group. Before the dialogue, Mike’s photograph was used as the group’s profile picture since he was celebrating his birthday the previous day. The next day, he comes onto the page to thank the participants for all their warm wishes. In thanking them, he switches from English to Akan at a clausal boundary. Another participant of the group also switches from Akan to English at a clausal boundary. 66 University of Ghana http://ugspace.ug.edu.gh Mike: Hmm...I still can see ma [my] pic[picture] as d [the]grp[group] dp [profile picture] n[and] I can't hide my joy. I feel proud to belong here. U [you] have made yesterday n[and] today soooo colorful. Thank u [you] all for making ma[my] day. Onyankopon nhyira mo [God bless you all] Kwaku: Yoooo yate [ok we hear]. Ur pix [picture] has overstayed. The dialogue in example 6 shows code-switching that occurs at the sentence boundary in which there is a switch from Akan to English and vice versa. Example (6b): In this example (6b) below, SAM sends a link to the group and this leads to Dave-RUSSIA commenting on politics in Ghana. SAM: http://mobile.ghanaweb.com/ http://mobile.ghanaweb.com/ GhanaHomePage/NewsArchive/I-got- GhanaHomePage/NewsArchive/I-got- zero-where-my-pastor-voted-Edward- zero-where-my-pastor-voted-Edward- Mahama-493947 Mahama-493947 Dave-RUSSIA: And so? (Sorry if it looks disrespectful, And so? (Sorry if it looks disrespectful, but don’t perceive it as such) The fact but don’t perceive it as such) The fact that he is ur Pastor doesn't guarantee u a that he is your pastor doesn't guarantee vote. Ebei, when will we understand you a vote. Why! When will we politics! Even ur wife or husband has understand politics? Even your wife or the choice to vote for or against u. This husband has the choice to vote for or isnt news kraaa. against you. This isn’t news at all. Ghanaweb has hit below the belt. Ghanaweb has hit below the belt. Ka as3m a, me s)fo anto amma me��. Instead of something important, you Enti di3n? are saying, “My pastor didn’t vote for It shows how people see u in the whole me”. And so what? It shows how people political race. Why should he waste it if see you in the whole political race. Why he think there is a better candidate? should he waste it (his vote) if he thinks Mekᴐ kraaa😊 there is a better candidate? I am even going. In his comments on the politics in Ghana, Dave-Russia uses the English language which is the default code of WhatsApp. However, he switches from the English language to Akan at a clausal boundary at a certain point. In the example above, the Akan clauses are underlined. Example (6c): In this extract, participants of the KNUST Organa Alumni Family discuss marriage. Then, Izzy questions another participant of the chat group why she has refused to 67 University of Ghana http://ugspace.ug.edu.gh accept marriage proposals from some men. Izzy begins his comment with an Akan clause and then switches to English. IZZY: Nyame ne Hene s3 s3n? When God has lined up God is King indeed like how? When big men before you. What again do you want from God has lined up big men before you, God. Or you want God to say "yes I will marry what again do you want from God? Or you" for you you want God to say “yes I will marry you” for you? In Izzy’s utterance, he switches from Akan to English at a clausal boundary. Therefore, the example above illustrates inter-sentential CS. 3.1.2.3 Intra-sentential CS Intra-sentential CS is a term used for switches that occur within the same clause. In the data, the distribution of Akan discourse markers in English constructions also exemplify as intra- sentential CS. Moreover, they contribute the bulk of intra-sentential CS (as will be shown in chapter 4). However, they are separated from intra-sentential CS and would be discussed independently (under section 3.1.2.4) as already indicated in the types of CS. This is because discussing the discourse markers with the intra-sentential CS will blur the fact that intra- sentential CS hardly occur on this electronic platform. This is because of the nature of communication on WhatsApp whereby the English language is the default code. Below are some cases to illustrate the point. These examples were taken from different chat groups. 68 University of Ghana http://ugspace.ug.edu.gh Example (7): switches that occur within sentences: (7a) Mills: Who will toaso today...?? Who will continue today...?? (7b) KOFA: Anelka koraaaaaaaaa adEn ne Anelka too why this jealousy? Girls, let's keep jealosy yia. Girls let's keep boiling. boiling. Hei Charlie where are you? Are you Hei char wer r u. R u teachings? teaching? Caro, the day is for me; you hear. Caro, D day is fr me wate. (7c) KOFA: Ooooooo ma colapse mpo I have even collapsed (7d) ROCK Nti momma me time kakra wae So give me some little time (7e) OK: Hwan ne small no Who are you calling small In example (7a), it is realised that an insertion is made from Akan into an English-based bilingual structure. As stipulated by the Matrix Language Frame model (Myers-Scotton, 1993a), the language which plays a more dominant role in intra-sentential CS is the Matrix Language (ML) while the other one is the Embedded language (EL). The ML provides the grammatical structure that allows for insertions from another language. Therefore in example (7a), the English language becomes the ML. However, it is very important to note that in spoken CS, hardly do such examples like (7a) occur. English is the default language on WhatsApp and it is expected that in intra-sentential CS, English would carry the structure into which insertions from the local language would be made. However, the data shows that there are few instances where English- based bilingual structures carried insertions from the Ghanaian local languages. This is because in spoken CS, such insertions are not the norm. Therefore what is realized is that although English is the default language, chatters still manoeuvre their way to use intra-sentential CS. When intra-sentential CS is used, Ghanaians 69 University of Ghana http://ugspace.ug.edu.gh usually stick to intra-sentential switches where the pattern is Akan-based bilingual structure in which Akan functions as the ML. As a result, there are many occurrences of Akan-based intra- sentential CS structures where English words are inserted into them. As shown in examples (7b, 7c, 7d, 7e), Akan provides the grammatical structure and serves as the ML, while the English words are inserted. In comparing this analysis to Marjie’s (2010) analysis of Swahili based texts, a sharp contrast is noticed in relation to how Swahili speakers use language in chats. This is because to Swahili speakers, the default code for both spoken and written communication is Swahili. So, they transfer their spoken code-switching into writing. Therefore, when communicating on a computer-mediated platform like WhatsApp, they transfer their spoken CS onto the platform. However, they have a choice to make whether to treat an English word as borrowed or code- switched depending on the medium (print or electronic writing). On the other hand, Ghanaians’ use of language in the spoken and the written differs. This is because on WhatsApp, chatters are required to use English and not their local languages. They move from the default local languages in spoken communication to English in written communication. Therefore, it is expected that insertions would be made into English structures, which is not the case since oral CS does not permit such switches. Because of this, chatters cannot just make insertions from the local languages into English. This results in the minimal use of intra-sentential CS. Whereas it is easier to find many instances of intra-sentential CS than inter-sentential CS in spoken data, the WhatsApp conversations revealed few instances of intra- sentential CS. This is because in spoken language, the local languages function as the matrix language, the default language into which insertions are made from English and not the other way round. So on WhatsApp, the default language has become what should have been embedded 70 University of Ghana http://ugspace.ug.edu.gh in spoken language. Hence, there is the expectation of the reversal of the structural roles of the languages involved. This limits the flow of intra-sentential CS, which could be attributed to the fact that intra-sentential CS is rule-bound and Ghanaians cannot create intra-sentential CS structures in which English is the ML into which insertions are made from the local languages since it is not expected in spoken CS. As a result, they then transfer their spoken intra-sentential CS into WhatsApp conversations by speaking Akan and inserting English words as shown in examples 7b, 7c, 7d and 7e above. 3.1.2.4 Single Word Switches as Discourse Markers In the data, some Akan single word switches that occur in English constructions function as discourse markers (DMs). Though these Akan discourse markers in English structures exemplify intra-sentential CS, they are discussed as a type of CS. The Akan discourse markers are not sentences integrals like other lexical words. They do not participate directly in the sentence frame. That is why anytime they occur in English based structures, English is actually the Matrix Language. Schiffrin (1987a: 31) defines discourse markers as “sequentially dependent elements that bracket units of talk, i.e. nonobligatory utterance – initial items that function in relation to ongoing talk and text”. Fraser (1988:22) explains that discourse markers are “lexical adjuncts to and are independent of an already well-formed sentence”. Brinton (1996:33) and Jucker & Ziv (1998b:3) argue that discourse markers have the following features: a. They are mostly a feature of spoken rather than of written discourse. b. They occur frequently in oral discourse. c. They undergo phonological reduction. 71 University of Ghana http://ugspace.ug.edu.gh d. Their syntactic position within discourse is not fixed and as a result may occur at the initial, medial or final position. e. They are considered to have little or no propositional meaning, or at least to be difficult to specify lexically. f. Though discourse markers may be found outside or loosely attached to a sentence, they have no clear structural function. Thus, they appear to be optional rather than obligatory features of discourse. According to Fraser (1988:22), the absence of discourse markers “does not render a sentence ungrammatical and/or unintelligible”. g. They possess the characteristic of multifunctionality. Thus, they function on the local and global levels concurrently; however, it is problematic to distinguish a pragmatically motivated from a non-pragmatically motivated use of the form. In the data, Akan discourse markers are employed in constructions which are predominantly written in English. The English language allows the insertions of these Akan DMs because they do not interfere with the English structure. Also, because of the peripheral nature of these DMs, the English language provides the slots for their insertions. These Akan discourse particles used in the conversations of the participants of each community of practice show the level of informality among chatters of the WhatsApp group chat. Examples 8a, 8b, 8c: The examples below show the peripheral nature of the DM Saa which is found in structures written in English: Item: Saa Example(8a)PAUL: Lol That yellow icon has been loading saa Laughing out loud. That yellow icon has been loading repeatedly 72 University of Ghana http://ugspace.ug.edu.gh Example(8b) LEE: Hehehehh.......heaven laugh saaa Hehehehh.......heaven laughs repeatedly Example(8c) JOE: I laugh saaaaaaaaaaa,heerrrrhhhh Sammy I laugh continually Sammy The excerpt above shows the use of saa as a discourse marker. Saa in these examples is optional and can be removed without distorting the meaning of the utterance. Moreover, it is found on mutual grounds i.e. final and medial position and does not interfere with the structure of the utterance. Therefore, this DM in example (8c) above shows how the participant laughed. It shows that the laughter was continuous. In the examples (9a, 9b, 9c) below, the discourse marker item koraa is found in constructions written in English. It also illustrates the peripheral nature of the Akan DMs. Item: Koraa Example(9a) If you have any Gh girlfriend saying I If you have any Ghanaian girlfriend QUAO-US: will think abt it........., leave her saying I will think about it...leave her koraa........ The black American girls immediately... The black American will arrive next week. girls will arrive next week 😂😂😂😂😂😂😂😂 Example(9b)MONS: Massa.... I laf enter mugabe polling Master...I even laughed and entered station koraaa Mugabe’s polling station Example(9c)IMMA: He was not discerning to know koraaa. He was not discerning to know at all. He wanted me to loose my precious wife. He wanted me to lose my precious Na lie! wife. It is a lie. Koraa in these extracts fulfils the function of a discourse marker since its absence does not render the construction meaningless. The above examples show that the use of Akan discourse markers are largely dependent on the context and the sequence of the conversation. 73 University of Ghana http://ugspace.ug.edu.gh 3.1.2.4.1 Functions of Akan Discourse Markers in English Constructions When these Akan DMs occur in English constructions, they usually serve as intensifiers. They intensify the effect of what is said. Moreover, they perform different functions depending on the context in which they appear. Likewise since the chat conversations are usually between peers, the DMs found in the data perform several interpersonal functions such as backchannel signals, response and reaction markers, and confirmation markers. In this section, the functions of some of the Akan DMs that occur in the data are discussed. 3.1.2.4.1.1 Saa This DM produces different meanings in different contexts. Below are examples to illustrate this. Example (10): This dialogue takes place in KNUST Organa Alumni Family chat group. In the conversation, HN tells the other participants of the group about his lost phone. This DM is used as a backchannel in the example below. Example (10) HN: I lost ma phone to robbers I lost my phone to robbers HN: Ve nw gotten anoda one I’ve now gotten another one TL: Fast ooo.I'm all ears Fast ooo. I’m all ears KOA: Oh saaaaaa? Oh really LSA: Saaaaaaa was it IPhone or yam Really, was it IPhone or yam? It is evident from the dialogue that as HN tells the others about his lost phone, KOA and LSA employ the DM saa to indicate to HN that they are interested to know the type of phone that got missing. In this context, saa performs a back-channel function which illustrates that the listeners are attentive to the speaker. Saa in a different context may not perform a backchannel function but only act as an intensifier. Below are examples 11a, 11b, 11c. In these examples, saa is used to intensify what the speaker says. 74 University of Ghana http://ugspace.ug.edu.gh Example(11a)DAVE- Eiii Ivy... Edward and I have Eiii Ivy...Edward and I have been looking RUSSIA: been looking for you saaa for you repeatedly Example(11b)EMMA: Masa ur designing for people Master you’re designing for people saa continually. Example (11c): TILLY Laugh saa cos more questions r Laugh more because more questions are coming coming In examples 11a, 11b, 11c, it could be realised that the use of saa in these constructions shows an activity which has been repeated. Moreover in example (11c), the speaker who is cautioning another participant uses saa to inform him to continue laughing since he has got more questions to ask him. 3.1.2.4.1.2 Wai Wai is often used as a closing frame marker and is very useful in the organization and assignment of turns during interaction. The original form of wai is w’ate [you hear]. However, the expression has now undergone elision process and users of the Akan language usually say wai instead of w’ate to the extent that in the WhatsApp chats, it is predominantly seen as a DM. In the examples 12a, 12b, 12c, 12d, 12e below, this DM is usually found at the end of these constructions: Example(12a)TIM: Why are you even talking? Why are you even talking? You're You're ineligible to vote so keep quiet wai! ineligible to vote so keep quiet you hear Example (12b)FII: We are liking it wai. Guys for you. We like it you hear. Guys for you Example(12c)LEE: Mills listen to this was....the said name Mills listen to this … the said above is wrong wai name above is wrong you hear Example(12d)FII: Massa enjoy your stay in Hong Kong wai. Master enjoy your stay in Hong Kong you hear Example(12e)IZZY: I'm starting to believe you don't follow I'm starting to believe you don't issues well here in the group much more follow issues well here in the Ghana politics. Please kam and vote wai group much more Ghanaian 75 University of Ghana http://ugspace.ug.edu.gh politics. Please come and vote you hear In the examples above, wai usually occur at the end of the constructions; thus, serving as a closing frame marker. 3.1.2.4.1.3 ooo Another DM that is usually used by Ghanaians in their chats is ooo4. This DM shows the Ghanaianisms portrayed by members of the chat groups. This DM is not peculiar to members of a particular ethnic group but usually used by Ghanaians during their chats. It is an intensifier and the most frequently used DM in the data. Here are some cases of its usage in examples 13a, 13b, 13c, 13d, 13e. Example(13a)ERIC-HK: You see ooo..the NPP is a family You see that the NPP is a family thing thing. Example(13b)IZZY: Hw3, that's what I'll be using to Look, that's what I'll be using to vote ooo, 1 man 1 woman 1 vote, 1 man 1 woman 1 concubine 1 concubine 1 sidechick... Dassooor side chick... that’s all Example (13c)DQ: We are kicking ooooo We are kicking Example(13d)TILLY: We have our preferences ooo We have our preferences Example(13e)ERIC-HK: Sammy protect your voters' id ooo Sammy protect your voters' ID In the above examples, the use of ooo brings out the ‘Ghanaianness’ of the conversations. The use of this DM can be attributed to the fact that Ghanaians usually use ooo when speaking their local languages and thus, transfer this practice into WhatsApp conversations. 4 Ooo is usually used by Ghanaians in spoken language. However, it is used pervasively in our data on WhatsApp chats. 76 University of Ghana http://ugspace.ug.edu.gh 3.1.2.4.1.4 Koraa Koraa also functions as an intensifier and it is one of the most frequently used DMs on WhatsApp. This DM may produce different effects in different contexts. Consider the following examples in (14a), (14b) and (14c). In these examples, the use of koraa produces different effects in different contexts. Example (14a): This is a dialogue between participants of the Ahwiaa R/C & L/A School chat group. Doris has been offline and just comes online and realizes that there has been a discussion before she got online. The dialogue revolves around the fact that she would have many messages to read. MARVIN: Den u ve 1000000000 messages to read Then you have 1000000000 messages to read DORIS: Yes oo Yes ooo MARVIN: Long essay Long essay DORIS: 😜😜😜 😜😜😜 DORIS: Have run out of battery koraaa I have even run out of battery In the interaction above, Doris acknowledges the fact she has many messages to read. However, she alerts the other participants of the group that she has run out of battery. In alerting the group members, she uses koraa to inform the others that she cannot read all the messages since she has run out of battery. The use of this DM shows the intensity of the problem which could lead to her going offline since her phone could go off. In the example (14b) below, the use of koraa produces another effect. In the example, Rich uses koraa as he complains that the chat group is not as active as it used to be. Example (14b): RICH: Comrades of late the platform is not Comrades, of late the platform is not busy koraaaaa busy at all 77 University of Ghana http://ugspace.ug.edu.gh It is evident in the example (14b) above that as koraa is used to describe the situation, it shows the low patronage on the platform. Below is example (14c) which also illustrates another use of koraa. This example is taken from Zighiss WhatsApp chat group. Participants of the group are informed about an upcoming wedding and Linda makes this comment. Example (14c) LINDA: I have even sowed my ataar koraaaa I have already sewed my attar In the example (14c), Linda’s use of koraa in her utterance shows that she is very ready to attend the wedding. 3.1.2.4.1.5 Papa The Akan DM papa is also used as an intensifier. Consider the following example. Example (15a) is a dialogue between some participants of the KNUST Organa Alumni Family chat group. In this dialogue, some of the participants talk about the love of money. In this discussion, Anna uses the DM papa when she talks about another member of the group. 1.ANNA: Cause u can easily love money Because you can easily love money 2.ANNA: Esp Dr Terror Especially Dr Terror 3.QUOFI: Hahahaha Hahahaha 4.ANNA: He loves money papa He really loves money 5.QUOFI.: Joy how is my brother doing Joy, how is my brother doing? 6.QUOFI: Lol Laughing out loud 7.QUOFI: Money itself is a good thing Money itself is a good thing 8.QUOFI: Is the love of it that ......... Is the love of it that ......... 9.QUOFI: Hahahaha Hahahaha 10.ANNA: Dr Terror loves money papa Dr Terror really loves money 78 University of Ghana http://ugspace.ug.edu.gh It is seen in the example above that Anna uses the DM papa to intensify her claim that Terror loves money in turn 4. She emphasizes her claim by repeating the same utterance in the last turn. The use of papa shows the intensity of Terror’s love for money. 3.1.2.4.1.6 Paa This DM functions as an intensifier on WhatsApp and can produce different meanings in different contexts. Consider the following examples below. Examples (16a), (16b), (16c), (16d), (16e) & (16f) are taken from the various groups and they illustrate the different implications for the use of the DM paa. Example(16a) KWESI: Chatty is true paaaa Chatty it’s really true Example(16b) KWESI: We should hv stopped fr Zagidi to We should have stopped for Zagidi beat him paaaa to beat him Example(16c) KWESI: I should hv served commissioner dat I should have served commissioner day paaa that day Example(16d) CLE-US: Eish counting paaaaa Eish counting? Really? Example(16e) Dorcas: S h e did well yesturnyt paaaa She did well last night Example(16f) ANNA: Becareful paaaaaa Be careful The use of paa in the above varies from example (16a) to the last example (16f). In (16a), paa emphasizes what the speaker says. Thus, the speaker uses this DM to indicate to the other participants of the group that what he is saying is nothing but the truth. In (16b), another function of the use of paa is seen. Its use in (16b) shows the speaker’s regret for not allowing Zagidi to beat that person. Before this utterance was made, Kwesi was recounting his past experiences to the members of the group and made this comment. In (16c), the speaker uses paa to emphasize 79 University of Ghana http://ugspace.ug.edu.gh his regret for not serving the commissioner on a particular day. In (16d), the use of paa shows the speaker’s doubt. In this context, the participants of the group were contemplating about why the election5 results had not been released. Then, one participant suggested that the Electoral commission was still counting the ballots. It was at this point that CLE-US came in with this utterance where he used the DM paa to show his doubt about the counting. Example (16e) also gives us another effect of the use of paa. In this example, it is used as an intensifier. Prior to this utterance, there had been a ‘hot seat’6 session in Ahwiaa R/C & L/A chat group and the member who was hosted did a great job in answering all the questions posed her; therefore, the next day, Dorcas informs the others of that person’s outstanding performance. In so doing, she uses the DM paa to intensify her claim. In the last example (16f), paa is also used as an intensifier. In this context, it intensifies Anna’s warning to another participant who posted a joke on the group page. 3.2 Marked versus Unmarked codes on WhatsApp The data revealed that the dominant languages on the WhatsApp group chat are English and Akan. The Markedness Model states that in any bilingual speech event, one code is the expected (unmarked) choice while the other is the marked choice. 