4. Is it possible to have a society in which everyone is completely bilingual in the same two languages and there is no diglossia? How stable would such a situation be?
5. Some communities regard bilingualism as a serious threat; it has even been referred to as a ‘Trojan horse,’ initially attractive but ultimately fatal. Why might this be so? (Consider the experience of migration and also the sorry state of many minority languages in the world.)
if the family is Christian, but in Hokkien if Buddhist or Taoist. The language of government employment will be formal Singapore English but some Man- darin will be used from time to time; however, shopping will be carried on in Hokkien, informal Singapore English, and the ‘bazaar’ variety of Malay used throughout the region. (See Platt and Platt, 1975, pp. 91– 4, for a fuller discus- sion.) The linguistic situation in Singapore offers those who live there a wide choice among languages, with the actual choice made on a particular occasion determined by the kinds of factors just mentioned. (It may even be possible to characterize the total linguistic situation in Singapore as a complicated diglossic one if we accept Fishman’s view of diglossia.)
We may also ask what happens when people from a multilingual society, people who are themselves multilingual, meet in a ‘foreign’ setting: what language or languages do they use? Tanner (1967) reports on the linguistic usage of a small group of Indonesian graduate students and their families living in the United States. Among them these students knew nine different languages, with nearly everyone knowing Indonesian (Bahasa Indonesia), Javanese, Dutch, and English.
They tended to discuss their academic work in English but used Indonesian for most other common activities. Unlike Javanese, ‘Indonesian . . . , whether the official or the daily variety, is regarded as a neutral, democratic language. A speaker of Indonesian need not commit himself to any particular social identity, nor need he impute one to those with whom he converses’ (p. 134). The students also used Dutch, but mainly as a resource, e.g., for vocabulary, or because of the place it necessarily held in certain fields of study, e.g., Indonesian studies.
Local languages like Javanese tended to be used only with intimates when fine shades of respect or distance were necessary, particularly when in the presence of important older people. Tanner’s findings conform to an earlier prediction made by Geertz (1960, p. 259): ‘Indonesian appeals to those whose sense of political nationality as Indonesians rather than as Javanese is most developed, to those who are interested in the cultural products of the new Indonesia’s mass media . . . and those who wish to take leadership positions in government and business.’ He adds that, ‘although the use of Indonesian for everyday con- versation is still mostly confined to the more sophisticated urbanites, and its use suggests something of an air of “public speaking” for most Javanese, it is rapidly becoming more and more an integral part of their daily cultural life and will become even more so as the present generation of school children grows to adulthood.’ Javanese will continue to be used ‘in certain special contexts and for certain special purposes.’
Situations such as those just described are not uncommon. In Kenya, local languages, Swahili, and English all find use and choosing the right language to use on a particular occasion can be quite a delicate matter. Whiteley (1984, pp. 74–5) describes the kind of situation that can occur between a member of the public and members of the government bureaucracy:
A man wishing to see a government officer about renewing a licence may state his request to the girl typist in Swahili as a suitably neutral language if he does not know her. To start off in English would be unfortunate if she did not know it, and on her goodwill depends his gaining access to authority reasonably quickly.
She may reply in Swahili, if she knows it as well as he does and wishes to be co-operative; or in English, if she is busy and not anxious to be disturbed; or in the local language, if she recognises him and wishes to reduce the level of formality.
If he, in return, knows little English, he may be put off at her use of it and decide to come back later; or, if he knows it well, he may demonstrate his importance by insisting on an early interview and gain his objective at the expense of the typist’s goodwill. The interview with the officer may well follow a similar pattern, being shaped, on the one hand, by the total repertoire mutually available, and on the other by their respective positions in relation to the issue involved.
Trudgill (1995, pp. 108–10) describes a situation in Kampala, the capital of Uganda, which is similar in many respects.
The actual choice of code in a setting clearly marked as bilingual can be a difficult task. As Heller (1982) has observed, language plays a symbolic role in our lives, and when there is a choice of languages the actual choice may be very important, particularly when there is a concurrent shift in the relationship between the languages, as is occurring in Montreal between English and French.
