A Case Study of Code- Switching in Multilingual Namibian Keyboard-To

Home , Namlish, World Englishes

A Case Study of Code- 1. Introduction In a linguistically diverse setting like Nami- Switching in Multilingual bia, it is not surprising that code-switching

Namibian Keyboard-to- he Republic of Namibia in southern constitutes a widespread practice among the Africa cannot only be characterized population and is not limited to face-to-face Screen Communication communication – especially in the light of T by its cultural diversity and multi- Journal Article ethnicity, the population’s de facto multi- devices like smart phones and other new

Frederic Zähres lingualism seems ubiquitous (cf., e.g., Busch- technologies versus the related new forms of feld & Kautzsch 2014: 122-123; Kautzsch & communication, which have emerged only in Despite its multilingual setting, Namibia’s Schröder forthc.: 1). Interestingly, the coun- recent times and have spread so fast around sociolinguistic situation has attracted little attention by researchers at this point. try has maintained a monolingual language the globe that (linguistic) research only hardly keeps pace with it (cf. Brock & While English has been the sole official policy – with English as the sole official lan- language for over 20 years, at least 10 guage – ever since its independence in 1990 Schildhauer fc.; Dürscheid & Frick 2014). other languages can be encountered in the (cf. Frydman 2011; Wallace 2011: 309). This Keyboard-to-screen communication – either southern African country, whereas English in the form of text messaging or, more is seldom acquired as L1. mostly ideologically-based decision, how- Keyboard-to-screen communication ever, does not reflect the linguistic daily life recently, represented by the smart phone (KSC), i.e. messaging services such as SMS of Namibians, neither in the past nor cur- application WhatsApp – is a popular means of and WhatsApp, is omnipresent in our rently. Also, the consequences of this communication, also in Namibia. current daily lives. As the medium fre- monolingual language policy on sectors like This case study aims at catching a quently exhibits elements of spoken language, the language used on WhatsApp education have attracted the attention of glimpse of Namibia’s insufficiently-re- can be an interesting source for pheno- researchers more recently to investigate, searched current sociolinguistic situation by mena that often occur in oral language use. among other aspects, language use and analyzing multilingual WhatsApp messages This case study aims at shedding light attitudes (cf., e.g., Buschfeld & Kautzsch provided by Namibian informants. Also, re- on the sociolinguistic situation of Namibia search on the digital medium WhatsApp still by analyzing occurrences of codeswitching 2014; Kautzsch & Schröder fc.), language within KSC of Namibians. Findings include contact phenomena (cf., e.g., Kamati 2011; is at an early stage and observations could the preference of English over the L1 Stell 2014a; Simasiku et al. 2015), and shed light on typical features of WhatsApp during casual conversation with discourse- potential nativization effects on the English communication. For this, the linguistic related code-switching to the L1 for spoken in Namibia (cf., e.g., Buschfeld & situation of Namibia will be briefly outlined emotional or urgent matters. Additionally, code-switching appears to be a non- Kautzsch 2014; Schneider & Schröder 2015; in the next section, followed by a summary of marked feature of KSC in Namibia. Kautzsch & Schröder fc.). the current state of research on Keyboard- to-screen communication – with a focus on Student Paper

1 0 p l u s 1 10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics L I v I n g L I n g u I s t I c s Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 32

WhatsApp. Then, the multilingual Namibian Basters (cf. ibd.). Also, European missi- which lasted until independence in 1990, the WhatsApp messages and, subsequently, the onaries found their way to the area during oppressive apartheid system was employed analysis will be provided and discussed. A the nineteenth century (cf. Wallace 2011: by the South African National Party in the concluding section will summarize the 54-56). This pre-colonialization contact zone middle of the twentieth century. Soon, findings. would not only connect the speakers of the resistance against the South African Bantu and the Khoisan languages, respect- oppressors emerged in form of the South 2. Namibia tively, but also introduced Indo-European West African People’s Organisation, languages to the country. SWAPO in short, and – literally – fought for 2.1 History of Namibia Not even fifteen years later, Germany Namibian independence, which was colonized South-West Africa until World ultimately gained in 1990 (cf. Buschfeld & Namibia’s history can be divided into several War I. The German rule of German South- Kautzsch 2014: 126-127; Wallace 2011: eras defined by distinct periods of language West Africa was not a peaceful period but 309). The drafted constitution includes, contact that have shaped the country’s served as a major phase of language contact among several other important changes, a current cultural and linguistic landscape. between the Germans and the South-West monolingual language policy that lists Before colonization, the area of present- Africans as German was an official language English as the sole official language of the day Namibia was home to several groups of of the period (cf. Frydman 2011: 182) – and Republic of Namibia as “Namibia’s liberation indigenous Khoisan peoples – the Nama, the remnants of that period can still be found in movement […] deemed it necessary to Damara, and the San – who were joined by present-day Namibia, especially when loo- replace Afrikaans, the ‘language of the different Bantu-speaking settlers – the king at Namibian place names, for example oppressors,’ and to establish a language Herero, the Ovambo, and the Kavango – in Lüderitz or Mariental. policy in preparation for an independent the seventeenth century (cf. Buschfeld & After the era of German South-West Namibia” (Frydman 2011: 182). Kautzsch 2014: 124-125). In the following African had ended in 1915, the country centuries, ethnically mixed groups immi- became a League of Nations mandate 2.2 Sociolinguistic Situation in grated to the territory: the first group governed by South Africa (cf. Wallace 2011: Present-Day Namibia consisted of Oorlams from Cape Colony who 205). German lost its status as an official migrated towards South-West Africa in the language and was replaced by both Today, the status of English as the sole late 1700s and early 1800s. Around 1870, a Afrikaans and English – with the latter never official language is not reflected in the second group came from South Africa to reaching the status of a de facto official diverse population, which is mostly multi- settle down in the Rehoboth territory, language (cf. Frydman 2011: 182). During lingual (cf. Buschfeld & Kautzsch 2014: 127). coming to be known as the Rehoboth the South African administration period,