5 The 2016 presidential and parliamentary elections in Ghana 6 Hot seat is a question and answer session on most WhatsApp chat groups when a participant of the group is ‘hosted’ and expected to answer every question posed to him/her truthfully. 80 University of Ghana http://ugspace.ug.edu.gh 3.2.1 Unmarked CS on WhatsApp An unmarked code refers to the language that is expected in a given context. Since a particular code may be expected in a particular context, it becomes the language of default. The analysis of the data revealed three instances of unmarked code choice. These are discussed below: 3.2.1.1 English as the unmarked code choice on WhatsApp In WhatsApp chat conversations, it is expected for Ghanaians to employ the English Language as the unmarked code. This is because the English language is the language of literacy in Ghana. Chatters on WhatsApp usually choose it because of the ease that comes with its usage. Another motivation for the use of the English language is the ‘multi-tap’ technique during chats which makes using English less time-consuming. Moreover, the keypad of mobile phones supports the English language usage and thus makes word suggestions for the user as s/he types the messages. This aids in fast typing. Therefore, the use of the English language offers the fastest, easiest, and the most convenient way of saying something with the least waste of time, effort, and resources. Besides the convenience in its use, some Ghanaian WhatsApp users choose it as the unmarked code because of their familiarity and identification with the language. Chatters usually use the English language because of their perceived proficiency in it when they are very poor in relation to writing their local languages. Therefore, chatters who cannot write their local languages would always choose the English language since they can identify themselves with it. The example (17) below illustrates the use of English as the unmarked code choice. Example (17): Before this conversation ensues, a participant of the group sends a picture to the page. The discussion revolves around the creativity of the person who took the picture and the language used is English, the default language. 81 University of Ghana http://ugspace.ug.edu.gh MIKE: Eld [Elder]Rock I see you FII: Could this be photoshop? KEN: It sure is @233249366938 IZZY: It could be Photoshoped IZZY: Whoever did it too is very good FII: Wow. Whoever did it is very good. NANA: that's gh [Ghana] KEN: Original pix [picture] @233249366938 FII: Thanks IZZY: Aha! KWAKU: Z o om in to his lips and you will see the straw is just on the lips ... Not entering his mouth ... IZZY: Ken didn't wanted [want] to understand IZZY: This Photoshop is too easy IZZY: The picture is very appropriate and convenient KEN: Yh [yeah] PHAT: of cos [of course].....too easy It is clear from this interaction that the English language is the default code. This is because all the participants who come onto the page respond in this language. Moreover, the speaker turns in the conversation prove that all the participants have had a great exposure to the English language to the extent that they are able to use shorthand for some particular words they use in their interaction. 3.2.1.2 Pidgin as the unmarked choice on WhatsApp Pidgin becomes an unmarked code when it is the default code of communication for a particular group of chatters. In a WhatsApp chat group that is made up of only males, it is probable that the expected choice of language is pidgin. As projected by Dako (2002), about 80-90% of Ghanaian males below the age of 50 with secondary or tertiary education speak pidgin. In the data, the chats of Fraser Boys WhatsApp group indicate that Pidgin is the unmarked code of the group. Pidgin is an in-group language which provides an identity for its users. Thus, though members of that group are no longer school mates, they still use the code which enables them to identify themselves as peers. This is echoed in the following use of unmarked Pidgin. 82 University of Ghana http://ugspace.ug.edu.gh Example (18): This dialogue is between six participants of the Fraser Boys WhatsApp chat group. In this conversation, JJ informs the others about the food he is eating. As they talk about the food he is eating, they change their topic of discussion from food to women. In the entire conversation, it is only in turns 20 and 23 that are English utterances. Apart from these, the rest of the turns is in pidgin which shows that pidgin is the default language of communication and thus written in italics. PARTICIPANTS CONVERSATION TRANSLATION 1.JJ: The thing dey ova me The thing is over me 2.JN: Wey i de hung fr here The way I am hungry over here 3.JJ: So mke I set give u? So let me set it for you 4.JN: Yh Yeah 5. JN: U whet my appetite waaa You just whet my appetite 6.FS: Akora u boorrr oo Akora, are you angry? 7.FS: See the way u cut cut See the way you are drinking (probably an alcoholic beverage ) 8.K: What he Dey chop? What is he eating? 9.FS: Kokonte Konkonte (A kind of food in Ghana) 10.JJ: Eba+shitor oil .... Eba+shitor oil ....(a kind of food) 11.AS: Haaaaahaaaaaa Haaaaahaaaaaa 12.AS: nfodzo cut de alomo bottle show we Nfodzo, snap a picture of the Alomo bottle to show us 13.AS: eba no dey make man sweat so Eba cannot make you sweat this way 14.JN: Lolz laugh out loud 15.AS: e hard oo messi too en appetite dey It is indeed surprising that Messi’s fi make wet appetite is also whet 16.AS: *whet *whet 17.JN: Hahaha Hahaha 18.SL: Em say no bra dey He said there is no bra 19.AS: Wow Wow 83 University of Ghana http://ugspace.ug.edu.gh 20.AS: a moda breastfeeding her children A mother breastfeeding her children 21.B-US: I sure say these ladies be Nigerians I am very sure that these ladies are Nigerians. 22.AS: I know dem 2 dey stay abeka I know that the two of them stay at Abeka 23.B-US: Really ? Really? 24.B-US: Them be Gh girls ? Are they Ghanaian girls? 25.JN: Make u no mind am Don’t mind him 26.AS: yeah 2 of dem Yeah two of them 27.SH: There noooo u appear It is there you appeared 28.SL: Eeei lair... der nor jeff go talk say e Eeei liar… it is there Jeff ate one inside chop one for inside 29.JJ: Hahahahha...TI shotto get u Hahahahha...TI Shotto has gotten you so cheaaap... 😂 cheap. 30.B-US: Lol Laugh out loud 31.AS: Haaaaahaaaaa Haaaaahaaaaa Aswear I swear 32.AS: de timing bi perfect The timing is perfect 33.B-US: Agbamey where U go hide? Agbamey, where will you hide? 34.B-US: We search you for like one month I have been looking for you for close to this one month. 35.B-US: We figa say the gods kidnap U We want to say that the gods kidnap you 36.AS: ebi more dan 1 month It is more than 1 month 37.JN: Why sleepo always de talk lair Why is Sleepo always spelling lair instead of Liar instead of Liar 38.B-US: For nogokpo shrine At Nogokpo Shrine It is obvious from the first turn to the last turn in this interaction that these participants are aware that all the others in the group are conversant with pidgin. In view of this, each speaker is able to reply others in the same language. Even when AS replies SL in English in turn 20, he returns to the default code pidgin in turn 22. Note that these participants would not have responded to the turns in pidgin were they not familiar with it. This shows that these individuals treat pidgin as their default medium of communication. Since they are all males, they use it as a ‘guys-code’ to mark their solidarity. 84 University of Ghana http://ugspace.ug.edu.gh 3.2.1.3 CS as an unmarked code choice on WhatsApp Myers-Scotton (1993b) maintains that there are instances where the use of two codes could represent an unmarked choice. Some participants of the WhatsApp chat groups are Akan-English bilinguals who are competent in using two codes. A typical example is the participants of the KNUST PENSA Organa Alumni Family. This can be linked to the fact that Akans dominate this group and that their use of Akan-English CS influences others to also use CS as an accommodation strategy. Because of this, the use of Akan-English CS on the platform has become habitual. Usually, there is no attempt by either of them to pay special attention to any of the switches used in conversations. Example (19): This is a sample chat of the KNUST PENSA Organa Alumni Family. In this speech event, seven chatters discuss marriage and relationships. Their discussion revolves around disappointments in relationships. The chatters involved are SAM, PHAT, QWAKU, JOE, PW, KO and MAY. This extract begins with a switch from Akan to English in turn 1. It is followed by unilingual Akan which is also followed by Akan-English CS. After this, Joe comes in with unilingual English in turn 5. Then, other participants who come into the discussion use unilingual English/Akan or Akan-English CS in subsequent turns. PARTICIPANTS CONVERSATION TRANSLATION 1.K O: wo no no, love nurse...now u turn That’s you, (KO is suggesting that the relationship dictionary person is noted for being interested in relationship talk) Love nurse...now you have turned a relationship dictionary. 2.PHAT: Saa na metease3 that's how I understand it 3.MAY: Continue wai Continue you hear 4.PW: M a policewoman ooo I take first I’m a policewoman ooo I only take first responses p3....subsequent ones responses...subsequent ones aren’t ain't considered considered 5.JOE: Will term it broken hearted I’ll term it broken hearted 6.PW: 😂😂😂😂😂😂😂😂 r'ship 😂😂😂😂😂😂😂😂 relationship dictionary dictionary paaa indeed 7.MAY: Lol Laughing out loud 85 University of Ghana http://ugspace.ug.edu.gh 8.PHAT: Meka y3!!! I said it!!! 9.MAY: Broken whole body Broken whole body 10.PW: Awwwww wate mmaa no Awwww he has broken those women’s akoma💔 heart 11.K O: Sometimes u these guys will give ur Sometimes you these guys will give your hearts to the ladies and give them hearts to the ladies and give them hammers, wat do u want them to hammers, what do you want them to use use the hammer for?? the hammer for? 12.JOE: Daabi No 13.MAY: Vice versa rather Vice versa rather 14.PW: Woy3 sure Tilly Are you sure Tilly 15.PHAT: )mo maa no 💔💔 they gave him broken heart 16.PW: He said he abandoned de projects He said he abandoned the projects oooo oooo 17.JOE: Tilly, u r really following Tilly, you are really following 18.MAY: Daabi ok. Thats wat he is saying No okay. That’s what he is saying 19.SAM: Ok am bk lets continue Okay, I’m back let’s continue 20.MAY: Mummy woho te sen Mummy, how are you? 21.PW: Plz hu z nt following 😂😂😂😂 Please who is not following 😂😂😂😂 22.JOE: U all are You all are 23.PHAT: happened before de abandonment happened before the abandonment 24.K O: The broken heart gave him The broken heart gave him convulsion and convulsion and he had to abandon he had to abandon the project the project 25.PW: Eeeeeiiiii masere saaaaa....wala Eeeeeiiiii I have laughed boi ampa continually...wala boy indeed 26.JOE: Lol Laugh out loud 27.QWAKU: Bonnke crusade is on Bonnke crusade is on 28.PHAT: de ladies had to suffer for de 💔💔 The ladies had to suffer for the broken heart. 29.SAM: Bt my boss if wer jilted bf starting But my boss if were jilted before starting dos 2 abandoned projects deaa den those 2 abandoned projects, then you have u v bn in 3 diff relationships bf 19th been in 3 different relationships before 19th oooo anaa ebi ka ho ooo or there are others more The switching between English and Akan in the turns without attaching any importance to those switches indicates that the chatters were using CS as their default code for communication. These chatters are bilinguals and are aware of their bilingual status. Therefore, those switches in the turns do not come as a surprise to members of this group since it is expected. CS indicates these participants’ solidarity of being Akan-English bilinguals. 86 University of Ghana http://ugspace.ug.edu.gh 3.2.1.3.1 Akan-English CS as used to display multiple identities in WhatsApp group chats When CS is used as the unmarked choice, it is usually used in in-group conversations. Myers- Scotton (2006: 167) argues that CS itself as the unmarked code choice is used to show the multiple identities of the participants. Such switches inform other participants of the ability of the participants to speak two languages. Here is example (20) to support the discussion just made above. Example (20): This conversation takes place in KNUST PENSA Organa Alumni Family. ROCK sends his wedding invitation to the group. The interaction occurs at the point where SER, BIG, IZZY, SILVA and NAANA congratulate ROCK. This dialogue begins with a turn in unilingual Akan. It is then followed by a turn which is in Akan-English CS. As the conversation progresses, the participants switch between English and Akan from turn 9 to turn 14. PARTICIPANTS CONVERSATION TRANSLATION 1.SER: Nyame y3da wo ase God, we thank you 2.ERIKA: Congrats Papa Rock. Nyame Nhyira Congrats Papa Rock. God bless w’aware your marriage 3.EA: Terror can't read this message today wai Terror can't read this message 😜😂😂💃� today you hear 😜😂😂💃� 4.COBY: Doc Doc Doc Doc 5.EA: Not forgetting ken😂😂😂😜 Not forgetting Ken😂😂😂😜 6.SER: 😂😂😂😂 😂😂😂😂 7.ERIKA: 😁😁😃 😁😁😃 8.NAANA: Congrats Papa Rock Congrats Papa Rock 9.DAVE-RU: I perfectly agree. Y3b3 sesa y3 Assemblies I perfectly agree. We will change ooo. Bibiaa y3 change� our Assemblies too. Everything is change 10.SILVA: Congrats rock. Congrats Rock. 87 University of Ghana http://ugspace.ug.edu.gh 11.IZZY: Tetelestai... 3y3 me k)n) dodo Tetelestai...it is appetizing 12.ROCK: Y3da mo ase bebree�� We thank you so much 13.ROCK: Pray with us and prepare to grace the occasion Pray with us and prepare to grace for us🙏�🙏� the occasion for us🙏�🙏� 14.BIG: Hehehe...saa Saaaa Saaaa.... Hehehe Yes, yes, yes In the above dialogue, the ROCK, SER, BIG, IZZY, SILVA and NAANA negotiate their identities by means of code-switching. This could also be for stylistic purposes since these chatters feel comfortable switching between Akan and English. 3.2.1.3.2 CS as used to negotiate the development of a conversation Code-switching on a WhatsApp group chat can be used to negotiate the development of a conversation. When CS is used in the development of a conversation, it maintains turns. In the example below, the participants switch between English, pidgin and Akan just to maintain turns in the dialogue. Example 21: The dialogue is between four male participants who are watching a football match in their various homes, but come online to discuss the match. This conversation displays a sequence of unmarked codes. In this context, Akan, English and Pidgin are unmarked codes. In the turns, English is in normal font, Akan is bold while pidgin is italicised. The first three turns begin in the English language which is followed by a switch to Akan-English CS in turn 4. After this, the next participant comes in with the English language in turn 5. As the conversation get to turn 9, there is a switch from the English language to pidgin which is followed by a switch to unilingual Akan in turn 10. As the turns in the conversation develop, the participants switch to the unilingual use of Akan, English and pidgin. In turns 15, 17 and 35, note the unilingual use of Akan. Pidgin use is also seen in turns 16, 19, 20, 23, 38, 43, and 44. However, note that 88 University of Ghana http://ugspace.ug.edu.gh participants adopt unilingual English massively than the other unmarked codes available. It is adopted in 25 turns in the conversation. This can be attributed to the nature of interaction on this medium. This is because WhatsApp in the Ghanaian context facilitates the use of the English language than other languages. Akan-English CS also occurs in turns 11, 30, and 46. The dialogue ends with turn 47 in the English language. Example (21) 1. ALEX: Atletico all de way 😜😜😜😘😘😘 Atletico all the way 😜😜😜😘😘😘 2.RICH: Clatenburg will spoil the match Clatenburg will spoil the match 3.RICH: The refree The referee 4.RICH: Against ani bɛ wu Those against will be ashamed 5.OWURA: Poor messi is now glued 2 his TV set Poor Messi is now glued to his TV set watching gud soccer from two best watching good soccer from two best team in d world teams in the world 6.RICH: Hahahahahaah Hahahahahaah 7.OWURA: 😂😂😂 😂😂😂 8.ALEX: U don't know wat u aa saying�� You don't know what you are saying�� 9.OWURA: Eii Rubin no malice, y u go hammar Eii Rubin no malice, you will hammer my ma head? head? 10.OWURA: Atanfo no mmo hyire3 Enemies be happy 11.OWURA: Na mo kae s3 penalty aba nti 3no a So you people thought there is a no no penalty so that is it 12.OWURA: I don't feel d heat of a final game in d I don't feel the heat of a final game in the ongoing match, r u wid me on-going match, are you with me 13.RICH: Yes true Yes true 14.OWURA: 😉😉 😉😉 15.CLEMENT-US: Osey atletico Hurray Atletico 16.OWURA: Atletico no go do tee Atletico can’t do anything 17.CLEMENT-US: Yoo I hear 18.OWURA: Wait till we increase d score to 2 Wait till we increase the score to two 19.CLEMENT-US: For the where At where 20.OWURA: Shine ur eyes u go see now Be vigilant; for you will see what will happen 21.CLEMENT-US: Goaaaaaaaaaalllll Goal 22.OWURA: Yes I c Yes I see 23.CLEMENT-US: Shine ooooo micky Be vigilant Micky 24.CLEMENT-US: History History 25.OWURA: Now we r zooming into d game Now we are zooming into the game 89 University of Ghana http://ugspace.ug.edu.gh Proper proper 26.OWURA: D heat is on The heat is on 27.OWURA: Wear ur seat belt cos we r abt 2 take Wear your seat belt because we are about flight to take flight 28.CLEMENT-US: Let's see Let’s see 29.ALEX: Mike u aa now watching football Mike you are now watching football 30.ALEX: No hasty conclusion wate!!! No hasty conclusion you hear 31.OWURA: No shaking No shaking 32.ALEX: In our childhood u ddnt play football In your childhood, didn’t you play b4 football before 33.OWURA: Na lie It is a lie 34.OWURA: I didn't play well bt no Hw 2 watch I didn’t play well but I know how to watch 35.ALEX: Tweaaaaa (it is not true){this is not its meaning but in this context that is the meaning implication it gives us} 36.ALEX: U may lack some technicalities You may lack some technicalities 37.CLEMENT-US: Mourinho never played before but Mourinho never played before but he’s he's one best coaches lol one of best coaches. Laughing out loud 38.OWURA: U hear rubin Did you hear? Rubin 39.CLEMENT-US: Where is CR7 Where is CR7? 40.OWURA: Hes out of d game nt literally He’s out of the game not literally 41.OWURA: Messi has bewitched him nt 2 make Messi has bewitched him not to make history today history today 42.CLEMENT-US: Hahahaha Hahahaha 43.CLEMENT-US: I beg let messi be lol I beg you let Messi be. Laughing out loud 44.OWURA: Ah e no b him hu dey do am Ah he is not the cause 45.OWURA: Tonyt is nt CR7 nyt Tonight is not CR7 night 46.ALEX: No Leave Messi out wae No, leave Messi out you hear 47.ALEX: Don’t worry Bro. Don't worry... In the example above, the participants use CS as part of the strategy to maintain their turns in order to keep the flow of the conversation. Thus in the conversation, the participants choose three unmarked codes which are alternated in sequence. Therefore as they switch from one code to another, they signal their competence in the use of those codes. 90 University of Ghana http://ugspace.ug.edu.gh 3.2.2 Marked CS on WhatsApp The marked code refers to the language that is unexpected in a given context. The Markedness Model further suggests that speakers use the marked code when they want to send other intended meanings to their interlocutors. The unexpected code choice is marked because it contradicts what the hearer expects; let us consider French. It is the second international language of Ghana. However, not many Ghanaians can speak or write in it. Therefore depending on the kind of participants and context it is used, hardly do Ghanaians use it in their WhatsApp communication. So its use could increase or decrease social distance depending on the context. Hence when some chatters decide to use French on a platform in which only few people understand, then they tend to narrow the conversation to only those who understand the language so that the others cannot share their views on the topic. Consider the Example (22) below: In KNUST PENSA Organa Alumni WhatsApp group, two participants suddenly switch to French during a discussion, but another participant tries all possible means to stop them from using a marked code. In the dialogue, MPK and EA use French which is a marked code in turns 1, 2 and 3. QA, who does not understand French, also in turn 5 tries to chat in a language (gibberish) which does not exist just to stop MPK and EA from using a marked code on the platform. When DQ protests, it forces EA to use English in turns 7, 9 and 10. DQ does not stop using the gibberish language though EA changes her code to the expected code in turn 12. She (DQ) confuses them in turn 12 through to turn 20 with her gibberish language to the extent that EA and MPK try to figure out what she is trying to say. When MPK and EA realise that they cannot understand DQ, they go back to using French in turns 22 and 23. As soon as they resort to using French, DQ resumes her gibberish language in turn 24. 91 University of Ghana http://ugspace.ug.edu.gh Participants Conversation Translation 1.MPK: Quoi? What? 2.EA: Quel mansonge??. What a lie?? 3.MPK: Bien évidemment mais la femme semble un Obviously good, but the woman seems a peu FBI sur son mari bit FBI on her husband 4.EA: 😒😡😣 😒😡😣 5.DQ: Mansonge eele quie? [trying to also speak (Gibberish) French which does not make sense] 6.DQ: 😒😡😣 😒😡😣 7.EA: Eeeei Diana,what language is Eeeei Diana, what language is that??😁😁😁😂😂😂 that??😁😁😁😂😂😂 8.DQ: French French 9.EA: I swear😂😂😂😂 I swear😂😂😂😂 10.EA: From which country.?? From which country? 11.MPK: Republic of kumasi Republic of Kumasi 12.DQ: Paddone koraa lemme e tai [trying to speak (Gibberish) French which does not make sense] 13.DQ: 😂😂😂😂👍💃� 😂😂😂😂👍💃� 14.EA: I cannot see that?? Eele quie (This is a quote I cannot see that??(gibberish) of what DQ said) 15.DQ: It means mansonge is sleeping? It means lie is sleeping 16.EA: Hahaha Hahaha 17.EA: But who taught you that But who taught you that 18.MPK: 🚨🚑🚨🚨 🚨🚑🚨🚨 19.EA: 😁😁😁😂😂😂😂😂 😁😁😁😂😂😂😂😂 20.DQ: Akwasi Nyarko Akwasi Nyarko 21.EA: Huh cheri, Ta's traduit ca pour Elle,??? Huh darling, did you translate this for her??? 22.MPK: Avec quel logiciel? With which software? 23.DQ: Capour Elle ellei aaa [trying to also speak (Gibberish) French but it is meaningless] 24.DQ: 🍦if you are confuse take this 🍦if you are confused take this 25.DQ: U will understand my french you will understand my French 26.EA: Diana, attention sinon!!�🔫💣🔨 Diana, be careful or else!! 27.DQ: 😇 😇 28.DQ: Then English please Then English please 29.EA: Je sais meme pas I don’t really know 92 University of Ghana http://ugspace.ug.edu.gh 30.DQ: How can your mother be on this page? How can your mother be on this page? 31.DQ: 😳😳😳😏 😳😳😳😏 32.EA: Diana STP laisse moi tranquille Diana please let me be 33.DQ: 😡😡😡 😡😡😡 34.DQ: Stop confusing the noble people on this page Stop confusing the noble people on this ooo page ooo 35.EA: 😂😂😂😂😂💃�💃�💃� 😂😂😂😂😂💃�💃�💃� 36.MPK: R u da spokesperson of da noble people? Are you the spokesperson of the noble people? 