THE LIFE AND DEATH OF A CREOLE

Of all the curious things which the Macanese can offer a European observer, none is of greater interest than the language they use amongst themselves. It is a kind of dialect based on XVth century Portuguese, a mixture of Chinese expressions and English phrases. One could say of the people who are most adapted to our way of life that they speak a passable Portuguese with us although their pronunciation has been influenced by the changes which most Latin languages undergo in tropical regions. The people, and in particular the women, use a language so curious and capricious that we Europeans can hardly decipher its meaning at all.

Bento da França: Macau e os seus Habitantes, 1897

BENTO DA FRANÇA'S paternalistic, slightly pedantic description of Macanese creole in the late XIXth century is not unusual. It is but an extreme example of a purist's preoccupation - that of the traveller in the East confronted with a corrupted, perhaps 'strange' version of his mother-tongue.

Since the Middle Ages, there have been references to pidgins and creoles in descriptions of voyages, letters, missionaries' reports, governors' and administrators' correspondence, grammar manuals, translations of religious texts and essays. Because of their lexical similarities to the European languages from which they were derived, and, in contrast, the 'strange' structure and phonetics which distinguish them from the same, these creoles have always been regarded as 'adulterations', 'strange and macaronic jargons', 'infantile and primitive corruptions', nothing more than mutilated versions of the master language spoken by groups or communities who have traditionally been seen as racially, socially and culturally inferior.

Western opinion concerning these languages has been filtered and distorted by prejudice and stereotypes. This can be seen in a process ranging from Gil Vicente's interpretation of the speech of the XVth century negro (Clérigo da Beira, Nau de Amores, Frágua d'Amor) for whom not even a lighter shade of skin can open the magical door of integration into white society, because his 'negro speech' defines him unmistakably as inferior, to the Encyclopedia Britannica's contemporary definition of pidgins and Creoles as an unruly bastard jargon, filled with nursery imbecilities, vulgarities and corruptions (1)and the not too distant past when the view of the academic world was that the origin of creoles and pidgins lay in an intentional simplification of the master language on the part of the dominant class to facilitate communication with the dominated class, by implication primitive and childish (Jespersen, Schuchardt).

In the last few decades, however, the progress of creole studies has contributed towards a shift in attitudes which, while it may not be detected in the layman's world, has at least been apparent in academia. The study of creoles has become fundamental to the study of changes in language and to understanding the complexities of communication between humans.

WHAT IS A CREOLE?

Creoles are mother-tongues of comparatively recent origin (when set in contrast to European languages and the majority of non-Western languages). They have arisen from 'linguistic encounters' in specific historical, social, political and linguistic situations. It is important to note, however, that their relative youth in no way prevents them from gaining the same degree of complexity, subtlety or expressive power that can be found in all other languages.

As a mother-tongue, any and every Creole must be able to respond to the communicative, expressive and intellectual demands made of it by its speakers. It will be used for doing business, praying, discussing, courting, seducing, playing, singing, scolding, controlling, submitting to control, insulting, conveying knowledge and traditions, convincing, arguing and imploring. Any language which fails to allow its speakers to carry out the range of human activities for which language is required will undoubtedly be rejected, neglected and forgotten. The list of languages which are now dead proves this, although obviously we cannot forget the factors which led these languages to fall into disuse. This latter point will be discussed later in the article.

It could be argued, nevertheless, that some creoles are 'poorer' than others in certain semantic areas when they are compared to the Western language model. It could be said that there are no essays or works of literature, philosophical, legal or technical/scientific studies in the known creoles. For the moment, however, we shall examine the specific socio-cultural and political conditions of creole communities. These are, as a general rule, discriminated minorities, whose language has been classified as inferior either because there is a dominant class which speaks a different language (the case of Malacca's Papiá Kristang) or because there is a prestige language, the language of a motherland, which is used by, and as a model of, the social classes in power (língu Makista).

From the moment when, for historical or political reasons, a creole gains prestige, traditionally sophisticated forms appear and the creole finds its expression in literature (Cape Verde), journalism (Guinea-Bissau, Sierra Leone), and maybe even in the bureaucratic-legal field, for instance when it is adopted as an official language. Inevitably, if educational and scientific development allows it, the creole will also find suitable forms for these functions. A glance at archaic Portuguese texts provides an excellent example of this.