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 33

Despite being labeled as the ‘language of the found that, in classrooms, code-switching is gence towards Afrikaans rather than English, oppressors’, Afrikaans is commonly en- often employed to facilitate understanding it seems that Afrikaans does not possess the countered in Namibia – and, to a lesser ex- and to overcome a language barrier that is connotations of an ‘oppressor language’ in tent, even German is still spoken. However, clearly present (cf. Kamati 2011; Simasiku et present-day Namibia. English has gradually spread among the al. 2015). However, more recently, researchers Namibians since independence: media such Stell (2014a: 98, 102-105) has dis- have found potential signs of nativization of as television, radio, and news-papers are covered that English only functions as a English within the speech of Namibians (cf. available in the English language and also means of interethnic communication for Buschfeld & Kautzsch 2014; Kautzsch & international programs are broad-cast (cf., non-whites – most notably the native Bantu- Schröder fc.; Schneider & Schröder 2015): e.g., Buschfeld & Kautzsch 2016: 5-6; language speakers, whereas Afrikaans re- This includes potential mergers and splits on Ejikeme 2011: 69-70). Also, the English mains a more general lingua franca within the phonological level, elements in the language has gained ground among the groups also including Coloureds and lexicon, morphosyntactical constructions, population and is regarded as a prestigious Whites1. In Namibia, Afrikaans still is the and also pragmatic features that are either language (cf. Buschfeld & Kautzsch 2014: native tongue of several ethnic groups closely related to other South African 141-143, 147-148). (including Coloureds and Rehoboth Basters) varieties of English or might even constitute However, the mainly ideologically- and a second language of various other unique features of a Namibian variety of based monolingual language policy is viewed groups (e.g., Namas, Damaras, Hereros) who English – future research will shed more light controversially by parts of the Namibian grew up with a Bantu or Khoisan language on this. Namibians themselves call their population (cf. Frydman 2011). Even today, themselves. Thus, as groups show conver- variety of English Namlish (cf. Buschfeld & English is usually not acquired as a native Kautzsch 2014: 27), which indicates a language but rather learned in schools as a certain linguistic awareness regarding the 1 The Bantu-speaking groups in Stell’s (2014a) study second language, where it eventually be- were represented by Ovambos and Hereros. English spoken in Namibia. In any case, all of comes the primary medium of instruction Especially Ovambos would primarily make use of this inevitably implies that English has from Grade 4 or even Grade 1 onwards (cf. the English language – in intra- and interethnic gained much ground in Namibia during the communication. Whether this is linguistically or Kamati 2011: 1; Stell 2014b: 227). Unfor- politically/ideologically motivated seems to be an last twenty years and is indeed present in tunately, though, in reality, this does not interesting question – the Ovambos in Stell’s study the linguistic everyday life of Namibians. stated not to be proficient in Afrikaans, which seem to correlate with the actual language could have geographical reasons (a possible That being said, language practice in proficiency of either the teachers or the consequence of the apartheid regime) or a Namibia has not been thoroughly students (cf., e.g., Kamati 2011; Kisting 2011, statement to advocate and support the SWAPO’s documented and researched yet. The language policy. This is, however, speculation and 2012; Simasiku et al. 2015: 8). It has been would need to be addressed in further research. present work aims at doing exactly that and