In such circumstances, as Heller observes, ‘negotiation in conversation is a play- ing out of a negotiation for position in the community at large’ (p. 109). Heller studied the uses of the two languages in a Montreal hospital during the summer of 1977. Which language was used varied as circumstances changed. What is particularly interesting is that the pattern that has evolved of asking which language someone wishes to use in a public service encounter (‘English or French, Anglais ou Français?’) is not very effective. The reason is that too many other factors are involved to make the choice that simple (p. 118):
the negotiation of language has to do with judgments of personal treatment, that is, how one expects to be treated in such a situation. But such judgments are dependent upon social knowledge, knowledge about group relations and bound- aries and ways of signalling them, and knowledge about other social differences, e.g., status differences.
. . . This negotiation itself serves to redefine the situations in the light of ongoing social and political change. In the absence of norms, we work at creating new ones.
The conventionalization of the negotiating strategies appears to be a way of nor- malizing relationships, of encoding social information necessary to know how to speak to someone (and which language to speak is but one aspect of this).
Most of Heller’s examples show how the conventionalization to which she refers – i.e., asking the other which language is preferred – often does not work very well in practice. Social and political relationships are too complicated to be resolved by such a simple linguistic choice.
We can see still other examples of how a speaker may deliberately choose to use a specific language to assert some kind of ‘right.’ A bilingual (in French and English) French Canadian may insist on using French to an official of the fed- eral government outside Quebec, a bilingual (Catalan and Spanish) resident of Barcelona may insist on using Catalan, a bilingual (Welsh and English) resident of Wales may insist on using Welsh, and so on. In these cases code choice becomes a form of political expression, a move either to resist some other power, or to gain power, or to express solidarity.
We are therefore turning to the issue of what brings a speaker to choose variety X of a language A rather than variety Y, or even language A rather than language B. What might cause a speaker to switch from variety X to variety Y or from language A to language B? A number of answers have been suggested, including solidarity, accommodation to listeners, choice of topic, and perceived social and cultural distance. In other words, the motivation of the speaker is an important consideration in the choice. Moreover, such motivation need not be at all conscious, for apparently many speakers are not aware that they have used one particular variety of a language rather than another or sometimes even that they have switched languages either between or within utterances.
Equating in this instance code with language, we can describe two kinds of code-switching: situational and metaphorical. Situational code-switching occurs when the languages used change according to the situations in which the con- versants find themselves: they speak one language in one situation and another in a different one. No topic change is involved. When a change of topic requires a change in the language used we have metaphorical code-switching. The inter- esting point here is that some topics may be discussed in either code, but the choice of code adds a distinct flavor to what is said about the topic. The choice encodes certain social values. Linguists have found it very difficult to explain precisely when, linguistically and socially, code-switching occurs, i.e., what all the constraints are. However, there is broad agreement about the general principles that are involved.
Instances of situational code-switching are usually fairly easy to classify for what they are. What we observe is that one variety is used in a certain set of situations and another in an entirely different set. However, the changeover from one to the other may be instantaneous. Sometimes the situations are so socially prescribed that they can even be taught, e.g., those associated with ceremonial or religious functions. Others may be more subtly determined but speakers readily observe the norms. This kind of code-switching differs from diglossia. In diglossic communities the situation also controls the choice of variety but the choice is much more rigidly defined by the particular activity that is involved and by the relationship between the participants. Diglossia reinforces differ- ences, whereas code-switching tends to reduce them. In diglossia too people are quite aware that they have switched from H to L or L to H. Code-switching, on the other hand, is often quite subconscious: people may not be aware that they have switched or be able to report, following a conversation, which code they used for a particular topic.
