Only a few centuries ago, the English language consisted of a collection of dialects spoken mainly by monolinguals and only within the shores of a small island. Now it includes such typologically distinct varieties as pidgins and Creoles, ‘new’ Englishes, and a range of differing standard and non-standard varieties that are spoken on a regular basis in more than 60 different countries around the world (Crystal 1985). English is also, of course, the main language used for communication at an international level.

Such diversity of form and function within what is nevertheless still thought of as a single language offers a unique opportunity to analyse and document the linguistic variation and change that is occurring on a far greater scale – as far as we know – than has ever happened in the world's linguistic history. It also allows us to investigate the relationship between language and the community in which it is used from a broader perspective than is usual. Academic disciplines tend to fragment into separate specialist fields: dialectology, bilingualism, pidgin and creole studies, and sociolinguistics, for example, are often treated as if they are relatively self-contained areas of study. All four of these fields, however, share the problems of describing and explaining linguistic variation, though the nature of the variation may differ; and all four fields investigate essentially the same social and educational issues arising from community attitudes that assign high prestige to some languages, or varieties of a language, and low prestige to others (see also Rickford 1988).

English, Maori and minority languages

New Zealand is one of the world's most monolingual nations. English is the first language of 95 per cent of the 3.4 million population – and the only language of 90 per cent, most of whom are of British descent. English dominates all public domains – media, education, government, law – despite efforts to increase the use of the language of the indigenous Maori people (cf. Benton 1987).

Maori are the largest minority group, constituting about 12 per cent of the population. Maori (a Polynesian language) has gained increasing official recognition, although the change appears too late to reverse a century of neglect and opposition which has brought it to the edge of extinction as a language of everyday interaction. Now less than 25 per cent of Maori people (and still fewer younger Maori) can speak their language fluently (Benton 1979a). Even in isolated rural areas it has virtually been replaced by English (Benton 1979b). Maori is therefore following the typical pattern of an indigenous tongue overwhelmed by an imperial language, and has reached a point from which few languages have recovered. Nevertheless, language revival efforts are underway, with bilingual schooling the most promising initiative (Hirsh 1987). While Maori may eventually be lost as an everyday tongue, it may survive as the language of formal speech events in Maori culture.

Plural marking in Liberian English

Liberian English, the range of English from pidgin to standard spoken in Liberia, is characterised by vast variation in the marking of semantically plural nouns. Some speakers never mark the plural, while others mark it most of the time. Of the 21 speakers examined in the study that forms the basis for the present article, three of them mark the plural 2 per cent of the time or less, while two others mark it 70 per cent or more. Speakers also vary as to how they mark the plural, whether by a postposed free morpheme, dεn (as in 1 and 2), or by an allomorph of suffixal -z.

1. (1) ma frεn dεn

2. ‘my friends’

3. (2) di gε dεn

4. ‘the girls’

This variation, both in frequency and choice of markers, is subject to disparate factors: social, phonological, and syntactico-semantic.

The impact of these factors on the frequency of plural marking was analysed by use of the varbrul programme. The data comprised 2,039 semantically plural nouns drawn from sociolinguistic interviews and conversations, with a maximum of 100 tokens taken from any one speaker.

Social factor groups

Singler (1984) has as its central point that Liberian English – as it extends from pidgin to Liberian standard English – is a continuum of the type proposed by De Camp (1971). The position of the speaker's output along the Liberian continuum correlates with the speaker's background.

Introduction

Newfoundland English

The English spoken in Newfoundland has long been recognised as a distinct variety of North American English (see, for example, Bailey 1982: 163; Chambers, this volume). Unlike much of the rest of eastern Canada, Newfoundland received little if any settlement of United Empire Loyalists from the eastern United States in the post-revolutionary period. Rather, most present-day Newfoundlanders are the descendants of immigrants from two highly concentrated areas in the British Isles: southwest England (particularly the counties of Dorset, Devon, Somerset and Hampshire – see Handcock 1977) and southeast Ireland (especially the counties of Waterford, Wexford, Kilkenny, Tipperary and Cork – see Mannion 1977a). While the beginnings of permanent settlement in Newfoundland can be traced back to the early seventeenth century, the peak period of immigration to the island came two centuries later; Irish immigration was particularly high in the periods 1811–1816 and 1825–1833 (Mannion 1977a: 7).

Within Newfoundland, the Irish were to settle primarily on the Avalon peninsula, south of the present-day capital, St John's (see figure 7.1). Much of the remainder of the long coastline of the island was inhabited by those of West Country English stock. As a result of sparse overall settlement, as well as the lack of inland transportation links until well into the present century, many of the small fishing communities or outports that dotted the coast remained fairly isolated from the outside world; some remain so even today.