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 34 presents a piece of the country’s linguistic Dürscheid & Frick 2014: 150; Hinz 2015). message. This, according to Dürscheid & puzzle with a focus on the use of code- Jucker & Dürscheid (2012) as well as Frick (2014), potentially causes real-time switching practices within keyboard-to- Dürscheid & Frick (2014: 161-177) have response pressure, which subsequently screen communication. A brief excursion to described WhatsApp and directly compared leads to a higher number of misspellings the current state of research in the area of it to SMS, while also establishing the term despite the more sophisticated virtual keyboard-to-screen communication seems keyboard-to-screen communication.2 Main keyboard of a smart phone as compared to necessary to put these findings into context. differences between both forms of commu- the numerical keyboard of older mobile nication include the different types of under- phones. This also leads to a fundamental 3. Keyboard-to-Screen lying constraints and multimedia capabili- difference in the communication: In SMS, Communication ties: SMS operate on a space constraint typically, single communicative acts are while WhatsApp seems to be constrained by conveyed whereas WhatsApp communica- During the previous 20 years, mobile phones time – i.e. in SMS communication, messages tion involves a higher degree of dialogicity 3 and, more recently, smart phones have be- are billed in 160 character intervals, due to the presence of communicative act come a constant companion in our every-day whereas WhatsApp is, as of 2016, a free sequences (cf. ibid.: 172). In other words, lives. With the devices, new forms of service that even allows up to 4,000 similar to oral conversation, a whole string of communication have also become ubiqui- characters per single message. However, related utterances is produced during tous, like the frequently-used short messa- WhatsApp represents a potentially more WhatsApp communication rather than SMS ging service (SMS) or, with the rise of smart synchronous form of communication than communication. Further, WhatsApp not only phones, the instant-messaging application SMS and resembles instant messaging allows group chats with up to 50 people but WhatsApp – the latter having already over- software on a computer that displays also enables users to enrich their messages taken the use of SMS in terms of the number whether a contact is online or typing a in various ways without any additional costs: of messages sent per day and year (cf., e.g., From pictograms – usually referred to as 2 Fischer 2015; Sparkes 2015). Also, Whats- The introduction of the term keyboard-to-screen emojis – to photos and videos to voice communication seemed necessary to distinguish recordings, WhatsApp users are able to App has only very recently reached more communication forms using phones from other than 1 billion users worldwide (cf. Statt forms of digital communication that are typically overcome the spatial distance between them 2016). addressed with computer-mediated communica- and their contacts and share their tion, even though, of course, smart phones are While WhatsApp is widely used, resear- technically computers. For further information, see experiences (cf. Arens 2014: 87-101; König chers have only started to study the medium, Jucker & Dürscheid (2012) and Dürscheid & Frick & Bahlo 2014: 8-9) almost effortlessly – (2014). 3 whereas SMS users only have very limited, also within the field of linguistics (cf. Today, SMS flatrates have made this constraint rather obsolete, however. rather complicated and expensive options in

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 35 this regard. This, bundled with the general 4. Data & Methodology left side, is a 25-year old teacher, while her oraliteracy found in keyboard-to-screen younger sister informant B is a 21-year old communication, allows for much more 4.1 Data Collection & Informants medical student. Their native language is expressive and emotional communication German and both informants have been (cf. Arens 2014: 101), the implications of The data investigated here consists of a exposed to English and Afrikaans from an which still needs to be better understood via corpus of 173 turns of sequential WhatsApp early age on due to their multilingual envi- more research. messages from five separate conversations ronment – formal language education, Within the field of keyboard-to-screen between the same two informants amoun- however, started from Grade 1 for English communication, linguistic studies addressing ting to roughly 850 words.4 The informants and from Grade 5 or 6 for Afrikaans. While code-switching can mostly be found for SMS are sisters and native Namibians with informant A uses German, English, and communication: Researchers worldwide German-speaking parents who were Afrikaans on a daily basis, informant B appa- have already compiled big corpora of lingu- contacted via email and asked whether they rently does not use Afrikaans as frequently. istic keyboard-to-screen communication would like to contribute to a research The informants did not make use of any data (cf., e.g., sms4science) with various project by providing the researcher with kind of autocorrect function during their research goals. For WhatsApp communica- screenshots of WhatsApp conversations of conversations, as is also evident from the tion, however, only small case studies seem their own choosing that, at some point, amount of misspellings within the excerpts. to exist and focus on writtenness of dialects contain more than one language.5 The data or group identities, respectively (cf., e.g., has been transcribed and anonymized. 4.2 Framework for Analysis Hernández 2015; Weber & Schürmann Additionally, informant A has agreed on 2014). Linguistic studies investigating key- answering questions regarding both infor- There are numerous approaches to code- board-to-screen communication in southern mants’ linguistic background and specific switching, especially distinguishing the African countries are unknown to the aspects of the WhatsApp conversations. formal and the functional perspective on author. Informant A, whose messages in the code-switching. For the purpose of this case following examples will be displayed on the study, the formal grammatical aspects of code-switching will not be the primary focus. The social aspect of code-switching will be 4 While the corpus and number of speakers is fairly small, it is still appropriate for a qualitative pilot the focal point of this work, even though it study, which will be followed up by a larger constitutes a rather complex matter that can investigation. 5 I would like to thank Prof. Dr. Anne Schröder for be fuzzy at times (cf. Stell & Yakpo 2015: 4). making this exchange possible in the first place.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 36

