Introduction

Sociolinguistic research that acknowledges the importance of viewing language as a human problem attempts to reconcile the facts of linguistic variation with those of social identity and inequality (Hymes 1973). In this paper I present a case study of Hawaii which examines this relationship in a Pacific English creole continuum and, more specifically, calls attention to its dynamic nature.

A history of cultural diversity

Hawaii, with a population of about one million, is the only American state in which no single ethnic group is a numerical majority, and where most of the people are of Asian and Pacific rather than European or African origin (Nordyke 1977). The population of the seven inhabited islands is roughly a quarter Japanese and a quarter Caucasian. Still another quarter is racially mixed (about 16 per cent part-Hawaiian), and the remaining quarter is comprised of a number of groups, including Filipinos, Chinese, Blacks, Koreans, Hawaiians, Samoans, and other Pacific Islanders (Schmitt 1982).

Hawaii's cultural diversity is largely the result of massive labour importation, triggered by the development of sugar plantations by north Americans during the late nineteenth and early twentieth centuries. The islands were transformed from a Hawaiian kingdom with a subsistence agricultural economy into a plantation economy in which sugar became ‘king’ (Fuchs 1961; Kent 1974, 1983). Political incorporation into the US began with the overthrow of the native Hawaiian monarchy in 1893 and annexation by the US in 1898, and was completed with statehood in 1959.

Introduction

Descriptions of the language situation in Guyana have tended to treat it as being made up of two maximally distinct language varieties, standard Guyanese English, on one hand, and Guyanese Creole, on the other, linked by a series of intermediate varieties (Allsopp 1958; Bickerton 1975). This kind of description, while capturing a particular aspect of the language situation, misses others. One such is the diglossic nature of the Guyana language situation. Varieties approximating to standard Guyanese English tend to be employed in the more public and formal situations of interaction. This contrasts with the more private and informal situations in which varieties approximating to Guyanese Creole tend to be employed. Another aspect not captured by an analysis which treats the Guyana situation as one involving a Creole-to-English continuum, is that speakers normally control more than one language variety. Such persons code-switch between the language varieties within their linguistic repertoires, depending on the social factors present within the speech event and the social functions associated with each of these varieties.

Standard Guyanese English operates as the sole official language of the country. What this means is that those speakers who do not have this variety within their repertoires or whose competence in the variety is limited, are effectively excluded from all forms of official communication. In these circumstances, it is clear that an alternative official language policy is necessary if one is to avoid the continued exclusion of non-English speakers from official communication.

Previous studies of the sex variable

Few linguistic studies have dealt systematically with sex differentiation in language use, and none in the context of Creole societies, even though the latter provide ideal terrain for this type of research because of recent changes in gender roles. The following claims have been made in respect to men and women's use of language, and their roles in the diffusion of linguistic change.

(1) Women are more likely to use standard or prestige forms, and to upgrade their speech patterns in formal situations than men of the same age, social class and education level. This appears to hold true for all social classes, and is interpreted as a strategy to compensate for the general subordination to which women are subjected in social structures (see, however, Milroy, this volume).

(2) Men's speech patterns are more directly related to their socioeconomic status. Nonstandard speech has positive connotations of masculinity and signals male solidarity, mostly for working classes, but for others as well. Thus, in formal situations working class men do not deviate from low status variants as much as women do. But lower middle class men are so conscious of the social value of standard speech that they often hypercorrect, a sign of linguistic insecurity, according to Labov (1972).

(3) Women are innovators in linguistic change, but only in the direction of standard speech, whereas men lead in the use of new vernacular forms.

Background

In the East African region of Kenya and Tanzania, a complex multilingual situation exists. It comprises nearly 140 indigenous languages, over 100 in Tanzania and about 40 in Kenya, spoken by over 40 million people. These languages belong to four major African language families: Bantu, Nilo-Saharan, Cushitic and Khoisan. The Khoisan family is represented by tiny remnants spoken by a few thousand speakers in central Tanzania. In Tanzania more than 90 per cent of the languages and speakers belong to the Bantu group, a linguistic fact that has made the acceptance of Kiswahili as the national and official language there relatively easier than in Kenya. In Kenya over 75 per cent of the population speak languages belonging to the Bantu family, about 20 per cent speak Nilo-Saharan languages, which include Luo, Maa and Kalenjin, and the rest speak Cushitic languages which include Somali, Oromo and Rendille. These ethnic languages serve as languages of group identity at the subnational level, as Schmied observes in his paper in this volume.

