What is meant by code-switching? Discuss the main types and functions of code-switching that have been identified.
In many societies, people use several different languages in conversations. In Montreal, Canada, it is not uncommon to hear two people speaking English and French to each other in an effortless way. Even in a country that is mostly monolingual, such as England, people must choose which language variety they will use. People from the north of England will often decide to speak in a manner that is closer to RP English when they move south. These phenomena are examples of code-switching. In this essay, I will explain what code-switching is. I will also discuss the main types and functions of code-switching.
First, it is important to establish what a code is. A code is a neutral term that is used to talk about language varieties. According to Wardhaugh, the term ‘code’ “can be used to refer to any kind of system that two or more people employ for communication” (Wardhaugh 1998: 86). It is not easy to outline the difference between a language and a dialect. The mainland Scandinavian languages, Swedish, Norwegian and Danish, are very closely related and often mutually intelligible. Norwegian has several different dialects, and some of them are extremely hard to understand for most Norwegians.
A Norwegian that speaks the dialect used in Oslo would find it easier to understand a Swede from Stockholm than a Norwegian who is from Inner Telemark (South-Eastern Norway). Yet Swedish and Norwegian are different languages, and the dialects in Oslo and Inner Telemark are varieties of Norwegian. Social and political factors seem to count more than mutual intelligibility when it comes to distinguishing languages from dialects. A language is a great part of a people’s cultural identity: a Norwegian would be insulted if one told him that what he spoke was merely a dialect of ‘Scandinavian’. For those reasons, the word ‘code’ is used instead of ‘language’ or ‘dialect’.
Why does one switch between codes? In the next part of this essay, I will explain the different theories on code-switching, starting with Ferguson’s theory of diglossia.
According to Ferguson (1959), a diglossic situation occurs in a society where two codes are used. These two codes are two distinct varieties of the same language, one being the ‘high’ variety (H) and the other one being the ‘low’ variety (L). The codes serve different functions and there is no overlapping between the two. Code-switching occurs because people must use the two varieties for different purposes – speaking only one variety is not enough. The H variety is the most prestigious one and is used in relatively formal situations, such as broadcasting the news on television and radio, delivering sermons at church and speaking in parliament. The H variety is often used in education, especially at a higher level. Textbooks are usually written in this variety, and dictionaries and grammar books in the L variety are virtually non-existent. Poetry and fine literature is more likely to be written in the H variety.
The L variety, on the other hand, has a lot less prestige. In some societies, some speakers of the L variety will even deny that they even speak it (Wardhaugh 1998: 88). The reality is, though, that all children learn the L variety at home. As Wardhaugh states: “the H variety is ‘taught’, whereas the L variety is learned” (Wardhaugh 1998: 89). The L variety is widely used in informal situations, such as chatting with friends or family, and can be heard on television in soap operas or on talk shows. In kindergarten and primary school, teachers will use the L variety for storytelling, for example. Folk literature is often written in this variety.
Societies in which diglossic situations occur are found all around the world. In Haiti, there is Standard French (H) and Haitian Creole (L). In Quï¿½bec, there is Quï¿½bï¿½cois French – or joual – (L) and ‘Standard French’ (H). In the Arabic countries, there is Classical Arabic (H) and the regional varieties (L). Attitudes to the H and L varieties are varied. In Haiti, some people do not recognise Haitian Creole as a language of the country, even though they use it every day in conversations with friends. Other people feel they can only express themselves properly in Haitian Creole (Holmes 2001).
In Quï¿½bec, people who speak joual were long regarded as less socially attractive and less intelligent by their own peers than people who spoke Standard French, as Lambert’s matched guise study showed (Wardhaugh 1998: 112; Swann, in Mesthrie 2005: 149). This study was conducted in the late 1950s; since the 1960s, Quï¿½bï¿½cois French has enjoyed a revival and has gained a lot of prestige. Quï¿½bï¿½cois French words are added to the dictionary every year, and Quï¿½bï¿½cois authors write in joual more and more. Standard French (which is spoken with a slight Quï¿½bï¿½cois accent) is often viewed as pompous, but is still used in news broadcasts and political speeches. The situation in Quï¿½bec is still diglossic, but things could be very different in fifty years from now. In Quï¿½bec and in many other diglossic societies, the L variety is used as a solidarity marker, whereas the H variety can be used to distance oneself from the people.
