A Sociolinguistic Study of the Use of Code Mixing in Social Media by Egyptian Bilingual Users

1. Introduction

            Recently, the wide spread of social media such as Facebook, Twitter and WhatsApp has dramatically changed the traditional form of written and spoken language. Crystal (2001) and Herring (2001) argue that we can no longer differentiate between written and spoken language though the former is more formal. Language has witnessed a shift of colloquial features of spoken language into written language. Examples of these features are the use of hesitation and reaction words (e.g., um, you see) and words and symbols reflecting emotion (e.g., oh my god) (Gupta, 2016).

In this study, some attempts to differentiate code switching and code mixing are discussed: for example, Muysken (2000), Cardenas-Claros and Isharyanti (2009), Vyas, Gella, Sharma, Bali, & Choudhury (2014) among others.  A sample of 23 social media texts (2 to 3 sentences each) was collected from three main social media: WhatsApp, Twitter and Facebook. Participants were divided into three main groups based on the degree and extent of their exposure to English as their second language. For the analysis of data, I followed Muysken (2000) in avoiding using the term code switching for the general process of mixing. With code mixing, some linguistic units such as morphemes, words and phrases of one language is embedded into an utterance of another language (Arabic/ English code mixing).  Lexical borrowing is also differentiated; it is defined as a process of transfer or copying where a native speaker adopts an element from other language into the recipient language  or a non-native speaker imposes properties of his/her native language onto a recipient language. Lexical borrowing takes place when second language speakers can’t avoid phonological and syntactic interference from their native language, but they can avoid using words from their native language (Haspelmath 2009). In this study, lexical borrowing and insertion are used as two dominant models that underpin code mixing. Dwijayanti and Wahyana (2012, p. 199) argue that a main characteristic of insertion is that “it is similar to spontaneous borrowing in which well-defined lexical item is inserted into a sentence that belongs to the first language”; the following is an example of insertion based on Asy’ari (2009):

-          Mudah-mudahan isi majalah  up to date terus

Hopefully the contenet of the magazine is always up to date.

                                                                     (taken from Dwijayanti and Wahyana 2012, p. 200) 

1. Internet Language

Due to the wide spread of social media such as Twitter, Facebook, Instagram, Snapchat, WhatsApp) and the availability of internet connections offered at reasonable prices, people have moved to a virtual world where actual interaction was replaced by written chat that has become people’s means of expressing views, ideas and emotions. Every aspect of life has been converted into a virtual world. For example, students exchange materials, discussions, assignments; internet advertisements have become more popular; work meetings and arrangements are organized through these social media. All these tasks and activities are carried out through social media. Accordingly, linguists’ attention has been directed towards the impact of the new written language employed in social media texts on spoken language. The difference between written and spoken language has raised a considerable attention in literature. For example, Crystal (2001, pp. 7-8) proposes the following features that highly characterize written language:

(1)   graphic features: distinctive typography, page design, spacing, use of illustrations, and color

(2)   orthographic (or graphological) features: the writing system of an individual language, defined in terms of such factors as distinctive use of the alphabet, capital letters, spelling, punctuation, and ways of expressing emphasis (italics, boldface, etc.)

(3)   grammatical features: the many possibilities of syntax and morphology, defined in terms of such factors as the distinctive use of sentence structure, word order, and word inflections

(4)   lexical features: the vocabulary of a language, defined in terms of the set of words and idioms given distinctive use within a variety

(5)   discourse features: the structural organization of a text, defined in terms of such factors as coherence, relevance, paragraph structure, and the logical progression of ideas

A phenomenon that accompanied this dramatic shift to Internet Language is switching from one language to another or mixing a language with another language. In the following section, a distinction between code switching and code mixing is established.

1. Code-Switching or Code- Mixing?

In literature, there are many attempts to differentiate between code-switching and code-mixing. Cardenas-Claros and Isharyanti (2009, p. 68) argue that code-switching occurs when a bilingual speaker uses more than one language in a single utterance above the clause level to appropriately convey his/her intents. Cardenas-Claros and Isharyanti (2009, p. 69) provide the following example of English/Spanish code-switching where participant B interacted in English during most of the conversation and then switched into Spanish: 

A: The picture looks so cool.