For the purposes of this article, Southeast Asia includes the following five countries: Indonesia, Malaysia, the Philippines, Singapore and Thailand. Apart from being members of the same political and educational organisations such as ASEAN (Association of Southeast Asian Nations) and SEAMEO (Southeast Asian Ministers of Education Organisation), all these countries have been studied as a region for the varieties of English spoken in them (Noss 1983; Llamzon 1983; Noss 1984). It seems reasonable, therefore, to include them as a single area. Hong Kong, consisting of the island of Hong Kong itself, Kowloon and the New Territories, fits less easily into the region although certain similarities and differences between Hong Kong, Singapore and Malaysia, have been pointed out by Platt (1982a: 384).

The aim of this article is to present an overview of the work done on selected sociolinguistic aspects of English in Southeast Asia and Hong Kong. This will be done by (a) highlighting common themes found in the published research on varieties of English in Southeast Asia; (b) discussing in greater depth the descriptions and analyses of the variety of English spoken in each of the countries mentioned above.

The focus throughout is on research already carried out rather than on what needs to be done. Accordingly, it is only in the conclusion that gaps in the current research are discussed and directions for future research on English in the region suggested.

Historical and political background

Since 1921, Ireland has been divided into two political entities: the independent 26-county Republic of Ireland (earlier the Irish Free State, capital city Dublin) and the six-county statelet of Northern Ireland (capital Belfast) which is currently part of the United Kingdom. Although the latter is frequently referred to as Ulster, it excludes three of the counties which comprise the historical province bearing that name.

The long-standing dispute over the political status of the Six Counties reflects the divided loyalties of its population. Broadly speaking, the nationalist community (largely Roman Catholic) desires a united independent Ireland, while the unionist community (largely Protestant and descended from Scottish and English settlers who arrived in the seventeenth century) professes allegiance to the British crown. The present ascendency of the unionist cause stems from the fact that the boundaries of the Six Counties were originally drawn in such a way as to ensure a Protestant majority. The current ratio of Protestants to Catholics in Northern Ireland is roughly two to one.

The English language first established a foot-hold in Ireland with the Anglo-Norman invasions of the twelfth century. For several centuries after this date, it made little headway against the Irish Gaelic of the indigenous population. In fact the influence of English waned to such an extent that by 1600 it was more or less restricted to a small enclave on the east coast.

Introduction

Cane Walk is a pseudonym for a Guyanese village within half an hour's drive of Georgetown, the capital. In the mid 1970s, when the data for this study were collected, approximately 3,650 people lived there. About 97 per cent of these were East Indian, descendants of indentured labourers brought from India between 1838 and 1917 to replace and supplement Africans (emancipated in 1838) as the sugar industry's labour force. The Cane Walk community was created by the nearby LBI (La Bonne Intention)/Ogle Sugar estate in the 1950s to provide alternative housing for its workers, after the barrack-like ‘logies’ in which they had housed them, on the estate itself, were condemned. The community's stratification into two classes, which we will refer to as ‘Estate Class’ (EC) and ‘Non-Estate Class’ (NEC), reflects in part a sugar industry distinction between ‘laboureres’ and ‘junior class’ employees (see Jayawardena 1963: 28–52). Most EC members work as cane-cutters, weeders and in other labouring capacities in the canefields behind the village. Some NEC members are junior supervisors on the estate, but most work as shopowners, contractors, clerks and in similar ‘lower middle class’ jobs off the estate, some in Georgetown.

Data

In this paper, I will summarise some of the key findings about sociolinguistic variation in this community, drawing on an earlier study of pronominal usage in a judgement sample of 24 Cane Walkers (Rickford 1979).

Introduction

Canada occupies a vast territory, embracing several regions distinguished from one another by climate, topography, network ties, orientation, and all the other factors that naturally accrue to geophysical spread. Sociocultural perspectives are further complicated by the existence within the Canadian boundaries of two long-standing national consciousnesses which simultaneously share Canadian nationality and maintain their own. Québec is the power base for the francophone minority, equal partners in Confederation since its inception in 1867. Newfoundland joined Confederation only in 1949 after centuries of colonial ties to Britain and self-government.

Linguistically, their presence affects Canadian English (hereafter CE) in interesting and very different ways. Québec's location interrupts the continuity of the English-language majority, splitting the Atlantic provinces from the central and western provinces, and it perpetuates bilingual buffer zones in the adjacent provinces of New Brunswick on the east and Ontario on the west. Newfoundland, though overwhelmingly anglophone, did not share mainland Canadian settlement history, and her political autonomy gave rise to an indigenous standard accent which is only now beginning to reflect the influence of mainland CE. Any generalisations that might be hazarded about CE must necessarily be qualified – implicitly or explicitly – by their presence.