Thus, the analysis of the WhatsApp con- whether turns contain elements of more shows the distribution of languages versation data follows a socio-functional than one language or are monolingual. according to the turns both informants take. conversational analysis approach, which Generally, informant A is a little more active assumes that conversational code-switching 5. Results than informant B as informant A has sent constitutes a deliberate and functional 103 turns to B while informant B has only choice of a speaker who takes into account 5.1 Overview sent 70 turns to A. Informant A clearly linguistic and cultural backgrounds of ad- prefers the English language during her dressees (cf. Auer 2009; Gumperz 1982; The languages that are used within the WhatsApp messages – even when code- Myers-Scotton 2009: 476; Weber & WhatsApp conversations between both switching within turns, English always Schürmann 2014: 196). Distinctions are informants are English, German, and appears in the message. While informant B made between insertional and alternational Afrikaans. Occasionally, the languages are mostly uses English as well (to a lesser de- forms of code-switching and discourse- and switched within the same turn. More gree than her sister however), she frequently participant-related switching – even though common is, however, the alternation resorts to sending very short and, within this both informants know each other extra- between turns. Several turns have been categorization, linguistically ambiguous ordinarily well and would, assumingly, not counted as other in this overview as they messages (n = 21). Additionally, informant B feel the necessity to enter negotiation contained ambiguous content – examples for hardly uses any Afrikaans – only two sequences, which are the defining element of this are emojis, interjections or ambiguous occurrences of intra-turn switching between participant-related code-switching (cf. Auer words. Ok, for instance, appears frequently English and Afrikaans can be found in the 2009: 500). The analysis should aim at in several spelling variations like ok, okay, oki data. Compared to the use of English and investigating the interactional component of or oku, often constitutes a whole turn, and German, informant A does not frequently code-switching without decontextualizing cannot clearly be attributed to one of the compile messages in Afrikaans, either. single occurrences of switching. For three occurring languages. Table 1 illustrates Table 1 Distribution of Languages According to descriptive purposes, terminology of the the distribution of languages that the Individual Informants (in Turns) formal Matrix Language Frame model by individual informants use, while Table 2 Myers-Scotton (cf., e.g., 2009: 484-485) will also be used. Additionally, to get a first impression, the turns have been quantified according to languages and code-switching status, i.e.,

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 37

Both informants clearly prefer the English language during WhatsApp conversations among each other (cf. Table 2): about 53% of all turns (n = 94) are monolingual and in Eng- lish. Under 15% of all turns (n = 21) contain bi- or multilingual content. This, however, does not necessarily reflect the amount of code-switching within the data – the intra- turn switches, in most cases, represent intra- sentential and also inser-tional code- switching.6 The instances of code-switching will be addressed more thoroughly in the next section. Table 2 Distribution of Languages According to Turns Taken by Both Informants 5.2 Conversational Analysis an appropriate dress and accessories and, thus, asks A whether she could send her Within the data, different types of code- something via mail. The following examples 6 This assumption is based on the findings of a switching can be found. These include se- show the frequent switching between corpus study by Schnitzer as presented in Dürscheid & Frick (2014: 169): the average amount veral forms of insertional and alternational mostly German and English, both in intra- of characters per WhatsApp message amounts to code-switching. This section presents ex- and inter-turn contexts. As previously 34.4 characters (while the average message in SMS communication is 110 characters long). This would cerpts from the WhatsApp conversations be- mentioned, informant A’s messages will be imply that a turn in WhatsApp communication tween both informants and provides an displayed on the left side while informant B’s rather represents phrases or short sentences and, messages are on the right side. Also, within thus, would contain forms of intra-sentential code- analysis of the respective code-switching switching. Whether insertions represent the most situation. the excerpts, the languages are coded as frequent type of intra-sentential switching (in The first conversation, which is also the follows: German, English, Afrikaans, Other. WhatsApp communication) and whether WhatsApp turns usually do not contain more than one longest conversation overall with 61 turns, is Translations are provided by the author and sentence would have to be addressed by further thematically based on a formal dance with a are displayed in parentheses. research. For the data at hand, this appears to be the case, though. masquerade ball theme, for which B needs