Fishman (1967) extended Ferguson’s definition of diglossia by stating that, in some societies, two different languages have a H / L relationship. Such is the case in Paraguay, where Spanish is the H variety and Guaranï¿½ is the L variety. This is known as ‘broad’ diglossia (Swann, in Mesthrie 2005: 41-42). Fishman is most famous for his domain theory (1972), which states that people’s code choices are habitual, and that different codes are associated with different ‘domains’. A domain is defined in terms of its situational factors (setting, topic, participants, function) and its social dimensions (solidarity scale, status scale, formality scale, affective/referential scale). It is a regular social situation. It can be, for example, ‘family’, ’employment’, or ‘education’ (Swann, in Mesthrie 2005). It is worth noting that there is a phenomenon known as ‘domain leakage’, where the use of one language in one domain ‘leaks into’ another domain.
The domain theory resembles the diglossia theory, as people choose one code in one situation, and another code in another situation. The main difference seems to be that in the diglossia theory, the code choice seems to be forced upon the people by the norms of society. The lack of textbooks in Haitian Creole force people in higher education to function in Standard French. The fact that Standard French is used in most formal situations also serves to make it more prestigious. In the domain theory, the H / L relationship between the languages may be still there, but it is less apparent. People seem to choose their codes purely out of habit, depending on the situation they are in and their relationship to the people with which they are speaking.
Wardhaugh (1998) develops this idea further. According to him, there are two types of code-switching.
Situational code-switching means that people use one code in some situations and another code in other situations. This is similar to Ferguson’s theory of diglossia, but Wardhaugh points out that in diglossic societies, people are very aware of the way they use the different codes. In the case of situational code-switching, people switch codes and they often do not notice they have done so. He also argues: “Diglossia reinforces differences, whereas code-switching tends to reduce them.” (Wardhaugh 1998: 103) People switch codes depending on the conversation’s setting and participants. In this respect, Wardhaugh’s theory is much like Fishman’s domain theory.
One important point in situational code-switching is that the topic of a conversation is not considered as being a situational factor. This is different from the domain theory. Holmes (2001) also writes about situational switching, and concurs with Wardhaugh for the most part. However, she also includes ‘topic’ as a situational factor. An example that illustrates the difference between Wardhaugh’s and Holmes’ definitions of situational code-switching is the following:
Two men from Liverpool are working in the same company in London. Most of the time, they moderate their accent to make it sound more like Received Pronunciation. Joe and Jack are discussing work matters, in RP. Jack then asks Joe if Everton won the football match last night. He does so in Scouse. Jack answers and the two talk about the match in Scouse for a few minutes. Then, the two resume talking about work and switch back to RP.
The fact that the two men decide to speak RP at work is an example of situational code-switching. Both Holmes and Wardhaugh would agree on this. Jack and Joe are in a different setting – at work in London instead of at home in Liverpool. When they change the conversation topic and switch codes, Holmes (2001) would argue that it is a case of situational code-switching. According to her, when the topic of the conversation changed, the relationship between the two men also changed. They went from ‘colleagues’ to ‘friends’. Wardhaugh (1998) would say that as soon as the topic of a conversation is changed, the code-switching becomes metaphorical, not situational.
Metaphorical code-switching is the second type of switching Wardhaugh describes. People switch from one code to another to discuss different topics. This “… adds a distinct flavour to what is said about the topic.” (Wardhaugh 1998:103) Metaphorical code-switching has affective functions. When one uses metaphorical code-switching, it is to redefine the social dimensions of the conversation. These social dimensions are the same ones that Fishman uses in his domain theory: solidarity scale, status scale, formality scale, and affective/referential scale. In Holmes (2001: 38), there is one example of metaphorical code-switching that illustrates this very well.
An English Black girl, Polly speaks both English and Patois, a type of Jamaican Creole. One day, her teacher had offended her by criticising her use of Patois in a story she had written. She then proceeded to abuse him in Patois. This scared him, and he did not criticise her again. By using Patois, Polly distanced herself from her teacher on the solidarity scale. She was in effect saying that the two of them were not the same kind of people. What she said to him conveyed affective rather than referential meaning. The teacher understood that she was mad at him, but did not know exactly what she was saying. Polly switched codes to make a point, and to achieve a special effect.