While the term 'creole' is traditionally applied only to languages of recent formation, it is worth looking at a parallel example of linguistic development. An examination of the evolution of Latin into the Romance languages can help to clarify the stance of those who would claim a mature status for creoles.

Polomé (1982: 132) suggests the following conditions for creolization to be a determining factor in the linguistic changes to which a language may be subjected:

1) The existence of clear proof of a hiatus in the continuity of linguistic development;

2) The existence of indications of the linguistic characteristics of creolization (stages of simplification and restructuring);

3) The existence of adequate evidence of the socio-economic conditions reflecting the process of the decline in the culture of the non-speakers and/or the process of acculturation of non-speakers as they acquire the dominant language.

The Barbarian invasions instigated a breach in the usual process of the development of Latin. During the transition to the Romance languages, we find a stage of structural simplification in which the 'barbarian' languages interact with a simplified form which they reorganize. A stage of diglossia is attained with the higher classes speaking classical Latin, the prestige variant, while the growing population of invading Barbarians increasingly speak a 'creolized' Latin.

The normal linguistic development of Latin would have led to a gradual change anyway. This could be seen in the fact that spoken Latin (Vulgar Latin) was already beginning to differ from Classical Latin. However, the continuity in Latin's linguistic development was abruptly interrupted by the political and social chaos of the Vth century and the subsequent collapse of the Roman administrative structures which led to the decline of the Empire. Not even the tardy romanization brought about by conversion to Roman Catholicism, introduced via Latin, could limit the sweeping linguistic restructuring affecting the entire crumbling Empire. The strengthening of latent dialectal features in Latin itself, the persistent influence of pre-Roman languages and the socio-cultural characteristics of the different provinces in the Roman Empire were to lead to the formation of proto-Romance dialects.

We can thus claim that, in the broadest sense of the word 'creole', the Romance languages, whose position as 'normal' and complex has never been questioned, also began as creoles.

In the case of presently accepted creoles, the conditions set by Polomé still hold. The abrupt, mass movements of populations which took place from the onset of the imposition of slavery by European nations in the XVth century, created the required linguistic interruption and decline in the slaves' original cultures. Transportation of slaves involved not only speakers of different African languages which were mutually incomprehensible, but also the common sense approach that dispersing slaves of a single culture and language would diminish or eradicate the risk of subordination.

Thus we have an example of 'linguistic encounter' in which manpower recruited by coercion and forced into a way of life set by their owners suffers a sudden sense of displacement from a familiar cultural and linguistic environment to a situation of linguistic pluralism, in which the cohesive model for contact is, inevitably, the masters' language.

These are the typical circumstances under which 'plantation creole', the linguistic product of 'plantation communities', develops. These are situations in which a large number of speakers of different languages live in segregation under the command of a minority which enjoys linguistic cohesion through the language of the mothercountry. The typical origin of these creoles is slavery and it is slavery which accounts for almost all of the creoles in the Atlantic region -Surinam, Curação, Haiti, Jamaica, Belize, São Tomé, Príncipe, Ano Bom and Cape Verde.

I interpret 'plantation' as being:

1. A piece of land in a tropical or sub-tropical region

2. which is used for commerical cultivation,

3. intended for profitable distribution in the markets of Europe

4. in which labour is provided by non-European slaves

5. and in which the land and other means of production belong to Europeans who are politically dependant on a European nation.

If we examine the plantation zones established by the Portuguese in the Atlantic region we can identify these characteristics right from the onset of Portuguese overseas expansion. In Madeira, which was uninhabited at the time of the Discoveries, just like the other islands, the importation of slave labour from the Canary Islands allowed the production of sugar cane.

When the Cape Verde Islands were being colonized, however, the population of the Canary Islands had already been decimated and the Portuguese were bringing their slaves from the African mainland. Later on, in São Tomé and Príncipe, the plantations were to depend heavily on the importation of African slaves. (2)

Although they were actively excluded from the social life of the owners and plantation masters, they were still required to behave like the Portuguese and to conform to Portuguese cultural habits. As a result of these contradictory pressures, a creole identity, culture and language in which African languages and cultures were incorporated began to appear as early as the late XVth century.