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 38

This first excerpt shows the starting point of communication – visualized through the consideration, i.e. the following turns, this the main part of the first conversation and is timestamps next to the message content, two-turn sequence would rather seem like already very representative for the inform- which represent one complete and sent discourse-related alternational CS: The very ants’ behavior during the rest of the message – and both turns have been sent in informal greeting to A is given in English, the conversation – not only in terms of code- rapid succession. As the first turn constitutes actual matter of importance for B, then, is switching: Numerous misspellings can be a salutation, one could argue for a case of introduced in German instead, possibly for found (e.g., *dide instead of dude, *Somewhee emblematic insertional switching here, emphatic or organizational reasons. A then instead of somewhere, etc.), typical non- which would, then, suggest that B has answers in English, which hints at the standard spellings and constructions are embedded English elements into a German possibility that this whole excerpt does not used (e.g., *wat instead of what, omission of matrix. Also, taking the further context into signify a linguistic negotiation sequence but determiner in *Hast du Abendkleid? – literally *Do you have evening gown?), apostrophes in contractions are ignored, blank spaces at word boundaries are frequently omitted, and capitalization is almost exclusively found at the start of a sentence (esp. in speaker A’s messages), which could be the result of an automation process that numerous modern (smart) phones possess. Interesting to note is also the spelling of the German determiner ein, or in this case *ei, which could either be a typing error or non-standard spelling that can be found in dialectal variation. Focusing on the code-switching aspects of Example (1), several interesting passages should be closely examined. The sequence of turns starts with two turns by informant B: the first is an informal English greeting while the second states B’s request in German. There is a clear turn boundary in this form of

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 39 rather presents a case of unmarked code- switching as both informants know each other well and communicate frequently. B’s next turn is ambiguous: while the latter part of the answer is definitely English, the for- mer two words (Okay. Cocktail? [as in cocktail dress or Cocktailkleid]) are found in German as well – both words are borrowings from English, though. It is clear that, within this turn, a switch to English is taking place; depending on the interpretation, however, the switch could be regarded as insertional or alternational. In any case, the next turns by A include the German insertion ja into A’s rather stable English frame. From the whole sequence, it seems impossible to determine any pragmatic function of this switch, which, again, suggests an unmarked and possibly conversationally non-functional code- switching situation, i.e. code-mixing. The element wat, on that note, is another ambiguous element as it could be Afrikaans, a non-standard spelling of English what, or a non-standard spelling of the German was primarily found in low German dialects, which both signify the same concept, however. B replies in German and, thus, uses the same language with which the main point of the sequence has already been B to talk about the benefits of her part-time English, which could be indicative of the introduced. After a while passes, A messages job in English to which B also replies in emphatic function of German for informant

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 40

Example (4) also shows another interesting instance of switching: It passes an den körper an n takes […]. The message starts in English, contains a German verb phrase and then switches back to English. The verb, however, contains English inflection attached to the German verb stem; the argument of the verb phrase is also given in German. As the English frame can still be recognized, both 7 B while English is the preferred choice for sent in the meantime. The alternational the intra-word switch and the intra-sen- casual conversation. Informant A’s last turn switch to German by A most likely functions tential switch could be considered cases of in Example (1) can be seen as another in- as another case of discourse-related insertional code-switching. On a functional stance of code-mixing, which A seems to be switching that organizes the conversation level, speaker A possibly uses the linguistic using more frequently than B. and, in this case, might draw the attention common ground between her and her sister Example (2) shows an excerpt from the from another issue back to the topic of the to overcome specific lexical gaps – the continuation of the same conversation, dress. Even after this excerpt, A continues informants might feel more comfortable in which takes place on the next day. The first using a German frame for a number of turns their native tongue with parts of the word turn of B in this example reflects her until she starts using more lexical insertions field clothing or A felt that the German language choice for the whole day: Even from English and also Afrikaans (cf. Example concept expresses her thoughts more though she is still asking A for accessories (3)) before she switches to a monolingual precisely, i.e., another case of unmarked that fit to the dress for the formal dance, she English message again (cf. Example (4)). The code-mixing. does not use German anymore. A, however, frequent multilingual insertions, this time The second conversation between A and employs more switching now. While she only into the German frame, suggest that code- B does not focus on one particular issue inserts an interjection from Afrikaans and a switching for both frames is an unmarked unlike the previous one. The conversation German quantifier into the English frame of choice for speaker A. consists of several sequences (58 turns) her first reply, she later exclusively compiles discussing different smaller topics, such as German messages. This alternation occurs church, one sister visiting the other, one after more than three hours have passed and 7 The informants have omitted this part from their sister feeling sick, et cetera. Example (5) different messages have apparently been conversations. Thus, it is not known to the author shows an excerpt from that conversation. what happened in the WhatsApp conversation during that timeframe. There, A and B message each other regar-

