Introduction

On ten of the 18 inhabited islands of Torres Strait, lying between Cape York Peninsula and the mainland of Papua New Guinea, and in the Cape York communities of Bamaga and Seisia, where Torres Strait Islanders predominate, the traditional languages are being or have been replaced by an English-based creole. Its speakers call this language Broken (from ‘Broken English’), Pizin (from ‘Pidgin English’), Big Thap (‘Big Piece’) or, among some younger speakers, Blaikman (‘Blackfellow’) or Ailan Tok (‘Island Talk’), but it is never called Langgus, a name reserved for the indigenous languages and English. Here I shall refer to it as Torres Strait Creole (TSC).

Today, the children of Erub, Hammond Island, Masig, Mer, Moa, Purma, Thursday Island, Ugar, Waraber and Yam (see map 11.1) speak TSC as their first language, although in some communities the children can still understand their traditional language. On Masig, Mer, Moa, Purma and Waraber only elderly speakers of the traditional languages remain. The creole has been the first language of the majority of inhabitants of Erub, Ugar and St Paul's Community, Moa, for four generations.

The number of first-language speakers of the creole in the Torres Strait islands and in the Cape York settlement of Bamaga is probably around 3,000 out of a total Islander population of about 4,000.

Indigenous languages of Australia and their speakers

The term ‘Australian languages’ can be used to refer to the indigenous languages of Australia including the languages of the Western Torres Strait and, perhaps, Tasmania. Most of the land mass including offshore islands through territorial affiliations has been linked to speakers of an indigenous language variety. Australia therefore contrasts with nearby Papua New Guinea, where significant areas of land have been uninhabited and unaffiliated with language groups (Wurm and Hattori 1981).

Australian languages are both numerous and diverse (see map 1.1). It has been estimated that around 250 distinct languages were spoken at the time of the first (significant) European contact in the late eighteenth century. Usually each language would have a number of dialects so that the total number of named varieties would have run to many hundreds (see e.g. Sutton, this volume). These varieties were spoken by a population of around 300,000 according to estimates generally accepted until fairly recently. However, Butlin (1983) suggests that previous estimates did not sufficiently allow for the devastating effects of introduced diseases such as smallpox: he therefore proposes an original population of around 900,000. A population of this size requires a rethinking of views on the relationship between language varieties and their speakers.

Introduction

When migrants enter their new country, they are immediately faced with the sometimes daunting task of finding a place to live. Studies of urban settlement patterns record that migrants often are concentrated in certain parts of cities; Little Italy and Chinatown in New York City may be among the better-known ethnic communities in the world, but they are far from unique. Ethnic neighbourhoods give migrants a place in the new country where they can speak their own language and obtain the goods and services they need to maintain to some degree the way of life they grew up with.

Ethnic communities are not entirely separate entities, however, built on the edge of town. They are created within the confines of the host community, occupying areas once exclusively the domain of the host community or areas vacated by other, often more upwardly mobile ethnic minorities. Initially, the hosts may be attracted by some of the innovations brought in by the migrants, notably the food and the festivals, but it is rare for the hosts to hold the ethnic communities in high regard. Nevertheless, whether the host is attracted or repelled, in finding a place for themselves, migrants bring about quite often dramatic changes in the host community.

Introduction

Plain English, that is the use of ordinary everyday English in official documents, is not an innovation of recent decades. There have been repeated pleas for it down the centuries (see, for example, Sir John Cheke, 1557) and even more recently in well-known works such as Sir Ernest Gowers' Complete Plain Words (1954). The current drive for Plain English, however, dates from the 1970s and has its mainspring in the consumer movement. The recognition that consumers had rights and were not just a source of income for manufacturers and sellers not only led to better products but also eventually spread to aspects of language used in the documents which described products and services. Indeed, one of the first plain English documents came from a market survey conducted by Sentry Insurance in 1974. Responding to the consumer movement, the Company sought to discover the desires of its customers for insurance. One request, unexpected by the Company, kept emerging: the desire for a comprehensible policy. The Sentry Insurance car policy and the Citibank loan agreements which appeared in early 1975 are generally recognised as the first manifestations of the current plain English movement.

Australia was not far behind these developments in the United States.

Introduction

Although research projects have been carried out on several Slavic languages in Australia (see, e.g. Kouzmin 1988), very little has been written on the language and speech of Serbian and Croatian immigrants in general or in particular states. This chapter presents the results of a survey of first-generation speakers living in Queensland. It seeks to determine which factors facilitate the maintenance of their native tongue or, on the contrary, lead to a shift to Australian English.

Is Serbo-Croatian a language?

Some 135 years after the Vienna Convention, where a draft for the creation of a common Serbo-Croatian language was drawn up, the battle among language varieties and literary languages still rages today inside Yugoslavia and outside wherever Yugoslav-born subjects have settled. Its origin is both political and linguistic. Although Macedonian, which is very close to Bulgarian, qualifies as a distinct language in the 1974 Federal Constitution of Yugoslavia for external political reasons, the same document only implicitly recognises that Serbian and Croatian are separate languages. Article 269 says that ‘Federal laws and other general acts of the organs of the Yugoslav Federation are to be promulgated in the languages of the nations recognised by the Constitutions of the Federated Republics’.