On the African mainland, however, it was only towards the end of the XVIth century that the first creoles began to appear. Even then (in Guinea Bissau and Senegal) it was only a result of the introduction of the Cape Verde model following the later migration from those islands in the archipelago to the Gulf of Guinea.

The Portuguese occupation of the African mainland started off as a commercial and religious venture. Factories and trading posts were established where the Portuguese could obtain much sought after goods and negotiate deals with the local chiefs who were occasionally elevated to symbolic positions in the Portuguese hierarchy. This behaviour allowed the native populations of the African coast a certain amount of autonomy wherby they could maintain their traditional languages and customs, so long as they recognized Portuguese sovereignty and conformed to the religious and economic dictates which were imposed on them. This explains why no Angolan creole developed.

In the Far East, the contacts with local populations were completely different. The routes which were opened up in the search for spices, sandalwood, silks and porcelain placed the Portuguese in contact with populations who lived in complex social-political societies with ancient cultures and features of civilisation which, although they were exotic, could still be compared with European civilisations.

The policies behind the Portuguese settlements in the Orient were, from the very beginning sexual policies intended to form multi-racial societies and secure residence through the creation of a mixed population. (3)

Under the auspices of Afonso de Albuquerque, a policy of miscegenation flourished, a policy for which the way had been prepared physically and psychologically, in the skin and soul of the Portuguese nation, in their flesh (from their Mozarabic roots) and in their spirit (from their universalist impulse). (4)

While an absence of racial prejudice in their encounters with other peoples facilitated and stimulated miscegenation, the lack of available European women in the Far East also made this a 'common sense' policy. The men who came had to withstand the demands of a long voyage on board a caravelle or nao:

tiny ships, rolling like corks in the trough of the waves, the deck space cluttered up with men whose language would be purple and whose manners rough, coarse food to eat, accommodation cramped, without security of any kind, and always a good chance of shipwreck on some savage shore. (5)

Each ship transported five, six or even seven hundred people as well as weapons, tools and slaves. Mutiny was commonplace. Between departure and arrival many of the passengers would die, only to be eaten by the rats. (6)

Because of these conditions, few women cared to brave the dangers of the voyage.

In addition to this practical reason, however, there was a political motive which led Albuquerque and his successors to support miscegenation.

The Portuguese population in the XVth century was, in comparison to the scale of expansion which was being pushed forward, pitifully small. In 1527, the first census, ordered by King João III, placed the population at between 1,100,000 and 1,400,000.

The call of the Indies was, as Camões wrote in his classic epic, the dream which wrecks all peace of soul and body, leads men to forsake and betray their loved ones, subtly yet undeniably consumes estates, kingdoms, empires, at the price of depopulating and weakening this ancient kingdom and squandering its resources. (7) However great the numbers of men leaving the old country, they were still unable to provide the human labour required to dominate the maritime and commercial routes and maintain and defend their trading posts in the Far East. The shortage of labour had to be made good one way or another in order to make that dream come true. Proselytizing and intermarriage created an in loco population, whose loyalty was based on shared religion and blood.

Recognition of the importance and the urgency with which this policy had to be implemented meant that the King's representatives, the Vice-roys, and also Royal Charters, were to encourage and protect the marriage of Portuguese men with native women. This led to a special linguistic and cultural integration and remarkable mutual adaptation of which the Portuguese creoles in the Orient are an excellent example.

WHEN we discuss creoles, particularly those of Portuguese origin spoken in the Orient, we constantly find ourselves using terms such as 'extinct' or 'dying out' rather as if we were dealing with biological specimens, plants or animals. Given that plants and animals are indeed biological specimens, they follow a predictable, unalterable cycle from birth, through infancy to maturity and eventual decline, culminating inexorably in death.

The biological metaphor was popular amongst XIXth century language scholars and it was on this basis that the German linguist Franz Bopp (8) claimed that:

Languages are to be considered organic natural bodies, which are formed according to fixed laws, develop as possessing an inner principle of life, and which gradually die out.