Another example in Holmes (2001) depicts another type of metaphorical code-switching. This time, Polly is using Patois (in italics) and English in the same utterance, saying things such as “With Melanie right you have to say she speaks tri different sort of language when she wants to.” (Holmes 2001: 39) Holmes (2001) and Wardhaugh (1998) also call this kind of metaphorical code-switching ‘code-mixing’. It occurs when people switch back and forth between codes within a single utterance. The codes are not mixed in a random way: by switching codes in the appropriate places, the speaker can transmit affective meaning in addition to referential meaning. It can be used as a rhetorical tool, and can be very effective in enriching the conversation. Bilinguals will often switch codes to mark their solidarity with one of the two groups to which they belong, sometimes for personal gain.
For example, it is very common for politicians in Quï¿½bec to switch back and forth between Standard French and joual during a speech, especially while they are campaigning. They hope that showing their solidarity with the ‘average Quï¿½bï¿½cois’ will earn them a few votes. Attitudes to code-mixing differ. For bilinguals, it is a great way of communicating, as they can use both languages to convey the exact meaning of what they are thinking. It is also a way of expressing one’s own identity as a bilingual person. Monolinguals will sometimes refer to code-mixing with pejorative terms, such as Spanglish (Spanish and English in the USA) and Svorsk (Swedish and Norwegian in Norway). Code-mixing is not to be confused with lexical borrowing, where one does not know a word in one language and must use its equivalent in another language. People who mix codes are usually very competent, if not fluent, in both codes they are using.
The last theory of code-switching I will be writing about is Myers-Scotton’s (1993) markedness model of conversational code-switching. Myers-Scotton classified the choices people make when deciding which code to use as either ‘marked’ or ‘unmarked’. An unmarked choice is the choice which would be expected in a given situation. For example, a Haitian that is being interviewed for a job will most likely decide to speak Standard French. A marked choice is when someone makes the decision to use a code which is not usually used in a given situation. If a Swiss German speaks Swiss German instead of English or Standard German to tourists, he is making a marked choice.
Myers-Scotton did many studies in Africa, especially in Kenya and Nigeria. After analysing her data, she came to the conclusion that there were four distinct code-switching patterns. They are: “code-switching as a series of unmarked choices between different languages; code-switching itself as an unmarked choice; code-switching as a marked choice; and code-switching as an exploratory choice.” (Swann, in Mesthrie 2005: 167) The first pattern she describes can include both Wardhaugh’s (1998) situational and metaphorical code-switching. When the setting, participants, function or topic of a conversation changes, the people make an unmarked choice to change codes. Myers-Scotton pushes the concept of code-switching further and argues that in some multilingual communities (which there are a lot of in Africa) code-switching itself is expected. It is perfectly normal to switch between codes – in fact, it would be abnormal not to do so. Using both codes together makes the conversations more meaningful, and gives a sense of community to the people who switch codes in the same way.
Myers-Scotton’s third pattern can also include Wardhaugh’s metaphorical code-switching. The example of Polly abusing her teacher in Patois is a case of code-switching as an marked choice. The teacher did not expect to hear Patois, and this made Polly’s insults more effective. Finally, Myers-Scotton states that code-mixing can be used in an exploratory way. When two Norwegians meet each other for the first time, they will probably decide to speak in a manner that resembles the Oslo dialect, or Bokmï¿½l. They might use some words and phrases from their own dialect, to let the other person know where they are from. This could result in the two Norwegians realising that they are both from Inner Telemark, and adopting the Inner Telemark dialect for the remainder of the conversation.
In conclusion, it is difficult to ascertain how many types of code-switching there are. Theorists have different ways of viewing the problem. Ferguson (1959) speaks of diglossia, where one language is more prestigious than the other; Fishman (1972) argues that certain codes are associated to certain domains; Wardhaugh (1998) and Holmes (2001) write about situational and metaphorical code-switching; and Myers-Scotton (1993) has her markedness model of conversational code-switching. People code-switch for a variety of reasons, such as: to adapt to their environment, to mark solidarity, to distance oneself from the rest, to enrich conversations, and to express their identity.
Holmes, J. (2001) An introduction to Sociolinguistics. 2nd edn. London: Longman
Mesthrie, R. et al. (2000) Introducing Sociolinguistics. Edinburgh University Press
Wardhaugh, R. (1998) An Introduction to Sociolinguistics.3rd edn. Oxford: Blackwell