This paper focuses on English in Kenya and Tanzania, which are the subjects of the case study papers in this volume. For discussion of English in other East African countries, see Hancock and Angogo (1984).

Background

Demography

Nigerian Pidgin (hereafter NP) is spoken as a second language in all parts of Nigeria and as a first language by a growing number of people in the southern part of the country (and perhaps in the urban centres of the North as well). Although no official figures are available, a conservative estimate of the number of speakers of NP at present would fall somewhere between 30 and 35 million, a number which can be expected to rise significantly in the near future, due to the rapid spread of NP among young adults and children, who together constitute well over 50 percent of the national population.

History

About 400 languages are spoken in Nigeria. Historical researchers are uncovering more and more evidence indicating the existence of a complex network of very vibrant mercantile (and often highly urbanised) cultures in all parts of the country. This linguistic diversity, alongside the need for interethnic communication in societies where speakers of different languages are in constant contact with one another due to geographic proximity, intermarriage, trade, travel, and the growth of cities and towns, makes it very likely that pidginised languages, learned as a common second language by people of different linguistic backgrounds who need to communicate with one another, have existed in Nigeria since ancient times.

Introduction

This is a study of two ways in which adult males in New Zealand become members of groups. It is, in other words, a study in solidarity. The models which will be put forward both achieve the same ends: induction into a group through the acquisition of certain routines, specifically certain linguistic routines. These routines are most clearly seen in the acquisition and use of a set of routine formulae (Coulmas 1979, 1982) which exemplify the strategies the two groups use to make men members of their groups. The two strategies are quite different. In one group, formulae are used to save face; specifically, the face of players in a recreational volleyball squad. In the other group, sexual humiliation is used as a means of creating group solidarity through the loss of face the individuals who belong to the group suffer.

Preliminary remarks

If I use the New Zealand English formula Gidday with a wink and a characteristic quick southeast to northwest movement of the head, the conditions of use for such a greeting are relatively specific; it is an informal greeting to one with whom one is not intimate. It is not middle class but characteristically working class, and/or rural. It is often used when no response is expected.

Introduction

Background

It is generally acknowledged that the duration of the contact between English and West African languages on a sustained basis is essentially coterminous with the colonial era, namely circa 1880–1960 (cf. Kirk-Greene 1971; Angogo and Hancock 1980; Bokamba 1984a). During much of that period and up to the late 1960s English was, as Spencer (1971) notes, an institutional link rather than a community language: It was primarily learned in school and used in educational, administrative, and commercial institutions.

While this institutional character remains largely true today, the development of Sierra Leonean Krio (c. 1800) and that of other more recent Pidgin English varieties (e.g., Cameroonian, Liberian, Nigerian), combined with the phenomenal expansion of pre-university education into rural areas in the 1970s, have significantly extended the domains of English in the region. In particular, established West African Pidgin Englishes have become the link languages par excellence for many communities in the region (cf. Shnukal and Marchese 1983; Jones 1987; Faraclas 1984 and this volume). ‘Educated West African English’ or the variety of English spoken by educated West Africans, has itself assumed an increasingly preponderant role as the intra-African and inter-continental language of communication for the sub-region. In view of these functional roles, it is reasonable to assume that the increased contact between West African languages and English in the past century or so has had important structural effects on the latter.

Variation in society – the background and its impact

Differing perceptions of the language, different emphases

In all of the several countries of South Asia, the English language, sustained by English-using élites of essentially the same kind, has continued, even after Independence, to play comparable crucial roles. In spite of this, and in spite of the further relevant fact that these countries share a great deal in the way of history, culture, interests and so on, and constitute a natural ‘linguistic area’ (D'souza 1987; Masica 1976) within what, moreover, is a natural geographical area, the attention paid within them to the English language has varied greatly, both in depth of coverage as well as in the range of issues looked at. India dominates the field in both respects, with some contributions coming from Sri Lanka and considerably fewer from Bangladesh and Pakistan.

Evidently, the unique developmental processes within each of these countries as they pursued their common task of post-colonial reconstruction have defined for them very different perceptions of the language and of their relationship to it, perceptions that do not lead them all alike to the kind of preoccupation with it that makes such obvious sense to the increasing band of scholars studying New Varieties of English (NVEs).