Introduction

South African Indian English (SAIE) illustrates the rich syntactic variation characteristic of an English dialect that has arisen from a process of language shift. Once a second or third language spoken by a group of indentured and merchant Indians in Natal, SAIE is today, 125 years after the first immigrations, a first language sharing a great deal with other English varieties of South Africa (SAE), while having a host of linguistic features of its own. Despite many features in common with the English of India, SAIE has, on the whole, the characteristics of a new dialect born on South African soil, rather than those of a transplanted Indian English variety (Mesthrie 1988).

Sources of variation in SAIE

The following typology of Englishes according to the (decreasing) degree of variability in syntax can be posited:

The break in tradition between ancestral languages and the new variety of English is less rapid in language shift than in pidginisation, and one may consequently expect less variation than that reported for creole and post-creole Englishes. I believe that the transfer of terms like basilect, mesolect, acrolect from creology is, however, appropriate in the study of language shift varieties, perhaps more so than for L2 Englishes, to which the extension was first made by Platt (1977).

In studies of non-native varieties of English (hereafter NNVE's), with few exceptions (D'Souza 1987; Kachru 1982, 1983, 1986; Smith 1983), not much attention has been paid to what may be called the pragmatic aspects of language use. By pragmatic aspects, I mean topics such as how certain speech acts, such as informatives, directives, commissives, etc. (cf. Austin 1962; Searle 1969) are performed in these varieties. Since NNVE's differ from native varieties in the performance of speech acts more than in formal properties, it is possible that the pragmatic approach may succeed in capturing the uniqueness of a NNVE where structural analyses fail to do so.

Recent studies have shown that there are important cross-cultural and cross-linguistic differences in the way the ‘same’ speech act is performed in different languages. For example, Olshtain and Cohen (1983) have pointed out that an apology in Hebrew is less likely to include a ‘promise of forebearance’ or ‘an offer of repair’ (for the damage) that is there in English. They also observe that the two languages differ in both the range of offences for which an apology is offered and the intensity of the expression of regret. (For other cross-cultural differences see Apte 1974; Loveday 1982; Sridhar and Sridhar 1986, Sridhar in press; Wolfson and Judd 1983).

Like many other non-standard varieties of English, Inner-Sydney English (ISE) allows variation in its subject–verb agreement patterns, especially with the auxiliary verb do (+ not), and the verb be (as a copula and an auxiliary).

Such variation has been investigated in British English (Trudgill 1974; Cheshire 1982) and American English both in Black English (Labov et al. 1968; Wolfram 1969) and in various white varieties (Labov et al. 1968; Wolfram and Christian 1976; Feagin 1979). In Australian English, such variation has been noted by Eagleson (1976) in Sydney, Dines et al. (1979), Bradley (1979) in Melbourne, and Shnukal (1978) in Cessnock, but only Bradley and Shnukal provide any quantitative analysis.

This paper will attempt to explore systematically the factors which influence variation in subject–verb agreement patterns in Inner-Sydney English, focusing on the two variables, do and be, in order to determine the linguistic and non-linguistic constraints on the variation apparent.

Methodology of present study

This study is based on the speech of 40 adolescent residents of the Inner-City area of Sydney. All the informants were long-term residents of the area and were Australian-born of Australian-born parents. This was important, given the high migrant population of the Inner-City area of Sydney, in order to eliminate the possibility of language transfer.

Introduction

The islands of the Caribbean stretch over a thousand-mile arc from the top of the Florida archipelago to the mainland of South America. They are usually classified according to their (latest) affiliation to the European powers that colonised them from the fifteenth century onwards, and also in terms of the official languages they recognise. For the most part, these two criteria coincide. The major groupings include the Spanish-speaking nations, the French-official (francophone) Caribbean, the Dutch-official territories, and the English-official (anglophone) Caribbean. The largest group of islands belongs to the last of these categories.

The English-official Caribbean countries together have a total population of about five and a half million speakers. They include the vast majority of the islands outside of the Greater Antilles group (see figure 37.1), as well as two mainland nations – Belize in Central America, and Guyana in South America. These communities differ substantially in their sociolinguistic structures and particularly in terms of the relationship that English bears to the other varieties of language in use.

The vast majority of these countries are communities in which English as the official language coexists side by side with a lexically related creole language as the vernacular, with other varieties either non-existent or severely restricted in use. Such communities make up the so-called creole continua of the Caribbean. A list of the countries, their area, population and languages is provided in table 37.1.