This somewhat simplistic view of the death of languages as a 'natural' process which is 'biologically determined' is now, one and a half centuries on, unacceptable.

On the one hand, the gradual change which a language undergoes over the centuries is sometimes, for historical reasons, given another name. 'Dead' languages, described as such because no longer does anybody speak them as a mother-tongue, are still in fact 'living' although in a radically different form. This is the case with Latin. Latin underwent a process of change and divergence over the centuries which altered its appearance and name and yet it still subsists in the form of its direct descendents - the Romance languages.

On the other hand, we must take cognizance of the fact that some languages do 'die' or disappear, not as the result of a slow, gradual process of linguistic transformation but in a much more dramatic way over the space of two or three generations. This dramatic, radical disappearance is what we see in the case of many creole languages.

Why should a language die? It is in this area that modern linguists distance themselves from the biological metaphor of which XIXth century linguists were so fond. A language does not die because it has reached the final stage in its life cycle, a stage of linguistic senility determined by an internal principle, or by fixed, organic principles. A language dies because it has been pushed out by another language for social and political reasons. 'Extinct' creoles or creoles which are 'dying out' are exemplary cases which confirm the process of social and political cause and effect as an explanation of the death of languages.

THE DEATH OF CREOLES

The life cycle of a 'normal' language spans centuries. The process of transformation is almost imperceptible to each generation of its speakers. Centuries of subtle, continual shifts have taken Portuguese from the vulgar Latin in which the first Portuguese documents were written, to the Portuguese found in the Medieval Cantigas de Amigo, from the language used by Gil Vicente to that of João de Barros and Camões, from Alexandre Herculano to Eça de Queirós, from Jorge de Sena to the Portuguese spoken in our own times. The ordinary speaker is rarely aware of these changes although they are occasionally proved and contested, as can be seen in the attacks made by grammarians in the XVth century and the 'grammar surgeries' which can be seen on Portuguese television and in Portuguese newspapers.

If we examine a 'normal' language - for instance Portuguese - and trace its slow phonological, semantic and syntactic development from generation to generation, we are confronted with a slow, detailed task like a long film projecting centuries onto the screen of our attention. The film of a creole's life cycle has, on the contrary, been speeded up and social and political factors play the most important roles.

Creoles, comparatively young languages which often undergo extremely rapid development in their initial phase, changing from a contact language (pidgin) to aid basic communication between speakers of different languages, to the status of native language of a community in the space of a couple of generations, seem to suffer the same acceleration in their final stage.

Portuguese creoles in the Orient, which are now regarded as extinct, were still alive in the last century. Those which remain, the Papiá Kristang of Malacca, Portuguese creoles in Sri Lanka and Indonesia, creoles in Daman and Kerala in India, all show signs of more or less imminent disappearance.

The basic alternatives seem clear enough. A creole can continue indefinately without substantial change, as Haitian French seems to be doing. It may become extinct, as Negerhollands and Gullah are doing. We say that it may further evolve into a 'normal' language, though we are hard put to find documented examples of this, and even harder to define what we mean by a 'non-creole' or 'ex-creole' language. Finally, it may gradually merge with the corresponding standard language as is happening in Jamaica. (9)

The first example, the French creole of Haiti, reveals the persistence of social and political factors which created a situation of diglossia (Ferguson: 1959) - a situation in which two mutually incomprehensible variants, one 'High' (French) and one 'Low' (creole), existed side by side in the same community. Each of them had a defined role with its own set of functions:

In one set of situations only H is appropriate and in another only L, with the two sets overlapping only very slightly.

L would be used amongst family members, friends and in informal situations; H would be used in school, in publishing and in more formal situations.

For an educated Haitian speaker the change from one variety to another, depending on the social situation (formal, informal, with relations or superiors, on the street or at work) constitutes an alternate use of two completely different languages, namely the mother-tongue and a foreign language.

This must have been. the situation in the Portuguese creole communities whenever a creole existed alongside another dominant language with greater social prestige, whether or not it was Portuguese.