37.DQ: Concerned citizen Concerned citizen In the above, French is a marked code. Consequently, its use distances MPK and EA from other participants in the WhatsApp group but narrows the social contact between them. As DQ distracts them, they switch to the unmarked code which signals clearly that they acknowledge their use of a marked code. However, their use of a marked code may probably have been unintended to side-line or exclude the others from the discussion. This may also suggest that probably the two participants usually use French when they chat outside the group page and as a result, may have unconsciously brought that code onto the group page. Usually in WhatsApp group chat, some people transfer their informal relationships unto the platform forgetting that they are not the only people in the group. It is important to note that it is only on the group page that French becomes marked. Had they used French in their private chat session, it wouldn’t have been marked. This buttresses the point that it is the context that determines whether a code is marked or unmarked. 3.2.2.1 Local languages as marked codes on WhatsApp Since the English language is the default language on these platforms, the use of other local languages is marked. Myers-Scotton argues that bilinguals make a conscious bid for new Rights 93 University of Ghana http://ugspace.ug.edu.gh and Obligations (RO) set when they use a marked code. When Ghanaians use their local languages on WhatsApp, it is a challenge to them since the spelling system of mobile phones does not support those languages. Sometimes, the spelling system of the mobile phones changes what is being typed into something else. Chatters who want to communicate certain meanings through their local languages would have to retype the message in order to get the intended message to the group. Sometimes, they use mathematical symbols just to write in their local languages. Because of the struggles and challenges that people go through just to switch from the English language to their local languages, they switch to communicate certain meanings. Examples (23a) and (23b) illustrate some of the difficulties that some Ghanaians endure in order to write in Akan. In example (23a), the chatter employs mathematical symbols just to write in Akan: Example (23a) Participant Message Translation JM Congrats Rock......y3b3hyia w) Kum n'ase Congrats Rock….we will meet in Kumasi 3 has been B racket has used in place of ‘ԑ’ been used in place of ‘ᴐ’ 94 University of Ghana http://ugspace.ug.edu.gh Example (23b): This example also illustrates some of the problems chatters go through to write in Akan on WhatsApp. In this example, the Akan word menua changed into something else (menus) because the keypad does not support the typing of Akan alphabet. Therefore, Sam had no choice than to rewrite the word again and send it to the group. SAM: Medie medan s3 me Maame wawo na ose 3y3 menus All I know is that my mother has given birth and she has showed the baby to me that s/he is my sibling SAM: Menua* Sibling So when Sam rewrites the correct word to the group, he attaches a star symbol to it to indicate to the other participants that what he wrote first was wrong and that he has rewritten the correct word. Note that the above Akan expression by Sam has a deeper meaning and it is the context of its usage that brings out the intended meaning. The use of marked CS on WhatsApp has several implications. Moreover, it is worthy to note that these implications of marked CS bring out the factors motivating chatters’ use of marked CS on WhatsApp. Usually chatters use marked code-switching to achieve their communicative objectives which include their inability to express their thoughts in English, to quote others, to be emphatic and to create fun. 3.2.2.1.1 Factors motivating chatters’ use of code-switching in the WhatsApp group chat The analysis of the data has shown that there are a number of practical factors that might be responsible for CS between Akan/any Ghanaian local language and English in WhatsApp group chats. Among these are the following: 3.2.2.1.1.1 Inability to express thoughts in English 95 University of Ghana http://ugspace.ug.edu.gh Chatters switch between codes because they believe an expression in one language carries what they want to say better than in another language. This implies that an expression or a lexical item from ‘one language fills a pragmatic gap in the other language’ (Myers-Scotton 2006: 143). It also indicates that participants would switch from one language to another when there is no translation equivalent to the word that would best describe their thoughts, experiences and feelings. For instance, the expression “w'asum y3 wo d3” and the word “nketenkete” in examples (24a) and (24b) below are switched in the middle of the sentence as there were no corresponding English words that could replace them. Example (24a): Five participants are chatting in English but KOA suddenly switches to Akan because it conveys what he wants to say better than English. KOA: Mr Paido needs her to survive Mr Paido needs her to survive KOA:😄 😄 ADJOA: I was jux passing thru I was just passing through ADJOA: Gudnite fam Good night family KOA:We understand We understand KOA:Someone needs you Someone needs you ADJOA: Sure😉 Sure😉 KOA: Good night dear Good night dear KOA:Pass through more times Pass through more times SIKA: Still waiting for the good news ooo @ Still waiting for the good news ooo from Eld.Rock Eld.Rock KOA: Eiiiii Nancy w'asum y3 wo d3 ooo Eiiiii Nancy[literally it means ‘your ears are sweet to you’] You like to listen to gossip G’fua: Don't worry😊 Don't worry😊 G’fua: It would definitely be someone we know It would definitely be someone we know SIKA: Papa Papa From the example, KOA’s switch to Akan enables him to better express his thoughts. Therefore, he would not have been able to express himself better in English had he not switched to Akan. Example (24b): In this dialogue, KEN switches to Akan (intra-sentential) because it seems he does not know the English word for the Akan word, nketenkete. 96 University of Ghana http://ugspace.ug.edu.gh KEN: My grandfather often tells a story My grandfather often tells a story of a of a heavily polygamous who died of heavily polygamous man who died of nketenkete [literally means ‘small small’] extensive hunger at night ANNA: 😂😂😂😂 😂😂😂😂 ANNA: 😂😂😂😂😂😂 😂😂😂😂😂😂 KEN: Because all his wives thought this Wife Or that will Because all his wives thought one wife cook for him..... or the other would cook for him..... ANNA: I laugh saaaaaaa get nketenkete I have laughed continuously that I have gotten extensive hunger at night KEN: Hehhehehe Hehhehehe ANNA: Hehehehehe Hehhehehe Since Ken does not know the English word for the Akan word, he simply says it as he knows it in Akan and this enables him to convey his message to the other participants. Moreover, some of the Akan expressions in the data are proverbs and idiomatic expressions. Chatters use these expressions because they could not have written those proverbs and idiomatic expressions in English. Besides, it is better for them to be expressed in their original language than to be translated into another language. This is because an attempt to translate those proverbs into English would not convey the exact meaning. However, the meaning implication of an Akan proverb can be conveyed in the English language. The following examples are some of the proverbs found in the data. Example (25a): The chatters in this dialogue discuss the 2016 elections of Ghana. In their anticipation of the election results to be declared, this conversation ensues. In the dialogue, RUTHY makes an attempt to use an Akan proverb but because she is aware that English is the default language, she translates the proverb into English in turn 5. Later, she realizes that the translation does not convey the exact message so she then writes the exact proverb in Akan which is its appropriate language in turn 7. 97 University of Ghana http://ugspace.ug.edu.gh 1.FII: They can't declare all at once They can't declare all at once 2.FII: They will declare one after the other. They will declare one after the other. 3.Joel: They are jux talking too much. They are just talking too much. We want dem to win must they must relax and We want them to win; must they relax wait for the EC and wait for the Electoral Commission? 4.RUTHY: Bcoz the leaders are not acting ryt Because the leaders are not acting right 5.RUTHY: They are running mampam race! They are running very slow race! (literally mampam means alligator) 6.Joel: Exactly Exactly 7.RUTHY: )mo tutu mampam mirika [wasted efforts] When Ruthy rewrites the proverb in its original language in turn 7, she recognizes the fact that some expressions are better said in their original language than to be translated into another language. Example (25b): In this conversation, the participants discuss the defeat of the National Democratic Congress (NDC) in the 2016 elections of Ghana. Then, Fii uses an Akan proverb to describe the actions of the defeated political party. LEF: I have missed him I have missed him FII: Me hearing it from Awuku today? Me hearing it from Awuku today? FII: Anyway I understand you. Anyway I understand you. FII: Aboa biara bewu a )pre [literally it means: every animal struggles for breathe before it dies] ERIK-HK: Yes Yes ERIK-HK: Smh Shaking my head FII: Let's leave it here Anyway I understand you. In example (25b), Fii switches to Akan in order to quote an Akan proverb. This is probably because the Akan proverb best describes what he wants to communicate. This proverb implies 98 University of Ghana http://ugspace.ug.edu.gh that though the political party tried every possible means to win the election but failed, its members are finding it difficult to accept the defeat. Example (25c): In this example, one participant of the group is on the ‘hot seat’. Participants ask many questions but to Dor, the questions are not enough. Her comment leads Rich to reply her in Akan. OC-Italy: Sincerely Sincerely CLEMENT-US: Eish Eish DOR: It shuld be more hotter dan dis It should be more hotter than this RICH: Wowe ofoɔ nsa na wo hwɛ wo deɛ (literally: when you chew a chimpanzee’s hand, then you watch yours)What you do unto others shall be done unto you. OK: @ Richard to learn harder @ Richard to learn harder KOFI: I'm here @ oyibo I am here whiteman OC-Italy: You want to remind her not to You want to remind her not to forget forget you clement you Clement In the example above, the Akan proverb implies reciprocity. Therefore, Rich is telling Dor that what she wants someone to experience shall be meted out to her. Rich decides to use the proverb since it best explains his thoughts to the participants of the group. 3.2.2.1.1.2 Quoting someone else When chatters want to quote someone else, they can switch codes on WhatsApp. After participants quote someone, they then continue the story in the language which was used first in the conversation. Participants usually quote others during chats because they want to convey what is being said as vividly as possible by trying to retain the original and precise words used 99 University of Ghana http://ugspace.ug.edu.gh by the speaker whom they are quoting. Moreover, the use of direct quotations makes a bilingual’s conversation more factual. In this study, this was found to be true as the participants switched codes when quoting someone else and when transmitting famous expressions. The explanations and examples of such situations can be seen below: In the conversation below, Anna first speaks in Akan-English CS and then, Lizzy also responds in Akan. However, Coby switches from English to Akan when he wants to quote the exact word Anna had said in Akan. Therefore in turn 6, Coby encloses the Akan word in quotation marks to show that he is referring to what Anna had said initially in turn 2. Example (26a) 1.ANNA: Hahahahahahah Hahahahahahah (laughter) 2.ANNA: Mpena twie sweet u paaaaaa ooooo You really enjoy dating (literally: dating sweet you) 3.ANNA: 😂😂😂😂😂😂😂😆😆😆😆😆 😂😂😂😂😂😂😂😆😆😆😆😆 4.LIZZY: Hahaha! Hahaha! 5.LIZZY: Papa Rock ahunu amane! Papa Rock is in trouble 6.COBY: That word "mpena" is very heavy oo That word “dating” is very heavy oo From the example, we realize that Coby would not have used an Akan word in an English based sentence had he not intended to quote what Anna had earlier said. This establishes the fact that Ghanaians may switch from an unmarked code to a marked code when they want to quote what others have said. In this example (26b) below, Sam switches to Ga when he quotes what the former President Mahama of Ghana said during one of his campaigns. He encloses these quotes in quotation marks. Example (26b) SAM: When I say Merry Christmas you say "No When I say Merry Christmas, you say Ababas33" 3 times. Then you make your “No going backwards” 3 times. Then hands like Usain Bolt and do you make your hands like Usain Bolt and 100 University of Ghana http://ugspace.ug.edu.gh "Hweeeeeeeew" do "Hweeeeeeeew" (it is a sound and it implies the act of moving very fast) From the example, Sam makes the other participants aware that he is delivering the exact words of the former President that is why he consciously encloses them in quotation marks. Note that he would not have switched from English to Ga had he not intended to quote the exact words of the former President. Example (26c): In this conversation, Phat switches to Akan to quote a common expression which was a slogan for the National Democratic Congress (NDC), a political party in Ghana, during the 2016 elections of Ghana. In this quotation, he does not enclose the word in quotation marks but decides to write the word in capital letters in order to show that it is a quotation. This distinguishes the Akan word from the English words in the sentence. Joel: Ghanaians will forget soon Ghanaians will forget soon Jacob: Maybe Maybe PHAT: If this is the way workers are gonna get salary If this is the way workers are going to increments, den der shouldn't be any TOASO for any get salary increments, then there party..... shouldn't be any continuation for any party..... In the example above, Phat writes the Akan word in capital letters in order to show that it is the slogan of a political party. Therefore, Phat would not have switched from English to Akan had he not intended to quote this slogan. 3.2.2.1.1.3 Being Emphatic 101 University of Ghana http://ugspace.ug.edu.gh Another reason is that when chatters want to be emphatic about something, they adopt CS. When they switch codes, they are able to say whatever they intend in a strong way without any doubt. They switch probably because the code-switched expressions may have greater effect on their interlocutors. Consider the following examples; In example (27a) below, the participants deliberate about the politics in Africa. In being emphatic about why the government should increase salaries some weeks to leave office, FII switches from English to Akan. Example (27a) FII: If really he meant to do all these things he could If really he meant to do all these things, he have done it earlier. Why increase salaries two could have done it earlier. Why increase weeks to your exit of office. salaries two weeks to your exit of office. Adwen b)ne nkoaa. Only evil motives. And someone too will Obi nso nk)ka a politics. say (attribute it to) politics. We will start We will start killing people in the name of politics killing people in the name of politics In the example above, one finds that the switch from English to Akan displays the speaker’s emotional involvement in what he is saying. In conversation (27b) below, Rock switches from his second language, English, to Akan when he explains to the group members of the need for them to attend the wedding ceremony of fellow participants of the group. Example (27b) ROCK: Had we not tried small, they would have outnumbered Had we not tried a little more, they us for Listowell's wedding would have outnumbered us at Listowell's wedding DUBY: Hmm Hmm ROCK: Yes ooo. Thanks for reminding me of them Yes ooo. Thanks for reminding me of them ROCK: Sorry for behaving like a nagging wife😜 Sorry for behaving like a nagging wife😜 MAT: Lol Laughing out loud ROCK: But yesterday for instance, from my But yesterday for instance, from my interaction with Jona, he felt a bit interaction with Jona, he felt a bit disappointed because he had made food disappointed because he had made 102 University of Ghana http://ugspace.ug.edu.gh reservation for us and we couldn't make it. food reservation for us and we Enti 3y33 no ya kakra couldn't make it. It indeed hurt him a little ROCK: But thanks to you all for understanding me But thanks to you all for understanding me In (27b), Rock becomes emphatic when he tells participants of the group that their failure to attend weddings is very disappointing to the couple in question. 3.2.2.1.1.4 Humour Chatters on WhatsApp platform also switch codes for amusement purposes. In this case, the switches contain words or jokes in different languages in order to make fun or entertain other participants on the platform. The data reveals that there are several instances when participants switched codes just to create humour. In the following examples, CS is used to portray this poetic function: Example (28a): The following conversation takes place when the researcher asks permission from the group members in order to study them. However, KO pretends that he would not agree for the group to be studied. So Jake gives his number to the researcher. The researcher chats him privately to ask him to agree; he then comes onto the group page to say these in example (28a). As KO talks, he switches from pidgin to English, then to Akan in order to create humour 1.K O: Oh jake paa, why you take ma number give the Oh Jake too, why did you give my lady that I said I won't agree??? number to the lady whose [plea] I said I would not accept? 2.K O: Ah u paa Ah you too 3.ANNA: 🙅�🙅�🙅�🙅�🙅�💪�💪�💪�💪�💪� 🙅�🙅�🙅�🙅�🙅�💪�💪�💪�💪�💪� 4.K O: Jake I bore koraa, am leaving the page. Jake I am so angry so I am leaving the I'll join after the switching. page. I’ll join after the switching. 5.K O: See you, abi u make hapi, u want me to leave so See you, hope you are happy. You want u can switch codes...me boa am not going me to leave so you can switch codes...I anywhere, me de t33 wo. am lying; I’m not going anywhere, I was 103 University of Ghana http://ugspace.ug.edu.gh only kidding From the example, we notice that KO intentionally switched to Akan in turn 5 just to create humour to calm the tension that had been inflamed earlier. In example 28b below, there is a conversation between Okoro, EA and Rock. Rock replies Okoro in Akan and then, he later realises that the researcher is also on the platform. He then creates humour in Akan-English CS in turn 4, which results in laughter among the participants of the group. Example (28b) 1.OKORO: I have no idea oo Boss...only I have no idea Boss...only God will know..lol God will know… laughing out loud 2.EA: 👍� 👍� 3.ROCK: Nyame aaa ne wo boafo� � 🙌�🙌�🙌� Only God is your helper� � 🙌�🙌�🙌� 4.ROCK: Eiiii, may3 code switching😂😂😂 Eiii, I have done code switching 5.OKORO: 🙏🙏 🙏🙏 6.SLY: Hahaha Hahaha From the example, we notice a deliberate attempt by Rock to create humour on the page. Rock’s switch from unilingual Akan to Akan-English CS erupts laughter among the participants of the group. Example (28c): Tilly had said something in an earlier conversation so Izzy replies her in this dialogue in Akan-English CS by repeating a particular Akan word ‘adeԑ’ [thing] in order to 104 University of Ghana http://ugspace.ug.edu.gh create a poetic effect. This action of Izzy serves the purpose of providing amusement on the platform. ANNA: Terror Terror SAM: So r u ppl saying one party "is mad" or "will go So are you people saying one party “is mad soon" mad” or “will go mad soon”? IZZY: Oh Tilly paa, gyae saa koraa na boys p3 ade3 as) Oh Tilly too, stop that because if boys ade3 so ay3 ade3 ma no ay3 ade3 ama y3 nya ade3 want something to top up something to do something for it to become something in order to get something SAM: Plss explain further wai Please explain further you hear IZZY: The next government will continue The next government will continue PAT: I like that. The ade3 rhyming I like that. The something rhyming ANNA: Hahahahaha Hahahahaha The repetition of ‘adeԑ’ in Izzy’s utterance is done consciously in order to create a rhyming effect which in turn creates humour on the page. 3.3 INTERVIEWS This section discusses the motivations behind code-switching on WhatsApp platform using interviews conducted with some participants from the various groups used for this study. The goal of the interviews was to investigate why chatters choose a particular code over another. It must be noted that chatters were interviewed whenever they switched from an unmarked code to a marked code during group chat sessions. Therefore, the researcher contacted them privately to inquire of their reasons for a particular linguistic behaviour. The analysis of the interviews was grouped in terms of the reasons chatters gave to the interviewer. 105 University of Ghana http://ugspace.ug.edu.gh 3.3.1 Inability to find an appropriate expression in the unmarked code Chatters informed the interviewer that they usually switch to their local languages when they cannot find an appropriate expression in the unmarked code to communicate their thoughts. Thus, they resort to communicating their thought in the appropriate language. Below is an example of a private chat between the interviewer and Sam, a participant: Example 29: An interview with a member of KNUST PENSA Organa Alumni Family Sam is a 24-year old man who is a member of the KNUST PENSA Organa Alumni Family WhatsApp group. He has attained a university education and is currently on national service. He is Akan but speaks English and Pidgin. In a conversation with the interviewer, he describes his reason for switching to Akan while the other chatters were using English. Interviewer: Hello Interviewer: Please I need your views for my research. SAM: Hiiii Interviewer: Why would you say “Diana nam ntete noa saa na wate asi ago prama noso [continue hearing that till you get to the field].....Fine case” Interviewer: Instead of writing English SAM: Hahahahahaha SAM: It’s difficult for me [to] express sm[some] of de [the] proverbs in English dats [that’s] y[why] most often I mix dem[them] with de [the] local dialect Interviewer: OK. Thanks very much Interviewer: So what you said is a proverb right SAM: U[you are] welcome SAM: Yes plss[please] SAM: Its [it’s] like she is trying 2 hide [to hide] sm [some] moves to get away..... Interviewer: OK. Thanks very much for your assistance SAM: Thnx u [thank you] too....we hope to get ur findings n[and] if possible recommendations. Interviewer: Yes ooo. 106 University of Ghana http://ugspace.ug.edu.gh From Sam’s utterances, it is evident that he often employs CS on WhatsApp and his reason for the switch is that he could not have expressed an Akan proverb in English so he needed to write it in its original language. 