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 41 ding a number of friends or family members who want to go to a local church. Whenever a friend’s or family member’s name is mentioned, the message is compiled in German, whereas most other parts of the conversation sent by A are, as usual, in English. The topic of friends and family could serve as a trigger to German – their native English. Even in a sleepy state, both parties The next section will discuss the analysis’ language, associated with said friends and keep using several languages, which, again, findings in the greater context of the lingu- family – as a form of discourse-related code- shows how comfortable both parties are istic situation of Namibia. switching. The same can also be observed for with the usage of a multilingual code. the other conversations. In summary, the following types of code- 6. Discussion The last example (cf. Example (6)) is an switching can be accounted for in the data: excerpt taken from the same conversation. The WhatsApp conversation data of both Even though it does not show any new 1) Instances of bi- and even trilingual informants present interesting insights into features of the informants’ code-switching insertional and alternational code- the linguistic situation of Namibia, which will behavior, the context makes this excerpt switching with a speaker-related be thoroughly discussed in the following. noteworthy: This part of the conversation function: both informants know each While this cannot be as clearly observed occurs at night when both informants are other’s personal and similar linguistic for informant B, informant A frequently uses about to go to sleep as they both are rather background well and can exhaust their English during her WhatsApp conversations tired, which they do not only explicitly state multilingual proficiency to a big extent despite her native tongue being German. in the sequence of messages but it is also without having to expect any problems During follow-up questions, A has also suggested by the amount of typing errors in their communication. expressed that she generally prefers using that A produces. Despite the late hour and 2) Discourse-related code-switching English in conversations. Her sister, infor- their tiredness, both informants – but occurs when specific topics are brought mant B, has stated that she has several especially A – alternate between two or up or require emphasis: the informants German friends in and outside of Namibia, three languages, respectively. A alternates seem to switch to German, their native which she regularly keeps in touch with. This from English to German to Afrikaans and language, to discuss important issues, could explain why B’s messages contain back to English, while B initially sends one such as requests of urgency or when more German. The few occurrences of message in German and then switches to talking about friends and family.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 42

Afrikaans within the data can be explained by the informants’ background: Even though both informants have been in contact with Afrikaans from an early age and have learned it formally in school, only A currently uses the language regularly in her job. The majority of the few turns that actually include Afrikaans are also compiled by A. From this, it is obvious that there is an imbalance in proficiency of Afrikaans between both informants, which they both are aware of and, hence, Afrikaans gets rarely used during their conversations – except for a few brief insertions that might be formulaic in nature or could be considered borrowings. Afri- kaans has had a long linguistic influence on the country (cf. section 2), so this would not be surprising. However the case may be, English is the language of choice for both informants for the majority of conversations within the data – and not their first language German. It would be interesting to see whether this represents a common phenomenon in Namibia: Do Namibian families, siblings, or friends with a very similar linguistic background actually con- verse in an L2 among each other on a daily basis? Further research would need to address this.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 43

aim at investigating whether this case of conversationally unmarked code-switching, i.e., code-mixing, is a representative case for Namibia or whether it satisfies lexical needs, e.g., the compensation for lexical gaps possibly caused by lacking proficiency. Namibians frequently refer to what they consider their variety of English as Namlish – whether signs for conventionalized code- mixing can be found is another interesting question that would deserve attention. However, for the latter question, the current data will in all probability not be representative as German is not widely spoken in Namibia anymore, even though it is The insertional code-switching found in the to-screen communication (cf. Section 3). This certainly present in parts of Namibian data is assumed to possess a speaker-related allows for constructions like the intra-word society. function. As the time codes next to messages switch presented earlier. Furthermore, the On that note, German, English and suggest, the messages have usually been switches can rather not be seen as linguistic Afrikaans belong to the same language compiled in rapid succession, so intra-turn negotiation sequences as they are usually family of the Indo-European languages and code-switching seems to happen effortlessly. not met by any form of special response and share similar phonemes and morphological Both informants have stated that their profi- still occur frequently. At least for informants as well as syntactical features. As there are ciency in German and English is equally high. A and B, this implies that frequent code- several other languages spoken in Namibia So, both siblings know that, among each switching does not constitute an unusual that do not belong to this language family other, they can efficiently exhaust their, in phenomenon during conversation and can, and possess different features, the formal this case, bilingual possibilities, i.e., they can thus, be considered an unmarked choice. As perspective of code-switching could be compile their messages even containing previously stated, code-switching seems to worth investigating with more data from several switches without causing any be a wide-spread phenomenon in Namibia Khoisan and Bantu languages. information loss while still satisfying the (cf. section 2), as it is in many language The pragmatic function of alternational constraints of quasi-synchronous keyboard- contact situations. Further research should code-switching is another matter of interest.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 44