Introduction

This paper reports on research involving the fine-grained qualitative analysis of interactions between Afrikaans and English speakers in South Africa, in the medium of English. The goal of this research is to see whether the explanation of the sources and consequences of miscommunication between Zulu and English speakers offered in Chick (1985) holds for Afrikaans and English speakers as well. The essence of this explanation is that a mis-match of culturally-preferred interactional styles contributes to asynchrony (see Erickson 1975, 1976, 1978), in the context of which participants misinterpret and misevaluate one another, and that repeated miscommunication of this sort generates negative cultural stereotypes. Asynchrony is the antithesis of conversational synchrony which, Erickson (1975) explains, is the rhythmic patterning of conversationalists' coordinated behaviour that enables them to judge the occurrence in real time of significant ‘next moments’. This information they need in order to accomplish conversational inferencing. As a great number of studies have shown (e.g. Gumperz 1982a, 1982b; Pride 1985), intercultural communication is frequently characterised by asynchrony in which the participants look, sound and feel clumsy, and often miss one another's signals because they occur at unexpected moments. In this paper I present evidence which suggests that, indeed, a mismatch of interactional styles is partly responsible for the asynchrony in the encounters analysed, and I suggest what some of the distinctive features of these putative styles are.

Introduction

The term Southern Africa is in fact primarily a political rather than a geographical description of the countries south of the Equator which have a historical link with white South Africa. Traditionally, Southern Africa has included English-speaking countries (South Africa, Lesotho, Swaziland, Botswana, Zimbabwe, Zambia and Malawi) which are former colonies of Britain, and the Portuguese-speaking countries of Angola and Mozambique, which are former colonies of Portugal. Namibia, which was a mandated territory of South Africa, is also English-speaking.

Needless to say, the linguistic geography of Southern Africa is such a complex phenomenon that it would take a paper of its own for it to be treated with justice. Despite that complexity, however, Southern Africa – with the exception of Angola and Mozambique – enjoys language unity through English. This point is important in understanding the context in which the case study papers in this section are set. Apart from perhaps Lesotho, Swaziland and Botswana, all the countries in the region are multi-ethnic and multilingual. In each country, there are serious clashes of interest when the question of local languages is at issue. Yet, for some reason, solace is often found in the fact that English is there to act as a common language.

Background

Philippine English, as one of the New Englishes, has been carefully studied in terms of its structural features (Llamzon 1969; Alberca 1978; Gonzalez and Alberca 1978). Llamzon studied oral and written varieties, while Alberca concentrated on oral and written varieties in the mass media, using data from radio and TV and from print (newspapers and weekly periodicals). Llamzon compared different styles of oral English (formal, i.e., the equivalent of a formal reading style; and informal); so did Alberca in drawing a distinction between newscasts and talk-shows on TV as oral varieties, and between different journalistic styles as written varieties. Newscasts were considered to be formal oral varieties (equivalent to careful reading style) and talk-shows to be informal oral varieties. Of the written varieties, news stories were considered to be formal varieties, whereas feature articles and columns were considered to be informal varieties.

In these seminal works on style, some attempt was made at frequency counts of the incidence of certain features, to demonstrate stylistic differences empirically. Subsequently, Gonzalez (1982) reported a more careful study of style, pointing out qualitative differences and using examples from different types of journalistic writing.

Introduction

In the Indian urban context the use of English is becoming increasingly widespread, with English acquiring more functions now than ever before. This was not envisaged, however, by the framers of the Constitution before independence. The Indian Constitution recognises 15 national languages, with Hindi as the official language and English as the associate official language, with a directive that English was to be replaced by Hindi in a period of 15 years. However, this did not happen, for various social and political reasons, and English has flourished even more after independence.

The national impact of Hindi, on the other hand, has not been able to equal that of English and has led to English–Hindi rivalry, with each language continuing to compete for recognition as a pan-Indian language (Kachru 1979). Here, English has certain advantages over Hindi. Since it is not the language of any major group, it does not threaten any group's ethnic identity and hence is politically more acceptable. It is the main language of education, administration, the mass media, science and technology, and it has provided appropriate and stable registers in these areas. Hindi, despite government support, is still in the process of standardising and codifying the specific registers. Therefore, it cannot provide much professional and linguistic mobility to its users. Any language which aspires to replace English at the national level needs to acquire the functional load of English (Kachru 1979).