3.3.2 Switching to create humour Interviews held with some of the participants revealed that they switch languages to create humour. Thus, chatters intentionally choose a particular code just to create fun on WhatsApp. An example from an interview that shows that chatters use CS to create jokes is given below: Example 30: Interview with a participant of Ahiwaa R/C Primary WhatsApp chat group King is in his late twenties and a member of the Ahiwaa R/C WhatsApp chat group. He is Akan and he speaks Akan, pidgin and English. He is a medical doctor by profession. He describes his reason for switching on the group platform. Interviewer: Hello. I want to ask a few questions Hello. I want to ask a few questions Interviewer: Plz why did you say “Oh as3m paaa Please why did you say “oh what an nni.... 3nny3 moaa.... a certain arsene issue...it is not your fault...a certain Arsene Wenger” instead of speaking English Wenger” instead of speaking English? King: Hello.... Hello... Interviewer: Hi Hi King: For a research? For a research? Interviewer: Yes Yes King: Ok.... Okay... Interviewer: Cos I thought it is easier and faster to Because I thought it is easier and faster to write in English than in twi on Whatsapp write in English than in Twi on WhatsApp King: True True Interviewer: So what is your reason for writing in twi So what is your reason for writing in Twi? King: But we want to localize whatsapp u know But we want to localize WhatsApp you know ��� Interviewer: Ok Okay Interviewer: How How King: S3 watease3 Hope you understand Interviewer: Aanne Yes King: Saaa saaa saaa Yes yes yes 107 University of Ghana http://ugspace.ug.edu.gh Interviewer: But I don't get your reason well ooo But I don't get your reason well ooo King: Oh.... just for fun.... Oh...just for fun Interviewer: Ok Okay Interviewer: Mate wai I hear you okay Interviewer: Medase paa Thanks very much King: 3nnaa ase Don’t mention it Interviewer: Plz do you also use pidgin on Whatsapp Please do you also use pidgin on WhatsApp? King: But maybe if u had some options that i But maybe if you had some options that I could choose one.... could choose one.... King: Yeah... sometimes Yes...sometimes King: Mostly to some specific friends Mostly to some specific friends Interviewer: When do you use pidgin When do you use pidgin? King: Some of my friends, I only speak pidgin For some of my friends, I only speak pidgin with them..... with them... Interviewer: What specifies those friend that you speak What specifies those friends you speak pidgin with pidgin with? Interviewer: Why do you speak only pidgin with them Why do you speak only pidgin with them? King: For those friends, that's how it's always For those friends, that's how it's always been been King: Like Gabby Silva Like Gabby Silva Interviewer: Are they only boys Are they only boys? King: Yep Yes Interviewer: So is it possible for me to say that pidgin So is it possible for me to say that pidgin has has become a guy's code become a guy's code? King: For Gabby, he's always speaking pidgin For Gabby, he's always speaking pidgin King: So everyone speaks pidgin with Gabby So everyone speaks pidgin with Gabby Interviewer: Ok Okay The comments by King show that he switches from English to Akan when he wants to create humour. So on the group page, he said “Oh as3m paaa nni.... 3nny3 moaa.... a certain arsene Wenger” [“oh what an issue...it is not your fault...a certain Arsene Wenger” ] in order to create humour. Moreover, he disclosed to the interviewer that he sometimes use solely Pidgin on WhatsApp. His comments show that the person he interacts with on WhatsApp determines his choice of code. So in a WhatsApp group made up of only males, he would frequently use pidgin. Hence he even mentions one person whom he always speaks pidgin with. 108 University of Ghana http://ugspace.ug.edu.gh 3.3.3 Switching for convenience sake Some chatters also told the interviewer that they usually switch to their local language because of the convenience it affords them. Therefore, when they switch from the English language to a local language, they believe the other parties feel the impact of the message sent to them. Such opinion is expressed in example 31. Example 31: Interview with a member of MyChisco WhatsApp chat group Rose is 30 years and graduated from KNUST. She is from Ga and Ewe ethnic background. She speaks English, Akan, Ewe and Ga. She lives in the Ashanti region of Ghana so she usually uses Akan in her chats. She is a teacher and a member of the MyChisco WhatsApp chat group. Interviewer: Hello Rose, I want to ask you a question Interviewer: Why would you say: Interviewer: ‘Happy B-Day Awuley. Nyame nhyira wo bebree na ɔnyɛ w'akomaso ebisadeɛ nyinaa mma wo[May God bless you and fulfil all your heart desires for you ]🎂🍷� � � � 🙌� 🙏� ’ Interviewer: Instead of using English Rose: You are still doing your research🙈🙈 Interviewer: Yes ooo Rose: I decided to type in Twi as a matter of preference Interviewer: Ok Rose: And sometimes too, you want your other party to feel the impact of what you want to say. So it's convenient in using your local dialect Rose: That's why I had to switch to Twi Interviewer: OK. Thanks very much for your time Rose: You are welcome dear Rose’s comment reveals that she often employs code-switching when chatting on WhatsApp. Like Rose, Mike who is another participant of another chat group confirms that he also adopts the local language in WhatsApp because of preference and convenience. This is because to him, the use of the local language makes others feel the impact of what is being communicated. 109 University of Ghana http://ugspace.ug.edu.gh 3.3.4 Switching to include and exclude others in a conversation Some of the chatters who were interviewed told the interviewer that they switch so that they either include or exclude others from a conversation. Therefore on the WhatsApp platform, they intentionally switch from the unmarked code to another language so that only those who understand a particular language can contribute to the conversation while those who do not understand the language become disadvantaged of it. Such opinion is expressed in the example below; Example 31: Interview with a member of the KNUST PENSA Organa Alumni Family Earthangel is a female teacher. She studied French in KNUST and also travelled to France to continue her education. She is an Akan and speaks Akan, English, Ga, and French. She is a member of the KNUST PENSA Organa Alumni Family. She makes these remarks about her use of French on the group platform. Interviewer: Hello Interviewer: Plz [please] I want to ask a question Interviewer: Do you speak French Earthangel: Yes pls[please] Earthangel: Yes, (I) studied French at knust [KNUST] and schooled in France too Interviewer: OK. I want to know why you and Monsieur usually speak French on the page although you are aware that others don't understand {the language}. Earthangel: Because I know he understands what I say so my conversation are strictly directed to him Interviewer: Then why don't you chat him privately but choose to speak to him alone on the page Earthangel: They are things we all discuss but just some thoughts we share Interviewer: OK. Thanks very much for your time Earthangel: Don't mention Her comments indicate that she deliberately use French on the platform just to exclude those who do not understand French. Therefore to her anytime she speaks French, it is directed to only one person in the WhatsApp group. 110 University of Ghana http://ugspace.ug.edu.gh 3.3.5 Switching in order to accommodate others Some chatters disclosed that they switch on the platform in order to accommodate the code- choice of other participants in the group. Thus when the participants they are chatting with suddenly switch from the unmarked code to a marked code, they also adjust themselves to use that marked code. Such opinion is expressed in the chat below. Initially, the example below was not an interview but a mere chat between the interviewer and her friend. In the course of the chat, the interviewer switched to Akan-English CS and her friend did same. So then, the interviewer asked her why she also switched. Example 32: An interview with a member of the Zighiss WhatsApp Group Araba is a post-graduate student of KNUST studying animal science. She is Akan and speaks Akan and English. She is a member of the Zighiss WhatsApp chat group Interviewer: The bursary forms are in oo Interviewer: Have you filled yours Araba: Yeah Interviewer: Sweet Araba: Eeeiii calling me sweet[Eeeiii you are calling me sweet] Araba: My head sweet me [You are flattering me] Interviewer: Nti menfr3 wo sweetie da [So haven’t I called you sweetie before?] Araba: Wo afr3 me daa mmom [You have called me that before ]but today its sweet Interviewer: Why did you answer in twi Interviewer: Pls [please] I need answers it would be useful for my Thesis Araba: Because u[you] spoke in twi Araba: * spoke twi [Twi] so I had to also answer with same language [You spoke Akan so I had to also answer with the same language.] Interviewer: OK. That is what I perceived From the interaction, it can be noted that Araba would not have switched to Akan had the interviewer not done same. Araba adjusted her code-choice in order to suit the interviewer’s code. 111 University of Ghana http://ugspace.ug.edu.gh The analyses of the various interviews conducted on the reasons for CS on WhatsApp platforms have proven that chatters have various reasons for employing code-switching. Whereas some chatters employ code-switching to create humour, others employ it for convenience sake. Similarly, others employ it because they do not know how to express their thoughts in English but would have to use the local language to express those thoughts and finally, some chatters switch codes just to exclude other participants from a conversation. These reasons given by the chatters themselves have established and confirmed the functions of code-switching given in this chapter. 3.4 SUMMARY This chapter sought to do a qualitative analysis of the data, which revealed the social motivations of code-switching on WhatsApp. In view of this, the chapter was divided into various sections. The first section of the chapter analysed the code-switching structures found in the data. The analyses of the code-switching structures led to the discussion of the types of code-switching found in the data. These are Tag-switching, Inter-sentential switching, Intra-sentential switching and Discourse markers. Moreover, this section also discussed the Akan discourse markers usually employed in utterances that are written in the English language. Finally, it also looked at the functions of these Akan discourse markers in those utterances. Section 2 presented an analysis of the data through the Markedness Model proposed by Myers- Scotton. Participants’ choice of one code over others during chat sessions was discussed. It was discovered that the English language is the unmarked code on WhatsApp but chatters employ marked codes for various reasons. These reasons enabled the socio-pragmatics of CS in WhatsApp group chats to be discussed. 112 University of Ghana http://ugspace.ug.edu.gh The findings indicated that participants of various WhatsApp chat groups employ marked code- switching for various reasons either consciously or subconsciously. The analyses showed that chatters employ marked code-switching when quoting someone else, when being emphatic, when the unmarked code cannot better express the participant’s thoughts, when creating humour and when an individual wants to exclude or include others in a conversation. Therefore, the argument is that chatters employ code-switching as an instrument to enhance their communication on WhatsApp. Likewise chatters use it as a strategy to effectively interact and negotiate meanings. The last section of the analyses dealt with the interview aspect of the data collection. The qualitative analysis of the interview responses also proved that Ghanaians switch codes when they cannot express their thoughts better in English, when they want to create humour, when they want to emphasize what is being said and also to accommodate other chatters on the WhatsApp platform. These findings from the interviews confirm the analyses of the data in relation to the social motivations for CS on WhatsApp. 113 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR VARIATIONIST ACCOUNT OF LANGUAGE USAGE 4.0 INTRODUCTION This chapter investigates the variations in the use of CS between participants and across the chat groups used for this study. These variations are discussed in relation to social variables like gender, socioeconomic status and situational context. Firstly, it looks at the variations in the distribution of CS types and then discusses variation in terms of the social categories. This chapter centres on quantitative analysis of CS structures and the social categories as championed by the Labovian Sociolinguistics model. 4.1 VARIATIONS IN CS STRUCTURES ACROSS CHAT GROUPS 4.1.1 Distribution of the Types of CS across the WhatsApp chat groups 114 University of Ghana http://ugspace.ug.edu.gh From the previous chapter (section 3.1.2), it was established that four types (tag-switching, inter- sentential, intra-sentential and discourse markers) of CS were found in the data. The frequency of occurrence of these types was counted manually across individual chat groups. The table below displays the distribution of the types of switches found in the posts. Note that the distribution of discourse markers are separated from other types of CS and would be discussed separately. Table 1: Distribution of CS in WhatsApp chat P O S T S TOTAL PERCENTAGE ORIGINAL FORWARDS NUMBER OF OCCURRENCES TYPES OF CODE- 5019 49 % SWITCHES INTER-SENTENTIAL 2533 58 2591 51% INTRA-SENTENTIAL 2481 46 2527 49% TAG-SWITCHING 5 0 5 0.10% 115 University of Ghana http://ugspace.ug.edu.gh 5123 100% Table 1 is a summary of the types of switches found in the data. As has been observed, the posts sent to the WhatsApp chat groups are grouped into forwards and original posts. The findings reveal that only few forwards contained code-switched elements while the actual conversations of the participants of the groups contained many code-switched elements. For the forwards, a single one could contain many types of switches as seen in section 3.1.1.1 in the previous chapter. From the table, inter-sentential CS is the predominant type of CS representing 51%. Intra-sentential CS follows with 49%. Tag-switching recorded the lowest frequency. It appeared only 5 times in the posts, representing 0.10%. The dominance of inter-sentential CS may be attributed to the fact that chatters usually switch at some key points in their conversations and as a result, switch between sentences. This is as a result of English being the unmarked code of WhatsApp chats and an H (high) variety. Therefore, there is the highest possibility that it would be spoken monolingually. However, the data presented in this study shows that Akan is gradually finding its way onto the WhatsApp platform. As a result, we notice that the social status of Akan and English is in opposition since English is considered an H variety and Akan the L (low) variety. Though English is the default code on WhatsApp, it cannot carry the bilingual structure into which insertions are made from the local languages. Therefore, chatters transfer spoken CS onto WhatsApp where the local languages accept insertions from English. This is attributed to the diglossic relationship that exists between English and the local languages so that it is basically an L variety that plays host 116 University of Ghana http://ugspace.ug.edu.gh to an H variety. It is not the norm for a prestigious language to receive insertion from an L variety because it is perceived to be self-sufficient. The L variety is always considered to be deficient; that is why it allows for insertions from the H variety. This is how CS has been around the world. So in naming the languages in CS, the L variety precedes the H variety. The second language in mention is the H variety. Hence, there are names like Akan-English CS, Yoruba- English CS, Ewe-English CS etc., but never English-Akan CS. These names imply that when people speak the L variety, they switch to an H variety. Likewise, it is not regular for the H variety to absorb the L variety. This is because of the social reality that the H variety is self- sufficient. People who speak the L variety always want to upgrade their status by grafting parts of the H variety into the L. In Ghana for instance, speaking an L variety with the H variety is a style of speaking which is in tune with educated people. Another reason for the dominance of inter-sentential CS is because inter-sentential CS does not permit the grammars of the two languages to come into contact and this enables chatters to freely use it on WhatsApp. Note that it is easier to find many instances of intra-sentential CS than inter- sentential CS in spoken data. This is because in spoken language, the local languages function as the matrix language, the default language, into which insertions are made from English and not the other way round. So on WhatsApp, the default language has become what should have been embedded in spoken language. So we expect a reversal of the structural roles of the languages involved. This limits the flow of intra-sentential CS. That is why there is the predominance of inter-sentential CS where the grammars of the two languages used in CS do not meet. Moreover, the data showed that chatters use particular types of switches depending on the communicative intent they want to perform. 117 University of Ghana http://ugspace.ug.edu.gh In looking at the types of CS found in the data, the focus was on the distribution of CS across the chat groups used for the study. The frequencies of the distribution of the types of CS were manually counted by the researcher without any digital tools. Though discourse markers were categorized as a type of CS in the previous chapter, it has been separated from the types of CS in this section. Below is Table 2 which presents a distribution of the types of CS with respect to the WhatsApp chat groups used for the study. Table 2: frequency of types of CS across chat groups Types of CS WhatsApp chat groups Inter- Intra-sentential Tag- sentential switching Total Ahiwaa R/C Primary 332 270 0 WhatsApp chat group 602 KNUST PENSA Organa 1528 1589 5 Alumni Family 3122 Zighiss WhatsApp Group 310 269 0 578 MyChisco WhatsApp 274 235 0 510 118 University of Ghana http://ugspace.ug.edu.gh chat group Fraser Boys WhatsApp 140 162 0 chat group 302 ND 1-14 7 2 0 9 As shown on the table, members of KNUST PENSA Organa Alumni Family chat group frequently use CS in their casual conversations. This may perhaps be the group culture to always switch codes during conversations. Also, in this group, switching between Akan and English is not met with any hostility. This may be attributed to the fact that Akans dominate other participants from other ethnic groups and because of this, members from other ethnic backgrounds hardly have a choice than to adjust their code choice in order to suit the majority. Because Akan is a dominant language in Ghana, many people from other ethnic backgrounds speak it as a second language. Such people in this chat group also switch to Akan to suit the Akan speakers. Therefore in this group, Akan-English CS is always expected. However, for those who cannot communicate in Akan, they always use English on the platform. Though there are instances of Ga-English CS on this platform, other members do not reply to such messages in Ga-English but would respond either in Akan-English or English. The data showed that participants in ND 1-147 chat group less frequently use CS during conversations because of the nature of the group. One peculiar thing about this chat group is that CS is not expected. In this group, the expected language choice is English. Although the students dominate, their choice of language is influenced by the hall tutors in the group. Also, the students 7 ND 1-14 chat group is excluded from the analysis of code choice in relation to gender because it has very few occurrences of CS patterns. 119 University of Ghana http://ugspace.ug.edu.gh in this group are of different social status. Whereas some are undergraduates, others are MPhil students while the rest are lecturers at other institutions pursing their PhD programmes. It is important to note that the switches that were found in the data of this group were not actual conversations of the participants of the group. The participant who sent that message created a message in which there was a conversation between two ladies who were talking about an upcoming hall week celebration. The intention of the participant who sent the message to the group was to inform them about the hall week celebration. 4.1.2 Distribution of Discourse Markers found of the data Discourse markers are sentence peripherals and are not more integrated into the sentence frame. They are usually set apart at clausal final positions and can be dropped. The use of DMs on Ghanaian WhatsApp platform is dependent on the kind of chat community. A very homogenous community allows for the pervasive use of DMs during chat sessions. However, the same may not be found in a very heterogeneous community that is made up of participants from different ranks. Even when DMs are used in a heterogeneous community, they occur minimally. The data used for this study showed that five groups out of the six groups (Ahiwaa R/C Primary, Zighiss, Fraser Boys, KNUST PENSA Organa and MyChisco) were homogenous while one group (ND 1-14 chat group) was heterogeneous. Therefore, there was the reoccurrence of DMs in the homogenous groups while only one was found in the heterogeneous group. The table below illustrates the distribution of DMs found in the communities of practice used for this study. These DMs were manually counted. 120 University of Ghana http://ugspace.ug.edu.gh Table 3: Distribution of DMs across chat groups WhatsApp Chat groups Frequency of DMs Ahiwaa R/C Primary 237 KNUST PENSA Organa Alumni Family 970 Zighiss WhatsApp Group 886 MyChisco WhatsApp chat group 1436 Fraser Boys 158 ND 1-14 1 Total 3688 As shown on the table, ND 1-14 chat group recorded only one occurrence of DMs though the group had existed for more than a year. This observation is attributed to the fact that the chat group is made up of both Senior Common Room (SCR) and Junior Common Room members (JCR). Moreover though the JCR members are dominant, they are always moved by the fact that senior members are also group members and as result, the JCR members choose their language to suit them. So in this group, the use of DMs is marked. It could also be that the interactions on the group page are not as extensive as it is observed in a group that is made up of only peers. The table also reveals that in homogeneous chat groups, the use of DMs is unmarked. That is why the homogenous chat groups recorded many instances of the use of DMs. For Fraser Boys chat group, they also recorded few instances of DMs because of their unmarked code choice (pidgin). This is because pidgin is a mixed code and has its own structure. Similarly, it is noteworthy to realise that Akan DMs are used when English carries the bilingual structure. Therefore, these DMs could not have been pervasively employed because of the nature of the Fraser Boys WhatsApp chat group. 121 University of Ghana http://ugspace.ug.edu.gh Table 4 presents a summary of the frequencies of the various Akan discourse markers usually employed by Ghanaians during their conversations on WhatsApp. Table 4: Frequency of Akan Discourse markers found in English constructions Akan Discourse Markers Frequency Percentage % Ooo 2907 79% Wai 264 7.2% Paa 250 7% Koraa 131 4% Waa 32 0.9% Saa 24 0.7% Aaba 19 0.5% Papa 13 0.4% Tweaa 8 0.2% 5 0.14% deԑ 5 0.14% Biaa Agyeii 5 0.14% Papaapa 5 0.14% kԑkԑ 4 0.11% Kwa 2 0.05% 2 0.05% yᴐᴐ 122 University of Ghana http://ugspace.ug.edu.gh Aaa 2 0.05% 1 0.03% ԑԑԑ 1 0.03% Bam 1 0.03% Ebei 1 0.03% pԑԑ 1 0.03% sᴐᴐ 1 0.03% Ah 1 0.03% Ai 1 0.03% Naa 1 0.03% Yee 1 0.03% Kai Total 3688 100 In all, 3688 Akan discourse markers were identified. Out of this number, ooo8 is predominately used by chatters, occurring 2907 times and representing 79% of the total. Wai follows with a total of 264 occurrences and represents 7.2% of the total. Paa is next with a total occurrence of 250 representing 7%. Koraa follows with 4% of the total. Others like waa, saa, aaba, papa and tweaa made up 0.9%, 0.7%, 0.5%, 0.4% and 0.2% respectively of the entire data. As shown in the table, it is very interesting that some of the DMs occurred only five times in the data (deԑ, papaapa, agyeii, biaa). kԑkԑ also appeared four times in the data while some occurred only twice (kwa, yᴐᴐ, aa) in the data. Finally, the table shows that some Akan DMs (naa, kai, yee, ai, sᴐᴐ, ah, pԑԑ, ebei, bam, ԑԑԑԑ) occurred just once in the data. The use of these DMs shows the level of 8 The frequency of the occurrence of ooo was counted with Microsoft word tool 123 University of Ghana http://ugspace.ug.edu.gh familiarity chatters have established among themselves. This is because in-groups (homogenous groups) usually adopt these DMs to show the informality of their interactions and their group culture. Also, the frequently used DMs are very common to the participants and so, they are familiar to their usage. For the less-frequently used DMs, certain circumstances lead to their usage. 