While the informants in this study seem to spelling that could hint at a dialectal insertions or as alternating sequences. The only switch to their L1 when discussing variation that is used by both informants. bi- and trilingual insertions are produced and topics that seem to represent importance, This study, as a small-scale case study, perceived effortlessly by the informants and e.g., friends and family or matters of does not aim at providing a representative do not seem to represent a marked choice in temporal urgency, would other Namibians description of code-switching in Namibia. their communication. Alternational sequen- organize their discourse in a similar manner? However, with the presented data, several ces to German are found as discourse- It would seem like a plausible choice to potential approaches to further research related code-switching when the informants employ this type of code-switching, e.g., to have been highlighted. address a topic of greater importance, such minimize information loss. Additionally, for a as friends and family or urgent matters. keyboard-to-screen communication medium 7. Conclusion Whether the findings are in any way re- like WhatsApp, code-switching always presentative for the whole country needs to represents a distinct visual stimulus that is The present case study has shown that the be investigated with further research. sent to the recipient of the message – non- linguistic contact zone in Namibia still As the study has shown, in the future, Indo-European languages would be even deserves more attention from researchers. linguists should investigate the addressed more salient in this regard. The complex language contact situation that issues of language choice and potential code- Regarding features of keyboard-to- has been shaped by the country’s history and mixing as well as the pragmatic functions of screen communication, both informants diversity has only started to be investigated. code-switching in conversations. Data from have not shown unusual behavior: both The briefly-introduced concept of keyboard- other informants, especially native speakers sisters moderately use typical means of to-screen communication constitutes ano- of Bantu or Khoisan languages, would pro- shortenings, e.g., abbreviations like lol or xx ther field that has only started to gain vide great insight into this matter. (signifying kisses), as well as keyboard attention by researchers but can provide character-based emoticons or the more linguists with interesting insights into the References recent emoji-pictograms (cf. Example (1), linguistic everyday lives of people. The data Arens, K. (2014). WhatsApp: Kommunikation 2.0. Example (2) and Example (4)). The amount of have revealed that, at least for the two Eine qualitative Betrachtung der multimedia- misspellings and typing errors as well as the current informants, English functions as a len Möglichkeiten. In K. König & N. Bahlo fast succession of turns seem to confirm popular L2 for casual communication despite (Eds.), SMS, WhatsApp & Co: Dürscheid’s & Frick’s (2014) assumption that a shared German L1 background. German Gattungsanalytische, kontrastive und WhatsApp communication seems to be based and Afrikaans, however, are observed in the variationslinguistische Perspektiven zur Analyse on a time constraint. Interesting to note in messages, either in the form of smaller mobiler Kommunikation (pp. 81-106). Münster: this regard is the non-standard German Monsenstein und Vannerdat.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 45