Introduction

The English language is now completing 200 years of continuous usage in Australia. In that time it has supplanted the original languages of the continent, and recruited most descendants of non-English speaking immigrants, so that today it is the overwhelmingly dominant tongue throughout Australia. Several features of the Australian situation yield a unique insight on the development and diversification of English: its geographic isolation, its social origins as a penal colony, and its recent wave of non-English speaking immigrants. Australian English (AE) has experienced language and dialect contact, but for most of the last two centuries Australia may have had the highest proportion of monolingual English speakers of any country in the world, aside from England itself. This bicentennial survey will hopefully serve to illuminate the Australian branch of ‘English around the world’ – in this case, about as far around the world as it could go.

The status of English in Australia: the national language

The status of the English language in Australia is today, and has been since British colonisation, that of the national language. It overwhelmingly dominates the linguistic landscape, both demographically and functionally. This is not to say that Australia is a monolingual country; on the contrary, a large number of languages are spoken within its borders.

Introduction

English is the national language of Fiji, used in government, business, and education. It is also the lingua franca among the many ethnic groups of the country: indigenous Fijians, Fiji Indians, Europeans, Chinese, Rotumans, other Pacific Islanders, and part-Europeans (as people of mixed race are called in Fiji).

A distinctive local variety of English is characterised by certain non-standard linguistic features. This ‘Fiji English’, like Singapore English (Platt 1975; 1978), is a linguistic continuum with variation according to the frequency of occurrence of these nonstandard features, here called ‘markers’. Speech at the lower end of the continuum contains the highest frequency of these markers, and speech at the upper end contains the lowest frequency. Speakers of Fiji English may differ both according to where on the continuum their speech is located and the range of the continuum over which they have competence.

The linguistic markers themselves differ according to their range in the continuum. Lexical markers seem to have the widest range, with some commonly used items from Fijian and Fiji Hindi (such as tanoa ‘bowl used for making kava’ and roti ‘Indian flat bread’) being found from one end of the continuum to the other. Certain phonological markers range from the lower through the middle of the continuum. These include the absence of certain consonant clusters and the presence of intonation patterns similar to those of Fijian – for example, questions beginning at a higher pitch than in standard English and ending with falling rather than rising intonation.

Introduction

Perhaps because of the interest and controversy which has been associated with the link between the Irish language and Hiberno-English, purely synchronic studies in which the direct influence of Irish can be effectively discounted are very rare. Classic descriptive studies such as those of Henry (1957, 1958) concentrate on rural dialects, while structural analyses such as those of Bliss (1972), Henry (1960–61), and others have tended also to look at Hiberno-English only in comparison with Irish. Although some early writers (e.g., Hayden and Hartog 1909) have also given space to a consideration of the role of earlier English forms in the development of Hiberno-English, these writers too were concerned only with rural varieties in which both Irish influence and dialectal ‘conservatism’ might be strongest. Recent studies by Harris (1983, 1984) have examined aspects of Hiberno-English more rigorously, while still using a contrast between this variety and ‘standard English’ or Irish as a point of departure.

We examine here a well-known syntactic feature of Hiberno-English, the use of after as a perfective marker, whose origins, though still not well documented, may lie in the large-scale contact between English and Irish speakers in the seventeenth century (see Bliss 1979 and Kallen 1986). What is of concern, though, is the set of factors which governs the contemporary use of after in Dublin, a large urban centre in which the Irish language can effectively be ruled out as a direct influence on the modern use of English.

The following analysis is a feasibility study for a research project on ‘English in East Africa: an independent African means of communication?’, which is the contribution of English Linguistics to the Special Research Programme (SFB) on ‘Identity in Africa’ carried out at the University of Bayreuth. I will therefore briefly explain the background to this study in relation to the notion of identity, before explaining the methodology and problems of the analysis and the results of this microsociolinguistic study on the co-variation of pronunciation, social and contextual variables.

English and Kenyan identities

The basic assumption of this research project is that attempts to define specific African national identities must rest on various distinct concepts of cultural identity as well as on various overlapping regional identities. In this overall framework language is seen as a means of expressing, together with a message, a personal and/or a group identity, which is chosen by the speaker and interpreted by the hearer. In modern ethno-psychology personal identity is often seen as the sum of heterogeneous identities. Thus if a market woman in a market in Nyeri responds in English to a white man's question in Swahili, she expresses part of her identity, just as when she talks in Kikuyu to her market neighbours. Similarly, a Luo hotel manager may talk in basilectal English to his Kikuyu cleaners and in acrolectal English to his foreign guests.