As the term itself suggests, metaphorical code-switching has an affective dimen- sion to it: you change the code as you redefine the situation – formal to informal, official to personal, serious to humorous, and politeness to solidarity. In a number of places Gumperz (particularly 1982a) cites examples of metaphorical code- switching from three sets of languages (Hindi and English, Slovenian and German, and Spanish and English) to show how speakers employ particular languages to convey information that goes beyond their actual words, especially to define social situations. What happens in each case is that one language expresses a we-type solidarity among participants, and is therefore deemed suitable for in- group and informal activities, whereas the other language is they-oriented and
is considered appropriate to out-group and more formal relationships, particularly of an impersonal kind. The we–they distinction is by no means absolute, so fine- shading is possible in switching; i.e., certain topics may be discussed in either code, and the particular choice made itself helps to define the social situation or to shift that definition, as the case may be. Woolard (1989) provides a good example of this kind of shift from Barcelona. Catalans use Catalan only to each other; they use Castilian to non-Catalans and they will even switch to Castilian if they become aware that the other person is speaking Catalan with a Castilian accent. Catalan is only for Catalans. It also never happens that one party speaks Catalan and the other Castilian even though such a conversation is theoretically possible since all Catalans are bilingual.
A particular group of people may employ different kinds of code-switching for different purposes. In their account of how the population of Hemnesberget, a small Norwegian town of 1,300 inhabitants located close to the Arctic Circle, use a local northern dialect of Norwegian, Ranamål, and one of the standard varieties, Bokmål, Blom and Gumperz (1972) show how both situational and metaphorical code-switching are used. Situational switching occurs when a teacher gives some kind of formal lecture in Bokmål but the discussion that follows is in Ranamål. Metaphorical switching is a more complicated phenomenon. One type tends to occur when government officials and local citizens transact busi- ness together. Although the variety generally used in such circumstances is Bokmål, it is not unusual for both parties to use the occasional Ranamål expression for special effect. Blom and Gumperz also discovered that, while most locals thought they used Ranamål exclusively in casual conversations and reserved Bokmål for use in school and church and on formal occasions, such was not the case. Tape recordings revealed switches to Bokmål to achieve certain effects. Moreover, the participants were not conscious of these switches, and even after such switching to Bokmål was pointed out to them and they declared they would not do it again, they continued to do so, as further tapings revealed.
Such persistence suggests that metaphorical code-switching in such situations is deeply ingrained and that it serves subtle but strong functions. Not only do natives of Hemnesberget find the existence of two varieties of Norwegian useful to them in demonstrating we-ness (Ranamål) and they-ness (Bokmål), but they also are able to employ both varieties together in such ways as to express fine gradations of feeling for others, involvement with the topic, politeness to strangers, and deference to officials.
Gumperz (1982a, pp. 44–58) also reports on an interesting situation in the Gail Valley of Austria near the borders of the former Yugoslavia and Italy, which shows how two languages (Slovenian and German) are used, what kinds of code-switching occur, and what changes appear to be in progress. Slovenian has long been spoken in the valley, but the valley is part of Austria so German is the prestige language. The Slovenians of the village are bilingual. However, the bilingual population tends to use Slovenian and German for quite different purposes. Gumperz explains (p. 47) that there are ‘three speech varieties: a formal style of standard Austrian German, the regional German dialect, and the village variety of Slovenian. To interact in accordance with the village communicative conventions, a speaker must control all three of these.’ Children are encouraged
to learn Standard German in school so as to maximize their opportunities for employment. It is regarded ‘as impolite or even crude to use Slovenian in the presence of German-speaking outsiders, be they foreigners or monolingual Ger- mans from the region.’ Slovenian itself is reserved mainly for use in the family and informal local friendship circles. However, ‘there is by no means a one-to-one relationship between extralinguistic context and language use.’
Gumperz’s analysis of code-switching in the community reveals that the situ- ation is quite complex because of the number of possibilities that are available, with the ‘right’ choice highly dependent on the social context and intent of the speaker. A further complication is that the need to maintain the in-group (Slovenian) versus out-group (German) difference has weakened considerably in recent years, mainly as a result of economic development brought about by increased tourism in the area. The use of Slovenian is decreasing. The young use less and less Slovenian and code-switch much less than their elders, preferring to use German almost exclusively among themselves. While older bilinguals do not speak German exactly like monolingual Germans, many young bilinguals do. There is therefore a shift in progress in the functions of the two languages, one which also affects code-switching since it involves the values attached to the languages and therefore a change in norms.