Auer, P. (2009). Bilingual Conversation. In N. Cascadilla Proceedings Project. Retrieved Kisting, D. (2012, January 10). Namibia’s Coupland & A. Jaworski (Eds.), The New from http://www.lingref.com/ Language Policy is ‘Poisoning’ its Children. Sociolinguistics Reader (pp. 490-511). cpp/acal/40/paper2574.pdf. Retrieved from Basingstoke: Palgrave Macmillan. Gumperz, J. (1982). Discourse Strategies. http://www.theguardian.com/education/201 Brock, A. & Schildhauer, P. (fc.). Communication Cambridge: CUP. 2/jan/10/namibia-english-crisis. Form: A Concept Revisited. In A. Brock & P. Hernández, N. (2015). Bilingual Encounters König, K. & Bahlo, N. (2014). SMS, WhatsApp & Schildhauer (Eds.), Potentials and their Use: Online: Constructing a Common Linguistic Co. – Forschungsstand und New Perspectives on Communication Forms, Code in German-English WhatsApp. Talk at Analyseperspektiven. In K. König & N. Bahlo Affordances and What Users Make of Them. Anglistentag 2015, Paderborn, September (Eds.), SMS, WhatsApp & Co: Frankfurt am Main et al.: Peter Lang. 23-25. Gattungsanalytische, kontrastive und Buschfeld, S. & Kautzsch, A. (2014). English in Hinz, L. (2015). The Functions of Emoticons and variationslinguistische Perspektiven zur Analyse Namibia: A First Approach. English World- Pictograms in Instant Messengers. 10plus1: mobiler Kommunikation (pp. 1-16). Münster: Wide, 33.2, pp. 121-160. Living Linguistics, 1, 92-103. Monsenstein und Vannerdat. Buschfeld, S. & Kautzsch, A. (2016). Towards an Jucker, A. H. & C. Dürscheid (2012). The Myers-Scotton, C. (2009). Code-Switching. In N. Integrated Approach to Postcolonial and Linguistics of Keyboard-to-Screen Coupland & A. Jaworski (Eds.), The New Non-postcolonial Englishes. World Englishes. Communication: A New Terminological Sociolinguistics Reader (pp.473-489). doi:10.1111/weng.12203. Framework. Linguistik online 56.6. Retrieved Basingstoke: Palgrave Macmillan. Dürscheid, C. & Frick, K. (2014). Keyboard-to- from http://www.linguistik- Schneider, K. P. & Schröder, A. (2015). Variational Screen-Kommunikation gestern und heute: online.org/56_12/juckerDuerscheid.html. Pragmatics and the Specificity of English in SMS und WhatsApp im Vergleich. Networx, Kamati, A. M. (2011): A Study of Code Switching in Namibia. Talk at IAWE 2015, Istanbul, 64, pp. 149-182. Junior Secondary Physical Science Classrooms in October 8-10. Ejikeme, A. (2011). Culture and Customs of Selected Schools in the Oshana Education Simasiku, L. et al. (2015). Barriers to Code Namibia. Santa Barbara: Greenwood. Region. Master of Education thesis, University Switching in English Second Language Fischer, D. (2015, November 2). WhatsApp vs. of Namibia. Medium Classrooms. International Journal of SMS: Kurznachrichten boomen in Kautzsch, A. & Schröder, A. (fc.). English in English Literature and Culture, 3.1, pp. 7-13. Deutschland. Retrieved from Multilingual and Multiethnic Namibia: Some Sms4science (n.d.). Retrieved from http://www.smartdroid.de/ Evidence on Language Attitudes and on the http://www.sms4science.org/. whatsapp-vs-sms-kurznachrichten-boomen- Pronunciation of Vowels. In C. Ehland et al. Sparkes, M. (2015, January 12). WhatsApp Over- in-deutschland/. (Eds.), Anglistentag 2015 Paderborn. takes Text Messages. Retrieved from Frydman, J. (2011). A Critical Analysis of Proceedings. WVT: Trier. http://www.telegraph.co.uk/technology/news Namibia’s English-only Language Policy. In E. Kisting, D. (2011, November 9). 98% of Teachers /11340321/WhatsApp-overtakes-text- G. Bokamba et al. (Eds.), Selected Proceedings not Fluent in English. Retrieved from messages.html. of the 40th Annual Conference on African http://www.namibian.com.na/index.php?id=8 Statt, N. (2016, February 1). WhatsApp has Linguistics: African Languages and Linguistics 7893&page=archive-read. Grown to 1 Billion Users. More than Doubling Today (pp. 178-189). Somerville MA: since Facebook Acquired it. Retrieved

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics Frederic Zähres | A Case Study of Code-Switching in Multilingual Namibian Keyboard-to-Screen Communication 46

http://www.theverge.com/2016/2/1/108895 34/whats-app-1-billion-users-facebook- mark-zuckerberg. Stell, G. (2014a). Social Identities in Post- Apartheid Intergroup Communication Patterns: Linguistic Evidence of an Emergent Nonwhite Pan-Ethnicity in Namibia? International Journal of the Sociology of Language, 230, pp. 91-114. Stell, G. (2014b). Uses and Functions of English in Namibia’s Multiethnic Settings. World Englishes, 33.2, pp. 223-241. Stell, G. & K. Yakpo (2015). Elusive of Self- Evident? Looking for Common Ground in Ap- proaches to Code-Switching. In G. Stell & K. Yakpo (Eds.), Code-Switching between Structural and Sociolinguistic Perspectives (pp. 1-16). Linguae & litterae 43. Berlin: de Gruyter. Wallace, M. (2011). A History of Namibia. London: C. Hurst & Co. Weber, K. & T. Schürmann (2014). Funktionen unterschiedlicher Codes in niederdeutscher SMS-Kommunikation von L1-Sprechern. In K. König & N. Bahlo (Eds.), SMS, WhatsApp & Co: Gattungsanalytische, kontrastive und variationslinguistische Perspektiven zur Analyse mobile Kommunikation (pp. 193-218). Münster: Monsenstein und Vannerdat.

10plus1: Living Linguistics | Issue 2 | 2016 | Contact Linguistics