B: Which picture?

A: The one you have in your messenger.

B: Ah…Si, me gusto mucho. (Ah…Yes, I liked it a lot. 

 Vyas, Gella, Sharma, Bali, & Choudhury (2014), though they use the term code mixing to imply code switching, argue that code switching refers to the exchange of passages of speech from two different grammatical systems whereas code mixing refers to the embedding of part of language (phrases, words or morphemes) into a structure that belongs to another language Code mixing is a recurrent phenomenon in multilingual user-generated content in social media texts. Maharjan, Blair, Bethard, & Solorio (2015) view code- mixing as a process of alternation between two or more languages while speaking or writing. Cardenas-Claros and Isharyanti (2009) view code-mixing as a result of the user’s attempt to change from a language to another, below clause level, within the same conversation. Cardenas-Claros and Isharyanti (2009, p. 70) discuss insertion as a main type of code-mixing as illustrated by the following example of Indonesian/English insertion:

B: Tergantung team, terus juga tergantung event.

 (It depends on the team and on the event.)   

Muyskn (2000) uses the term code-mixing to refer to cases where a speaker uses both lexical items and grammatical features of two or more languages in just one sentence. Barman, Das, Wagner, & Foster (2014) argue that multilingual participants, in social media conversations, switch between languages; they view code switching and code mixing as referring to the same phenomenon. Gullberg & Couto (2016) hold the same view that code switching and code mixing include inserting linguistic units, such as phrases and words, of one language in the subsystem of a different language.

In this study, I argue against the claim that the two terms of code switching and code mixing can be used interchangeably. Following Muysken (2000), the use of the term code switching to refer to the general process of mixing is avoided. Muysken (2000, p. 4) rejects the use of the term code-switching as it is less neutral in two perspectives: first, code-switching is closer to the process of alternation, not to insertion. He argues that switching or alternation can take place between clauses and within the clause itself. Second, code switching rules the general process of code mixing out of the phenomenon of borrowing from one language or interference between two languages. So, switching and alternation have to be embedded under the umbrella of code mixing. Hence, the general frame of the theory of code mixing that will be followed in this study can be summarized as follows:

1. Code mixing involves alternation and insertion
2. Alternation and switching are the same
3. Switching is an alternational type of mixing
4. What is encountered in social media texts is code mixing rather than code switching.
5. Alternation is taken to be “a clear example of code-switching which takes place between utterances in a turn or between turns” (p. 5).

1. Theoretical Framework of Code- Mixing

Having settled on referring to the phenomenon under investigation as code mixing rather than code switching, we move to the different processes that characterize code mixing and also refer to the phenomenon of Transliteration that forms a basic writing system that is frequently used by social media users in Egypt.

4.1 dominant models of code mixing

Muysken (2000, pp. 3-4) proposes the following three way classification of the types of code mixing:

(a)    Insertion: This includes inserting a material such as a lexical item or an entire constituent from a language into a structure from another language. It is viewed as a type of borrowing. It entails the insertion of a lexical item or a phrasal category from one language into a certain structure from another language (e.g., a noun or a noun phrase can be inserted). The following examples of insertion are based on Muysken (2000, pp. 4-5):

(1)   kalau dong tukan bikin dong tukan bikin

when they always make they always make

voor acht personen dek orang cuma nganga dong makan

for eight persons and then people only look they eat       

           ‘When they [cook], it is always for eight people, and then they only look at it, they eat. . .’

(Moluccan Malay/Dutch; Huwae 1992)

(2)   na’iish-crash la

lsg:pass out-crash EM PH

‘I am about to pass out. ’                                  (Navaho/English; Canfield 1980: 219)

  (3)   Yo anduve in a state of shock por dos dias.

 ‘1 walked in a state of s h o c k for two days.’   (Spanish/English; Pfaff 1979: 296)