Demographic situation

There are some 75 to 100 ‘immigrant’ or ‘community languages’ in use in Australia, the exact number depending on one's criteria for the term ‘language’. The term ‘community language’ is employed to denote the socalled ‘immigrant’ languages used within the Australian community to emphasise the legitimacy of their continuing existence. These languages have been part of the Australian scene since early in the history of white settlement. However, there has been discontinuity between the earlier and contemporary use of community languages.

The 1986 Census data provides the most detailed large-scale corpus of information on home language use. Of the Australian population 13.6 per cent reported employing a language other than English at home. (However, there are many who now use a community language not in their own but in their parents' homes). This rose to 18.5 per cent in Victoria, and was lowest in Queensland (5.9 per cent) and Tasmania (3.6 per cent), the states least affected by postwar migration. The community languages most widely used in the home were Italian (415,700 speakers), Greek (277,500), Serbo-Croatian (140,500), Chinese (139,100), Arabic (199,200), and German (111,300). Other languages with over 50,000 regular users were Dutch, French, Polish, and Spanish. With the large number of South-east Asian refugees since 1976, Vietnamese has also now joined this group of languages.

Introduction

Most definitions and descriptions of Aboriginal English confine themselves to formal linguistic structures, for example, phonology, syntax, and vocabulary. Although some Aboriginal speakers of English may share a common grammar with speakers of Standard Australian English (see e.g. Kaldor and Malcolm, this volume), they do not necessarily share meanings and interpretation because of the sociocultural context of Aboriginal ways of speaking. For example, constraints on the social dimensions of the use of question structures by Aboriginal speakers of English (discussed in 4.4.1 below), transcend grammatically defined dialectal boundaries, and are part of widespread Aboriginal communicative strategies, regardless of the language variety used. Such pragmatic evidence (and further evidence discussed later in this chapter) indicates that our definitions of Aboriginal English need to look beyond grammatical features and include aspects of communicative strategies. I therefore use the term ‘Aboriginal English’ to refer to Aboriginal varieties of English, which in some instances may differ from Standard Australian English primarily in features of pragmatics (and minimally in grammar).

On communicative strategies

This chapter is concerned with the patterned ways in which Aboriginal people use dialects of English as part of their social interaction. The focus is on ‘socially constituted linguistic’ study (Hymes 1974), which relies particularly on ethnographic method, and which reverses the typical structural linguistic tendency of isolating referential structure and then posing questions about social functions (see e.g. Gumperz 1972 on ‘correlational sociolinguistics’).

Background to the Australian survey

A pilot study on regional variation in the lexicon of Australian English (Bryant 1985) found clearly defined usage regions in Australia, namely, Western Australia, south-east South Australia, the Victorian usage region which included Tasmania and part of southern New South Wales, and New South Wales and Queensland (which form a fairly homogeneous region). The borders of these regions do not coincide with the State borders. The evidence for regional variation in Australia was sufficiently strong to warrant a full-scale survey.

This chapter reports an innovative methodology necessitated by constraints of time and money as well as by conditions peculiar to Australia. It covers the first two stages of the full-scale survey, making comparisons with surveys done in other countries, and examines some early results.

The Australian survey compares most closely with the United States surveys, as conditions in Australia are more like those in the United States than elsewhere. For example, both have fairly recent European settlement, with the first settlers drawn from the same country, comparable geographical size, and larger usage regions than in other countries. Chambers and Trudgill (1980: 108) note that ‘in more recently settled regions, like inland North America and Australia, it is becoming apparent that dialect features tend to be shared over relatively great distances when the settlement history goes back only one or two centuries’.

Warlpiri and its social setting

Warlpiri is spoken in a number of desert communities north-west and north of Alice Springs. It belongs to the Pama-Nyungan group of languages, which encompasses 90 per cent of the Aboriginal languages of the continent (see Walsh, this volume). It has a number of structural features typical of Pama-Nyungan languages, for example, agglutinating word formation and case-marking on nouns. Warlpiri has three distinct case-marking patterns for different transitive verbs, but the dominant case frame for transitive sentences is ergative-absolutive. Its syntax follows a nominative-accusative pattern, in that subject agreement is found with subjects regardless of their case form. It also has one of the freest word orders in the Pama-Nyungan languages; elements within complex noun phrases and non-finite clauses need not be contiguous. Like other Pama-Nyungan languages, Warlpiri has a highly developed system of pronouns and demonstratives: it distinguishes singular, dual and plural number, as well as first-person inclusive and exclusive. As is the case for some Pama-Nyungan languages, Warlpiri utilises bound pronouns that cross-reference subjects and objects. Typical of Pama-Nyungan languages, Warlpiri has few vowels and a large number of articulatory positions for consonants: there are five for oral and nasal stops and three for laterals and rhotics; it has no voiced-voiceless contrast and no fricative phonemes; words are always at least two syllables long and have primary stress on the first syllable; all words and syllables begin with a consonant.