Code-switching is not a uniform phenomenon; i.e., the norms vary from group to group, even within what might be regarded as a single community. Gumperz (1982a, p. 68) has pointed out, for example, that:
In a relatively small Puerto Rican neighborhood in New Jersey, some members freely used code-switching styles and extreme forms of borrowing both in everyday casual talk and in more formal gatherings. Other local residents were careful to speak only Spanish with a minimum of loans on formal occasions, reserving code- switching styles for informal talk. Others again spoke mainly English, using Span- ish or code-switching styles only with small children or with neighbors.
He adds that ‘each communicating subgroup tends to establish its own con- ventions with respect to both borrowing and code-switching,’ and that factors such as region of origin, local residence, social class, and occupational niche are involved in defining the norms. Moreover, bilinguals in such communities are aware not only of the norms that apply within their own sub-groups but also of some of the norms that other bilinguals observe (p. 69):
Residents of such large Spanish–English-speaking communities as San Francisco or New York, which include immigrants from many Latin American regions, in fact claim that they can tell much about a person’s family background and politics from the way that person code-switches and uses borrowings. What the outsider sees as almost unpredictable variation becomes a communicative resource for members.
Since bilingual usage rules must be learned by living in a group, ability to speak appropriately is a strong indication of shared background assumptions. Bilinguals, in fact, ordinarily do not use code-switching styles in their contact with other bilinguals before they know something about the listener’s background and atti- tudes. To do otherwise would be to risk serious misunderstanding.
The following example from Chinua Achebe’s novel No Longer at Ease (1975b, pp. 66–7) shows how code-switching can be used within a situation involving a creole to imply wrong-doing. Obi is a British-educated Nigerian and Joseph a minister in the government. Clara is a friend of both.
When Obi got back to Joseph’s room it was nearly eleven o’clock. Joseph was still up. In fact he had been waiting all the afternoon to complete the discussion they had suspended last night.
‘How is Clara?’ he asked . . .
‘She is fine,’ he said. ‘Your Nigerian police are very cheeky, you know.’
‘They are useless,’ said Joseph, not wanting to discuss the police.
‘I asked the driver to take us to the Victoria Beach Road. When we got there it was so cold that Clara refused to leave her seat. So we stayed at the back of the car, talking.’
‘Where was the driver?’ asked Joseph.
‘He walked a little distance away to gaze at the lighthouse. Anyway, we were not there ten minutes before a police car drew up beside us and one of them flashed his torch. He said: “Good evening, sir.” I said: “Good evening.” Then he said: “Is she your wife?” I remained very cool and said: “No.” Then he said: “Where you pick am?” I couldn’t stand that, so I blew up. Clara told me in Ibo to call the driver and go away. The policeman immediately changed. He was Ibo, you see. He said he didn’t know we were Ibos. He said many people these days were fond of taking other men’s wives to the beach. Just think of that. “Where you pick am?” ’
‘What did you do after that?’
‘We came away. We couldn’t possibly stay after that.’
In addition to the police officer’s switch from the standard Is she your wife? to the creole Where you pick am?, we have the reported shift to the use of Ibo, which has the immediate effect of changing the police officer’s attitude because he too speaks Ibo. Further on in the novel (p. 100), Achebe describes the speech of still another character in the following way:
Whether Christopher spoke good or ‘broken’ English depended on what he was saying, where he was saying it, to whom and how he wanted to say it. Of course that was to some extent true of most educated people, especially on Saturday nights.
But Christopher was rather outstanding in thus coming to terms with a double heritage.
(The ‘double heritage’ referred to is the local combination of African and British influences.)
On p. 85, I pointed out a kind of code-switching that occurs among certain young people of Caribbean descent in England. These youngsters speak the local variety of English natively but also have a creole-based variety of their own which they switch to for purposes of solidarity. What is interesting is that on certain occasions they will allow youths of their acquaintance from other ethnic groups to switch to that variety too, as they show their willingness to extend some kind of solidarity to them. Rampton (1995) discusses some of the con- sequences of this ‘crossing’ behavior of members of these other groups, i.e., their deliberate adoption of stigmatized linguistic forms.