The third alternative also seems to be determined by social and political factors. The creole must become the dominant prestige language before it can evolve into a 'normal' language by which we mean a language which does not suffer from the social stigma usually attributed to creoles.

Cape Verde creole seems to be developing in this direction despite the fact that Portuguese is still the official language. The tendency for this creole to be classified as 'normal' is reflected in the appearance of literary works in creole, its use in the press and a change in the attitude of its speakers towards their mother-tongue. They no longer regard it as inferior, nor the recourse of a community relegated to the lowest social stratum because of a colonial mentality. Now it is used as an element in a national cultural identity which is specifically from Cape Verde.

Nevertheless, it is the other two alternatives which are of most interest to us in this particular study: the convergence with the language of the motherland, and extinction.

SUICIDE OR MURDER?

Jean Aitchison (1981, p.208-222) gives the suggestive terms 'suicide' and 'murder' to the two alternatives presented by Decamp for the death of a language:

The first possibility is that speakers of the old language will continue speaking it, but will gradually import forms and constructions from the socially dominant language, until the old one is no longer identifiable as a separate language. This is in reality an extreme form of borrowing. The language concerned seems to commit suicide. It slowly demolishes itself by bringing in more and more forms from the prestige language, until it destroys its own identity.

The second possibility is more dramatic. In some circumstances, the old language simply disappears. We are dealing not so much with the natural passing away of a language, but a case of murder-murder by the dominant language as it gradually suppresses and ousts the subsidiary one. (10)

A CASE OF SUICIDE

The disappearance of lingu Makista is an excellent example of suicide. The prolonged geographic coexistance with the official, prestige language of the motherland in which it is rooted is one of the determining conditions for 'suicide' and it is present here.

Another necessary condition for survival of the creole is mutual incomprehensability between the two variants. The similarities between the two are flagrant:

Que pena eu nôn pôde escrevê portuguez assim galante como aquele bulicioso de manjor Rua; mas mesmo cusa tudo portuguez-portuguez que tá bem curtido já com nosso lingu de Macáo, lôgo entendê esse rabucenga que eu escrevê.

The creole contains a vast lexical component adopted from Portuguese although the divergence may be more obvious at a syntactic and phonological level. It is thus extremely easy for model patterns to be imported.

In the third place, there was no rigid social stratification in Macau between the creole community and the motherland Portuguese, as had been the case with the Atlantic creoles in which creole was associated with the lower classes (usually descended from slaves). Even though the higher positions in the administration were exclusively reserved for Portuguese, there was always enough social mobility to motivate large numbers of patois speakers to bring their speech closer to the linguistic model.

Finally, we find in Macau, particulary in this century, an educational programme which is strong enough to exert its corrective influence on creole and thus bring it nearer to Portuguese:

Vós lôgo sintí grandi diferença na minha modo di escrevê. Eu já aperfeiçoá bastante neste um pôco temp. Tudo este escóla novo de machu e femia, e aquela gazeta Ta-ssi-yang-kuo já fazê indretá bastante nosso lingu.

Ta-ssi-yang-kuo, 1st series, 5th of January, 1865

Some Macanese mention punishments meted out to children in the thirties and forties if they were caught using patois at school.

The corrective pressures did not have a uniform effect on all patois speakers. The extent to which acculturation took place depended on various factors such as age, gender, social class, access to education and the educational level attained as well as the frequency and intensity of contact with Portuguese speakers.

The process by which the creole becomes more and more similar to the model language is, in other aspects, fairly slow. This is why we can speak of a process of decreolization during which we can find evidence of a continuum, a synchronic gradation normally associated with the insertion of speakers into various social classes, age groups and gender which thus determines not only level of access to education but also the intensity of contact with model Portuguese.