Language and setting

Kalaw Kawaw Ya (KKY) is a dialect of the Western Torres Strait language, which is considered to belong to the Pama-Nyungan family, the largest of the Australian subgroupings (Dixon 1980). Kala Lagaw Ya, Kala Lagaw Langgus, Yagar Yagar and Mabuiag are other names which have been applied to the Western Torres Strait language or another of its dialects. Speakers have agreed that the language as a whole should be called Kala Lagaw Ya, the ‘Western Island Language’.

The Western Torres Strait region, whose islands and waters lie mainly within Australia's international boundaries, includes four dialect zones, marked by broken lines on map 7.1. The description presented here comes from Saibai Island speakers and is representative of the northern zone. Kalaw Kawaw Ya is their own name for the ‘Western Island Language’. Bani and Klokeid (1975 and 1976) have described the dialect of Mabuiag Island in the western central zone. The dialect of the eastern central zone is used only by older speakers; younger people have adopted Torres Strait Creole (see Shnukal, this volume) as their mother tongue. The southern zone includes the town of Thursday Island, the administrative and service centre for the region, where groups of speakers from different dialect zones now live.

Introduction

The Aboriginal Land Rights (Northern Territory) Act of 1976 is intended to give Aboriginals in the Northern Territory legal title to their traditional lands, where these lands are unalienated from the Crown or where all interests in the lands are held by or on behalf of Aboriginals. When an area of land is being claimed, an Aboriginal Land Council prepares a claim on behalf of the traditional owners, often using contract researchers, in which it tries to ascertain the names and relationships of all the traditional owners, the location of sites on their land, and the nature of their affiliations (spiritual and economic) to the land. This evidence is presented before an Aboriginal Land Commissioner, who is a judge of the Supreme Court of the Northern Territory, in a quasi-legal hearing with lawyers representing the claimants and other parties with interests in the same lands, for example, mining and pastoral companies, the Northern Territory government, local councils. The proceedings are tape-recorded and an official transcript produced by the Commonwealth Reporting Service. The Aboriginal Land Commissioner issues his findings and recommendations in a published report to the Federal Minister for Aboriginal Affairs, who may then make a grant of the land to an Aboriginal Land Trust established to hold title to the land for the benefit of its traditional owners (see Maddock 1983 and Neate 1989 for a general discussion of the Land Rights Legislation).

Introduction

The original preoccupation with the description of the linguistic performance of immigrant bilinguals in their two languages in an attempt to discover universal linguistic features of language contact has often meant that questions regarding a bilingual's perception and awareness of contact phenomena have not received a lot of attention. Usually studies of language contact in the Australian context have done little more than mention the fact that bilingual informants, when questioned about their speech patterns, realised that some mixing of languages occurred (see e.g. Clyne 1967; Pauwels 1980; Bettoni 1981). The same studies often also recorded that this awareness of mixing one's languages triggered off a whole array of feelings in the bilinguals ranging from indifference or resignation to annoyance or even acute embarrassment. Recently, with the upsurge of studies of language attitudes, researchers have become interested in establishing to what extent bilinguals' perceptions of ‘language mixing’ have an impact on the process of language maintenance or language shift (see e.g. Chana and Romaine's 1984 study of the attitudes to mixing of Panjabi English bilinguals in Britain). That is, does the perceived linguistic quality — pure or mixed — of the ethnic language affect its status, usefulness and so on with respect to English? I will explore this issue in relation to Dutch in Australia.

Introduction

This chapter focuses on the main distinguishing structural properties of Kriol, with some mention of the variation that exists between dialects (see Sandefur 1986 and Sandefur and Harris 1986). Other brief general descriptions can be found in Graber (1986b), Sharpe (1985) and Steffensen (1977). Sociohistorical aspects of the language are discussed by Harris (this volume).

Phonological structure

The phonological structure of Kriol is very complex, due both to the origin of the language and to the effects of continued contact with English, the socially dominant language from which it was in part derived (for details of the orthographic structure of Kriol, see Sandefur 1984a).

The phonological system consists of a continuum of sounds which extend from what is referred to in the eastern dialects as hebi — ‘heavy’ Kriol to lait — ‘light’ Kriol with a span in between being prapa — ‘proper’ Kriol (see Sandefur 1982a, 1982b, 1985b, 1986). The extreme heavy end of the continuum reveals a subsystem that is virtually identical with that of traditional Aboriginal languages. There are, for example, no affricates, no fricatives, no contrastive voicing with stops, no consonant clusters within a syllable, but five points of articulation for stops and nasals. The extreme light subsystem, in contrast, includes virtually all the contrasts which occur in English.