4.1.3 Code choice by Gender One of the most important variables investigated in this thesis is gender. This research sought to find out how CS manifests gender differences on the various platforms; that is, how the code- switching patterns of male speakers differ from those of female speakers. Below is a table showing the frequency of the distribution of switches between the males and females in the data. Table 5 shows quantitatively the code choice of participants in the mixed WhatsApp chat groups. The table suggests that the males usually switch codes on WhatsApp more than their female counterparts. In the first group, the table reveals that the males frequently switch codes during conversations, representing 76% of the total switches while the females represent 24%. Moreover, the table displays that the males in the group use more inter-sentential CS, intra-sentential CS, pidgin and Akan discourse markers than the females. Table 5: Code choice among males and females in mixed WhatsApp chat groups Groups Gender Total Percentage Ahiwaa LA Primary WhatsApp chat group 124 University of Ghana http://ugspace.ug.edu.gh Male Female Inter-sentential CS 156 22 178 32% Intra-sentential CS 115 13 128 23% Discourse markers 140 97 237 42% Pidgin 17 0 17 3% Total 428 132 560 100% Percentage 76% 24% MyChisco WhatsApp chat group Inter-sentential 133 85 218 10% Intra-sentential 90 49 139 7% Discourse markers 957 479 1436 67% Pidgin 275 65 340 16% Total 1455 678 2133 100% Percentage 68% 32% KNUST PENSA Organa Alumni Family Inter-sentential 757 280 1037 32% Intra-sentential 828 211 1039 33% Discourse markers 666 304 970 31% Pidgin 89 16 105 3% Total 2340 811 3151 100% Percentage 74% 26% In MyChisco WhatsApp chat group, the same thing happens since the males employ more code- switching in their conversations than the females. The switches of the males represent 68% of the total code-switches while that of the females represent 32%. However in this group, the males employ more pidgin in their conversations. Because the males on this platform usually use pidgin, the females on the platform also adjust their choice of language just to use pidgin to the extent that the frequency of the use of pidgin among them is the highest when compared to the frequency of pidgin use among females in the other two WhatsApp chat groups. 125 University of Ghana http://ugspace.ug.edu.gh In KNUST PENSA Organa Alumni Family, the table shows that the males dominate women in the switching of codes on WhatsApp, representing 74% of the total switches found in the data. The switches of the females in the group represent 26%. In view of the statistics above, it can be argued that females usually use the expected code choice (English language) on WhatsApp when they are in mixed groups. This might be that since they are in a mixed group, the females are usually compelled to use the expected language. However, the table below shows otherwise when they are in an all-female group. Let us consider the table below. Table 6 displays the frequency of CS between sentences and within sentences, the use of Akan discourse markers in English constructions and the use of pidgin in conversations. Table 6: Code choice between Fraser Boys and Zighiss Switches Fraser Boys WhatsApp Zighiss WhatsApp Total Percentage chat group chat group Inter-sentential 140 255 Intra-sentential 147 163 Discourse markers 158 886 Pidgin (did not count it because 50 it is the unmarked code of the group) Total 445 1304(excluding pidgin frequency for the group) Percentage 25% 75% The table shows that the females employ more CS and Akan discourse markers in conversations more than their male counterparts. Whereas the females’ use of CS and Akan discourse markers represent 75% of the total number of switches, the males’ use of CS and Akan discourse markers represent 25%. However, it must be known that participants of the Fraser Boys WhatsApp chat group employ pidgin as their unmarked code in their chats and that is why there are few 126 University of Ghana http://ugspace.ug.edu.gh instances of CS involving Akan/Ga and the English language. This is because pidgin is the default language of these male chatters. There were instances of switching that involve the English language and Ga. In the Zighiss WhatsApp chat group, the use of pidgin occurs only 50 times in the entire chat sessions and it shows that females do not usually use pidgin when chatting on WhatsApp. Pidgin is a reduced and a non-standard language and because of this, it was distinguished from the English language. The findings on this table show that male chatters use more CS whenever they are in a mixed group with their female counterparts. This observation confirms Labov’s (1990)9 principle that women tend to use more standard forms than men. This is because females tend to use the expected language during chats. Males usually resort to the use of pidgin when they are in a WhatsApp group made up of only males since to them, Student Pidgin is an in-group language. On the other hand when the males are in mixed groups, they use little pidgin. The female chatters do not also use more CS when in mixed groups; however, they tend to use more CS when they are in a chat group made up of only females. 4.2 SUMMARY This chapter focused on the variations that exist in the code-switching patterns found across the WhatsApp chat groups. The findings revealed that the use of CS depends on the kind of community in which chatters find themselves. This is because an in-group (homogenous group) permit the use of CS structures while a heterogeneous group made up of senior and junior members does not permit the pervasive use of CS. 9 Gender Paradox is a Labovian principle which stresses that “women conform more closely than men to sociolinguistic norms that are overtly prescribed, but conform less than men when they are not” (Labov, 2001: 261) 127 University of Ghana http://ugspace.ug.edu.gh The analyses of the data proved that the use of code-switching in WhatsApp group conversations manifest gender differences. Thus, the gender of chatters play a role in the code-switching process. It was revealed that male chatters tend to use more code-switching than their female counterparts in mixed chat groups. However, the situation changes in a single sex group. When males are alone in a group, they frequently use pidgin. The females, on the other hand, use more code-switching than the males in a single sex chat group. These findings show that gender is a key factor to the code-switching patterns used. While male speakers use less pidgin when they are in a group with females, they resort solely to it when alone in a group. This shows that the biological differences between males and females account for how they view themselves. Therefore because of how they view themselves, they will use language differently when they are alone or mixed in a group. 128 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE SUMMARY AND CONCLUSION 5.0 INTRODUCTION In a community where many languages exist, individuals always have a choice in relation to the language they use in their everyday lives. They may either use one of the many codes available or switch between those codes. The habit of switching between codes within a single conversation has become a norm among many bilinguals and Ghanaians are no exception. Ghanaians usually switch between their local languages and the English language. When Ghanaian bilinguals switch codes, they convey special meanings whether consciously or unconsciously and these meanings are understood within the community’s Rules and Obligations (RO) sets. Although this linguistic phenomenon has attracted different thoughts from all over the world, it continues to gain grounds in all domains of interpersonal communication to the extent that it has found its way into digital communication especially WhatsApp. This current study aimed at investigating the use of code-switching on a computer-mediated communication medium. It sought to examine the socio-pragmatic factors for code-switching on 129 University of Ghana http://ugspace.ug.edu.gh the WhatsApp platform. It also examined the patterns of code-switching on WhatsApp platform. This chapter discusses the major findings and concludes the study. There are also recommendations for future related studies. 5.1 SUMMARY Chapter one introduces this study. The linguistic situation in Ghana and the background of the WhatsApp messenger are provided. The problem statement, objectives and research questions are also outlined. Chapter two reviews previous studies that are relevant to this study. It starts with a discussion on bilingualism and then presents previous studies relevant to this work. The previous studies on the socio-pragmatics of spoken code-switching in Ghana show that CS is expected in the Ghanaian context to the extent that it satisfies little or no pragmatic function farther than assessing speakers’ solidarity (Amuzu, 2012). Moreover, the discussion on CS in computer-mediated communication shows that communication on CMC bears semblance to spoken language and also influences language and brings out novel ways of writing (Georgakopoulou, 1997). Besides, previous studies demonstrate that CS is adopted by participants in order to improve their in-group loyalty and familiarity with one another. Similarly, Siebenhaar (2006) maintains that the use of CS on CMC is age related. 130 University of Ghana http://ugspace.ug.edu.gh The Community of Practice theory, Markedness Model and Variationist Sociolinguistics which serve as the theories behind the analysis of the data are explained. The Community of Practice theory by Lave and Wenger (1991) is a social theory of learning that describes how individuals gain knowledge when they join a community. When this theory was introduced into linguistics studies by Eckert & McConnell-Ginet (1992), it was used as a research tool for investigations into the relationship between language and gender. This theory is used in the data collection since it enables the researcher to get access to the data. Also, it enables the researcher to investigate the in-group practices of the chat groups. Myers-Scotton’s Markedness Model, on the other hand, is grounded on the concept that speakers make language choices because of the goals they want to achieve with their listeners. It also suggests that language users can recognise and distinguish between marked and unmarked codes. Thus, speakers would switch between either the marked or unmarked code when they want to communicate certain meanings. The model is chosen because of its ability to describe the reasons why speakers choose particular linguistic codes over others. This model is used in chapter three for the qualitative analysis of the data. Variationist Sociolinguistics suggests that variations exist in language use and these variations are determined by social variables. This theory is chosen because of its ability to illustrate the effects of social constraints on patterns of linguistic variations across the chat groups. This model is used in chapter four. Finally, the methods used for the data collection are outlined. These methods include interview surveys, participant observation and background checks. In chapter three, the study analyses code-switching structures found in the data. Tag-switching, intra-sentential switching, inter-sentential and discourse markers are identified as the four types 131 University of Ghana http://ugspace.ug.edu.gh of code-switching found in the data. The study also finds out that Akan discourse markers are usually employed by chatters even when they chat in English. It is observed that the Akan discourse markers usually intensify what is said and their functions depend on the context of their usage. The Markedness Model is used for discussing the socio-pragmatic factors that influence participants’ linguistic choices on WhatsApp. The findings of the study prove that on WhatsApp, the English language is the unmarked code choice because of the ease it affords chatters. Alternatively, the use of the Ghanaian local languages is marked on WhatsApp because of the struggles chatters endure just to write in those languages. Therefore, it is established that when chatters endure those struggles to write in their local languages, it means that they communicate special intentions. In the data, it is evident that a switch from the expected code to an unexpected code choice serves to increase or decrease social distance between the participants. Also, there is the discussion of the factors that motivate the use of marked CS on WhatsApp. The analysis shows that the factors that motivate the use of marked CS include the inability to express some thoughts in the English language, the need to quote another person, the quest to create humour and the need to be emphatic. Moreover, the reasons for CS in WhatsApp group conversations are explored using interview responses. The major finding is that chatters employ code-switching for the reasons which include accommodating others’ linguistic choice, creating humour, etc. The finding from the interviews confirms the factors outlined in the data analysis. Chapter four of this study analyses the variations found in code-switched structures across the WhatsApp chat groups. This analysis adopts the Variationist Sociolinguistic approach as the 132 University of Ghana http://ugspace.ug.edu.gh theoretical backdrop and as a result, based on the quantitative analysis of the data. Conclusions drawn from this analysis show that it is the nature of a community of practice which determines chatters’ linguistic choices. This is because when a subordinate is in the same chat group with his superiors, it would be very difficult for him to adopt informal forms of language in the group conversations. He may be compelled to always use standard forms of language while his superiors may use colloquial forms of language. The disparity between the homogenous chat groups and the heterogeneous chat groups proves this observation. Likewise, it is established that in a male-only group, the default language could be pidgin because it is considered an in-group code for males. Again, the study discovers that Akan- English CS can be an unmarked code choice on a WhatsApp platform. This occurs when the Akans in a WhatsApp group dominate others from other ethnic groups and those people are compelled to also switch from the English language to Akan in order to accommodate to the speech patterns of others. When Akan-English CS is used in WhatsApp conversations, it is used as a means to display the multiple identities of chatters (Myers-Scotton, 2000:138). Finally, code-switching is discussed in relation to gender. It is established that code-switching manifests gender differences. In mixed groups, male chatters use more code-switching than their female counterparts. However in single-sex groups, females use more code-switching than the males in relation to switching from the English language to the local languages. In a male-only WhatsApp chat group, there is the highest possibility that pidgin would be used as the default code choice. 5.2 SUMMARY OF FINDINGS 133 University of Ghana http://ugspace.ug.edu.gh As argued by Crystal (2001), the use of language on the internet signifies a new kind of discourse that is moulded by the creativity of its users. These users usually form communities of practice through which they establish their identity through the kind of language they use and the relationship they establish among themselves. The study has proven that although the English language remains a global language and a language of higher education in Ghana, the internet is gradually becoming multilingual. This is because the code-switching phenomenon has moved beyond the boundaries of oral communication to written communication especially to the Computer-mediated Communication medium. The advancement in new modes of Computer-mediated Communication like the WhatsApp Messenger has increased the practice of code-switching to the extent that Ghanaians have adopted ways of writing in order to switch codes. It has also resulted in novel ways of writing among Ghanaians. However, it is worthy to note that anytime English becomes the default language, there is minimal CS and this has been observed in the data presented in Albakry & Ofori (2011) and Quarcoo and Amuzu (2016). That is why it is very pertinent to understand how code-switching functions as a communicative strategy in this electronic environment. This study has proven that Ghanaians employ code-switching during their WhatsApp group chat sessions as a tool to improve their interaction. Thus, they switch from the English language to their local languages or Pidgin depending on the suitability of their interactional needs. Notwithstanding the various communicative functions of CS on this medium, CS serves to exclude or include others in a conversation and may serve as an accommodation strategy. 134 University of Ghana http://ugspace.ug.edu.gh This study has also shown that spoken CS in Ghana bears semblance to the written CS on the WhatsApp platform though there are few instances of novel patterns of CS in which chatters mixed languages in creative and skilful ways that might not have occurred in speech. Also, it was established that inter-sentential CS dominates intra-sentential CS on this platform because the English language functions as the default language on this medium. As a result, the pattern of written CS on WhatsApp may not reflect that of spoken CS. This is because there is the probability to find more intra-sentential CS in spoken language where the local languages function as the default language than in the written discourse on WhatsApp where English functions as the default language of communication. However when intra-sentential CS is used on WhatsApp, there is the dominance of Akan-based structures into which insertions are made from the English language. This is because English is an H variety and so, it is not the norm for it to receive insertions from an L variety since the H variety is considered self-sufficient. Thus, there are few instances of English-based bilingual structures taking insertions from the local languages in the data. Although Akan DMs are usually inserted into English-based structures, the argument is that those DMs are peripheral and not sentences integrals like other lexical words. Therefore, they do not participate directly in the sentence frame when inserted into English- based structures. This study has also established that code-switching in WhatsApp conversations is only possible depending on the kind of community in which participants find themselves. This is because the use of CS in a particular WhatsApp chat group might mean differently and be unacceptable in another WhatsApp chat group. This is attributed to the kind of practices that those communities of practice have accepted as conventional so that the departure from those conventions generates 135 University of Ghana http://ugspace.ug.edu.gh different reactions and interpretations. The conclusion drawn is that CS is much expected in homogeneous chat groups. The thesis has also confirmed that the gender of chatters plays a role in the process of code- switching. It has established that males have a stronger tendency to switch codes on WhatsApp than females in mixed groups. However, males usually use pidgin as their unmarked code on WhatsApp when in an all-male group and hardly switch from the English language to the local languages. Females, on the other hand, have a higher tendency to switch from English to the local languages in a female chat group. This finding confirms Jagero & Odongo’s (2011) study on the patterns of code-switching among males and females in different ranks in Nairobi Kenya. It reiterates their conclusion that the gender of speakers is a significant factor to the CS pattern used. The findings of this study contribute to the previous studies on bilingual interaction on Computer-mediated Communication. Specifically, this study sets the beginning of the investigation of language contact in WhatsApp conversations in Ghana since there has not been any known study on code-switching in WhatsApp interactions in the Ghanaian community. Also, this research reveals that code-switching on WhatsApp is used to fulfil certain communicative needs of Ghanaians; thus proving that code-switching is not only restricted to spoken communication but also found in written interaction. 5.3 FUTURE RESEARCH Although this study aimed at exploring the socio-pragmatics of code-switching in WhatsApp group chat among Ghanaians, there are still areas that need to be further explored in linguistic 136 University of Ghana http://ugspace.ug.edu.gh research since research on bilingualism in Computer-mediated Communication modes especially, WhatsApp has not been fully investigated in the Ghanaian community. For instance, this study looked at the conversations of only six WhatsApp chat groups which cannot be a perfect representative for all Ghanaian WhatsApp chat groups in relation to their practices, and the harmony that exists in those groups. However, there is the need for us to bear in mind that some WhatsApp chat groups are characterised by disagreements/conflicts among participants and these keep those groups going. Significantly, much research needs to be conducted on the orthography and the spelling used in writing the Ghanaian local languages used in the code- switching. Studies in this area will help to find out how far Ghanaians are literate in their local languages in terms of writing. Also, a comparative study can be done on the code-switching found in WhatsApp private chats and group chats. This will ascertain the modes of WhatsApp that create the environment for code-switching. Moreover, studies can be conducted to investigate Ghanaians’ attitude towards the language use and the use of code-switching in this medium. It is also important for studies to be carried out on the grammar of the switches found in WhatsApp chats. Finally, a systematic comparative study can be carried out on the code-switching found in different computer-mediated communication modes. This will help to bring about a better projection about the cross-linguistic manifestation of code-switching in CMC. 137 University of Ghana http://ugspace.ug.edu.gh REFERENCES Adika, G. S. K. (2012). English in Ghana: Growth, Tensions and Trends. International Journal of Language, Translation and Intercultural Communication 1, 151–166. Ahmed, E. S. (2014). Code Switching in WhatsApp Messages among Kuwaiti High School Students. (Unpublished MA Thesis). Manchester Metropolitan University. Albakry, M. A. & Ofori, D. M. (2011). Ghanaian English and code-switching in Catholic churches. World Englishes, 30, (4), 515-532. Ammon, U. (2004). Sociolinguistics / Soziolinguistik, Volume 1. NY: Walter de Gruyter. 138 University of Ghana http://ugspace.ug.edu.gh Amuzu, E.K. (1998). Aspects of Grammatical Structure in Ewe-English Codeswitching. (Unpublished MPhil. Thesis). University of Oslo, Oslo. Amuzu, E. K. (2005). Ewe-English Codeswitching: A case of Composite rather than Classic Codeswitching. (Unpublished PhD dissertation) Australian National University, Canberra. Amuzu, E. K. (2012). Socio-Pragmatics of Conversational Code-switching in Ghana. Ghana Journal of Linguistics 1.2:1-22. Amuzu, E. K., Kuwornu, A. E. A, & Opoku-Fofie, S. (Forthcoming). “I’ve been calling you saaa”: L1 Discourse Markers in L2 English-based WhatsApp conversations among Ghanaians. Achimbe, A. E. (2015). Code-switching: Between identity and exclusion. In G. Stell & K. Yakpo (Eds.), Code-switching between Structural and Sociolinguistic Perspectives, (pp. 139- 161). Anderson, J. & Wiredu, J. F. (2007). Code-Switching as a Linguistic Resource in Ghanaian Advertisements. A paper presented at the International Conference on Codeswitching in West Africa: Theory and Implications. Legon. Andoh, A. K. (1997). Code-switching in Christian Sermons. BA project, Department of Linguistics. University of Ghana. Androutsopoulos, J. (2013). Code-switching in computer-mediated communication. In S. C. Herring, D. Stein & T. Virtanen (Eds.), Handbook of the Pragmatics of CMC, (pp. 667- 694). Berlin/Boston: Mouton de Gruyter. 