Marques Pereira (op. cit., vol I, p.55) mentions three kinds which could be distinguished from the patois of his day and which he regarded as alive and well, if not spoken in public and in conversation with the 'rulers', at least in the heart of the family:

a) closed Makista or pure Makista (if it could be described as such) which is the most interesting. It is spoken by the lowest classes.

b) Makista modified by the tendency to approximate itself to current Portuguese. This is used by more educated people on contact with the Portuguese.

c) Makista spoken by the Chinese. (11)

In 1920 (12) we find the following reference which is probably based on Marques Pereira:

The Macaists speak a Portuguese patois, very unlike real Portuguese. There are three forms of this dialect: (1) as spoken by the lower classes, (2) a form that approximates more to pure Portuguese, (3) as spoken by the Chinese. There are several 'Portuguese' papers in the colony, said to be written in better Portuguese than those of Goa.

We can infer from these references that at the end of the last century there was a situation of diglossia - the language spoken in public and in conversation with the 'rulers' and the languages spoke in the heart of the family - and a decreolization for which the reason seems to be the social hierarchy.

Bento da França (1897, p.200) pointed out the same process of decreolization in which the difference in how the creole grew towards the model language was determined by the gender of its speakers:

One could say of the people who are most adapted to our way of life that they speak a passable Portuguese with us although their pronunciation has been influenced by the changes which most Latin languages undergo in tropical regions. The people, and in particular the women, use a language so curious and capricious that we Europeans can hardly decipher its meaning at all. In addition to the highly adulterated words which are sometimes pure invention, the topsy-turvy phrases and conventional greetings, the 'nhonhas' and 'nhonhonhas' sprinkle their chit-chat with little cries, laughs, exclamations and so on, which gives their 'babbling', as the ladies call it, a completely unique character.

The process of decreolization led to the extinction of patois although nowadays there are still traces of creole in the language used on a familiar level within the Macanese community:

Despite this (education), the Portuguese learnt at school is not wholly used on a familiar level. At this level certain expressions and constructions appear which the person knows to be incorrect and sometimes modifies when speaking to a Portuguese person. For instance, they say 'three pataca' but they know that it should be 'three patacas' and they know how to say it. (13)

In addition to morphological cases of the kind shown above, Batalha also found a plethora of creole traces both at a phonological and syntactic/semantic level.

Makista creole texts dating from the last century also seem to show that decreolization was taking place, a fact reflected by the varying input of creole in many of them. What is more interesting, however, are the allusions which reveal how the speakers themselves perceived their own language, their status as inferior and the corrective influence of the model language. These can be seen in the following three texts:

Sam divéra sintí unga ancusa pezado na coraçam quando uvi inglezada fazê chacota di nosso boboriça. Nôs ôtro, Sium cô eu, sam nôs já nacê na Macau; mas nôsso gente sam tudo de aqui, por isso nôs sintí vergonha olá este um pôco, vai pa alá sevandiziá com tudo chinachina. Qui sabe Sium lôgo entendê este linguazi ô nadi. Nôsso rancho nunca sã falá assim sa: mas eu já prendê com dôs chacha más véla de Macau, promor de sintí saião deçá cává ung-a lingu assi chistosa. Oze em dia tudo nhonhonha sã falá portuguezado; caregá na R corn acung-a Chente.

Estunga linguazi já servi pra eu anos trazado visti bobo na Quarentóna na más. Si Sium nom pôde intendê, eu lôgo falá môdo de agora, mêo portuguezado, mêo ingrezado.

(Marques Pereira, ob. cit., III, p. 781)

Sã linguaze de estudante

nôsôtro nunca entendê,

porque, quando elle criança,

algu cusa já prendê.

Nunca sâ cimo ôtro criança.

sua lingu caregado;

vai estudo, vên pâ casa

sâ falá portuguezado.

Nôs nun sã chómá porcobezo?

Elle nunca - siára óla:

elle fála percebejo,

tudu "r" caregado.

(Marques Pereira, ob. cit., I, pp. 59-60)

A. - Primo falla erradamente

Falla apenas um patois

J. F. - Masqui patuá tudo gente

Entendê cusa eu fallá.

(Marques Pereira, ob. cit., I)(14)

Marques Pereira also commented on the effects of education in Portuguese:

Circumstances have changed quite considerably over the last few years with the measures taken by Portugal to make the public education in the colony more Portuguese. This is neither the place nor the time to discuss whether these measures have been applied well, but there have certainly been results both in the way in which the Macanese, who have never been to the Motherland speak and in the way they write their newspapers, which are generally well written, in better Portuguese than the Indian ones. (15)

The slow process of convergence and the survival of traces of creole not only in the language but also in Macanese culture, the affection which the 'sons of the earth' display towards their ancestors' 'amusing language' all go to show that even if Makista commited suicide, it did not do so without pangs and resistance.