Background

Although speakers of northern hemisphere varieties of English perceive few, if any, differences between New Zealand English (NZE) and that of Australia (Wells 1982: 605; Trudgill and Hannah 1985: 18), New Zealanders have been aware of differences in their own speech for at least a century (Gordon 1983a, 1983b). All Kiwis share the stereotypes of the broad-accented cow cocky and the refined university professor. However, the notion that accent varies according to socioeconomic ‘class’ distinctions stands in direct conflict with the cherished belief that New Zealand is a ‘classless’ society, at least in theory (Sinclair 1980: 316–17; Ausubel 1960: 27ff). The economic difficulties of the years after 1973 have weakened the myth, but its influence may have contributed to a lack of interest in sociolinguistic variation in NZE.

A second factor which impressed me as an American immigrant to New Zealand in 1970 was an apparent feeling of inferiority about the NZE accent (cf. Bayard 1990a) – a belief that it was little more than a colonial, non-standard variety of RP. When I began this study in 1984, very little quantitative research into NZE sociolinguistics had been done. The phonetic research which had been carried out relied on ‘general NZE’ speakers (Hawkins 1973a, 1973b; Maclagan 1982), only roughly defined along the lines of Mitchell and Delbridge's (1965) ‘broad-general-culti-vated’ trichotomy in Australia (see Guy, this volume).

Singapore English is probably the classic case of the indigenisation of English as it performs such a wide range of functions, not only internationally but intranationally, not only interethnically but intraethnically, not only in the more public domains but increasingly in the more private domains of family and friendship as well. This does not, of course, mean that it has supplanted the various local languages: Chinese dialects, Malay and the various languages of the Indian communities. However, its use in various domains has been increasing annually.

English came to Singapore with the establishment of a trading centre by Sir Stamford Raffles in 1819. The functions of English increased quite rapidly, and with the establishment of English-medium schools came the beginnings of Singapore English.

Although a pidginised form of Malay, Bazaar Malay, was the general lingua franca throughout the colonial period and was used not only for interethnic communication between Chinese, Malays and Indians but also by Europeans communicating with the majority of the local population, English gradually became a kind of prestige lingua franca among the small but increasing English-medium educated section of the population.

As pointed out in Platt (1975: 365, 1977a: 83) and Platt and Weber (1980: 17–22), it was through English-medium education that an indigenised variety of English developed. A kind of fossilised interlanguage became a lingua franca in the English-medium schools among students whose home language might be one of the Chinese dialects (not all of them mutually intelligible), an Indian language or Malay.

This study of Vancouver English using instrumental procedures suggests the following sociophonetic generalisations: (1) the speech of men and women differs systematically in the particular vowels that function as social indicators, (2) individual vowels differ systematically in quality between working class speech and middle class speech, (3) class-based differences in vowel quality are realised as superimposed secondary articulations representing prosodic shifts in vocalic phonology, (4) classbased differences in speech extend to consonant articulation as an acoustic result of choice of habitual voice setting, and (5) subtle shifts in voice setting, which affect the acoustic realisation of vowel and consonant articulations, function as salient social indicators and potential markers of style.

English in Vancouver is consistent with the general characteristics of heartland Canadian English as far as the urbanised and relatively uniform social structure of its anglophone residents is concerned (Chambers 1979: 190). There is considerable representation of other varieties of English as well as other languages including German, Cantonese, Punjabi and Vietnamese, to the extent that up to 40–50 per cent of students in many school districts do not speak English as a first language. Varieties of English from the British Isles are in strong evidence and play a large role in the way people in Vancouver think about speech and its prestige value. The status of English as a second language also holds a place in how people think about language, in social interaction and in political rhetoric.

Introduction

More research has been carried out on the English that is spoken in the UK and, particularly, in the USA, than anywhere else in the world. It is impossible to survey this vast array of research within a single chapter, and I shall make no attempt to do so.

Nevertheless, as mentioned in the Introduction, a volume whose subject matter is English around the world can hardly exclude the two countries where English has been institutionalised longer than anywhere else, and whose standard varieties have, until very recently, held an unchallenged position as reference models for the teaching of English throughout the world. This chapter, therefore, briefly discusses some sociolinguistic aspects of English as it is used in the UK and the USA, keeping broadly to the format that the authors of the overview papers were asked to use, in order to make comparisons possible between different areas of the world.

Whilst admittedly very different, the UK and the USA share some important characteristics. Perhaps the most important of these is that both countries are overwhelmingly monolinguistic in their official orientation, even though throughout their history they have always been multilingual. A further similarity is that the standard varieties of English in the USA and in southern Britain each have well codified norms, enshrined in their different national dictionaries and grammars; and within each country the existence of a standard variety has given rise to similar sets of attitudes towards the uncodified non-standard varieties (which are spoken by the majority of the population) and has resulted in a similar set of social and educational problems.