139 University of Ghana http://ugspace.ug.edu.gh Anyidoho, A. & Dakubu, M. E. K (2008). Ghana: indigenous languages, English, and an emerging national identity. In A. Simpson (ed.), Language and National Identity in Africa, (pp. 141-157). Oxford: Oxford University Press. Apenteng, M. A. & Amfo, A. N. A. (2014). The Form and Function of English Loanwords in Akan. Nordic Journal of African Studies 23(4): 219–240. Arce, N. (2015). WhatsApp calling for android and IOS: How to get it and what to know. http://www.techtimes.com/articles/38291/20150309/whatsapp-calling-for-android-and- ios-how-to-get-it-and-what-toknow.htm [last accessed 27.10.16]. Asare-Nyarko, C. (2012). Akan-English Codeswitching in some selected churches in Accra. (Unpublished MPhil Thesis). Department of Linguistics, University of Ghana, Legon. Asilevi, F. (1990). English-Ewe Code-mixing in Conversational Discourse: A case of English as a second language in Ghana. (Unpublished MA Thesis). Department of Linguistics, University of Ghana, Legon. Associates for Change. (2010). The Status of French Language Teaching and Learning across Ghana’s Public Education System. Auer, P. (1984). Bilingual Conversation. Amsterdam: John Benjamins. Auer, P. (1995). The Pragmatics of code-switching: a sequential approach. In L. Milroy & P. Muysken (Eds.), One Speaker, two Languages: Cross Disciplinary Perspectives on Code-switching (pp.115-135). Cambridge: Cambridge University Press. Auer, P. (1998).Code-switching in conversation: Language, Interaction and Identity. London: Routledge. 140 University of Ghana http://ugspace.ug.edu.gh Bamgbose, A. (2011). African Languages Today: The Challenge of and Prospects for Empowerment under Globalization. In Eyamba G. & Bokamba et al (Eds.), Selected Proceedings of the 40th Annual Conference on African Linguistics. (pp. 1-14). Somerville, MA: Cascadilla Proceedings Project. www.lingref.com, document #2561. Belly, R.T. (1976). Sociolinguistics: Goals, Approaches and Problems. London: B.T. Batsford Ltd. Bentahila, A., & Davies, E. D. (1983). The syntax of Arabic-French code-switching. Lingua, 59, 301-330. Bhatia, T.K. & Richie, W.C. (2004). Bilingualism in South Asia. In T.K. Bhatia and W.C. Richie (Eds.), The handbook of Bilingualism. (pp.780-807). Malden, Mass: Blackwell. Blom, J. P. & Gumperz, J. J. (1972). Social meaning in linguistic structures: Codeswitching in Norway. In J. J. Gumperz & D. Hymes (Eds.), Directions in sociolinguistics: The ethnography of communication. (pp. 407- 434). New York: Holt, Rinehart, and Winston. Bokamba, E. (1988) Code-mixing, language variation, and linguistic theory: evidence from Bantu language. Lingua, 76, 21-62. Brinton, L. (1996). Pragmatic markers in English. Grammaticalization and Discourse Functions. Berlin/New York: Mouton de Gruyter. Brobbey, S. (2015). Codeswitching on Ghanaian Radio Talk-show: “Bilingualism as an Asset”. (Unpublished MPhil Thesis). Department of Linguistics, Literary and Aesthetic Studies, University of Bergen, Norway. 141 University of Ghana http://ugspace.ug.edu.gh Brown, A. (2016). WhatsApp quietly introduced a brand-new way to reply to your messages. Retrieved July 13, 2016 from http://www.express.co.uk/life-style/science technology/679325/WhatsApp-How-To-Quote-Message-In-Reply-iOS-Android-Chat. Brown, G. & Yule, G. (1983). Discourse analysis. Cambridge: Cambridge University Press. Bucholtz, M. (1999). “Why be normal?”: Language and identity practices in a community of nerd girls. Language in Society 28, 203–223. Bucholtz, M. & Hall, K. (2004). Theorizing identity in language and sexuality research. Language in Society 33, 469–515. Cambridge University Press. (2008). Cambridge online dictionary, Cambridge Dictionary online. Retrieved at April 23, 2017, from the website temoa: Open Educational Resources (OER) Portal at http://www.temoa.info/node/324. Chan, B. H. S. (2009). English in Hong Kong Cantopop: language choice, code-switching and genre, World Englishes 28(1), 107-29. Crystal, D. (2004). Language and the Internet, 2nd ed., Cambridge: Cambridge University Press. Church, K., & de Oliveira, R. (2013). What's up with WhatsApp?: comparing mobile instant messaging behaviors with traditional SMS. In Proceedings of the 15th international conference on Human-computer interaction with mobile devices and services (pp. 352- 361). ACM. Dako, K. (2002). Student Pidgin (SP): The language of the educated male elite. Research Review, 18, 53-62. 142 University of Ghana http://ugspace.ug.edu.gh Davis, B. L. (2005). Communities of Practice: Legitimacy, Membership and Choice. Journal of Sociolinguistics 9, (4), 551 – 581. December, J. (1996). What is Computer-mediated Communication? Retrieved October 7, 2016 from http://www.december.com/john/study/cmc/what.html. Devic, H. (2008). Code-Switching in Computer-Mediated Communication: A Case-Study of Croatian-English Discussion Forums. Hamburg: Druck Diplomica® Verlag GmbH. Du Bartell, D. (1995). Discourse features of computer-mediated communication: 'spoken-like' and 'written-like'. In B. Warvik, S-K. Tanskanen & R. Hiltunen (Eds.), Organization in discourse: Proceedings from the Turku conference, University of Turku, Finland: Anglicana Turkuensia 14, (pp. 231- 239). Dubé, L., Bourhis, A. & Jacob, R. (2005). The impact of structuring characteristics on launching of virtual communities of practice. Journal of Organizational Change Management. 18 (2): 145–166. Dürscheid, C. & Stark, E. (2011). SMS4science: An international corpus-based texting project and the specific challenges for multilingual Switzerland. In C. Thurlow & K. Mroczek (Eds.), Digital Discourse: Language in the New Media (pp. 299-320). Oxford and New York: Oxford University Press. Eastman, C. M. (1992). Codeswitching as an urban language contact phenomenon. Journal of Multilingual and Multicultural Development, 13, 1-17. Eckert, P., & McConnell-Ginet, S. (1992). Think practically and look locally: Language and gender as community-based practice. Annual review of Anthropology 21, 461-90. 143 University of Ghana http://ugspace.ug.edu.gh Eckert, P. (2008). Variation and the indexical field. J. Socioling, 12, 453–76. Eckert, P. (2012). Three waves of variation study: the emergence of meaning in the study of sociolinguistic variation. Annual Review of Anthropology, 41, 87–100. Emoji Collection [Online image]. Retrieved June 5, 2017 from http:/www.shutterstock.com. Forson, B. (1979). Code-switching in Akan-English Bilingualism. (Unpublished Ph.D Thesis.) University of California. Fraser, B. (1988). Types of English Discourse Markers. Acta Linguistica Hungarica, Vol. 38(1- 4), pp. 19-33. Georgakopoulou, A. (1997). Self-presentation and interactional alliances in e-mail discourse: The style-and code-switches of Greek messages. International Journal of Applied Linguistics 7: 141-164. Ghana Statistical Service (2017). 2010 Population and Housing Census: Summary of final results. Accra: Ghana Statistical Service. Goldbarg, N. R. (2009). Spanish-English Codeswitching in Email Communication Language@Internet, 6, 1-21, article 3. (www.languageatinternet.de, urn:nbn:de: 0009- 721398, ISSN 1860-2029). Gordon, R. G. Jr. (Ed.). (2005). Ethnologue: Languages of The World, (16th edn). Dallas, Tex: SIL International. Online version: http://www.ethnologue.com/. Goyvaerts, L.D. (1992). Codeswitching in Bukavu. In M.C Eastman (Ed.), Codeswitching (pp. 71-81). Philadelphia: Multilingual Matters Ltd. 144 University of Ghana http://ugspace.ug.edu.gh Gumperz, J. (1982). Discourse strategies. Cambridge: Cambridge University Press. Guerini, F. (2012). “ENGLISH koraa yɛde Twi MIXe!” Is the variety of Akan spoken by Ghanaian immigrants in Northern Italy developing into a mixed code? International Journal of Bilingualism 4 (1), 174–202. Halim, N.S. & Maros, M. (2014). The Functions of Code-switching in Facebook Interactions Procedia - Social and Behavioral Sciences 118, 126 – 133. Hård af Segerstad, Y. (2002). Use and adaptation of written language to the conditions of Computer-Mediated Communication. Department of Linguistics, Göteborg University. Haust, D. (1995). Codeswitching in Gambia. Eine soziolinguistische Untersuchung von Mandinka, Wolof und Englisch in Kontakt. (Sprachkontakt in Afrika, Band 1.) Köln: Rüdiger Köppe Verlag. Herring, S. C. (1999a), Interactional coherence in CMC, Journal of Computer-mediated Communication 4(4), 0-0 Retrieved 12 November 2016, from . Herring, S. C. (2001). Computer-mediated Discourse. In D. Schiffrin, D. Tannen and H. E. Hamilton (Eds.), Handbook of Discourse Analysis (pp. 612-34), Oxford: Blackwell. Holmes, J. & Meyerhoff, M. (1999). The community of Practice: Theories and methodologies in language and gender research. Language in Society 28, 173 – 183. Hudson, R. A. (1996). Sociolinguistics Second Edition. Cambridge: Cambridge University Press. 145 University of Ghana http://ugspace.ug.edu.gh Jagero, N. & Odongo, K. E. (2011). Patterns and Motivations of Code Switching among Male and Female in Different Ranks and Age Groups in Nairobi Kenya. International Journal of Linguistics, Vol. 3, (1), 1-13. Jisha, K, & Jebakumar (2014). Whatsapp: A Trend Setter in Mobile Communication among Chennai Youth. Journal Of Humanities And Social Science 19(9), 01-06. Jucker, A., & Ziv, Y. (1998). Discourse markers: Descriptions and theory. Amsterdam/Philadelphia: John Benjamins. Kachru, Y. (1983). Cross-cultural texts and interpretation. Studies in the Linguistic Sciences, 13 (2), 57-72. Kaplan, A. M. & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons, 53, 59—68. Karpisek, F., Baggili I., & Breitinger, F. (2015). WhatsApp network forensics: Decrypting and understanding the WhatsApp call signalling messages. Digital Investigation, 30: 1-9. Kietzmann, J. H., Hermkens, K., McCarthy, I. P., Silvestre, B. S., (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54, 241—251. Kietzmann, J., Plangger, K., Eaton, B., Heilgenberg, K., Pitt, L. & Berthon, P. (2013). Mobility at work: A typology of mobile communities of practice and contextual ambidexterity. Journal of Strategic Information Systems. 3 (4): 282–297. Labov, W. (1963). The social motivation of a sound change. Word 18:1–42. 146 University of Ghana http://ugspace.ug.edu.gh Labov, W. (1966). The Social Stratification of English in New York City. Washington, DC: Cent. Appl. Ling. Labov,W. (1972a). Language in the Inner City. Philadelphia: Univ. Penn. Press. Labov, W. (1990). The intersection of sex and social class in the course of linguistic change. Language Variation and Change 2, 205-254. Labov, W. (2001). Principles of Linguistic Change, Vol. 2: Social Factors. Malden, MA: Blackwell Publishers Inc. Lave, J. & Wenger, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press. Li Wei & Milroy, L. (1995). Conversational code-switching in a Chinese community in Britain: A sequential analysis. Journal of Pragmatics 23, 281-299. Li Wei (2000). Dimensions of bilingualism, In Li Wei (Ed.), The Bilingualism Reader (pp.1-21). London: Routledge. Mackey, W.F (2010). The description of bilingualism, In Li Wei (Ed.), The Bilingualism Reader, (pp.22-52). London: Routledge. Marjie, S. (2010). A study of English Items in Kisawahili-Based Texts. (Unpublished MPhil Thesis). University of Ghana, Legon. McClure, E. (2000). Oral and written Assyrian–English codeswitching, In R. Jacobson (Ed.), Codeswitching Worldwide II, (pp. 157–91), Berlin: Mouton de Gruyter. 147 University of Ghana http://ugspace.ug.edu.gh McGoogan, C. (2016). WhatsApp has finally added video calling 15 November 2016 • 10:54AM http://www.telegraph.co.uk/technology/2016/11/15/whatsapp-has-finally-added-video- calling/ Milroy, L. (1980). Language and social networks. Oxford: Basil Blackwell. Milroy, L & Muysken, P. (1995). Introduction: Code-switching and Bilingualism Research. In L. Milroy, & P. Muysken, (Eds.), One Speaker, Two Languages: Cross-Disciplinary Perspectives on Code-Switching, (pp. 1-14). Cambridge: Cambridge University Press. Milroy, L. & Gordon, M. (2003). Sociolinguistics: Method and Interpretation. USA: Blackwell Publishing Ltd. Montes-Alcalá, C. (2000). Written codeswitching: powerful bilingual images, In R. Jacobson (Ed.), Codeswitching Worldwide II (pp. 193–219). Berlin: Mouton de Gruyter. Montero-flet, A., Begoña, Anna I., Montesinos, L., Pérez-Sabater, C. & Turney, E. D. (2009). Computer mediated communication and informalization of discourse: The influence culture and subject matter. Journal of Pragmatics 41 (4), 770-779. Morris, L. (1998). The function of English in Contemporary Ghanaian Society. African Diaspora ISPs. Paper 52. Retrieved on November 25, 2015 from http://digitalcollections.sit.edu/african_diaspora_isp/52. Müller, S. (2005). Discourse markers in native and non-native English discourse. Amsterdam: John Benjamins. 148 University of Ghana http://ugspace.ug.edu.gh Mvan Gass, K. (2008). Language contact in computer-mediated communication: Afrikaans- English code-switching on internet relay chat (IRC). Southern African Linguistics and Applied Language Studies, 26(4): 429–444. Myers-Scotton, C. (1983). The negotiation of identities in conversation: A theory of markedness and code choice. International Journal of the Sociology of Language, 44: 115-136. Myers-Scotton, C. (1992). Comparing codeswitching and borrowing. Journal of Multilingual and Multicultural Development. 13(1-2): 19-39. Myers-Scotton, C. (1993a). Duelling Languages: Grammatical Structure in Codeswitching. Oxford: Clarendon Press. Myers-Scotton, C. (1993b). Social Motivation for Codeswitching: Evidence from Africa. Oxford: Clarendon Press. Myers-Scotton, C. (1998). Codes and Consequences. Oxford: Oxford University Press. Myers-Scotton, C. (2000). Code-switching as indexical of social negotiations, In Li Wei (Ed.), The Bilingualism Reader, (pp.127-153). London: Routledge. Myers-Scotton, C. (2002). Contact Linguistics: Bilingual Encounters and Grammatical Outcomes. UK: Oxford University Press. Myers-Scotton, C. (2006). Multiple Voices: An Introduction to Bilingualism. USA: Blackwell. Obeng, G. S (1997). An analysis of the linguistic situation in Ghana. African Languages and Cultures, 10:1, 63-81. 149 University of Ghana http://ugspace.ug.edu.gh Obiri-Yeboah, A. (2008). Codeswitching in University of Ghana Students’ Academic discussions. (Unpublished BA Long Essay). Linguistics Department, University of Ghana, Legon. Olusegun O. J. (2012). Roles of Code Switching in Multilingual Public Primary Schools in Ile- Ife, Nigeria. American Journal of Linguistics 1(3): 40-46. Opoku-Amankwa, K. & Brew-Hammond, A. (2011). ‘Literacy is the ability to read and write English’: defining and developing literacy in basic schools in Ghana. International Journal of Bilingual Education and Bilingualism 14 (1), 89-106. Paolillo, J. C. (1996). Language choice on soc.culture.punjab. Electronic Journal of Communication 6 (3). Retrieved October 18, 2016 from http://www.cios.org/www/ejc/v6n396.htm. Paoletti, I. (2011). Sociolinguistics. Sociopedia.isa, 1-14. Pérez-Sabater, C. (2015). Discovering language variation in WhatsApp text interactions. Onomázein 31, 113-126. Poplack, S. (1980). Sometimes I’ll start a sentence in Spanish y termino en español: toward a typology of code-switching, Linguistics 18: 581-618. Pride, J.B (1979). A transactional view of speech functions and code-switching. In McCormack & S. A.Wurm (Eds.), Language and Society. (pp.27-52). The Hague: de Gruyter. Quarcoo, M. (2009). Grammatical constraints on Twi – English codeswitching. (Unpublished MPhil thesis). University of Ghana, Legon. 150 University of Ghana http://ugspace.ug.edu.gh Quarcoo, M. (2013). Codeswitching in Academic Discussions: A Discourse Strategy by Students in the University Of Education, Winneba. (Unpublished PhD thesis). University of Ghana, Legon. Quarcoo, M & Amuzu, E. K. (2016). Code-switching In University Students’ Out-of-Classroom Academic Discussions. Issues in Intercultural Communication Volume 4, Number 1, 1- 14. Sánchez-Moya, A. & Cruz-Moya, O. (2015). Whatsapp, textese, and moral panics: discourse features and habits across two generations. Procedia - Social and Behavioral Sciences 173, 300 – 306. Sánchez-Moya, A. & Cruz-Moya, O. (2015). “Hey there! I am using WhatsApp”: a preliminary study of recurrent discursive realisations in a corpus of WhatsApp statuses. Procedia – Social and Behavioral Sciences 212, 52-60. Schiffrin, D. (1987). Discourse markers, Cambridge: Cambridge University Press. Sey, K.A.(1973). Ghanaian English. London: Macmillan. Siebenhaar, B. (2006). Code choice and code-switching in Swiss-German Internet Relay Chat rooms. Journal of Sociolinguistics 10/4, 481–506. Sridhar, S. N. and Sridhar, K. K. (1980). The syntax and psycholinguistics of bilingual code mixing. Canadian Journal of Psychology 34(4): 407-16. Statista (2017). Number of monthly active WhatsApp users worldwide from April 2013 to January 2017 (in millions) Retrieved April 16, 2017, from https://www.statista.com/statistics/260819/number-of- monthly-active-whatsapp-users/. 151 University of Ghana http://ugspace.ug.edu.gh Swigart, L. (1992). Practice and Perception: Language Use and Attitudes in Dakar. (PhD Thesis). Seattle: University of Washington. Tagoe, E. (2013) Is Whatsapp the Commonest Mobile App In Ghana? Retrieved November 9, 2016, from http://www.edwardtagoe.com/blog/2013/01/15/is-whatsapp-the-commonest- mobile-app-in- ghana/. Tannen, D. (1982). The oral/literate continuum in discourse, In D. Tannen (Ed.), Spoken and Written Language: Exploring Orality and Literacy. (pp. 1–16). Norwood, NJ: Ablex. Weber, T. (2010). Why companies watch your every Facebook, YouTube, Twitter move. Retrieved November 5, 2016, from http://www.bbc.co.uk/news/business-11450923. Wenger, E. (1998). Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press. Wenger E., McDermott, R., & Snyder, W. M. (2002). Cultivating Communities of Practice. Harvard: Business Press. Weinreich, U. (1953), Languages in contact: Findings and problems, (2nd printing in 1963). The Hague: Mouton. Weinreich, U., Labov, W., & Herzog, M. (1968). Empirical foundations for a theory of language change. In W. Lehmann (Ed.), Directions in historical linguistics. (pp. 97-195). Austin: University of Texas Press. Yevudey, E. (2009). Ewe-English codeswitching on radio. (Unpublished BA Long Essay). Linguistics Department, University of Ghana, Legon. 152 University of Ghana http://ugspace.ug.edu.gh APPENDICES 6.1 APPENDIX 1 6.1.1 SNAPSHOTS OF THE STATEMENT OF INFORMED CONSENT A. Statement of informed consent sent through research assistants to the chat groups 153 University of Ghana http://ugspace.ug.edu.gh B. Statement of informed consent sent to WhatsApp chat groups by the Researcher 154 University of Ghana http://ugspace.ug.edu.gh 155 University of Ghana http://ugspace.ug.edu.gh 156 University of Ghana http://ugspace.ug.edu.gh Some Approvals from participants of the WhatsApp chat groups 157 University of Ghana http://ugspace.ug.edu.gh 158 University of Ghana http://ugspace.ug.edu.gh 6.2 APPENDIX 2: Interview Responses Interview 1 Interviewer: Hello Hello Interviewer: Please I need your views for my Please I need your views for my research. research. SAM: Hiiii Hi Interviewer: Why would you say 'Diana nam Why would you say [quotation of what ntetenoa saa na wate asi ago the interviewee had said during a group prama noso.....Fine case' chat] Interviewer: Instead of writing English Instead of writing English SAM: Hahahahahaha Hahahahaahaha SAM: Its difficult for me express sm of de It’s difficult for me to express some of proverbs in English the Akan proverbs in English. That is dats y most often I mix dem with why most often I mix them with the de local dialect local language Interviewer: OK. Thanks very much Okay. Thanks very much Interviewer: So what you said is a proverb right So what you said is a proverb right SAM: U welcome You are welcome SAM: Yes plss Yes please SAM: Its like she is trying sm hide moves to get It’s like she is trying some hide moves away..... to get away... Interviewer: OK. Thanks very much for your assistance Okay. Thanks very much for your assistance SAM: Thnx u too....we hope to get ur findings n if Thank you too...we hope to get your possible recommendations findings and if possible Interviewer: Yes ooo. Yes ooo Interview 2 Interviewer: Hello. I want to ask a few questions Hello. I want to ask a few questions Interviewer: Plz why did you say “Oh as3m paaa Please why did you say “oh what an nni.... 3nny3 moaa.... a certain arsene issue...it is not your fault...a certain Arsene Wenger” instead of speaking English Wenger” instead of speaking English? King: Hello.... Hello... Interviewer: Hi Hi 159 University of Ghana http://ugspace.ug.edu.gh King: For a research? For a research? Interviewer: Yes Yes King: Ok.... Okay... Interviewer: Cos I thought it is easier and faster to Because I thought it is easier and faster to write in English than in twi on Whatsapp write in English than in Twi on WhatsApp King: True True Interviewer: So what is your reason for writing in twi So what is your reason for writing in Twi? King: But we want to localize whatsapp u know But we want to localize WhatsApp you know ��� Interviewer: Ok Okay Interviewer: How How King: S3 watease3 Hope you understand Interviewer: Aanne Yes King: Saaa saaa saaa Yes yes yes Interviewer: But I don't get your reason well ooo But I don't get your reason well ooo King: Oh.... just for fun.... Oh...just for fun Interviewer: Ok Okay Interviewer: Mate wai I hear you okay Interviewer: Medase paa Thanks very much King: 3nnaa ase Don’t mention it Interviewer: Plz do you also use pidgin on Whatsapp Please do you also use pidgin onWhatsApp? King: But maybe if u had some options that i But maybe if you had some options that I could choose one.... could choose one.... King: Yeah... sometimes Yes...sometimes King: Mostly to some specific friends Mostly to some specific friends Interviewer: When do you use pidgin When do you use pidgin? King: Some of my friends, I only speak pidgin For some of my friends, I only speak pidgin with them..... with them... Interviewer: What specifies those friend that you speak What specifies those friends that you speak pidgin with pidgin with? Interviewer: Why do you speak only pidgin with them Why do you speak only pidgin with them? King: For those friends, that's how it's always For those friends, that's how it's always been been King: Like Gabby Silva Like Gabby Silva Interviewer: Are they only boys Are they only boys? King: Yep Yes Interviewer: So is it possible for me to say that pidgin So is it possible for me to say that pidgin has 160 University of Ghana http://ugspace.ug.edu.gh has become a guy's code become a guy's code? King: For Gabby, he's always speaking pidgin For Gabby, he's always speaking pidgin King: So everyone speaks pidgin with Gabby So everyone speaks pidgin with Gabby Interviewer: Ok Okay Interview 3 Interviewer: Hello Rose, I want to ask you a question Hello Rose, I want to ask you a question Interviewer: Why would you say: Why would you say: Interviewer: ‘Happy B-Day Awuley. ‘Happy B-Day Awuley. Nyame nhyira wo bebree na ɔnyɛ God bless you and fulfil all your heart w'akomaso ebisadeɛ nyinaa mma wo desires for you 🎂🍷� � � � 🙌� 🙏� ’ 🎂🍷� � � � 🙌� 🙏� ’ Interviewer: Instead of using English Instead of using English Rose: You are still doing your research🙈🙈 You are still doing your research🙈🙈? Interviewer: Yes ooo Yes ooo Rose: I decided to type in Twi as a matter of I decided to type in Twi as a matter of preference preference Interviewer: Ok Okay Rose: And sometimes too, you want your other And sometimes too, you want your other party to feel the impact of what you want to party to feel the impact of what you want say. So it's convenient in using your local to say. So it's convenient in using your dialect local dialect Rose: That's why I had to switch to Twi That's why I had to switch to Twi Interviewer: OK. Thanks very much for your time OK. Thanks very much for your time Rose: You are welcome dear You are welcome dear Interview 4 Interviewer: Hello Hello Interviewer: Plz I want to ask a question Please I want to ask a question. Interviewer: Do you speak French Do you speak French? Earthangel: Yes pls Yes please Earthangel: Yes, studied French at knust and schooled in Yes, I studied French at KNUST and France too schooled in France too. Interviewer: OK. I want to know why you and monsieur Okay. I want to know why you and usually speak French on the page although monsieur usually speak French on the you are aware that others don't understand. page although you are aware that others don’t understand. 