A CASE OF MURDER

The murder of a language, in other words the process by which it becomes extinct, is a great deal more dramatic than suicide, which we have already discussed. First of all, 'murder' involves the geographic coexistence of two languages which are mutually incomprehensible. The socially and politically dominant language replaces the mother-tongues of a community over the span of a couple of centuries.

The first generation to be affected is usually bilingual during childhood, equally skilled in the mother-tongue learnt from the parents and the locally dominant language to which it is exposed on the street, in the market and in school.

Because creole communities generally have few economic resources, and their traditional occupations (fishing in Malacca, farming in Korlai and crafts in Daman) are threatened by economic development in the area where they live, they are forced to emigrate in order to survive and improve their economic and social situation. When they are located in a new linguistic environment, where their mother-tongue is absent, they rapidly forget it and only use it sporadically on visits to older relatives at home.

The creole can only be maintained over a few generations if the group of emigrants from a single place of origin is quite large and there are thus blood ties and a strong sense of group identity at a cultural, ethnic or linguistic level which can bind them together. This was the case with the Kristang in Singapore and the Macanese in Hong Kong.

The example of Hong Kong is particularly interesting as the survival of the creole which was taken there also seems to be due to the absence of the corrective influence of standard Portuguese. In the meantime, the dominant languages, English and Cantonese, were soon to replace the role of Portuguese with the result that the creole simply became the language spoken at home:

Mas, Minha Querida Chencha, eu amestê pedi ung-a cuza com vôs, pra quando escrevê ôtro vez nuncabom botá na carta inglez, ô francez, pra tem que incomodá vôsso Tio pra pôde sabe são que cuza. Vôs escrevê na carta "City Hall" - querê que eu divinhá que asnera de palavra são este?

(Barreiros, ob. cit., p. 129)

In the case of Singapore, the immigrant population has not been subjected to the influence of Portuguese in its place of origin (Malacca) for a very long time, other than in contact with the Portuguese missions in Malacca and Singapore. The proximity of Malacca, the constant influx of new immigrants and strong unifying influence of religion (seen both in movement to Malacca to celebrate religious festivals, and by the presence of Singaporean inhabitants during mass at the Portuguese mission) leads them to maintain their language of origin for some generations. Nowadays, however, the younger generations do not know how to speak Papiá Kristang.

For those who remain, the only option is to change employment. Fishermen, farmers and labourers become factory workers, civil servants, or work in the service sector, all activities which require a knowledge of the dominant language. Simultaneously, compulsory education, which involves the youngest sections of the population, is carried out in the official languages. The tendency which the young have to speak the dominant language amongst themselves and even at home is encouraged by the parents who, in the face of a bleak future for their children, try to make them fluent in the dominant language which will open the door to the best jobs and make them upwardly mobile.

This is why we find that most families within the Kristang community in Malacca use English and Malay at home. An analogous situation occurs with creoles in India where English and local languages become the mother-tongue of the creole communities. It is worth noting that these communities always were bilingual or even multilingual as they constituted part of a much wider linguistic society. The difference nowadays is that if several languages are spoken, it tends to be the creole which is dropped.

The process of extinction is usually quick but it may, in exceptional circumstances, be delayed. This happens in isolated communities where traditional occupations are maintained and the creole language and culture have undergone a slow restructuring, bringing them closer to the surrounding languages and cultures. This is the case with the creole from Korlai (a village to the south of Bombay with around seven hundred inhabitants). The proximity of a Maratha-speaking community, whose economic, fishing and commercial activities function in tandem with those of the Christian village in an almost symbiotic relationship, has effected a shift in cultural manifestations such as food and dress. It has also caused a restructuring in the language. Not only have Maratha words gradually been introduced into the creole to replace creole terms but there has also been a reorganization of the syntax. The usual syntactic order for Portuguese creole is subject followed by predicate followed by complement. This has changed to subject, complement, predicate, the typical Maratha order. The same reorganization of syntax has also occured in Sri Lanka creole (Smith, 1977), moving it, in this case, towards Tamil.