161 University of Ghana http://ugspace.ug.edu.gh Earthangel: Because I know he understands what I say so Because I know he understands what I my conversation are strictly directed to say so my conversation are strictly him�� directed to him�� Interviewer: Then why don't you private chat him but Then why don't you chat him privately choose to speak to him alone on the page but choose to speak to him alone on the page? Earthangel: They are things we all discuss but just some They are things we all discuss but just thoughts we share some thoughts we share Interviewer: OK. Thanks very much for your time Okay. Thanks very much for your time Earthangel: Don't mention Don't mention Interview 5 Interviewer: The bursary forms are in oo The bursary forms are in. Interviewer: Have you filled yours Have you filled yours? Araba: Yeah Yes Interviewer: Sweet Sweet Araba: Eeeiii calling me sweet Eeeiii you are calling me sweet Araba: My head sweet me You are flattering me Interviewer: Nti menfr3 wo sweetie da So haven’t I called you sweetie before? Araba: Wo afr3 me daa mmom but today its You have called me that before Sweet but today it’s sweet. Interviewer: Why did you answer in twi Why did you switch to Akan? Interviewer: Pls I need answers it would be useful for my Please I need answers. It would be Thesis useful for my thesis. Araba: Because u spoke in twi I switched because you spoke in Akan Araba: * spoke twi so i had to also answer with same You spoke Akan so I had to also language answer with the same language. Interviewer: OK. That is what I perceived OK. That is what I perceived Interview 6 Interviewer: Hello Hello Interviewer: Plz this is Sylvia Please this is Sylvia Interviewer: I want ask a question I want to ask a question. TILLY: Hello dear Hello dear TILLY: OK dear OK dear 162 University of Ghana http://ugspace.ug.edu.gh Interviewer: Why would you say Enoa na meka instead of Why would you say “that is what I’m speaking English saying” instead of speaking English Interviewer: On the group page On the group page TILLY: Hahaha Hahaha TILLY: Natn dear Nothing dear Interviewer: I want to know the reason behind that code I want to know the reason behind that choice code choice TILLY: We sometimes blend the two We sometimes blend the two languages Interviewer: With no reason in mind With no reason in mind? TILLY: No reason ooo No reason ooo TILLY: Oh no Oh no TILLY: Not at all Not at all Interviewer: OK thanks so much OK thanks so much TILLY: Uw dear You are welcome dear TILLY: We normally say those things to make fun ok We normally say those things to make fun ok Interviewer: OK. Thanks very much for your assistance OK. Thanks very much for your assistance TILLY: Uw dear You are welcome dear TILLY: Pls r u the one in the dp Please are you the one in the profile picture? Interviewer: Yes Yes TILLY: Ok Ok TILLY: Nice one der Nice one there Interviewer: Thanks Thanks TILLY: Uw You are welcome Interview 7 Interviewer: Hello. Plz this is Sylvia Hello. Please, this is Sylvia Interviewer: Plz I want to know why you used Twi in this Please, I want to know why you used sentence you sent to the group page Twi in this sentence you sent to the group page Interviewer: Eii, how kam you were all born on this day... “Eii, how come you were all born on this Beautiful ladies and my prodigal T. O. day... Beautiful ladies and my prodigal (father). Nyame Nhyira mo nyinaa T. O. (father). God bless you all.” NANA: Herhhh, so u still dey d page top Herhhh, are you still on the group page? Interviewer: Yes oo Yes oo Interviewer: Still on the study I am still undertaking the study. Interviewer: I want to know why you switched to twi I want to know why you switched to Twi. NANA: Okay, I thought you were done Okay, I thought you were done Interviewer: No No NANA: I'm very comfortable with that I'm very comfortable with that Interviewer: OK. Thanks OK. Thanks NANA: U w You are welcome 163 University of Ghana http://ugspace.ug.edu.gh NANA: Dassooor? That’s all Interviewer: Yes Yes NANA: Ok Ok Interview 8 Interviewer: Hello please this is Sylvia Hello please this is Sylvia. Interviewer: I want to ask a question I want to ask a question Interviewer: Why would you use two in wishing someone a Why would you use Twi in wishing happy birthday and not use English someone a happy birthday and not use English Interviewer: HBD Awuley..Onyame mfiri ne “Happy Birthday Awuley... From Kronkronbea nter3 wahye3 mu na )nte God’s holy place, may He enlarge nanim nkyer3 wo. Onyame asomdwie nka your coasts and may His face shine wo..🙏🙏 upon you. May the peace of God be with you.” OKORO: Haha, Sylvia you just want to worry me Haha, Sylvia you just want to worry me Interviewer: No. No. Interviewer: I want to know your reason for using twi I want to know your reason for using Twi Interviewer: Instead of English Instead of English OKORO: Papa Rock did same oo,plss did u see it? Papa Rock did same oo, please did u see it? Interviewer: I didn't check it I didn't check it Interviewer: So you used twi because Papa Rock did same So you used Twi because Papa Rock did same? OKORO: Oo not really, though I am a Twi man Oo not really, though I am a Twi man I don't use twi but I just felt within that I used I don't usually use Twi but I just felt wish him in twi within that I should wish him in Twi Interviewer: OK. So you don't really have a reason OK. So you don't really have a reason Right Right? OKORO: That's the reason That's the reason Interviewer: OK. Thanks very much for your time OK. Thanks very much for your time OKORO: I felt within that I should wish him in I felt within that I should wish him in twi...that's exactly what I wanted to say Twi...that's exactly what I wanted to say OKORO: Anyway, am happy u monitoring me..lol Anyway, I’m happy you are monitoring 164 University of Ghana http://ugspace.ug.edu.gh me.. laughing out loud Interviewer: I am actually monitoring the group I am actually monitoring the group OKORO: That's gud That's good OKORO: I hope there's nothing wrong with that, wishing I hope there's nothing wrong with that, someone in twi wishing someone a happy birthday in Twi Interviewer: No No OKORO: Will do same on ur birthday I will do same on your birthday Interviewer: OK. In my thesis I want to Establish the OK. In my thesis, I want to establish the reasons for such switches though we know that reasons for such switches though we English use is better in terms of Convience and know that English use is better in terms typing speed of Convenience and typing speed OKORO: Ok...that's very gud Ok...that's very good OKORO: 2nd degree right.. 2nd degree right? Interviewer: Yes Yes OKORO: Nice Nice OKORO: Will get you the reasons when time comes Will get you the reasons when time comes Interviewer: So what time So what time OKORO: Lol Laughing out loud OKORO: One, to catch attention or make ur One, to catch attention or make your msg/info/wish/whatever you want to put across message/information/wish/whatever you easily seen by everyone especially in groups want to put across easily seen by everyone especially in groups Interviewer: What about those who don't understand the What about those who don't understand Akan the Akan Language. OKORO: Two, to communicate exactly what one Wants Two, to communicate exactly what one to put across to avoid omission. Typing the wants to put across to avoid omission. same in English wouldn’t av been problem but Typing the same in English wouldn’t the likelihood of one omitting some words have been problem but the likelihood of one omitting some words. OKORO: Disadvantage of it. Disadvantage of it. Interviewer: Ok Ok OKORO: There r a lot, u can see it from other way round There are a lot; you can see it from the and you will get disadvantages too other way round and you will get disadvantages too OKORO: 1.Language barrier 1.Language barrier OKORO: 2.hatred(people) won't be happy y u used 2.hatred(people) won't be happy why different language you use a different language Interviewer: Ok Ok 165 University of Ghana http://ugspace.ug.edu.gh OKORO: Plsss do ur research Sofo maame Please do your research Sofo maame Interviewer: Yoo mate Ok I hear OKORO: Papa Rock did same, u can enquire from him Papa Rock did same, you can enquire as well from him as well Interviewer: I have asked him and he has given me his I have asked him and he has given me reasons his reasons Interviewer: Thanks very much Thanks very much OKORO: Okk..don't mention Ok…don't mention OKORO: Will get u additional reasons later I will get you additional reasons later OKORO: Nice day Nice day Interviewer: Thanks very much Thanks very much Interview 9 Interviewer: Hello Hello Interviewer: Pls this is Sylvia. I want you a question and it Please this is Sylvia. I will you a question is in relation to a research I am undertaking and it is in relation to a research I am undertaking. Interviewer: Pls what is your gender Please what is your gender FAM: Hi Hi FAM: Sylvia frm where? Sylvia from where? Interviewer: I am part of the organa chat group I am part of the Organa chat group. Interviewer: And I have been given the mandate to study And I have been given the mandate to the chats of the group study the chats of the group Interviewer: I just want to know whether you are a male or I just want to know whether you are a male Female or female FAM: kk Ok FAM: Male Male Interviewer: Thanks very much for your time Thanks very much for your time FAM: u welcome You are welcome FAM: is there anything more u want to know? Is there anything more you want to know? Interviewer: Ok. I want to know why you use twi on the Ok. I want to know why you use Twi on platform instead of English which is more the platform instead of English which is convenient more convenient FAM: Twi? Twi? FAM: How? How? Interviewer: In the chats, you have used both Twi and In the chats, you have used both Twi and English English Interviewer: An eg is "my dear 3ny3 by force wai" An example is “my dear it is not by force you hear Interviewer: You said this You said this FAM: Really? Really? Interviewer: So I want to know why you did that So I want to know why you did that 166 University of Ghana http://ugspace.ug.edu.gh FAM: When was that? When was that? Interviewer: 10/31/16 9:03pm 10/31/16 9:03pm FAM: When was that? When was that? FAM: dont get u? I don’t get you Interviewer: 31st October 2016 31st October 2016 Interviewer: Or you don't have any reason for switching Or you don't have any reason for switching codes codes? FAM: realli? what was tha chat about? Really? What was the chat about? Interviewer: OK. I just copied the portions of the chat I OK. I just copied the portions of the chat I need for my Work need for my work. Interviewer: So I don't have the whole chat in front of me So I don't have the whole chat in front of now me now FAM: kk Ok FAM: well it makes the chat interesting Well, it makes the chat interesting. Interviewer: Ok. Thanks very much for your time OK. Thanks very much for your time. FAM: u welki You are welcome. Interview 10 Interviewer: Hello. This is Sylvia. I want to ask a question Hello. This is Sylvia. I want to ask a question. Interviewer: Tilly wo sei nk)m) ooo instead of speaking Tilly, “you don’t make English conversation interesting” instead of speaking English. Interviewer: Why would you say that Why would you say that? IZZY: Ok Ok IZZY: Shoot Shoot IZZY: Ooooo sorry Ooooo sorry Interviewer: I want the reason behind that code choice I want the reason behind that code choice IZZY: There's no code ooo just that the twi carries the There's no code ooo just that the thought well for me Twi carries the thought well for me Interviewer: Hahahaa Hahahaa IZZY: I didn't know how to readily express that in I didn't know how to readily express English that in English. Interviewer: OK. Thanks very much OK. Thanks very much IZZY: You welcome You welcome Interviewer: Hello Hello Interviewer: Good morning Good morning Interviewer: Pls is Dubbies a male or female Please is Dubbies a male or female? 167 University of Ghana http://ugspace.ug.edu.gh Interviewer: Big six and lamb What about Big Six and Lamb Interviewer: Pls I want to their gender for my analysis Please I want to know their gender for my analysis? IZZY: Analysis? Analysis? IZZY: 👀 👀 IZZY: Well Lamb is lambert a guy Well, Lamb is lambert, a guy Interviewer: I am looking at bilingualism in relation to gender I am looking at bilingualism in relation to gender IZZY: But Dubbies am not too sure but I guess it would But Dubbies I’ m not too sure but I be a lady guess it would be a lady Interviewer: What abt big six What about Big Six IZZY: Eiiiish vasfo) Eiiiish university Interviewer: Hahaha Hahaha IZZY: Hehehe Hehehe Interviewer: What abt big six What about Big Six IZZY: Big six I guess is a guy he usually like to Big six I guess is a guy because he comment on politics and sports...and he is kind of usually likes to comment on politics muscline is the way he looks at things and sports...and he is kind of masculine is the way he looks at things IZZY: Big six is definitely not a lady Big Six is definitely not a lady Interviewer: Ok. Thanks very much for your time Ok. Thanks very much for your time IZZY: Anytime Anytime Interviewer: So grateful So grateful Interview 11 Interviewer: Hello Hello Interviewer: This is Sylvia This is Sylvia Interviewer: Plz I want to ask a question Please I want to ask a question Interviewer: What language have you used on the organa What language have you used on the page organa page Interviewer: “Kw333 Sergeant Lee.....otashi “Look Sergeant Lee... you are otwa teeeaaa..😂😂” comfortably seated and preparing tea” Interviewer: Plz can you explain to me why you decided Please can you explain to me why you to write in this language instead of English decided to write in this language instead of English 168 University of Ghana http://ugspace.ug.edu.gh PHAT: Good evening Sylvia Good evening Sylvia PHAT: Sorry for the long wait Sorry for the long wait Interviewer: Don't worry Don't worry PHAT: Well, dats Ga language Well, that’s Ga language Interviewer: What does it mean What does it mean PHAT: it means you're comfortably seated and It means you're comfortably seated and preparing tea preparing tea. Interviewer: OK. Then why didn't you say it in OK. Then why didn't you say it in English English PHAT: that's almost like whiles others are probably It simply means that whilst others are seriously doing something, you're probably seriously doing something, comfortably seated preparing tea you're comfortably seated preparing tea Interviewer: What are your reasons for using GA on the What are your reasons for using GA on platform the platform PHAT: I used it bcoz it rhymes with a "line" in a I used it because it rhymes with a line in a recently released song recently released song Interviewer: OK. OK. Interviewer: Thanks very much for your time Thanks very much for your time PHAT: of cos....anytime Of course...anytime Interviewer: Very grateful I’m very grateful PHAT: I tot you were done with you I thought you were done with your bilingualism.....or?? bilingualism.....or?? Interviewer: Nope. I am dating my data analysis No. I am doing my data analysis Interviewer: ****Doing ****Doing PHAT: good👍 Good PHAT: more power to you elbow den💪� more power to your elbow then💪� Interviewer: Thanks Thanks Interview 12 Interviewer: I want to know why you would use Twi on I want to know why you would use whatsapp though it is very comfortable to Twi on WhatsApp though it is very speak in English comfortable to speak in English. MIKE: I use it sometimes for convenience sake's I use it sometimes for convenience 169 University of Ghana http://ugspace.ug.edu.gh sake. Interviewer: How How MIKE: Sometimes too jx to make ma friends feel the Sometimes too, I use it just to make chat my friends feel the chat. MIKE: There r certain expressions in Twi that I may There are certain expressions in Twi not get their equivalent in English that I may not get their equivalence in English. MIKE: Again such words may come handy Again such words may come handy 6.3 Appendix 3: WhatsApp Data The following are some data from the WhatsApp chat groups. (1) SAM: Saaaa na afe aso.....if like u bi team aa like u b Really and also… if you were like a Black Terror...full of hope bt always 4th team like you would be Black place... Terror…full of hope but always 4th 🏃170🏃170🏃170🏃170🏃170🏃170🏃 place…🏃170🏃170🏃170🏃170🏃 awaaayyyy 170🏃170🏃 away EA: Ka na ka biom😏😏😏😏mouth lazt Say it and say it again😏😏😏😏 lazy mouth EA: Lazy# Lazy DAVE-RUSSIA: Oh Nana Abotr3😂😂😂 Oh Nana patience 😂😂😂 DAVE-RUSSIA: PhD, I believe, is not at any point considered PhD, I believe it is not at any point to understanding a human being by studying considered to understanding a human her (in that context or relationship/marriage) being by studying her (in that context of relationship/marriage) DAVE-RUSSIA: *of (not or) Jude: Mills Mills Jude: Is it true or it's a Joke Is it true or it's a joke EA: ���� ���� Jude: BEY: But April has 30 days. Lol But April has 30 days. Laugh out loud Jude: Mteeeewww (chuckling) BEY: 😂😂😂😂😂 😂😂😂😂😂 Jude: Thanx errhhh Thanks BEY: Welcome dear Welcome dear 170 University of Ghana http://ugspace.ug.edu.gh FRANK: I'm sure the writers point is "claiming to study I'm sure the writers point is "claiming to her" after 5 yrs. I will ve a lot of difficulty study her" after 5 yrs. I will have a lot understanding any man who dates a lady for 5 of difficulty understanding any man yrs and tells me he is studying. It sounds not who dates a lady for 5 years and tells been serious. A more better response such as me he is studying. It sounds not been "we are preparing to marry" sounds more serious. A much better response such as mature. "we are preparing to marry" sounds more mature. By the way, congrats Rockson. I have By the way, congrats Rockson. I ve booked ur booked a seat at your wedding, I'm wedding, I'm praying nothing hinders my praying nothing hinders my coming coming EA: 😂😂😂😂😂😂 😂😂😂😂😂😂 DAVE-RUSSIA: DAVE-RUSSIA: #Trumpanosomiasis #Trumpanosomiasis K O: Borga u no go slp??? Been-to won’t you sleep? QUOFI-US: As again oooo Ask again DAVE-RUSSIA: Lol Laugh out loud ANNA: Lol sen Why laugh out loud? (2) IZZY: *Asantefoɔ nhoma 1:1-2* Ashanti Book 1:1-2 *1*. Na ntɛm ara, wɔ tea mu bisaa sɛ hena ne Ɔbɛrima! Na ɛhɔ yɛɛ dinn 1. Immediately he proclaimed in a loud voice, ‘who is a man!’ and there was silence all over. 2. Then a little boy said! ‘a man is *2.* Na abɔfra ketoa bi kaa sɛ!, Ɔbɛrima ne nea ɔde nensa bɛsi ne bo so apue 3n3 anadwo nanka someone who can boldly go out this ɔbɛ te ne kaka nka. ��😂😂😂😂 evening; like he would see where power 🏃🏃🏃🏃🏃🏃🏃 lies’. (3) RICH: I was in church some time ago and the pastor said, "DO SOMETHING CRAZY FOR THE LORD!" I could see ladies removing their wigs. The pastor said again, "DO SOMETHING CRAZIER FOR THE LORD!" I saw people carrying plastic chairs. Now he was like, "can't you do something?...the CRAZIEST?" Someone from the back came and took the offering and was walking outside the church. The pastor shouted, "Hey! hey! gyimii! [Fool] this is not craziness; it’s MADNESS!!! OC.ITALY: Really hahahaha 171 University of Ghana http://ugspace.ug.edu.gh RICH: Cantona this is pure craziness or madness OC.ITALY: Stupidity OC.ITALY: How can you take offering away as a result of craziness (4) SLY: Somewhere, some place, right now, my future wife Somewhere, some place, right now, my is planning valentine wid some idiot. Father I pray future wife is planning valentine with some u cancel their plans..... idiot. Father I pray you cancel their S3 me b) me s3m 👏� 👏� 👏� 👏� na me plans..... b) mpayiea ogya oo If I clap my hands and pray, fire ooo fire 🔥🔥👏� 🔥🔥👏� 🔥🔥🔥🔥🔥🔥🔥👏� 👏� 👏� 🔥🔥🔥🔥👏� 🔥🔥🔥🔥👏� 👏� 👏� 🔥🔥😁😁😁 (5) BENE: Teacher: If I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven Sir. Teacher: Naaaaaa[No], listen carefully. If I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven. Teacher: oh...wo nso... [oh you too]Let me put it to you differently. If I gave you 2 apples, and another 2 apples and another 2, how many will you have? Abusua: Six. Teacher: Good. Now if I gave you 2 rabbits, and another 2 rabbits and another 2, how many will you have? Abusua: Seven! Teacher: oh...na wo nso wagyimi anaaa?[you too, are you that foolish?] seven no nso firi he?[where does the seven come from?] Abusua: aaah...Sir, Me wɔ rabbit baako wɔ fie dada oo...[Sir, I already have one rabbit at home] Teacher: ofui...� [Fool] crack some ribs�� 172 University of Ghana http://ugspace.ug.edu.gh (6) PHAT: I lost two laptop bags bcoz of kojapharma[escorts] I lost two laptop bags because of and still de ladies didn't even look at my face than to escorts and still the ladies didn't even even consider me.......I tot I was de only one, not look at my face than to even consider knowing I was only continuing a trend...... me.......I thought I was the only one, not In fact, Nyame ne hene ampa knowing I was only continuing a trend...... In fact, God is king indeed KEN: In fact wo hia Yesu In fact you need Jesus KEN: Happy birthday Esme..... in fact wo ho y3 me f3 Happy birthday Esme..... in fact you are papapapa. very beautiful. (7) DQ: We hear We hear KEN: Hehehe Hehehe KEN: That's right That's right PHAT: medaase wᴐ hᴐ! Thanks for that DQ: 2008????? 2008????? DQ: 😳😳😳😳😳 😳😳😳😳😳 DQ: Eeeiiiiiiii Eeeiiiiiiii DQ: Ebi womu ooooooooo it has been long oo G’fua: Diana r u surprised it's been 8yrs?? Diana, are you surprised it's Oh sorry..just assuming been 8yrs?? Oh sorry…just assuming DQ: 8long years 8 long years DQ: God is good God is good PHAT: ebi womu se sen?? Why is it long?? 173 University of Ghana http://ugspace.ug.edu.gh Joel: Long lasting freshest bachelor 🙋�♂🙈 Long lasting freshest bachelor 🙋�♂🙈 KEN: Jesus ᴐy3 ᴐhene k3see ampaa.... ᴐno nkotoo naa ᴐy3 It is true that Jesus is a great ᴐhene k3see ampaaaa ..... King...He alone is the greatest King indeed. It is true (8) SAM: http://mobile.ghanaweb.com/ http://mobile.ghanaweb.com/ GhanaHomePage/NewsArchive/I-got- GhanaHomePage/NewsArchive/I-got- zero-where-my-pastor-voted-Edward- zero-where-my-pastor-voted-Edward- Mahama-493947 Mahama-493947 Dave-RUSSIA: And so? (Sorry if it looks disrespectful, And so? (Sorry if it looks disrespectful, but dont perceive it as such)The fact but don’t perceive it as such) The fact that he is ur Pastor doesn't guarantee u a that he is your pastor doesn't guarantee vote. Ebei, when will we understand you a vote. Why! When will we politics! Even ur wife or husband has understand politics? Even your wife or the choice to vote for or against u. This husband has the choice to vote for or isnt news kraaa. against you. This isn’t news at all. Ghanaweb has hit below the belt. Ghanaweb has hit below the belt. Ka as3m a, me s)fo anto amma me��. Instead of something important, you Enti di3n? are saying, “My pastor didn’t vote for It shows how people see u in the whole me”. And so what? It shows how people political race. Why should he waste it if see you in the whole political race. Why he think there is a better candidate? should he waste it (his vote) if he thinks Mekᴐ kraaa😊 there is a better candidate? I am even going. (9) IZZY: Nyame ne Hene s3 s3n? When God has lined up God is King indeed like how? When big men before you. What again do you want from God has lined up big men before you, God. Or you want God to say "yes I will marry what again do you want from God? Or you" for you you want God to say "yes I will marry you" for you (10) a. Mills: Who will toaso today...?? Who will continue today...?? b. KOFA: Anelka koraaaaaaaaa adEn ne jealosy Anelka too why this jealousy? Girls, yia. Girls let's keep boiling. Hei char wer let's keep boiling. Hei Charlie where are 174 University of Ghana http://ugspace.ug.edu.gh r u. R u teachings? Caro, D day is fr me you? Are you teaching? Caro, the day is wate. for me you hear. c. KOFA: Ooooooo ma colapse mpo I have even collapsed d. KOFA: Gm small Good morning small e. OK: Hwan ne small no Who are you calling small (11) PAUL: Lol That yellow icon has been loading saa Laughing out loud that yellow icon has been loading repeatedly LEE: Hehehehh.......heaven laugh saaa Hehehehh.......heaven laughs repeatedly JOE: I laugh saaaaaaaaaaa,heerrrrhhhh Sammy I laugh continually Sammy (12) HN: I lost ma phone to robbers I lost my phone to robbers HN: Ve nw gotten anoda one I’ve now gotten another one TL: Fast ooo.I'm all ears Fast ooo. I’m all ears KOA: Oh saaaaaa? Oh really LSA: Saaaaaaa was it IPhone or yam Really, was it IPhone or yam? (13) QUAO-US: If you have any Gh girlfriend saying I If you have any Ghana girlfriend saying will think abt it........., dawg her I will think about it...leave her koraa........ The black American girls immediately... The black American will arrive next week. girls will arrive next week 😂😂😂😂😂😂😂😂 MONS: Massa.... I laf enter mugabe polling Master...I even laughed and entered station koraaa Mugabe’s polling station IMMA: He was not discerning to know He was not discerning to know at all. He koraaa. He wanted me to loose my wanted me to lose my precious wife. Is precious wife. Na lie! it a lie? (14) 175 University of Ghana http://ugspace.ug.edu.gh MARVIN: Den u ve 1000000000 messages to read Then you have 1000000000 messages to read DORIS: Yesoo Yes ooo MARVIN: Long essay Long essay DORIS: 😜😜😜 😜😜😜 DORIS: Have run out of battery koraaa I have even run out of battery (15) RICH: Comrades of late the platform is not Comrades of late the platform is not busy koraaaaa busy at all LINDA: I have even sowed my ataar koraaaa I have already sewed my attar (16) ANNA: He loves money papa He really loves money ANNA: Dr Terror loves money papa Dr Terror really loves money WINKY: Eiii.... baffour you have political eye papa Eiii.... Baffour you really have political eye SIKA: Eeeiii this super man is serious papa Eeeiii this superman is really serious ANNA: GOD HAS SAVED ME PAPA GOD HAS REALLY SAVED ME ERIC_HK: It is long papa...me am tired reading... will It is too long...me I’m tired of continue interesting though reading...I’ll continue. It is interesting though (17) KWESI: Chatty is true paaaa Chatty it’s true Paa gives emphasis to what is said KWESI: We should hv stopped fr Zagidi We should have stopped for Paa shows regret for not to beat him paaaa Zagidi to beat him allowing Zagidi to beat him KWESI: I should hv served commissioner I should have served Paa shows his regret dat day paaa commissioner that day CLE-US: Eish counting paaaaa Eish counting? Really? Paa shows the speaker’s doubt Dorcas: S h e did well yesturnyt paaaa She did well last night Paa intensifies the (intensifier) 176 University of Ghana http://ugspace.ug.edu.gh person’s performance ANNA: Becareful paaaaaa Be careful Paa intensifies the warning DAVE- Eiii Ivy... Edward and I have been Eiii Ivy...Edward and I have Saa shows a repeated RUSSIA: looking for you saaa been looking for you activity ROCK: I didn't know u had sweet words I didn’t know you had sweet Saa in this context saa. �� words shows surprise EMMA: Masa ur designing for Master you’re designing for Saa shows a repeated people saa people continually. action TILLY: Laugh saa cos more questions r Laugh more because more Saa serves as a caution coming questions are coming in this context (18) TIM: Why are you even talking? You're ineligible to vote Why are you even talking? You're so keep quiet wai! ineligible to vote so keep quiet you hear FII: We are liking it wai. Guys for you. We like it you hear. Guys for you LEE: Mills listen to this was....the said name above is Mills listen to this was....the said name wrong wai above is wrong you hear FII: Massa enjoy your stay in Hong Kong wai. Master enjoy your stay in Hong Kong you hear IZZY: I'm starting to believe you don't follow issues well I'm starting to believe you don't follow here in the group much more Ghana politics. Please issues well here in the group much kam and vote wai more Ghanaian politics. Please come and vote you hear (19) ERIC-HK: You see ooo..the NPP is a family thing You see that the NPP is a family thing. IZZY: Hw3, that's what I'll be using to vote ooo, 1 man Look, that's what I'll be using to vote, 1 woman 1 concubine 1 sidechick... Dassooor 1 man 1 woman 1 concubine 1 side chick... that’s all 177 University of Ghana http://ugspace.ug.edu.gh DQ: We are kicking ooooo We are kicking TILLY: We have our preferences ooo We have our preferences ERIC-HK: Sammy protect your voters' id ooo Sammy protect your voters' ID (20) OK: Awww. I wish u stay here for long but I can't Aww. I wish you stay here for long but stay at mummy's womb anymore. I need to come I can’t stay at mummy’s womb out soon. anymore. I need to come out soon. OK: U welkum twin brother You are welcome twin brother OC.Italy: Thank you Thank you DR: U re wlcm oooo You are welcome ooo OC.Italy: Yoooo Yes OK: U r 18th may You are 18th May OK: Rite? Right? OC.Italy: Sharp teeth kofi Sharp teeth Kofi OK: To November? To November OK: Hahaha Hahaha OK: Small boy Small boy OK: Y r u happy? Why are you happy? OC.Italy: You have changed your date of birth You have changed your date of birth no wonder E.C mpo asesaa won logo no wonder even E.C has changed their logo DR: Aaaaah Osei Charles!!! 😜😜😜😜😜 Aaaaah Charles!!! 😜😜😜😜😜 DON: We r happy that u are happy @bday boy We are happy that you are happy birthday boy (21) PARTICIPANTS CONVERSATION TRANSLATION JJ: The thing dey ova me The thing is over me JN: Wey i de hung fr here The way I am hungry over here JJ: So mke I set give u? So let me set it for you JN: Yh Yeah JN: U whet my appetite waaa You just whet my appetite FS: Akora u boorrr oo Akora, are you angry? FS: See the way u cut cut See the way you are cutting it (probably a phone call) K: What he Dey chop? What is he eating? FS: Kokonte Konkonte (A kind of food in 178 University of Ghana http://ugspace.ug.edu.gh Ghana) JJ: Eba+shitor oil .... Eba+shitor oil ....(a kind of food) AS: Haaaaahaaaaaa Haaaaahaaaaaa AS: nfodzo cut de alomo bottle show we Nfodzo, snap a picture of the Alomo bottle to show us AS: eba no dey make man sweat so Eba cannot make you sweat this way JN: Lolz laugh out loud AS: e hard oo messi too en appetite dey fi It is indeed surprising that Messi’s make wet appetite is also whet AS: *whet *whet JN: Hahaha Hahaha SL: Em say no bra dey He said there is no bra AS: Wow Wow AS: a moda breastfeeding her children A mother breastfeeding her children B-US: I sure say these ladies be Nigerians I am very sure that these ladies are Nigerians. AS: I know dem 2 dey stay abeka I know that the two of them stay at Abeka B-US: Really ? Really? B-US: Them be Gh girls ? Are they Ghanaian girls? JN: Make u no mind am Don’t mind him AS: yeah 2 of dem Yeah two of them SH: There noooo u appear It is there you appeared SL: Eeei lair... der nor jeff go talk say e Eeei liar… it is there Jeff ate one chop one for inside inside JJ: Hahahahha...TI shotto get u Hahahahha...TI Shotto has gotten cheaaap... 😂 you so cheap. B-US: Lol Laugh out loud AS: Haaaaahaaaaa Haaaaahaaaaa Aswear I swear AS: de timing bi perfect The timing is perfect B-US: Agbamey where U go hide? Agbamey, where will you hide? B-US: We search you for like one month this I have been looking for you for close to one month. B-US: We figa say the gods kidnap U We want to say that the gods kidnap you AS: ebi more dan 1 month It is more than 1 month 179 University of Ghana http://ugspace.ug.edu.gh JN: Why sleepo always de talk lair Why is Sleepo always spelling lair instead of Liar instead of Liar B-US: For nogokpo shrine At Nogokpo Shrine (22) PARTICIPANTS CONVERSATION TRANSLATION K O: wo no no, love nurse...now u turn That’s you,(KO is suggesting that the relationship dictionary person is noted for being interested in relationship talk) Love nurse...now you have turned a relationship dictionary. PHAT: that's how I understand it that's how I understand it MAY: Continue wai Continue you hear PW: M a policewoman ooo I take first I’m a policewoman ooo I only take first responses p3....subsequent ones ain't responses...subsequent ones aren’t considered considered JOE: Will term it broken hearted I’ll term it broken hearted PW: 😂😂😂😂😂😂😂😂 r'ship dictionary 😂😂😂😂😂😂😂😂 relationship dictionary paaa indeed MAY: Lol Laughing out loud PHAT: I said it!!! I said it!!! MAY: Broken whole body Broken whole body PW: Awwwww wate mmaa no akoma💔 Awwww he has broken those women’s heart K O: Sometimes u these guys will give ur Sometimes you these guys will give your hearts to the ladies and give them hearts to the ladies and give them hammers, wat do u want them to use hammers, what do you want them to use the hammer for?? the hammer for? JOE: Naaa No MAY: Vice versa rather Vice versa rather PW: Woy3 sure Tilly Are you sure Tilly PHAT: they gave him 💔💔 they gave him broken heart PW: He said he abandoned de projects He said he abandoned the projects oooo oooo JOE: Tilly, u r really following Tilly, you are really following MAY: Daabi ok. Thats wat he is saying No okay. That’s what he is saying 180 University of Ghana http://ugspace.ug.edu.gh SAM: Ok am bk lets continue Okay, I’m back let’s continue MAY: Mummy how be Mummy how be PW: Plz hu z nt following 😂😂😂😂 Please who is not following 😂😂😂😂 JOE: U all are You all are PHAT: happened before de abandonment happened before the abandonment K O: The broken heart gave him The broken heart gave him convulsion convulsion and he had to abandon and he had to abandon the project the project PW: Eeeeeiiiii masere saaaaa....wala boi Eeeeeiiiii I have laughed ampa continually...wala boy indeed JOE: Lol Laugh out loud QWAKU: Bonnke crusade is on Bonnke crusade is on PHAT: de ladies had to suffer for de 💔💔 The ladies had to suffer for the broken heart. SAM: Bt my boss if wer jilted bf starting But my boss if were jilted before starting dos 2 abandoned projects deaa den u those 2 abandoned projects, then you have v bn in 3 diff relationships bf 19th been in 3 different relationships before oooo anaa ebi ka ho 19th ooo or there are others more (23) Participant Message Translation JM Congrats Rock......y3b3hyia w) Kum n'ase Congrats Rock….we will meet in Kumasi Bracket has 3 has been been used in used in place of ‘ԑ’ place of ‘ᴐ’ 181 University of Ghana http://ugspace.ug.edu.gh (24) SAM: Medie medan s3 me Maame wawo na ose 3y3 menus All I know is that my mother has given birth and she has showed the baby to me that the baby is my sibling SAM: Menua* Sibling (25) Participants Conversation Translation MPK: Quoi? What? EA: Quel mansonge??. What a lie?? MPK: Bien évidemment mais la femme semble un peu Obviously good, but the woman seems FBI sur son mari a bit FBI on her husband EA: 😒😡😣 😒😡😣 DQ: Mansonge eele quie?[trying to also speak (Gibberish) French which does not make sense] DQ: � � EA: Eeeei Diana,what language is Eeeei Diana, what language is that??😁😁😁😂😂😂 that??😁😁😁😂😂😂 DQ: French French EA: I swear😂😂😂😂 I swear😂😂😂😂 EA: From which country.?? From which country?? MPK: Republic of kumasi Republic of kumasi DQ: Paddone koraa lemme e tai [trying to speak (Gibberish) French which does not make sense] DQ: 😂😂😂😂👍💃� 😂😂😂😂👍💃� EA: I cannot see that?? Eele quie (This is a quote of I cannot see that?? what DQ said) (gibberish) DQ: It means mansonge is sleeping? It means lie is sleeping EA: Hahaha Hahaha EA: But who taught you that But who taught you that MPK: 🚨🚑🚨🚨 🚨🚑🚨🚨 EA: 😁😁😁😂😂😂😂😂 😁😁😁😂😂😂😂😂 DQ: Akwasi Nyarko Akwasi Nyarko DQ: � � EA: Huh cheri,Ta's traduit ca pour Elle,??? Huh darling, did you translate this for her??? MPK: Avec quel logiciel? With which software? DQ: Capour Elle ellei aaa [trying to also speak (Gibberish) French but it is meaningless] DQ: 🍦if you are confuse take this 🍦if you are confuse take this DQ: U will understand my french you will understand my french 182 University of Ghana http://ugspace.ug.edu.gh EA: Diana, attention sinon!!�🔫💣🔨 Diana, be careful or else!! DQ: 😇 DQ: Then English please Then English please EA: Je sais meme pas I don’t really know DQ: How can your mother be on this page? How can your mother be on this page? DQ: 😳😳😳😏 😳😳😳😏 EA: Diana STP laisse moi tranquille Diana please let me be DQ: 😡😡😡 😡😡😡 DQ: Stop confusing the noble people on this Stop confusing the noble people on page ooo this page ooo EA: 😂😂😂😂😂💃�💃�💃� 😂😂😂😂😂💃�💃�💃� MPK: R u da spokesperson of da noble people? Are you the spokesperson of the noble people? DQ: Concerned citizen Concerned citizen (26) PARTICIPANTS CONVERSATION TRANSLATION SER: Nyame y3da wo ase God, we thank you ERIKA: Congrats Papa Rock. God bless your marriage Congrats Papa Rock. God bless your marriage EA: Terror can't read this message today Terror can't read this message la😜😂😂💃� today OK 😜😂😂💃� COBY: Doc Doc Doc Doc EA: Not forgetting ken😂😂😂😜 Not forgetting Ken😂😂😂😜 SER: 😂😂😂😂 😂😂😂😂 ERIKA: 😁😁😃 😁😁😃 NAANA: Congrats Papa Rock Congrats Papa Rock DAVE-RU: I perfectly agree. We go change our Assemblies I perfectly agree. We will change too. Bibiaa y3 change� our Assemblies too. Everything is change SILVA: Congrats rock. Congrats Rock. IZZY: Tetelestai... 3y3 me k)n) dodo Tetelestai...it is appetizing 183 University of Ghana http://ugspace.ug.edu.gh ROCK: Y3da mo ase bebree�� We thank you so much ROCK: Pray with us and prepare to grace the occasion Pray with us and prepare to grace for us🙏�🙏� the occasion for us🙏�🙏� BIG: Hehehe...saa Saaaa Saaaa.... Hehehe Yes, yes, yes (27) KOA: Mr Paido needs her to survive Mr Paido needs her to survive KOA: 😄 😄 ADJOA: I was jux passing thru I was just passing through ADJOA: Gudnite fam Good night family KOA: We understand We understand KOA: Someone needs you Someone needs you ADJOA: Sure😉 Sure😉 KOA: Good night dear Good night dear KOA: Pass through more times Pass through more times SIKA: Still waiting for the good news ooo @ Still waiting for the good news ooo from Eld.Rock Eld.Rock KOA: Eiiiii Nancy w'asum y3 wo d3 ooo Eiiiii Nancy[you always want to be informed] G’fua: Don't worry😊 Don't worry😊 G’fua: It would definitely be someone we know It would definitely be someone we know SIKA: Papa Papa (28) KEN: My grandfather often tells a story My grandfather often tells a story of a of a heavily polygamous who died of heavily polygamous man who died of nketenkete [extensive hunger at night] ANNA: 😂😂😂😂 😂😂😂😂 ANNA: 😂😂😂😂😂😂 😂😂😂😂😂😂 KEN: Because all his wives thought this Wife Or that will Because all his wives thought this one cook for him..... wife or the other would cook for him..... ANNA: I laugh saaaaaaa get nketenkete I have laughed continuously that I have gotten extensive hunger at night KEN: Hehhehehe Hehhehehe ANNA: Hehehehehe Hehhehehe (29) 184 University of Ghana http://ugspace.ug.edu.gh FII: They can't declare all at once They can't declare all at once FII: They will declare one after the other. They will declare one after the other. Joel: They are jux talking too much. They are just talking too much. We want dem to win must they must relax and We want them to win; must they relax wait for the EC and wait for the Electoral Commission? RUTHY: Bcoz the leaders are not acting ryt Because the leaders are not acting right RUTHY: They are running mampam race! They are running very slow race! (literally mampam means alligator) Joel: Exactly Exactly RUTHY: )mo tutu mampam mirika [wasted efforts] (30) LEF: I have missed him I have missed him FII: Me hearing it from Awuku today? Me hearing it from Awuku today? FII: Anyway I understand you. Anyway I understand you. FII: Aboa biara bewu a )pre [literally it means: every animal struggles for breathe before it dies] ERIK-HK: Yes Yes ERIK-HK: Smh Shaking my head FII: Let's leave it here Anyway I understand you. (31) OC-Italy: Sincerely Sincerely CLEMENT-US: Eish Eish DOR: It shuld be more hotter dan dis It should be more hotter than this RICH: Wowe ofoɔ nsa na wo hwɛ wo deɛ (literally: when you chew a chimpanzee’s hand, then you watch yours)What you do unto others shall be done unto you. OK: @ Richard to learn harder @ Richard to learn harder KOFI: I'm here @ oyibo I am here whiteman OC-Italy: You want to remind her not to You want to remind her not to forget forget you clement you Clement (32) 185 University of Ghana http://ugspace.ug.edu.gh ANNA: Hahahahahahah Hahahahahahah (laughter) ANNA: Mpena twie sweet u paaaaaa ooooo You really enjoy dating (literally: dating sweet you) ANNA: 😂😂😂😂😂😂😂😆😆😆😆😆 😂😂😂😂😂😂😂😆😆😆😆😆 LIZZY: Hahaha! Hahaha! LIZZY: Papa Rock ahunu amane! Papa Rock is in trouble COBY: That word "mpena" is very heavy oo That word “dating” is very heavy oo (33) SAM: When I say Merry Christmas you say "No When I say Merry Christmas, you say Ababas33" 3 times. Then you make your hands “No going backwards” 3 times. Then like Usain Bolt and do "Hweeeeeeeew" you make your hands like Usain Bolt and do "Hweeeeeeeew"(it is a sound and it implies the act of moving very fast) (34) Joel: Ghanaians will forget soon Ghanaians will forget soon Jacob: Maybe Maybe If this is the way workers are going to PHAT: If this is the way workers are gonna get salary get salary increments, then there increments, den der shouldn't be any TOASO for any shouldn't be any continuation for any party..... party..... (35) FII: If really he meant to do all these things he could If really he meant to do all these things, he have done it earlier. Why increase salaries two could have done it earlier. Why increase weeks to your exit of office. salaries two weeks to your exit of office. Adwen b)ne nkoaa. Only evil motives. And someone too will Obi nso nk)ka a politics. say (attribute it to) politics. We will start killing people in the name of politics We will start killing people in the name of politics (36) Had we not tried a little more, they ROCK: Had we not tried small, they would have outnumbered would have outnumbered us for us for Listowell's wedding Listowell's wedding DUBY: Hmm Hmm ROCK: Yes ooo. Thanks for reminding me of them Yes ooo. Thanks for reminding me 186 University of Ghana http://ugspace.ug.edu.gh of them ROCK: Sorry for behaving like a nagging wife😜 Sorry for behaving like a nagging wife😜 MAT: Lol Laughing out loud ROCK: But yesterday for instance, from my But yesterday for instance, from my interaction with Jona, he felt a bit interaction with Jona, he felt a bit disappointed because he had made food disappointed because he had made reservation for us and we couldn't make it. food reservation for us and we Enti 3y33 no ya kakra couldn't make it. It indeed hurt him a little ROCK: But thanks to you all for understanding me But thanks to you all for understanding me (37) K O: Oh jake paa, why you take ma number give the Oh Jake too, why did you give my lady that I said I won't agree??? number to the lady that I said I won’t agree??? K O: Ah u paa Ah you too ANNA: 🙅�🙅�🙅�🙅�🙅�💪�💪�💪�💪�💪� 🙅�🙅�🙅�🙅�🙅�💪�💪�💪�💪�💪� Jake I bore koraa, am leaving the page. Jake I am so angry so I am leaving the K O: I'll join after the switching page. I’ll join after the switching K O: See you, abi u make hapi, u want me to leave so u See you, hope you are happy. You want can switch codes...me boa am not going me to leave so you can switch codes...I anywhere, me de t33 wo. am lying; I’m not going anywhere, I was only kidding (38) OKORO: I have no idea oo Boss...only I have no idea Boss...only God will know..lol God will know… laughing out loud EA: 👍� 👍� ROCK: Nyame aaa ne wo boafo� � 🙌�🙌�🙌� Only God is your helper� � 🙌�🙌�🙌� ROCK: Eiiii, may3 code switching😂😂😂 Eiii, I have done code switching OKORO: 🙏🙏 🙏🙏 SLY: Hahaha Hahaha (39) 187 University of Ghana http://ugspace.ug.edu.gh ANNA: Terror Terror SAM: So r u ppl saying one party "is mad" or So are you people saying one party “is "will go mad soon" mad” or “will go mad soon”? IZZY: Oh Tilly paa, gyae saa koraa na boys p3 Oh Tilly too, stop that because if boys ade3 as) ade3 so ay3 ade3 ma no ay3 ade3 want something to top up something to ama y3 nya ade3 do something for it to become something in order to get something SAM: Plss explain further wai Please explain further you hear IZZY: The next government will continue The next government will continue PAT: I like that. The ade3 rhyming I like that. The something rhyming ANNA: Hahahahaha Hahahahaha (40) ALEX: Atletico all de way 😜😜😜😘😘😘 Atletico all the way 😜😜😜😘😘😘 RICH: Clatenburg will spoil the match Clatenburg will spoil the match RICH: The refree The referee RICH: Against ani bɛ wu Those against will be ashamed OWURA: Poor messi is now glued 2 his TV Poor Messi is now glued to his TV set set watching gud soccer from two watching good soccer from two best team in d world best teams in the world RICH: Hahahahahaah Hahahahahaah OWURA: 😂😂😂 😂😂😂 ALEX: U don't know wat u aa saying�� You don't know what you are saying�� OWURA: Eii Rubin no malice, y u go hammar Eii Rubin no malice, you will hammer my ma head? head? OWURA: Atanfo no mmo hyire3 Enemies be happy OWURA: Na mo kae s3 penalty aba nti 3no a So you people thought there is a penalty no no so that is it OWURA: I don't feel d heat of a final game in d I don't feel the heat of a final game in the ongoing match, r u wid me on-going match, are you with me RICH: Yes true Yes true OWURA: 😉😉 😉😉 CLEMENT- Osey atletico Hurray Atletico US: OWURA: Atletico no go do tee Atletico can’t do anything CLEMENT- Yoo I hear 188 University of Ghana http://ugspace.ug.edu.gh US: OWURA: Wait till we increase d score to 2 Wait till we increase the score to two CLEMENT- For the where At where US: OWURA: Shine ur eyes u go see now Be vigilant for you will see what will happen CLEMENT- Goaaaaaaaaaalllll Goal US: OWURA: Yes I c Yes I see CLEMENT- Shine ooooo micky Be vigilant Micky US: CLEMENT- History History US: OWURA: Now we r zooming into d game Now we are zooming into the game Proper proper OWURA: D heat is on The heat is on OWURA: Wear ur seat belt cos we r abt 2 Wear your seat belt because we are about take flight to take flight CLEMENT- Let's see Let’s see US: ALEX: Mike u aa now watching football Mike you are now watching football ALEX: No hasty conclusion wate!!! No hasty conclusion you hear OWURA: No shaking No shaking ALEX: In our childhood u ddnt play football In your childhood, didn’t you play b4 football before OWURA: Na lie It is a lie OWURA: I didn't play well bt no Hw 2 watch I didn’t play well but I know how to watch ALEX: Tweaaaaa (it is not true){this is not its meaning but in this context that is the meaning implication it gives us} ALEX: U may lack some technicalities You may lack some technicalities CLEMENT- Mourinho never played before but Mourinho never played before but he’s US: he's one best coaches lol one of best coaches. Laughing out loud OWURA: U hear rubin Did you hear? Rubin CLEMENT- Where is CR7 Where is CR7? US: OWURA: Hes out of d game nt literally He’s out of the game not literally OWURA: Messi has bewitched him nt 2 make Messi has bewitched him not to make history today history today 189 University of Ghana http://ugspace.ug.edu.gh CLEMENT- Hahahaha Hahahaha US: CLEMENT- I beg let messi be lol I beg you let Messi be. Laughing out loud US: OWURA: Ah e no b him hu dey do am Ah he is not the cause OWURA: Tonyt is nt CR7 nyt Tonight is not CR7 night ALEX: No Leave Messi out wae No, leave Messi out you hear Don’t worry Bro. Don't worry... 190