Inevitably, the fact that these creole languages are fenced in by local languages and the process of acculturation arising from social and economic change means that creoles will be almost entirely assimilated. The effort to retain a creole, which is par excellence the vehicle of the culture, religion and identity of these communities, has also become a daily fight for survival:

Sie kere canta

Canta dratoe purtieges

Numiste canta

Mallaiye landes

(If you want to sing

Sing in proper Portuguese

You shouldn't sing

In Malay or Dutch.) (16)

BIBLIOGRAPHY

AITCHISON, JEAN. 1981. Language Change: Progress or Decay?, London, ed. Fontana,

BATALHA, GRACIETE NOGUEIRA. 1959. "Estado actual do dialecto Macaense", off-print from Revista Portuguesa de Filologia, vol IX, 1958, Coimbra.

BARREIROS, DANILO L. 1943-44. "Dialecto Português de Macau", Renascimento, Macau.

DA FRANÇA, BENTO. 1897. Macau e os seus Habitantes, Lisbon, ed. Imprensa Nacional.

DECAMP, AVID. 1971. "Toward a generative analysis of a post-creole continuum" in Dell Hymes (ed.), Pidginization and Creolization of Languages, ed. Cambridge. University Press.

FERGUSON, C. A. 1959. "Diglossia", Word, 15:325;40.

JACKSON, DAVID K. (in the press) Canta sinvergonya. Oral Traditions of Indo-Portuguese Creole Verse.

LESSA, ALMERINDO. 1970. Anthropologie et Anthroposociologie de Macau, PhD thesis, University of Toulouse.

MARQUES PEREIRA, JOÃO FELICIANO, (ed). 1901. "Cancioneiro musical crioulo", Ta-Ssi-Yang-Kuo, IInd series, 1, pp. 239-43; 2, pp. 703-7, Lisbon.

POLOMÉ, EDGAR. 1983. "Creolization and Language Change" in Woolford, E. and W. Washabaugh (ed). The Social Context of Creolization, Ann Arbor.

SANCEAU, ELAINE. 1959. "Portuguese women during the first two centuries of expansion overseas" in Actas, III International Conference on Brazilian Studies, Lisbon, vol I.

SMITH, IAN RUSSELL. 1977. "Sri Lanka Creole - Portuguese Phonology", Cornell University dissertation, also published in International Journal of Dravidian Linguistics, 7:2 (1978), pp. 248-405.

WASHABAUGH, W. and S. GREENFIELD. 1983. "The development of Altantic creole languages" in Woolford, E. and W. Washabaugh (ed.), The Social Context of Creolization, Ann Arbor.

NOTES

(1) Quoted in Ian Hancock: Creolization and Change: The Development of the Field, p.2.

(2) The absence of a creole in Madeira or the Azores where slaves from the Canary Islands were also used is, perhaps, due to the fact that there were always a great many more settlers from the mainland than there were slaves.

(3) Almerindo Lessa, 1970, p.7.

(4) Ibidem, p.5.

(5) Elaine Sanceau, 1957, p.237.

(6) Almerindo Lessa, ibidem, p. 8.

(7) Luis Vaz de Camões: The Lusiads (trans. by William C. Atkinson), ed. Penguin Classics, Harmondsworth, 1987.

(8) Quoted in O. Jesperson: Language: Its Nature, Development and Origin, London, 1922, p.65.

(9) David Decamp, 1971, p.349.

(10) Jean Aitchison, 1981, p.209.

(11) Italics in original.

(12) Peace Handbooks, vol XIII, 81, p.4.

(13) Batalha, 1957, p.4.

(14) Transcribed in Barreiros, 1943, p.89.

(15) Marques Pereira, 1901, p.55.

(16) Jackson, (in press) p.158.

* Research fellow in the Department of Linguistics in the University of Austin (Texas); holder of a research scholarship from the Institute of Culture, Macau.